From sunshine at public.kherson.ua Fri Oct 1 08:32:30 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 00:32:45 2004 Subject: cursor count In-Reply-To: Message-ID: On 10/1/04 8:21 AM, "mdeh@earthlink.net" wrote: > Name: Michael > > Question regarding: v4rb Hi Michael, > Ruslan, > Is there a way of knowing how many cursors are open for a given query? > Tks I think you mean for a given database? Please check if exists Vdatabase.CursorCount If no then Hmm, this is my mistake. We can add it quite easy. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 1 08:41:53 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 00:42:04 2004 Subject: complex sql query: problem with "or" In-Reply-To: <415C8C92.2070506@nexbyte.com> Message-ID: On 10/1/04 1:45 AM, "nexbyte gmbh - Simon Anderegg" wrote: Hi Sam, Thank you very much for good words. Your application looks very nice! You are right -- good database backend allow to developers do miracles. I am always impressed what cool things developers can do with Valentina. I'd couldn't! :-) > hi ruslan > > time to say that valentina db is the greatest xtra for director i know! > valentina was the perfect choice to develop the application containings > really complex sql queries and more than 10 related tables. > > - more than 30 tables > - material table, more than 10000 entries > - dynamic db search (requery db after every key pressing, only possible > with the power of valentina) > - 3 different languages (handled in valentina) > - calculates machining tool data in different units > - creating pdf files on the fly (db content -> pdf) > > it was a challenge, but really funny. > ...and thanks again for your fantastic support! > > sam > Ruslan Zasukhin wrote: > >> On 8/27/04 1:33 AM, "s.anderegg@nexbyte.com" wrote: >> >> >>> Name: sam >>> >>> Question regarding: sql >>> >>> hi >>> >>> i have to reengineering a access project to macromedia director and had >>> buyed the valentina db (current version). everyting works fine exept the sql >>> queries containing "or" arguments: >>> >>> SELECT t1.recID, t1.name, t2.minval t2.maxval FROM t1, t2 WHERE t2.minval >>> <= 50 AND t2.maxval >= 50 AND (t1.recID = t2.ptr_possible_id_1 OR >>> t1.recID = t2.ptr_possible_id_2) >>> >>> on every other database (exept from valentina) my example query works fine. >>> the same query without "OR..." does it also. >>> >>> in your usergroup you had described several solutions to solve the problems >>> with "or", but every test failed with my complex sql query. >>> (compareing 2 times 2 variable values from 2 tables...) >>> >>> - do you have a solution (one sql statement) for my problem? - i think (like >>> other users in the ug) this is a very basic function of each sql database. >>> do you have a patch, a bugfix or a newer (and stable) version of valentina >>> where the "or" argument are implemented correctly? >>> >>> greetings from switzerland (by night) >>> sam >> >> >> Hi Sam, >> >> Valentina 2.0 which we finish now must work better with such quires. >> >> Your query even differ from others guys on list. >> You use OR between "link conditions"! >> >> I think you can try this form. >> >> Note, that you do join >> >> t1.recID = t2.ptr_possible_id_1 OR >> t1.recID = t2.ptr_possible_id_2 >> >> You want all records from T2 which have NON-zero pointer >> ptr_possible_id_1 or ptr_possible_id_2 >> >> >> So >> >> SELECT t1.recID, t1.name, t2.minval t2.maxval >> FROM t1, t2 >> WHERE T1.Recid IN ( >> SELECT ??? >> FROM T2 >> WHERE t2.minval <= 50 AND t2.maxval >= 50 and >> (ptr_possible_id_1 <> 0 or ptr_possible_id_1 <> 0) >> ) >> >> >> Problem is -- what to SELECT in the second subquery. >> >> Probably we can use here BaseObject method with formula >> >> if( ptr_possible_id_1 <> 0, ptr_possible_id_1, ptr_possible_id_2 ) >> >> As result in subquery we must get table with one column which contain values >> of T1.RecID >> >> Then we apply IN to this values. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From giv at tlc.kherson.ua Fri Oct 1 10:46:22 2004 From: giv at tlc.kherson.ua (Igor Gomon) Date: Fri Oct 1 02:46:36 2004 Subject: Bug in vServer with multiple databases? References: Message-ID: <001e01c4a78a$bf7e7e00$3b04a8c0@giv> >>>> When you have e.g. 2 databases in the DB folder, getDatabaseCount() >>>> will return "3", which is OK, since the master.db is #3. >>> Now when you >>>> stop the service, delete one database and restart the service, >>>> getDatabaseCount() still returns "3". > > But have you un-register db ? > > > Vserver watch into MasterDb but not into physical files. > > > Actually I think this is not logical Igor. > If masterdb refer some db, but it was not found, > Then this db must not be counted. I fixed this. Now VServer counts only existing databases - it checks each database entry in the "master" database producing warning message in case of any problem. The fix will be available with the next - a66 - build. -- Best regards, Igor Gomon ------------------------------------------------------------- e-mail: giv@tlc.kherson.ua web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://listserv.macserve.net/mailman/listinfo/valentina From fb at memedia.de Fri Oct 1 09:55:01 2004 From: fb at memedia.de (Florian Bogeschdorfer) Date: Fri Oct 1 02:55:18 2004 Subject: Bug in vServer with multiple databases? In-Reply-To: <001e01c4a78a$bf7e7e00$3b04a8c0@giv> Message-ID: OK, thank you Florian > > > > Actually I think this is not logical Igor. > > If masterdb refer some db, but it was not found, Then this > db must not > > be counted. > I fixed this. > Now VServer counts only existing databases - it checks each > database entry in the "master" database producing warning > message in case of any problem. > The fix will be available with the next - a66 - build. > > -- > Best regards, > Igor Gomon From sunshine at public.kherson.ua Fri Oct 1 14:52:53 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 06:53:02 2004 Subject: error 13 (OS X) and error 57 (OS 9) In-Reply-To: <20040930211431.31057.qmail@web52408.mail.yahoo.com> Message-ID: On 10/1/04 12:14 AM, "Chris Sheffield" wrote: > I have a customer who is trying to install VServer on > her Mac OS X server (running 10.2.8 I believe). > Everything goes fine and the service is running. But > after installing our software on the client > workstations, she is unable to connect to the database > on the server. She's getting error 13 under Mac OS X > (Panther) and error 57 under Mac OS 9. -57 noMacDskErr not a Macintosh disk (sig bytes are wrong) > Does anyone > know what these errors mean? Guys. When you will teach self do google for "MacOS Error codes" You will find several sites where these errors are listed. They are listed also in the docs of Apple that are located in the folder DEVELOPER on MacOS X On Version tracker exists utility "Apple Errors" which also contains error descriptions. > I know they're not normal Valentina errors, but something is causing them when > trying to access a Valentina database. > > I can't find either one when searching Google. A search for error 13 says > it's a type-mismatch under Windows, but can't find anything about Mac OS. > > Can someone help? Everybody from you have file on your OS X with errors of unix type. Path /usr/include/sys/errno.h #define EACCES 13 /* Permission denied */ -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 1 14:54:58 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 06:55:05 2004 Subject: error 13 (OS X) and error 57 (OS 9) In-Reply-To: Message-ID: On 10/1/04 2:52 PM, "Ruslan Zasukhin" wrote: >> I can't find either one when searching Google. A search for error 13 says >> it's a type-mismatch under Windows, but can't find anything about Mac OS. >> >> Can someone help? Aha, so you did search on goggle. You need search not for specific error code, but for Macos error codes Then you will find many places -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From vidal_olivier at yahoo.fr Fri Oct 1 15:58:00 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Fri Oct 1 08:49:06 2004 Subject: How to track down doubloons in valentina ? In-Reply-To: References: Message-ID: thank you Beatrix ! olivier Le 30 sept. 04, ? 11:59, Beatrix Willius a ?crit : > Please find below a quick example from Access: > > SELECT [PART_NUMBER], [Gross Value], [Phase Name] > FROM [epsilon vpps] > WHERE [PART_NUMBER] In (SELECT [PART_NUMBER] FROM [epsilon vpps] As Tmp > GROUP BY [PART_NUMBER] HAVING Count(*)>1 ) > ORDER BY [PART_NUMBER]; > > This query searches for duplicates in the Part_Number field and also > shows > Gross Value and Phase Name. If I want to search for duplicates in > something like First Name + Last Name, I would create a query with the > concatenated fields in it and base the duplicate search on this query. > > Cannot check now, if this works in Valentina. > > HTH > > > Mit freundlichen Gruessen > Beatrix Willius > Adam Opel AG > CE Cost Reduction - Database-Team - IPC 33-01 > Tel.: ++49-6142-7-50694 Fax: ++49-6142-7-63383 > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From vidal_olivier at yahoo.fr Fri Oct 1 16:16:01 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Fri Oct 1 09:07:05 2004 Subject: How to track down doubloons in valentina ? In-Reply-To: References: Message-ID: <6C38D1B6-13B4-11D9-8C9D-00039310B7DA@yahoo.fr> Hi Ruslan and list, > No. > > There general operation is GROUP BY, DISTINCT, > Which can compare order remove THE SAME values. I always study the problem for the doubloons of names of company but the example of Beatrix is interesting. It can be applicable for the family names. For example, if I want to make a search for doubloons on the name, zip, city: SELECT * FROM Customers GROUP BY Name, Zip, City HAVING COUNT(*)>1 It's good, simple and fast. It is damage that Valentina 1 does not understand ' HAVING '. There a trick for to obtain the same result but without HAVING? thank you olivier From sunshine at public.kherson.ua Fri Oct 1 20:58:43 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 13:05:18 2004 Subject: How to track down doubloons in valentina ? In-Reply-To: <6C38D1B6-13B4-11D9-8C9D-00039310B7DA@yahoo.fr> Message-ID: On 10/1/04 5:16 PM, "olivier" wrote: >> There general operation is GROUP BY, DISTINCT, >> Which can compare order remove THE SAME values. > > I always study the problem for the doubloons of names of company but > the example of Beatrix is interesting. > It can be applicable for the family names. > > For example, if I want to make a search for doubloons on the name, zip, > city: > > SELECT * FROM Customers GROUP BY Name, Zip, City HAVING COUNT(*)>1 > > It's good, simple and fast. Note, having cannot be used without GROUP BY > It is damage that Valentina 1 does not understand ' HAVING '. > > There a trick for to obtain the same result but without HAVING? Only using TMP Table I think. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 1 22:45:43 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 14:45:51 2004 Subject: V4RB, idea - StructureBrowser. Can somebody help? Message-ID: Hi All, Hi Frank, I think you have see that I have extract from viSQL the structure rowser window into SHARED folder. Thanks to this we all can now very easy add this window into our projects. So I think we all are interested in the improvement of it. I see the next possible improvements right now: 1) browser simply show field flags. But I suggest the next idea. Let it show check boxes. Let we can set ON/OFF that check boxes to change field flags without programming. Can somebody do this? Frank? 2) this Browser must become aware of new properties of Database Tables and fields. At least LocaleName Later CollationStrength -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 1 23:10:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 1 15:10:59 2004 Subject: V4RB, idea - StructureBrowser. Can somebody help? In-Reply-To: Message-ID: On 10/1/04 10:45 PM, "Ruslan Zasukhin" wrote: Ops, actually this is for Valentina beta list. > Hi All, > Hi Frank, > > I think you have see that I have extract from viSQL the structure rowser > window into SHARED folder. > > Thanks to this we all can now very easy add this window into our projects. > > So I think we all are interested in the improvement of it. > > I see the next possible improvements right now: > > 1) browser simply show field flags. > But I suggest the next idea. > Let it show check boxes. > Let we can set ON/OFF that check boxes to change field flags > without programming. > > Can somebody do this? Frank? > > > 2) this Browser must become aware of new properties of > Database Tables and fields. > At least > > LocaleName > > Later > CollationStrength -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From listmail at gearyweb.com Fri Oct 1 16:59:17 2004 From: listmail at gearyweb.com (michael geary) Date: Fri Oct 1 17:59:28 2004 Subject: 2.0 client for use in browser Message-ID: <85E47312-13FD-11D9-B159-000A95ACCDAA@gearyweb.com> Hi Ruslan & co. I've got a project with a _large_ database that needs to be Shockwave-enabled. I believe I understand correctly that V4MD client will work inside a browser, and I can simply point it to a remote server running VServer. Assuming I understand this correctly, can you give me any information about availability of this solution? If you can't talk about ship dates, I will have no choice but to implement a PHP/MySQL solution, at least for the interim. Thank you, michael geary From sunshine at public.kherson.ua Sat Oct 2 08:49:23 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 2 00:49:37 2004 Subject: 2.0 client for use in browser In-Reply-To: <85E47312-13FD-11D9-B159-000A95ACCDAA@gearyweb.com> Message-ID: On 10/2/04 1:59 AM, "michael geary" wrote: Hi Michael, > I've got a project with a _large_ database that needs to be > Shockwave-enabled. I believe I understand correctly that V4MD client > will work inside a browser, and I can simply point it to a remote > server running VServer. YES. > Assuming I understand this correctly, can you give me any information > about availability of this solution? I hope that about 1 Nov we will get V4MD_Client 2.0 SW for testing. > If you can't talk about ship > dates, I will have no choice but to implement a PHP/MySQL solution, at > least for the interim. But then you can use PHP + Vserver! We also have PHP for Valentina client. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 2 08:50:23 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 2 00:50:31 2004 Subject: 2.0 client for use in browser In-Reply-To: Message-ID: On 10/2/04 8:49 AM, "Ruslan Zasukhin" wrote: > On 10/2/04 1:59 AM, "michael geary" wrote: > > Hi Michael, > >> I've got a project with a _large_ database that needs to be >> Shockwave-enabled. I believe I understand correctly that V4MD client >> will work inside a browser, and I can simply point it to a remote >> server running VServer. > > YES. > >> Assuming I understand this correctly, can you give me any information >> about availability of this solution? > > I hope that about 1 Nov we will get V4MD_Client 2.0 SW for testing. > >> If you can't talk about ship >> dates, I will have no choice but to implement a PHP/MySQL solution, at >> least for the interim. > > But then you can use PHP + Vserver! > > We also have PHP for Valentina client. I mean we have it right now. You can find it on our beta pages -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Sat Oct 2 15:39:39 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Sat Oct 2 16:39:52 2004 Subject: V4RB, idea - StructureBrowser. Can somebody help? In-Reply-To: References: Message-ID: <9023B781-14BB-11D9-B445-0003939246BC@mindstarprods.com> Hi Ruslan, On Oct 1, 2004, at 1:45 PM, Ruslan Zasukhin wrote: > I think you have see that I have extract from viSQL the structure > rowser > window into SHARED folder. > > Thanks to this we all can now very easy add this window into our > projects. > > So I think we all are interested in the improvement of it. > > I see the next possible improvements right now: > > 1) browser simply show field flags. > But I suggest the next idea. > Let it show check boxes. > Let we can set ON/OFF that check boxes to change field flags > without programming. > > Can somebody do this? Frank? Unfortunately I lack the time to do this right now - I've just relocated and I'm looking for a new job. In any event, I don't see that as the purpose of ViSQL or the Structure Browser. After all, it is meant to simply *browse* (look but not change) the structure :^). I believe VStudio should do this. > 2) this Browser must become aware of new properties of > Database Tables and fields. > At least > > LocaleName > > Later > CollationStrength Yes, I totally agree with this and when I finally get around to updating ViSQL (probably just for Valentina 2), I will add these. Best regards, Frank From erichg at mcmm.com Sun Oct 3 20:18:32 2004 From: erichg at mcmm.com (Erich Geiersberger) Date: Sun Oct 3 13:19:04 2004 Subject: [V4MD] Updating a project from beta Message-ID: Hi Ruslan and list, I am currently writing an update for a client-server and single user project running both on PC and Macs The old versions used are: PC: VServer dll version 2.0.0.0 a55/344 V4MD_Client.x32 version 2.0.0.0 a53 V4MD.x32 version 1.10.0.0 Mac OSX: V4MD_Client_Carbon.Xtra version 2.0a50 V4MD_Carbon.Xtra version 1.10 Mac Classic: V4MD_Client.Xtra version 2.0a50 V4MD.Xtra version 1.10 There have been no problems so far with these installations. During this update I need to modify the database structure both on single place installations and on server installations. Should I stick with the old Xtras or should I use the newest ones? Are there any known issues with updating Databases using these versions to newer ones? Best regards -- Erich Geiersberger Media Connect Multimedia Gratzmuellerstr. 1 86150 Augsburg, Germany Tel: +49 821/ 34752 - 0 fax: +49 821/ 34752 - 49 eMail erichg@mcmm.com http://www.mcmm.com From sunshine at public.kherson.ua Sun Oct 3 21:41:10 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 3 13:41:29 2004 Subject: [V4MD] Updating a project from beta In-Reply-To: Message-ID: On 10/3/04 9:18 PM, "Erich Geiersberger" wrote: Hi Erich, > I am currently writing an update for a client-server and single user > project running both on PC and Macs > > The old versions used are: > > PC: > VServer dll version 2.0.0.0 a55/344 > V4MD_Client.x32 version 2.0.0.0 a53 > V4MD.x32 version 1.10.0.0 > > Mac OSX: > V4MD_Client_Carbon.Xtra version 2.0a50 > V4MD_Carbon.Xtra version 1.10 > > Mac Classic: > V4MD_Client.Xtra version 2.0a50 > V4MD.Xtra version 1.10 > > There have been no problems so far with these installations. > > During this update I need to modify the database structure both on > single place installations and on server installations. Should I > stick with the old Xtras or should I use the newest ones? > > Are there any known issues with updating Databases using these > versions to newer ones? Newer versions of LOCAL engine just fix few bugs. Newer versions of CLIENT/SERVER also fix few minor bugs and add some minor features that was requested. So you can upgrade to newer with no problems. Formats of files the same -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From daniel at monumental-i.com Sun Oct 3 17:48:22 2004 From: daniel at monumental-i.com (Daniel Crowder) Date: Sun Oct 3 16:48:32 2004 Subject: [V4MD] Index by Words (newbie question) In-Reply-To: References: Message-ID: I'm looking for ways to speed-up a product search. I currently am using LIKE to check to see if certain key-words exist in a field called TEXT. Reading over the documentation it says that using LIKE in a query does not take advantage of a field's index. I am now confused as to how to take advantage of the INDEX BY WORDS feature. How do I perform the fastest query of a text field that is using INDEX BY WORDS without using LIKE? Any suggestions appreciated! - Daniel From sunshine at public.kherson.ua Mon Oct 4 00:53:08 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 3 16:53:16 2004 Subject: [V4MD] Index by Words (newbie question) In-Reply-To: Message-ID: On 10/4/04 12:48 AM, "Daniel Crowder" wrote: Hi Daniel, > I'm looking for ways to speed-up a product search. > > I currently am using LIKE to check to see if certain key-words exist in > a field called TEXT. Reading over the documentation it says that using > LIKE in a query does not take advantage of a field's index. I am now > confused as to how to take advantage of the INDEX BY WORDS feature. How > do I perform the fastest query of a text field that is using INDEX BY > WORDS without using LIKE? You should do query of kind WHERE left( textFld, 3 ) = 'abc' This will use index -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From daniel at monumental-i.com Sun Oct 3 18:05:20 2004 From: daniel at monumental-i.com (Daniel Crowder) Date: Sun Oct 3 17:05:31 2004 Subject: [V4MD] Index by Words (newbie question) In-Reply-To: References: Message-ID: <512DCC44-1588-11D9-8B21-000D93326CD0@monumental-i.com> Thanks! If I wanted to search a complete word and not a partial word is there a more streamlined way of doing that? On Oct 3, 2004, at 5:53 PM, Ruslan Zasukhin wrote: > On 10/4/04 12:48 AM, "Daniel Crowder" wrote: > > Hi Daniel, > >> I'm looking for ways to speed-up a product search. >> >> I currently am using LIKE to check to see if certain key-words exist >> in >> a field called TEXT. Reading over the documentation it says that using >> LIKE in a query does not take advantage of a field's index. I am now >> confused as to how to take advantage of the INDEX BY WORDS feature. >> How >> do I perform the fastest query of a text field that is using INDEX BY >> WORDS without using LIKE? > > You should do query of kind > > WHERE left( textFld, 3 ) = 'abc' > > This will use index > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > > _______________________________ Daniel Crowder Partner Monumental Natural Interactive Solutions 828 Ralph McGill Blvd. W-2 Atlanta, GA 30306 AIM: danielmonumental Phone: 404.223.0303 Fax: 404.223.0309 monumental-i.com From daniel at monumental-i.com Sun Oct 3 18:08:01 2004 From: daniel at monumental-i.com (Daniel Crowder) Date: Sun Oct 3 17:08:06 2004 Subject: [V4MD] Index by Words (newbie question) In-Reply-To: <512DCC44-1588-11D9-8B21-000D93326CD0@monumental-i.com> References: <512DCC44-1588-11D9-8B21-000D93326CD0@monumental-i.com> Message-ID: nm. I was over-thinking that last question! SELECT * FROM product_attributes WHERE text = 'metal' and text = 'cast' Doing that on a field that is indexed by words returns the desired result. No need for anything else! On Oct 3, 2004, at 6:05 PM, Daniel Crowder wrote: > Thanks! > > If I wanted to search a complete word and not a partial word is there > a more streamlined way of doing that? > > On Oct 3, 2004, at 5:53 PM, Ruslan Zasukhin wrote: > >> On 10/4/04 12:48 AM, "Daniel Crowder" wrote: >> >> Hi Daniel, >> >>> I'm looking for ways to speed-up a product search. >>> >>> I currently am using LIKE to check to see if certain key-words exist >>> in >>> a field called TEXT. Reading over the documentation it says that >>> using >>> LIKE in a query does not take advantage of a field's index. I am now >>> confused as to how to take advantage of the INDEX BY WORDS feature. >>> How >>> do I perform the fastest query of a text field that is using INDEX BY >>> WORDS without using LIKE? >> >> You should do query of kind >> >> WHERE left( textFld, 3 ) = 'abc' >> >> This will use index >> >> >> -- >> Best regards, >> Ruslan Zasukhin [ I feel the need...the need for speed ] >> ------------------------------------------------------------- >> e-mail: ruslan@paradigmasoft.com >> web: http://www.paradigmasoft.com >> >> To subscribe to the Valentina mail list go to: >> http://lists.macserve.net/mailman/listinfo/valentina >> ------------------------------------------------------------- >> >> _______________________________________________ >> Valentina mailing list >> Valentina@lists.macserve.net >> http://lists.macserve.net/mailman/listinfo/valentina >> >> > _______________________________ > Daniel Crowder > Partner > > Monumental > Natural Interactive Solutions > > 828 Ralph McGill Blvd. W-2 > Atlanta, GA 30306 > > AIM: danielmonumental > Phone: 404.223.0303 > Fax: 404.223.0309 > monumental-i.com > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > > _______________________________ Daniel Crowder Partner Monumental Natural Interactive Solutions 828 Ralph McGill Blvd. W-2 Atlanta, GA 30306 AIM: danielmonumental Phone: 404.223.0303 Fax: 404.223.0309 monumental-i.com From wonderfef at noos.fr Mon Oct 4 14:58:18 2004 From: wonderfef at noos.fr (Eric Ferrer) Date: Mon Oct 4 07:58:32 2004 Subject: [V4RBclient] specifying Table in cursor.field Message-ID: Hello, My database may be badly designed, but the fact is I have fields in some tables that share the same name : "Name". When I do a SQL select such as : SELECT Clients.Name, Units.Name FROM Clients, Units, WHERE Clients.ParentUnitID *= Units.ID I can't populate my listbox with both Clients and Units names, because cursor.field("Name").getstring always return the name of the Client... I tried cursor.field("Units.Name").getstring but it raises a nil exception.... Maybe there is a workaround before changing my fields' name and spending hours updating my code... Thanks in advance, Eric From erichg at mcmm.com Mon Oct 4 16:33:29 2004 From: erichg at mcmm.com (Erich Geiersberger) Date: Mon Oct 4 09:37:01 2004 Subject: [V4MD] Updating a project from beta In-Reply-To: References: Message-ID: > > I am currently writing an update for a client-server and single user >> project running both on PC and Macs >> >> The old versions used are: >> >> PC: >> VServer dll version 2.0.0.0 a55/344 >> V4MD_Client.x32 version 2.0.0.0 a53 >> V4MD.x32 version 1.10.0.0 >> >> Mac OSX: >> V4MD_Client_Carbon.Xtra version 2.0a50 >> V4MD_Carbon.Xtra version 1.10 >> >> Mac Classic: >> V4MD_Client.Xtra version 2.0a50 >> V4MD.Xtra version 1.10 >> >> There have been no problems so far with these installations. >> >> During this update I need to modify the database structure both on >> single place installations and on server installations. Should I >> stick with the old Xtras or should I use the newest ones? >> >> Are there any known issues with updating Databases using these >> versions to newer ones? > >Newer versions of LOCAL engine just fix few bugs. > >Newer versions of CLIENT/SERVER also fix few minor bugs and add some minor >features that was requested. > >So you can upgrade to newer with no problems. > >Formats of files the same Thanks for clarifying. Erich From sunshine at public.kherson.ua Mon Oct 4 17:45:18 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 4 09:45:27 2004 Subject: [V4RBclient] specifying Table in cursor.field In-Reply-To: Message-ID: On 10/4/04 3:58 PM, "Eric Ferrer" wrote: > Hello, > > My database may be badly designed, but the fact is I have fields in some > tables that share the same name : "Name". > > When I do a SQL select such as : > > SELECT Clients.Name, Units.Name FROM Clients, Units, WHERE > Clients.ParentUnitID *= Units.ID > > I can't populate my listbox with both Clients and Units names, because > cursor.field("Name").getstring > always return the name of the Client... > > I tried > cursor.field("Units.Name").getstring > but it raises a nil exception.... > > Maybe there is a workaround before changing my fields' name and spending > hours updating my code... You can use cursor.Field( Index ) for this. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From j.peters at valentina-db.de Mon Oct 4 17:31:25 2004 From: j.peters at valentina-db.de (Jochen Peters) Date: Mon Oct 4 10:31:47 2004 Subject: Valentina question In-Reply-To: <41615D8B.2020901@roundboxmedia.com> References: <41615D8B.2020901@roundboxmedia.com> Message-ID: <73EEF9FF-161A-11D9-A2D0-000A959F6B0C@valentina-db.de> Hello! > I have some Valentina questions. I'm not sure exactly where to ask > these, so please let me know if this is the correct way for me to do > this. Yes - you can ask here - and - of course - also on the Valentina mailing list - please visit: http://listserv.macserve.net/mailman/listinfo/valentina to register for the list. It is very very usefull because many Valentina developers are listening to the list, offering help to each others. I have forwarded this mail to the list. > > Here is what we are trying to accomplish: > > We have data in a MySQL database on a Solaris box that we are trying > to load into a Valentina db. Unfortunately, it's not a straight data > transfer - we need to process the data slightly before it goes into > Valentina. Right now, we're retrieving the data from MySQL using a > Java program (on the Solaris box) that processes the data and then > creates a file of "INSERT INTO ......." sql statements. We're then > processing this sql file on Windows using a Director program that > creates a Valentina db and inserts the data. This works, but we need > to automate this process. Ideally, what we would like to do is to > retrieve the data and insert it into Valentina right on the Solaris > box. Hhmm - but Valentina does not run on Solaris. So - you must do this on Windows or Mac. > I have not found a way to do this using Java/JDBC. Do you have any > ideas for how we can automatically transfer data from MySQL to > Valentina on Solaris? If it's not possible on Solaris, what would be > your recommendation on the best way to do it on Windows? Well - your approach seems to work. What exactly is your problem? You have to do this on a regular base and wants to automate this? I think Valentina Studio would be the ideal candidate for this - problem now is that it does not allow processing of data during import. But i plan to integrate this into Valentina Studio. Studio already contains a JavaScript engine - so - in future it will be possible to do this task with Studio. Mike - what is your time frame for this task? Maybe i can implement this feature into vStudio right now.... > > > Also, I have had some trouble with the Load feature in Valentina > Studio (I'm using Ver. 1.0a66 Build 0116092004). I cannot get data > loaded from a SQL file into a Valentina db. Any suggestions will be > appreciated. Answer is easy: Valentina 1.x CAN'T load from SQL! (only from XML). This feature will be available for Valentina 2.x. > -- Best regards, Jochen Peters PIIT GmbH ------------------------------------ http://www.valentina-db.de From wonderfef at noos.fr Mon Oct 4 17:35:12 2004 From: wonderfef at noos.fr (Eric Ferrer) Date: Mon Oct 4 10:35:21 2004 Subject: [V4RBclient] specifying Table in cursor.field In-Reply-To: Message-ID: Ruslan, > You can use cursor.Field( Index ) for this. Yes, of course... Thank you one more time! Eric From rjb at rz.uni-potsdam.de Mon Oct 4 17:27:43 2004 From: rjb at rz.uni-potsdam.de (Robert Brenstein) Date: Mon Oct 4 10:35:37 2004 Subject: cursor count In-Reply-To: References: Message-ID: >On 10/1/04 8:21 AM, "mdeh@earthlink.net" wrote: > >> Name: Michael >> >> Question regarding: v4rb > >Hi Michael, > > >> Ruslan, >> Is there a way of knowing how many cursors are open for a given query? >> Tks > >I think you mean for a given database? > >Please check if exists Vdatabase.CursorCount > >If no then Hmm, this is my mistake. >We can add it quite easy. > > >-- >Best regards, >Ruslan Zasukhin [ I feel the need...the need for speed ] That would be quite useful for me (in vxcmd) as well. Robert From sunshine at public.kherson.ua Mon Oct 4 18:41:04 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 4 10:41:13 2004 Subject: Valentina question In-Reply-To: <73EEF9FF-161A-11D9-A2D0-000A959F6B0C@valentina-db.de> Message-ID: On 10/4/04 6:31 PM, "Jochen Peters" wrote: >> Also, I have had some trouble with the Load feature in Valentina >> Studio (I'm using Ver. 1.0a66 Build 0116092004). I cannot get data >> loaded from a SQL file into a Valentina db. Any suggestions will be >> appreciated. > Answer is easy: Valentina 1.x CAN'T load from SQL! (only from XML). > This feature will be available for Valentina 2.x. You can write self simple parser which find ; After each next SQL command and execute command by command. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon Oct 4 18:41:25 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 4 10:41:33 2004 Subject: cursor count In-Reply-To: Message-ID: On 10/4/04 6:27 PM, "Robert Brenstein" wrote: >> I think you mean for a given database? >> >> Please check if exists Vdatabase.CursorCount >> >> If no then Hmm, this is my mistake. >> We can add it quite easy. >> >> >> -- >> Best regards, >> Ruslan Zasukhin [ I feel the need...the need for speed ] > > That would be quite useful for me (in vxcmd) as well. Ok, then we will add this property -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Mon Oct 4 10:37:08 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Mon Oct 4 11:37:17 2004 Subject: New Valentina-REALbasic Tutorial In-Reply-To: References: Message-ID: I've entered the code in the tutorial, up to page 19 of the documentation, where it calls for the student to compile and run. Here are my observations. Errors(?) in the Tutorial: Page 9, line of code for instancing the ULong field reads: LongFld = new VULong( "ulong_fld? ) It should read as: ULongFld = new VULong( "ulong_fld? ) [Note: It's a typo; missing the leading "U".] Page 11 and 12, documentation says to make new menu handlers, but uses Sub FileNew() instead of RB's Function FileNew() when you create a menu handler to deal with the command for making a new file. Page 16, the documentation says that the Adding records function encoded into the boPersons object will be called by a menu item. However, the tutorial does not specify how that menu handler is supposed to read. Page 17, documentation calls for an item called mCallerWindow, however this item is not created anywhere prior to this line of code. Therefore, you get an error about property or method not existing. Page 17, documentation calls for making calls to the PopulatePanes method from Window1. However, this method is not being recognized as existing in Window2, despite the method being declared as "Public" in Window1. [Note: Should these have been generated at the App level?] The source code in the Add, Update and Delete buttons dragged over from Window1 into Window 2 is generating multiple errors. Most of them about method or property not existing in Window2, specifically PopulatePanes, CollectPanes. This despite the fact that these methods in Window1 are set to Public. [Note: Should these have been generated at the App level?] Also, with respect to the problem about source code for buttons from Window1 dragged and dropped into Window2, mDatabase is not recognized in Window2. Still working away on this. Hope you can advise me on what needs to be done to resolve these issues. Regards, Chuck Pelto From j.peters at valentina-db.de Mon Oct 4 19:19:47 2004 From: j.peters at valentina-db.de (Jochen Peters) Date: Mon Oct 4 12:20:06 2004 Subject: Valentina question In-Reply-To: <41617D6C.5090601@roundboxmedia.com> References: <41615D8B.2020901@roundboxmedia.com> <73EEF9FF-161A-11D9-A2D0-000A959F6B0C@valentina-db.de> <41617D6C.5090601@roundboxmedia.com> Message-ID: <97A2A440-1629-11D9-A2D0-000A959F6B0C@valentina-db.de> > Jochen - > > Thanks for all if the information. This was very helpful. My time > frame is fairly short - I'm looking for something over the next few > weeks. Ok - so let's see how we can help you.... > > Yes, we are trying to do transfer data from MySQL to Valentina on a > regular basis and we're tying to automate it. I guess my problem is > that right now I still don't see how this can be done. On Windows, > I've tried to do the whole thing from Java, but I haven't been able > connect to a Valentina database using JDBC. The current Valentina JDBC driver is old and will not ne updated. We will get a complete new JDBC driver for connecting to valentina Server databases in the future. > It seems like Studio can't be used to load data from a sql file and > even if it could, I'm not sure how I would automate a Studio function. > Can Studio functions be run and controlled from the command line? I can easily add this - no problem. Right now you can start Studio from the command line passing a database which it will open. Of course we can add some more commands here. On startup right now a JavaScript file called "Startup.js" is executed by Studio's JavaScript engine. By default this Script opens the Navigator and it can also open Valentina database files. > Could Studio possibly be used for the whole thing (retrieving from > MySQL also)? I think yes. Right now you can easily IMPORT data from MySQL (via ODBC) and store it in a Valentina database. You can use the Import Wizard for this. Note: This will do no processing of the data. You can only specify which field from the source database should go to which field in the destination database. I can improve the Import Wizard in the following way: 1) Adding a Script option This means that you can specify a JavaScript which will be executed for EACH record during import. 2) Adding field-access to JavaScript This means i must add JavaScript commands to access the fields in the current record. Get and Set methods. This way you would be able to do some data processing in JavaScript during the import - what you think? -- Best regards, Jochen Peters PIIT GmbH ------------------------------------ http://www.valentina-db.de From sunshine at public.kherson.ua Mon Oct 4 20:52:09 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 4 12:52:16 2004 Subject: New Valentina-REALbasic Tutorial In-Reply-To: Message-ID: On 10/4/04 7:37 PM, "Chuck Pelto" wrote: > I've entered the code in the tutorial, up to page 19 of the > documentation, where it calls for the student to compile and run. > > Here are my observations. > > Errors(?) in the Tutorial: > > Page 9, line of code for instancing the ULong field reads: > > LongFld = new VULong( "ulong_fld? ) > > It should read as: > > ULongFld = new VULong( "ulong_fld? ) > > [Note: It's a typo; missing the leading "U".] Right, typo -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon Oct 4 20:53:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 4 12:53:38 2004 Subject: New Valentina-REALbasic Tutorial In-Reply-To: Message-ID: On 10/4/04 7:37 PM, "Chuck Pelto" wrote: > Page 11 and 12, documentation says to make new menu handlers, but uses > Sub FileNew() instead of RB's Function FileNew() when you create a menu > handler to deal with the command for making a new file. I think you talk about the same. The word SUB REALbasic produce when EXPORT function to disk. Probably I need remove that garbage words, to not confuse people -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon Oct 4 20:54:41 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 4 12:54:47 2004 Subject: New Valentina-REALbasic Tutorial In-Reply-To: Message-ID: On 10/4/04 7:37 PM, "Chuck Pelto" wrote: > Page 16, the documentation says that the Adding records function > encoded into the boPersons object will be called by a menu item. > However, the tutorial does not specify how that menu handler is > supposed to read. > > Page 17, documentation calls for an item called mCallerWindow, however > this item is not created anywhere prior to this line of code. > Therefore, you get an error about property or method not existing. > > Page 17, documentation calls for making calls to the PopulatePanes > method from Window1. However, this method is not being recognized as > existing in Window2, despite the method being declared as "Public" in > Window1. [Note: Should these have been generated at the App level?] > > The source code in the Add, Update and Delete buttons dragged over from > Window1 into Window 2 is generating multiple errors. Most of them about > method or property not existing in Window2, specifically PopulatePanes, > CollectPanes. This despite the fact that these methods in Window1 are > set to Public. [Note: Should these have been generated at the App > level?] Also, with respect to the problem about source code for buttons > from Window1 dragged and dropped into Window2, mDatabase is not > recognized in Window2. > > Still working away on this. Hope you can advise me on what needs to be > done to resolve these issues. Thank you for points, Chuck, I will try resolve this ASAP. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Mon Oct 4 14:54:06 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Mon Oct 4 15:54:13 2004 Subject: Valentina question In-Reply-To: References: Message-ID: <87CD1BD2-1647-11D9-85F3-0003939246BC@mindstarprods.com> On Oct 4, 2004, at 9:41 AM, Ruslan Zasukhin wrote: > On 10/4/04 6:31 PM, "Jochen Peters" wrote: > >>> Also, I have had some trouble with the Load feature in Valentina >>> Studio (I'm using Ver. 1.0a66 Build 0116092004). I cannot get data >>> loaded from a SQL file into a Valentina db. Any suggestions will be >>> appreciated. >> Answer is easy: Valentina 1.x CAN'T load from SQL! (only from XML). >> This feature will be available for Valentina 2.x. > > You can write self simple parser which find ; > After each next SQL command and execute command by command. ViSQL already does this. Just open your database with it and cut and paste your SQL statements into the SQL window. It will go through and execute all of them. -Frank From listmail at gearyweb.com Mon Oct 4 16:31:18 2004 From: listmail at gearyweb.com (michael geary) Date: Mon Oct 4 17:31:31 2004 Subject: corrupted databases - we need a tool! Message-ID: <1C3245FE-1655-11D9-BAED-000A95ACCDAA@gearyweb.com> Hi Folks, I am a big fan of Valentina, and as long as things go smoothly, I don't have any problems. BUT, I think we need some tools to repair/rebuild databases when they go bad. Using the Valentina App (OS X) I just deleted 1000000 records from a table, emptying it. I then closed the DB and tried to re-open it. Nothing. It is broken. Nothing I can do at all. The database is well over 300 MB, so it's not terribly practical for me to store an XML dump on a frequent basis. Fortunately, I'm not really losing any data with this corruption, because I can reconstruct the DB from other sources, but I am losing time. This is not the first time this has happened. Not at all. When things are good, they are very good. But when things are bad, they are very bad. Has anyone had similar experiences? I really wish I didn't have to go through this. michael geary From devlist at mono.info Tue Oct 5 08:32:17 2004 From: devlist at mono.info (Paul Gaspar) Date: Tue Oct 5 01:33:32 2004 Subject: corrupted databases - we need a tool! In-Reply-To: <1C3245FE-1655-11D9-BAED-000A95ACCDAA@gearyweb.com> References: <1C3245FE-1655-11D9-BAED-000A95ACCDAA@gearyweb.com> Message-ID: <4D7A9226-1698-11D9-8A0F-000A95A60398@mono.info> Am 05.10.2004 um 00:31 schrieb michael geary: > Hi Folks, > > I am a big fan of Valentina, and as long as things go smoothly, I > don't have any problems. BUT, I think we need some tools to > repair/rebuild databases when they go bad. > > Using the Valentina App (OS X) I just deleted 1000000 records from a > table, emptying it. I then closed the DB and tried to re-open it. > Nothing. It is broken. Nothing I can do at all. > > The database is well over 300 MB, so it's not terribly practical for > me to store an XML dump on a frequent basis. Fortunately, I'm not > really losing any data with this corruption, because I can reconstruct > the DB from other sources, but I am losing time. > > This is not the first time this has happened. Not at all. When things > are good, they are very good. But when things are bad, they are very > bad. > > Has anyone had similar experiences? I really wish I didn't have to go > through this. Hi Michael, you have my vote for such a tool! Valentina databases are very stable, but there a few cases with problems, and then its hard work to repair it. Paul From sunshine at public.kherson.ua Tue Oct 5 09:37:48 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 01:37:55 2004 Subject: corrupted databases - we need a tool! In-Reply-To: <1C3245FE-1655-11D9-BAED-000A95ACCDAA@gearyweb.com> Message-ID: On 10/5/04 1:31 AM, "michael geary" wrote: Hi Michael, > I am a big fan of Valentina, and as long as things go smoothly, I don't > have any problems. BUT, I think we need some tools to repair/rebuild > databases when they go bad. > > Using the Valentina App (OS X) I just deleted 1000000 records from a > table, emptying it. I then closed the DB and tried to re-open it. > Nothing. It is broken. Nothing I can do at all. Bad. It looks that deletion of huge number of records can cause problems... > The database is well over 300 MB, so it's not terribly practical for me > to store an XML dump on a frequent basis. Fortunately, I'm not really > losing any data with this corruption, because I can reconstruct the DB > from other sources, but I am losing time. > > This is not the first time this has happened. Not at all. When things > are good, they are very good. But when things are bad, they are very > bad. > > Has anyone had similar experiences? I really wish I didn't have to go > through this. 1) I think or 2.0 we will prepare very intensive tests like your. To make sure it works correctly. 2) in Valentina 3.0 we plan add transactions. This is the most powerful way of protection. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From rjb at rz.uni-potsdam.de Tue Oct 5 11:01:41 2004 From: rjb at rz.uni-potsdam.de (Robert Brenstein) Date: Tue Oct 5 04:13:04 2004 Subject: corrupted databases - we need a tool! In-Reply-To: <4D7A9226-1698-11D9-8A0F-000A95A60398@mono.info> References: <1C3245FE-1655-11D9-BAED-000A95ACCDAA@gearyweb.com> <4D7A9226-1698-11D9-8A0F-000A95A60398@mono.info> Message-ID: >Am 05.10.2004 um 00:31 schrieb michael geary: > >>Hi Folks, >> >>I am a big fan of Valentina, and as long as things go smoothly, I >>don't have any problems. BUT, I think we need some tools to >>repair/rebuild databases when they go bad. >> >>Using the Valentina App (OS X) I just deleted 1000000 records from >>a table, emptying it. I then closed the DB and tried to re-open it. >>Nothing. It is broken. Nothing I can do at all. >> >>The database is well over 300 MB, so it's not terribly practical >>for me to store an XML dump on a frequent basis. Fortunately, I'm >>not really losing any data with this corruption, because I can >>reconstruct the DB from other sources, but I am losing time. >> >>This is not the first time this has happened. Not at all. When >>things are good, they are very good. But when things are bad, they >>are very bad. >> >>Has anyone had similar experiences? I really wish I didn't have to >>go through this. > > > >Hi Michael, > >you have my vote for such a tool! Valentina databases are very >stable, but there a few cases with problems, and then its hard work >to repair it. > >Paul > My vote too. This is actually not a new topic if you search the list archives. At some point, Ruslan even added a flag to check whether the db was closed properly but as far as I know it is still not used. At least we got lately a diagnose option, still not finished, but usable. Yes, with the race to add new features and improve operation, the maintenance aspects of Valentina have always gotten a short end of the stick. There is no testing that can prevent all problems. Even with transactions, one may need these since the files can get corrupted. And the stability of Valentina dbs is always very fragile. Like that corruption I recently reported. It came back again this week and that bo was was not even written to in the meantime. Luckily, my db is small enough that xml route gets me out of trouble. But I need to run diagnose regularly. More often than I care or should. Robert From vidal_olivier at yahoo.fr Tue Oct 5 11:30:59 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Tue Oct 5 04:22:03 2004 Subject: corrupted databases - we need a tool! In-Reply-To: References: Message-ID: <4481BC62-16B1-11D9-9E04-00039310B7DA@yahoo.fr> Hi Ruslan and list, Le 5 oct. 04, ? 08:37, Ruslan Zasukhin a ?crit : > Bad. It looks that deletion of huge number of records can cause > problems... ???? Interesting. From what number of erased recordings the problem is it already produced? (To part this one) olivier From vidal_olivier at yahoo.fr Tue Oct 5 11:42:16 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Tue Oct 5 04:33:15 2004 Subject: Valentina 2 Message-ID: Hi Valentina Team, I know that you work a lot on Valentina 2. Everything takes place well as expected ? Be you satisfied with the first betas ? Do you think that the version 2.0 will be ready before the end of the year? 'HAVING' (in 'SELECT') is foreseen in the version 2.0? Thank you very much for all your work. Olivier From sunshine at public.kherson.ua Tue Oct 5 12:56:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 04:56:50 2004 Subject: Valentina 2 In-Reply-To: Message-ID: On 10/5/04 12:42 PM, "olivier" wrote: Hi Oliver, > I know that you work a lot on Valentina 2. > > Everything takes place well as expected ? yes > Be you satisfied with the first betas ? yes > Do you think that the version 2.0 will be ready before the end of the > year? Yes. Even sooner > 'HAVING' (in 'SELECT') is foreseen in the version 2.0? yes > Thank you very much for all your work. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From vidal_olivier at yahoo.fr Tue Oct 5 12:16:46 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Tue Oct 5 05:08:14 2004 Subject: Valentina 2 In-Reply-To: References: Message-ID: perfect ! Le 5 oct. 04, ? 11:56, Ruslan Zasukhin a ?crit : > Even sooner From sunshine at public.kherson.ua Tue Oct 5 13:16:27 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 05:16:41 2004 Subject: corrupted databases - we need a tool! In-Reply-To: <4481BC62-16B1-11D9-9E04-00039310B7DA@yahoo.fr> Message-ID: On 10/5/04 12:30 PM, "olivier" wrote: > Hi Ruslan and list, > > Le 5 oct. 04, ? 08:37, Ruslan Zasukhin a ?crit : > >> Bad. It looks that deletion of huge number of records can cause >> problems... > > ???? > > Interesting. > > From what number of erased recordings the problem is it already > produced? (To part this one) I don't know, just 2-3 times was reported that after deletion at once of many records in table can be prolem -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue Oct 5 07:52:46 2004 From: jda at his.com (jda) Date: Tue Oct 5 06:52:52 2004 Subject: corrupted databases - we need a tool! In-Reply-To: References: Message-ID: > > > >> From what number of erased recordings the problem is it already >> produced? (To part this one) > >I don't know, just 2-3 times was reported that after deletion at once >of many records in table can be prolem > Ruslan, don't you plan to include a more developer-friendly diagnose at some point in 2.x? I mean where we can obtain the results and format our own reports to our users? And you've been talking about a repair function for years -- I thought that was a planned 2.x feature as well (not necessarily 2.0, but 2.x)? Jon From sunshine at public.kherson.ua Tue Oct 5 15:40:13 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 07:40:22 2004 Subject: corrupted databases - we need a tool! In-Reply-To: Message-ID: On 10/5/04 2:52 PM, "jda" wrote: Hi Jon, >>> From what number of erased recordings the problem is it already >>> produced? (To part this one) >> >> I don't know, just 2-3 times was reported that after deletion at once >> of many records in table can be prolem >> > > Ruslan, don't you plan to include a more developer-friendly diagnose > at some point in 2.x? I mean where we can obtain the results and > format our own reports to our users? You mean produce it as string? Remind about this on beta list About 2.0 diagnose I can tell, that I have add diagnose for indexes. > And you've been talking about a repair function for years -- I > thought that was a planned 2.x feature as well (not necessarily 2.0, > but 2.x)? I believe that REAPIR method is a big lie in the DBMS area. For example mySQL's command FIX do fix only indexes. Actually they simply rebuild them it seems. It is very and very often simply not possible to repair in full term of this word. Only real protection -- transactions and log files. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From rjb at rz.uni-potsdam.de Tue Oct 5 15:47:08 2004 From: rjb at rz.uni-potsdam.de (Robert Brenstein) Date: Tue Oct 5 08:53:05 2004 Subject: corrupted databases - we need a tool! In-Reply-To: References: Message-ID: >On 10/5/04 2:52 PM, "jda" wrote: > >Hi Jon, > >>>> From what number of erased recordings the problem is it already >>>> produced? (To part this one) >>> >>> I don't know, just 2-3 times was reported that after deletion at once >>> of many records in table can be prolem >>> >> >> Ruslan, don't you plan to include a more developer-friendly diagnose >> at some point in 2.x? I mean where we can obtain the results and >> format our own reports to our users? > >You mean produce it as string? >Remind about this on beta list > >About 2.0 diagnose I can tell, that I have add diagnose for indexes. > >> And you've been talking about a repair function for years -- I >> thought that was a planned 2.x feature as well (not necessarily 2.0, >> but 2.x)? > >I believe that REAPIR method is a big lie in the DBMS area. > >For example mySQL's command FIX do fix only indexes. >Actually they simply rebuild them it seems. > > >It is very and very often simply not possible to repair in full term of this >word. > >Only real protection -- transactions and log files. > > >-- >Best regards, >Ruslan Zasukhin [ I feel the need...the need for speed ] While true repair might not always be possible, such a tool could offer a rebuild function, which creates a new copy the db by extracting all that is possible to extract from the old one, thus rescuing as much as feasible, hopfully reaching deeper than xml dump can do. Ideally, these things should be part of kernel self, not a standalone tool, so we can use them directly in all environments, allowing us to include recovery options in our products. Trasactions are still distant future while a repair tool has been talked about since early 2002 at least (posts dealing with db corruption go back to 1999 upon quick search) and was expected for 1.x technology to follow on the heels of the diagnose coming to life. But diagnose self still has a few glitches and is not available as part of kernel functionality. As long as Valentina does not clean up and build up certain aspects that have been asked for over the years, it can't shake its shareware-only image and step into the ring of big boys. Robert From wonderfef at noos.fr Tue Oct 5 16:01:59 2004 From: wonderfef at noos.fr (Eric Ferrer) Date: Tue Oct 5 09:02:08 2004 Subject: [V4RBclient] cursor.setblank really necessary? Message-ID: Hello all, In one hand, I have a cursor containing a few records. In another hand, I have a string that contains ids of records, fields names and values. In a loop, I analyse this "updating" string in order to update the records in the cursor. The trouble is that my "updating" string may not contain exactly all the fields available in the cursor: Say my cursor contains "field1", "field2" and "field3" but my updating string only contains values for "field2". So, if I call "cursor.setblank", it will erase the values in "field1" and "field3", which is bad! The question is: Is "cursor.setblank" really necessary? Of course, I could examine first what's in my updating string in order to get the good fields from the cursor, but this is not possible in the way I use my updating string... Thanks Eric From sunshine at public.kherson.ua Tue Oct 5 17:04:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 09:04:59 2004 Subject: corrupted databases - we need a tool! In-Reply-To: Message-ID: On 10/5/04 4:47 PM, "Robert Brenstein" wrote: > While true repair might not always be possible, such a tool could > offer a rebuild function, which creates a new copy the db by > extracting all that is possible to extract from the old one, thus > rescuing as much as feasible, hopfully reaching deeper than xml dump > can do. Ideally, these things should be part of kernel self, not a > standalone tool, so we can use them directly in all environments, > allowing us to include recovery options in our products. Aha, I see idea, Robert! Yes this is good idea. Really good. It is possible to TRY recovery any data which are readable, And code which will read this should have many protection from failure. > Trasactions are still distant future while a repair tool has been > talked about since early 2002 at least (posts dealing with db > corruption go back to 1999 upon quick search) and was expected for > 1.x technology to follow on the heels of the diagnose coming to life. > But diagnose self still has a few glitches and is not available as > part of kernel functionality. As long as Valentina does not clean up > and build up certain aspects that have been asked for over the years, > it can't shake its shareware-only image and step into the ring of big > boys. Right. But as I have told, we have developer which have many years work with Sybase. He say that IF Sybase db corrupts, it also is not many chances to recover it. This is complex task, and it cannot be solved in 5 minutes. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue Oct 5 17:06:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 09:06:38 2004 Subject: [V4RBclient] cursor.setblank really necessary? In-Reply-To: Message-ID: On 10/5/04 5:01 PM, "Eric Ferrer" wrote: Hi Eric, > Hello all, > > In one hand, I have a cursor containing a few records. > In another hand, I have a string that contains ids of records, fields names > and values. > > In a loop, I analyse this "updating" string in order to update the records > in the cursor. > > The trouble is that my "updating" string may not contain exactly all the > fields available in the cursor: > > Say my cursor contains "field1", "field2" and "field3" but my updating > string only contains values for "field2". > So, if I call "cursor.setblank", it will erase the values in "field1" and > "field3", which is bad! SetBlank() should be used before ADD RECORD, but not before UPDATE > The question is: > Is "cursor.setblank" really necessary? Yes for ADD > Of course, I could examine first what's in my updating string in order to > get the good fields from the cursor, but this is not possible in the way I > use my updating string... -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From rjb at rz.uni-potsdam.de Tue Oct 5 16:35:35 2004 From: rjb at rz.uni-potsdam.de (Robert Brenstein) Date: Tue Oct 5 09:53:09 2004 Subject: corrupted databases - we need a tool! In-Reply-To: References: Message-ID: >On 10/5/04 4:47 PM, "Robert Brenstein" wrote: > >> While true repair might not always be possible, such a tool could >> offer a rebuild function, which creates a new copy the db by >> extracting all that is possible to extract from the old one, thus >> rescuing as much as feasible, hopfully reaching deeper than xml dump >> can do. Ideally, these things should be part of kernel self, not a >> standalone tool, so we can use them directly in all environments, >> allowing us to include recovery options in our products. > >Aha, I see idea, Robert! > >Yes this is good idea. >Really good. > >It is possible to TRY recovery any data which are readable, >And code which will read this should have many protection from failure. > > >> Trasactions are still distant future while a repair tool has been >> talked about since early 2002 at least (posts dealing with db >> corruption go back to 1999 upon quick search) and was expected for >> 1.x technology to follow on the heels of the diagnose coming to life. > > >> But diagnose self still has a few glitches and is not available as >> part of kernel functionality. As long as Valentina does not clean up >> and build up certain aspects that have been asked for over the years, >> it can't shake its shareware-only image and step into the ring of big >> boys. > >Right. > >But as I have told, we have developer which have many years work with >Sybase. He say that IF Sybase db corrupts, it also is not many chances to >recover it. > >This is complex task, and it cannot be solved in 5 minutes. Sure. But last more extensive discussion was in early 2002 and a few releases came out since then. DB integrity check was first delayed because it required file format change. After that happened, it kept waiting for diagnose to grow up. So, a couple years later, 1.x reached its life end so do speak and still can't check integrity when opening dbs and recovery is restricted to xml dump. Let me be clear that I/we have no problem accepting the fact that such a tool can't always be successful. It will be like using Norton Disk Doctor for diagnosing and repairing hard drives. Norton can correct few things in place, others have to be recovered to another drives, but sometimes it can't do anything useful. But at least it lets us try. The fact that Sybase can't do it well is not an excuse. I haven't used Sybase but spent many years with FileMaker and never had any serious corruption issues. And the few times I crashed, the built-in repair of FileMaker fixed the db up. I know that you are more of a visionary and concept-man and thus things like recovery, documentation, examples, etc tedious aspects are not exciting for you (and waste of your talents as well), but now that you have more people on your team, you can set someone up to it. Robert From joergp at three-2-one.com Tue Oct 5 20:07:40 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Tue Oct 5 13:07:53 2004 Subject: [V4RB] Adding unique relational records... Message-ID: <72103747-16F9-11D9-A47A-000A959A1396@three-2-one.com> Hi, I have a table "Tracks" pointing to a table "Artists" and use the following code in "Artists" to add a new unique artist: Function AddArtist(s as string) As integer dim cur As VCursor cur = database.SQLSelect("select RecID from Artists where Artist='" + escStr(s) + "' no_case") if (cur.recordCount = 0) then SetBlank Artist.value = s AddRecord return GetRecID else return cur.ULongField(1).value end if End Function In "Tracks" I call it like: mArtistPtr.value = musicDatabase(dataBase).mArtists.AddArtist(artist) Is there an easier, perhaps faster way to do this? Thanks for any comment, J?rg From sunshine at public.kherson.ua Tue Oct 5 21:36:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 13:36:53 2004 Subject: [V4RB] Adding unique relational records... In-Reply-To: <72103747-16F9-11D9-A47A-000A959A1396@three-2-one.com> Message-ID: On 10/5/04 9:07 PM, "Joerg Pressel" wrote: > Hi, > > I have a table "Tracks" pointing to a table "Artists" and use the > following code in "Artists" to add a new unique artist: > > Function AddArtist(s as string) As integer > dim cur As VCursor > > cur = database.SQLSelect("select RecID from Artists where Artist='" > + escStr(s) + "' no_case") > if (cur.recordCount = 0) then > SetBlank > Artist.value = s > AddRecord > return GetRecID > else > return cur.ULongField(1).value > end if > > End Function > > In "Tracks" I call it like: > mArtistPtr.value = musicDatabase(dataBase).mArtists.AddArtist(artist) > > Is there an easier, perhaps faster way to do this? I do not understand why you use string search for Artist ? > cur = database.SQLSelect("select RecID from Artists where Artist='" > + escStr(s) + "' no_case") Btw, no_case have sense only for LIKE. Must be where Artists = 145 ^^^^ number -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From joergp at three-2-one.com Tue Oct 5 20:52:35 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Tue Oct 5 13:52:47 2004 Subject: [V4RB] Adding unique relational records... In-Reply-To: References: Message-ID: > I do not understand why you use string search for Artist ? > >> cur = database.SQLSelect("select RecID from Artists where Artist='" >> + escStr(s) + "' no_case") > > Btw, no_case have sense only for LIKE. OK ? thanks. > Must be > > where Artists = 145 > ^^^^ number No. Table boArtist has one field Artist as VVarChar which holds the artist name. Table boTracks has mArtistPtr as VObjectPtr that points to boArtist. I want to add a new artist: so in boTracks I call my function in boArtist to add a new unique artist and get back its RecID: (add other fields here) mArtistPtr.value = musicDatabase(dataBase).mArtists.AddArtist(artist) AddRecord Is this there an easier way? From sunshine at public.kherson.ua Tue Oct 5 22:01:16 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 5 14:01:32 2004 Subject: [V4RB] Adding unique relational records... In-Reply-To: Message-ID: On 10/5/04 9:52 PM, "Joerg Pressel" wrote: >> I do not understand why you use string search for Artist ? >> >>> cur = database.SQLSelect("select RecID from Artists where Artist='" >>> + escStr(s) + "' no_case") >> >> Btw, no_case have sense only for LIKE. > > OK ? thanks. > >> Must be >> >> where Artists = 145 >> ^^^^ number > > No. > > Table boArtist has one field Artist as VVarChar which holds the artist > name. > Table boTracks has mArtistPtr as VObjectPtr that points to boArtist. > > I want to add a new artist: so in boTracks I call my function in > boArtist to add a new unique artist and get back its RecID: > (add other fields here) > mArtistPtr.value = musicDatabase(dataBase).mArtists.AddArtist(artist) > AddRecord > > Is this there an easier way? I think no -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From joergp at three-2-one.com Tue Oct 5 21:03:22 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Tue Oct 5 14:03:34 2004 Subject: [V4RB] Adding unique relational records... In-Reply-To: References: Message-ID: <3A07989A-1701-11D9-A47A-000A959A1396@three-2-one.com> >> Is this there an easier way? > > I think no I will keep it that way. Thanks, J?rg From cindy at kowhaiprogramming.com Wed Oct 6 08:25:21 2004 From: cindy at kowhaiprogramming.com (Cindy Brown) Date: Tue Oct 5 14:26:50 2004 Subject: Where Valentina is heading Message-ID: Hi Ruslan, We are currently working through a government contract at the moment and we have to provide some idea of the long-term viability of Valentina. I was hoping you would be able to answer a few questions to help us with this: How long has Valentina been going? How many staff are employed in programming Valentina, etc.? Roughly how many people are using the latest current version of Valentina? What sort of aims do you have for Valentina in the future and roughly when would you expect to accomplish them? E.g. v2.0, transactions, etc. The government wants to know that Valentina has a plan for the future. What sort of assurances can you give that Valentina will be here for a long time? I really hope you can give us some answers on these as, with government contracts, they like to know exactly where their software is heading and whether it is going to last for a while. Thanking you very much in anticipation. Cindy Brown New Zealand From pmccon at bigpond.net.au Wed Oct 6 10:52:07 2004 From: pmccon at bigpond.net.au (Peter McConachie) Date: Tue Oct 5 20:00:37 2004 Subject: Help on scripting languages Message-ID: Hi Everyone I know that a lot of subscribers to this list use Revolution, Director and other scripting languages. I'm trying to make a comparison of different scripting languages for a talk I'm to present in the near future. I would like some assistance on one point. It seems to me that Revolution and SuperCard are unique in the way they provide frontScript and backScript functionality to the message path. This allows some unique opportunities. Can anyone tell me the following: 1. Does this functionality exist native in any other scripting languages ? 2. It does not appear native in Director. Has anyone managed to create a workaround where one can have a script or object other than the normal cast members get first crack at system messages ? 3. Can anyone advance a reason why is it that more languages do not provide ? For example if I was to consider a Windows based language like Delphi is it simply that Delphi chooses not to implement or cant because of the way such an application handles messages passed to it from Windows ? With thanks Peter McConachie From rjb at rz.uni-potsdam.de Wed Oct 6 11:47:22 2004 From: rjb at rz.uni-potsdam.de (Robert Brenstein) Date: Wed Oct 6 04:48:50 2004 Subject: Help on scripting languages In-Reply-To: References: Message-ID: >Hi Everyone > >I know that a lot of subscribers to this list use Revolution, Director and >other scripting languages. > >I'm trying to make a comparison of different scripting languages for a talk >I'm to present in the near future. I would like some assistance on one >point. > >It seems to me that Revolution and SuperCard are unique in the way they >provide frontScript and backScript functionality to the message path. This >allows some unique opportunities. > >Can anyone tell me the following: > >1. Does this functionality exist native in any other scripting languages ? > >2. It does not appear native in Director. Has anyone managed to create a >workaround where one can have a script or object other than the normal cast >members get first crack at system messages ? > >3. Can anyone advance a reason why is it that more languages do not provide >? For example if I was to consider a Windows based language like Delphi is >it simply that Delphi chooses not to implement or cant because of the way >such an application handles messages passed to it from Windows ? > >With thanks >Peter McConachie > Peter, I am not sure whether Front and Backscripts can be really considered language features. I mean that message paths are really part of the runtime environment in my mind. Naturally, there is no clear border between the language and the runtime environment since the language has to support its functionality. The way I look at this is that message paths are part of the event-driven programming model more than a language feature per se. Some people may look at this as an instantion of inheritance. Robert From forgete at cafederic.com Wed Oct 6 14:46:07 2004 From: forgete at cafederic.com (Eric Forget) Date: Wed Oct 6 13:46:19 2004 Subject: [VSDK] Crash with Text search... Message-ID: Hi Ruslan, I found a new way to crash Valentina. It is reproducible even in ViSQL. You make a query like this: SELECT Name FROM MyTable WHERE FullSearch LIKE 'p' This is the Valentina crash stack: 0 com.paradigmasoft.vsdk 0xb00419f4 FBL_UseFormula::UseFormulaForTextMethod(BF_BLOB*) + 0x90 1 com.paradigmasoft.vsdk 0xb00411d0 FBL_UseFormula::Run() + 0x34 2 com.paradigmasoft.vsdk 0xb0029330 BF_TEXT::SearchLike(char const*, bool) + 0xe0 3 com.paradigmasoft.vsdk 0xb0014384 FBL_BaseField::MakeSearch(OBL_FieldConstraints*) + 0x2f4 4 com.paradigmasoft.vsdk 0xb002d19c FBL_IndexedTable::MakeSearch(OBL_FieldConstraints*) + 0x74c 5 com.paradigmasoft.vsdk 0xb00c1ef8 OBL_CursorTable_Base::SearchLeaf(OBL_Parser*, OBL_FieldConstraints*, FBL_Array*) + 0x80 6 com.paradigmasoft.vsdk 0xb00c1df0 OBL_CursorTable_Base::DoSelections(OBL_Parser*) + 0x12c 7 com.paradigmasoft.vsdk 0xb00c45c0 OBL_CursorTable_Indirect::BuildMultiplicationTable(OBL_Parser*) + 0x1c 8 com.paradigmasoft.vsdk 0xb00c356c OBL_CursorTable_Indirect::BuildTable(OBL_Parser*) + 0x3c 9 com.paradigmasoft.vsdk 0xb007f2f8 OBL_Cursor::BuildCursorTable() + 0x30 10 com.paradigmasoft.vsdk 0xb007ed10 OBL_Cursor::Execute() + 0x4c If you are interested, I may send you, directly, a database with just 2 records. Unfortunately, this is a show stopper for my next release... ?ric ___________________________________________________________________ Eric Forget Cafederic ForgetE@cafederic.com From snw at paradise.net.nz Thu Oct 7 10:52:06 2004 From: snw at paradise.net.nz (Sean Wilson) Date: Wed Oct 6 16:52:13 2004 Subject: Help on scripting languages In-Reply-To: References: Message-ID: <6.1.2.0.2.20041007104425.02e2beb0@pop3.paradise.net.nz> >It seems to me that Revolution and SuperCard are unique in the way they >provide frontScript and backScript functionality to the message path. This >allows some unique opportunities. > >Can anyone tell me the following: > >2. It does not appear native in Director. Has anyone managed to create a >workaround where one can have a script or object other than the normal cast >members get first crack at system messages ? I'm not familiar with Rev and the other card-based environments, so I'm not sure what frontScript and backScript are, but do they bear any relationship to Director's mouse(/key)Down(/Up)Script properties for 'intercepting' these events? -Sean. From pmccon at bigpond.net.au Thu Oct 7 09:00:40 2004 From: pmccon at bigpond.net.au (Peter McConachie) Date: Wed Oct 6 18:09:03 2004 Subject: Help on scripting languages In-Reply-To: <6.1.2.0.2.20041007104425.02e2beb0@pop3.paradise.net.nz> Message-ID: on 7/10/2004 7:52 am, Sean Wilson at snw@paradise.net.nz wrote: > >> It seems to me that Revolution and SuperCard are unique in the way they >> provide frontScript and backScript functionality to the message path. This >> allows some unique opportunities. >> >> Can anyone tell me the following: >> >> 2. It does not appear native in Director. Has anyone managed to create a >> workaround where one can have a script or object other than the normal cast >> members get first crack at system messages ? > > I'm not familiar with Rev and the other card-based environments, so I'm not > sure what frontScript and backScript are, but do they bear any relationship > to Director's mouse(/key)Down(/Up)Script properties for 'intercepting' > these events? > Sean, Thanks for the reply FrontScripts in Revolution and Supercard allow a programmer to intercept all system messages before they are passed to the object which would be their normal destination (target). These can be pre-processed and then passed to the normal target or rejected if inappropriate. If one wanted, for example, to ensure that certain characters could not be entered in fields (say because they were reserved delimiters used in processing strings or database fields) the simplest way would be to put a rawKeyDown handler in a frontScript to reject the message if it contained the reserved character(s) or otherwise pass the message to the normal target (possibly for further processing). Advantage one script protecting multiple objects + no chance that any objects created in the future can break the protocol. The Revolution rawKeyDown message is probably the same as Directors keyDown. The requirement to enter text and access databases is a task often performed in Director/Valentina. For this reason I believe the need is applicable to non-card based metaphors. I just dont know if the concept of frontScripting is possible in Director, and if so how it is done. Many thanks Peter McConachie From snw at paradise.net.nz Thu Oct 7 12:22:35 2004 From: snw at paradise.net.nz (Sean Wilson) Date: Wed Oct 6 18:23:01 2004 Subject: Help on scripting languages In-Reply-To: References: <6.1.2.0.2.20041007104425.02e2beb0@pop3.paradise.net.nz> Message-ID: <6.1.2.0.2.20041007121611.02dfdd70@pop3.paradise.net.nz> >FrontScripts in Revolution and Supercard allow a programmer to intercept all >system messages before they are passed to the object which would be their >normal destination (target). These can be pre-processed and then passed to >the normal target or rejected if inappropriate. > >If one wanted, for example, to ensure that certain characters could not be >entered in fields (say because they were reserved delimiters used in >processing strings or database fields) the simplest way would be to put a >rawKeyDown handler in a frontScript to reject the message if it contained >the reserved character(s) or otherwise pass the message to the normal >target (possibly for further processing). And if one wanted to do the same thing in Director you could use the keyDownScript and stopEvent/pass to filter keys so it sounds to me like similar functionality. But you could also do the same thing in with keyDown because the character hasn't been "entered" into the field yet. It depends more on at which level you want the filtering to occur. > Advantage one script protecting >multiple objects + no chance that any objects created in the future can >break the protocol. > >The Revolution rawKeyDown message is probably the same as Directors keyDown. > >The requirement to enter text and access databases is a task often performed >in Director/Valentina. For this reason I believe the need is applicable to >non-card based metaphors. > >I just dont know if the concept of frontScripting is possible in Director, >and if so how it is done. What you describe can certainly be done and as with lots of tasks in Director: there are many ways to skin a cat. Cheers, -Sean. From pmccon at bigpond.net.au Thu Oct 7 10:04:39 2004 From: pmccon at bigpond.net.au (Peter McConachie) Date: Wed Oct 6 19:13:03 2004 Subject: Help on scripting languages In-Reply-To: <6.1.2.0.2.20041007121611.02dfdd70@pop3.paradise.net.nz> Message-ID: >> FrontScripts in Revolution and Supercard allow a programmer to intercept all >> system messages before they are passed to the object which would be their >> normal destination (target). These can be pre-processed and then passed to >> the normal target or rejected if inappropriate. >> >> If one wanted, for example, to ensure that certain characters could not be >> entered in fields (say because they were reserved delimiters used in >> processing strings or database fields) the simplest way would be to put a >> rawKeyDown handler in a frontScript to reject the message if it contained >> the reserved character(s) or otherwise pass the message to the normal >> target (possibly for further processing). > > And if one wanted to do the same thing in Director you could use the > keyDownScript and stopEvent/pass to filter keys so it sounds to me like > similar functionality. But you could also do the same thing in with keyDown > because the character hasn't been "entered" into the field yet. It depends > more on at which level you want the filtering to occur. > >> Advantage one script protecting >> multiple objects + no chance that any objects created in the future can >> break the protocol. >> >> The Revolution rawKeyDown message is probably the same as Directors keyDown. >> >> The requirement to enter text and access databases is a task often performed >> in Director/Valentina. For this reason I believe the need is applicable to >> non-card based metaphors. >> >> I just dont know if the concept of frontScripting is possible in Director, >> and if so how it is done. > > What you describe can certainly be done and as with lots of tasks in > Director: there are many ways to skin a cat. > Sean I'm still not certain that we are discussing the same concept. In Revolution one inserts a front script into the message heirachy using the command "insert script of into front". The inserted script has its own keyDownScript which can stop/pass the event. The point is that all messages are sent to the inserted object *before* they are passed to the normal target and then up the message heirachy if passed. I cannot see any equivalent to the insert command. I am very new to Director and wonder if you could detail exactly how one goes about setting up a front object that varies the normal event entry point and so gets first crack at all events. Many thanks, Peter McConachie From snw at paradise.net.nz Thu Oct 7 14:27:44 2004 From: snw at paradise.net.nz (Sean Wilson) Date: Wed Oct 6 20:28:01 2004 Subject: Help on scripting languages In-Reply-To: References: <6.1.2.0.2.20041007121611.02dfdd70@pop3.paradise.net.nz> Message-ID: <6.1.2.0.2.20041007141632.036511a0@pop3.paradise.net.nz> >I'm still not certain that we are discussing the same concept. More and more I'm thinking we are :-) >In Revolution one inserts a front script into the message heirachy using the >command "insert script of into front". The inserted >script has its own keyDownScript which can stop/pass the event. > >The point is that all messages are sent to the inserted object *before* they >are passed to the normal target and then up the message heirachy if passed. > >I cannot see any equivalent to the insert command. I am very new to Director >and wonder if you could detail exactly how one goes about setting up a front >object that varies the normal event entry point and so gets first crack at >all events. I would tend to use a prepareMovie handler to set up a key filter. Something like: on prepareMovie the keyDownScript = "mFilterKeys" end on mFilterKeys case(the key) of -- example filtered key -- perhaps safer to use the keyCode here, but: "a": stopEvent otherwise: pass end case end I'm quite happy to discuss Director's methods etc. but as this is potentially not of interest to anyone else we should perhaps take it off-list. Cheers, -Sean. From sunshine at public.kherson.ua Thu Oct 7 09:21:34 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 01:21:47 2004 Subject: [VSDK] Crash with Text search... In-Reply-To: Message-ID: On 10/6/04 9:46 PM, "Eric Forget" wrote: Hi Eric, > I found a new way to crash Valentina. It is reproducible even in ViSQL. You > make a query like this: > > SELECT Name FROM MyTable WHERE FullSearch LIKE 'p' You mean FullSearch is method ? Yes send me db > This is the Valentina crash stack: > > 0 com.paradigmasoft.vsdk 0xb00419f4 > FBL_UseFormula::UseFormulaForTextMethod(BF_BLOB*) + 0x90 > 1 com.paradigmasoft.vsdk 0xb00411d0 FBL_UseFormula::Run() + 0x34 > 2 com.paradigmasoft.vsdk 0xb0029330 BF_TEXT::SearchLike(char const*, > bool) + 0xe0 > 3 com.paradigmasoft.vsdk 0xb0014384 > FBL_BaseField::MakeSearch(OBL_FieldConstraints*) + 0x2f4 > 4 com.paradigmasoft.vsdk 0xb002d19c > FBL_IndexedTable::MakeSearch(OBL_FieldConstraints*) + 0x74c > 5 com.paradigmasoft.vsdk 0xb00c1ef8 > OBL_CursorTable_Base::SearchLeaf(OBL_Parser*, OBL_FieldConstraints*, > FBL_Array*) + 0x80 > 6 com.paradigmasoft.vsdk 0xb00c1df0 > OBL_CursorTable_Base::DoSelections(OBL_Parser*) + 0x12c > 7 com.paradigmasoft.vsdk 0xb00c45c0 > OBL_CursorTable_Indirect::BuildMultiplicationTable(OBL_Parser*) + 0x1c > 8 com.paradigmasoft.vsdk 0xb00c356c > OBL_CursorTable_Indirect::BuildTable(OBL_Parser*) + 0x3c > 9 com.paradigmasoft.vsdk 0xb007f2f8 OBL_Cursor::BuildCursorTable() + > 0x30 > 10 com.paradigmasoft.vsdk 0xb007ed10 OBL_Cursor::Execute() + 0x4c > > If you are interested, I may send you, directly, a database with just 2 > records. > > Unfortunately, this is a show stopper for my next release... -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From wgregoire at paris.mediagerance.com Thu Oct 7 12:33:17 2004 From: wgregoire at paris.mediagerance.com (Gregoire William) Date: Thu Oct 7 05:30:31 2004 Subject: SQL query IFNULL In-Reply-To: Message-ID: <000901c4ac59$0ffc5b40$1d0a11ac@wgregoire> i bump this one year old message because i currently have the same prblem... i'm trying to do a SELECT IFNULL(name,'rien') but i read on ValentinaList you need a BaseObject methods I modified the "005 (Methods)" example : I removed the data of "frank" and modified nameUpr to have "jon roberts 26/06/2001 16:41:12 JON ROBERTS 2001 " " schima 27/06/1998 16:41:12 rien SCHIMA 1998 " 1/ with set fldRef = addField(gDB, ArtistRef, "nameUpr", #kTypeString, 20, 0, "IF(name is null, name,'rien')" ) , I have "jon roberts 26/06/2001 16:41:12 rien ROBERTS 2001 " " schima 27/06/1998 16:41:12 rien SCHIMA 1998 " 2/ with set fldRef = addField(gDB, ArtistRef, "nameUpr", #kTypeString, 20, 0, "IFnull(name,'rien')" ) , i have "jon roberts 26/06/2001 16:41:12 jon ROBERTS 2001 " " schima 27/06/1998 16:41:12 SCHIMA 1998 " what's wrong ?! thanks WilG -----Message d'origine----- De : valentina-bounces@lists.macserve.net [mailto:valentina-bounces@lists.macserve.net]De la part de Ruslan Zasukhin Envoye : mercredi 29 octobre 2003 14:35 A : valentina@lists.macserve.net Objet : Re: SQL query IFNULL on 10/29/03 15:29, Martin Kloss at martin.kloss@gmx.de wrote: > At 14:03 29.10.2003, you wrote: >> You can use functions only in BaseObject methods. >> You need create BaseObject method and use it in SELECT. > > ok, I created a method "has_comment" with the calculation: > > IF(comment != '', 'j', 'n') > > But when selecting the records: > > SELECT id, has_comment FROM data > > all I get is "j", but some records don't have any data in the > field, so I assume it's empty. > > IF(comment IS NULL, 'n', 'j') > > gives me the same result. Try to check on small db with VAPP. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From forgete at cafederic.com Thu Oct 7 08:53:22 2004 From: forgete at cafederic.com (Eric Forget) Date: Thu Oct 7 07:54:09 2004 Subject: [VSDK] Crash with Text search... In-Reply-To: Message-ID: > On 10/6/04 9:46 PM, "Eric Forget" wrote: > > Hi Eric, > >> I found a new way to crash Valentina. It is reproducible even in ViSQL. You >> make a query like this: >> >> SELECT Name FROM MyTable WHERE FullSearch LIKE 'p' > > You mean FullSearch is method ? Yep... > Yes send me db Done... ?ric ___________________________________________________________________ Eric Forget Cafederic ForgetE@cafederic.com From fb at memedia.de Thu Oct 7 15:31:52 2004 From: fb at memedia.de (Florian Bogeschdorfer) Date: Thu Oct 7 08:32:08 2004 Subject: 2 databases with V4MD server/client Message-ID: Hi Ruslan! I want to bring two products together. On is an adress/mailing system, the other one a kiosk terminal. I thought of just putting both existing databases in the server/databases folder and that is it. Am I right that I can not use different passwords for the two databases? One application will use both databases, is it right that I will need two connections, one for each database (x=openDatabase(1) and y=openDatabase(2)) ? Is there a way to open both DBs using one variable? Also what about licenses? When I do 2 connections from one program, will it count 2 licenses? TIA, Florian From milan at mook.co.uk Thu Oct 7 14:41:50 2004 From: milan at mook.co.uk (Milan) Date: Thu Oct 7 08:42:35 2004 Subject: Speed Issues From CD-ROM Message-ID: <200410071528.PAA04147@moo.mook.co.uk> Hi, I'm using the Valentina Xtra for Director on a CD-ROM to perform searches on a database of approx. 60mb. Unfortunately, I'm finding the first couple of searches painfully slow (I assume these first searches access the .dat file and cache it?). Is there any way I can speed up the search process, or compress the database to reduce the wait time? Milan Hawkins Mook a. Tuscany Wharf, 4a Orsman Road, London N1 5QJ e. milan@mook.co.uk w. www.mook.co.uk t. 020 7749 2525 f. 020 7749 2526 What's new at Mook? * Play our new game for the smart forfour at http://www.smartlings.timesonline.co.uk/?origin=source * Phil Jones, former vice chairman of EHS Brann, joins Mook as Chairman. * Major exhibition site launches for the V&A http://www.vam.ac.uk/encounters/ * Check out the nominees for the 2004 MTV Video Music Awards at http://www.mtv.co.uk/vma * Short film 'Flowerbed' featured in 8th international http://www.onedotzero.com festival. * Mook voted 'Top 10 most respected agency' in New Media Age Top 100. * Mook named 'Top 30 New Media company' in Televisual magazine industry survey. From sunshine at public.kherson.ua Thu Oct 7 17:12:59 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 09:13:07 2004 Subject: SQL query IFNULL In-Reply-To: <000901c4ac59$0ffc5b40$1d0a11ac@wgregoire> Message-ID: On 10/7/04 1:33 PM, "Gregoire William" wrote: Hi Gregoire, > i bump this one year old message because i currently have the same prblem... > > i'm trying to do a > SELECT IFNULL(name,'rien') > but i read on ValentinaList you need a BaseObject methods > I modified the "005 (Methods)" example : > I removed the data of "frank" and modified nameUpr to have > "jon roberts 26/06/2001 16:41:12 JON ROBERTS 2001 " > " schima 27/06/1998 16:41:12 rien SCHIMA 1998 " > > 1/ > with set fldRef = addField(gDB, ArtistRef, "nameUpr", #kTypeString, 20, 0, > "IF(name is null, name,'rien')" ) , I have > > "jon roberts 26/06/2001 16:41:12 rien ROBERTS 2001 " > " schima 27/06/1998 16:41:12 rien SCHIMA 1998 " > > > 2/ > with set fldRef = addField(gDB, ArtistRef, "nameUpr", #kTypeString, 20, 0, > "IFnull(name,'rien')" ) , i have > > "jon roberts 26/06/2001 16:41:12 jon ROBERTS 2001 " > " schima 27/06/1998 16:41:12 SCHIMA 1998 " > > > > > what's wrong ?! What you expect and what you get ? IFnull(name,'rien') will return 'rien' only if field name is NULL. You have remove Frank... Where and how ? you sure that you have set it to NULL ? may be it is empty string ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 7 17:15:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 09:15:52 2004 Subject: Speed Issues From CD-ROM In-Reply-To: <200410071528.PAA04147@moo.mook.co.uk> Message-ID: On 10/7/04 4:41 PM, "Milan" wrote: Hi Milan, > I'm using the Valentina Xtra for Director on a CD-ROM to perform searches on > a database of approx. 60mb. Unfortunately, I'm finding the first couple of > searches painfully slow (I assume these first searches access the .dat file > and cache it?). > > Is there any way I can speed up the search process, or compress the database > to reduce the wait time? Actually you can have the next problem: * you have db * but you have not index some fields which you try to search so Valentina spend time to build index in the TMP folder on HDD. If this is a case you need study what fields must be indexed. Or you can play with your db on HDD before put it on CD. If this is not a case, then probably yes, you see result of -- slow CD -- caching You can try do XML dump/load to get the most compact db. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From theguy at sympatico.ca Thu Oct 7 10:19:32 2004 From: theguy at sympatico.ca (Guy McLoughlin) Date: Thu Oct 7 09:19:53 2004 Subject: Speed Issues From CD-ROM In-Reply-To: <200410071528.PAA04147@moo.mook.co.uk> References: <200410071528.PAA04147@moo.mook.co.uk> Message-ID: <6.1.2.0.2.20041007101440.01fa70e8@pop6.sympatico.ca> Hi Milan, If your CD burning software allows this ( I use NERO which has this feature ), make sure your database file is burned last, so that it will sit at the outermost edge of the CD. If you have to add a big dummy file, which you write first, so that it forces everything else to the outer edges of the CD. There is a huge read speed difference between the inner and outer parts of a CD-ROM disc. With NERO you set the burn priority to LOW for the files you want burned first ( inner part of the CD ), and set the files you want to be burned last to either MEDIUM or HIGH, which will burn them last. ( outer part of the CD ) - Guy At 09:41 AM 07/10/2004, Milan wrote: > >I'm using the Valentina Xtra for Director on a CD-ROM to perform searches on >a database of approx. 60mb. Unfortunately, I'm finding the first couple of >searches painfully slow (I assume these first searches access the .dat file >and cache it?). > >Is there any way I can speed up the search process, or compress the database >to reduce the wait time? -------------------------- Guy McLoughlin New Media Developer Toronto Ontario Canada E: theguy@sympatico.ca --------------------------- From wgregoire at paris.mediagerance.com Thu Oct 7 16:44:04 2004 From: wgregoire at paris.mediagerance.com (Gregoire William) Date: Thu Oct 7 09:41:11 2004 Subject: SQL query IFNULL (V4MD) In-Reply-To: Message-ID: <001601c4ac7c$18826f20$1d0a11ac@wgregoire> here we go : set fldRef = addField(gDB, ArtistRef, "name", #kTypeString, 20 ) set fldRef = addField(gDB, ArtistRef, "lname", #kTypeString, 20 ) set fldRef = addField(gDB, ArtistRef, "method1", #kTypeString, 20, 0, "IFNULL(name, 'rien')" ) set fldRef = addField(gDB, ArtistRef, "method2", #kTypeString, 20, 0, "IF(name is null, 'rien', name)" ) SetBlank(PersonCursor) SetField(PersonCursor, "name", "jon" ) SetField(PersonCursor, "lname", "roberts" ) AddRecord(PersonCursor) SetBlank(PersonCursor) -- no name SetField(PersonCursor, "lname", "schima" ) AddRecord(PersonCursor) result : -- "jon roberts jon jon " -- " schima " and not -- "jon roberts jon jon " -- " schima rien rien " i first though it was the strings but i tried with a byte type field and i have the same result... SO, it seems SetBlank add a defaut value ("" for string, 0 for byte) and not NULL Am I wrong ? I can bypass it using IF(name='', 'rien', name) ? How does it work with ImportAscii ? If I import an ascii file with ImportAscii(myTextFile, "$", RETURN) who contains --- jon$roberts $schima --- will it add a NULL for "schima" name or a "" ? thanks WilG -----Message d'origine----- De : valentina-bounces@lists.macserve.net [mailto:valentina-bounces@lists.macserve.net]De la part de Ruslan Zasukhin Envoye : jeudi 7 octobre 2004 16:13 A : valentina@lists.macserve.net Objet : Re: SQL query IFNULL What you expect and what you get ? IFnull(name,'rien') will return 'rien' only if field name is NULL. You have remove Frank... Where and how ? you sure that you have set it to NULL ? may be it is empty string ? From sunshine at public.kherson.ua Thu Oct 7 18:11:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 10:11:43 2004 Subject: Speed Issues From CD-ROM In-Reply-To: <6.1.2.0.2.20041007101440.01fa70e8@pop6.sympatico.ca> Message-ID: On 10/7/04 5:19 PM, "Guy McLoughlin" wrote: > Hi Milan, > > If your CD burning software allows this ( I use NERO which has this feature ), > make sure your database file is burned last, so that it will sit at the > outermost > edge of the CD. If you have to add a big dummy file, which you write first, so > that it forces everything else to the outer edges of the CD. > > There is a huge read speed difference between the inner and outer parts of > a CD-ROM disc. Interesting TIP, Guy! We need add this somehow to docs/tips What is difference? 2? 3? 10 times? > With NERO you set the burn priority to LOW for the files you want burned > first ( inner part of the CD ), and set the files you want to be burned last > to > either MEDIUM or HIGH, which will burn them last. ( outer part of the CD ) -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From theguy at sympatico.ca Thu Oct 7 11:40:46 2004 From: theguy at sympatico.ca (Guy McLoughlin) Date: Thu Oct 7 10:41:07 2004 Subject: Speed Issues From CD-ROM In-Reply-To: References: <6.1.2.0.2.20041007101440.01fa70e8@pop6.sympatico.ca> Message-ID: <6.1.2.0.2.20041007112647.01fb5e70@pop6.sympatico.ca> Hi Ruslan, At 11:11 AM 07/10/2004, Ruslan Zasukhin wrote: >> There is a huge read speed difference between the inner and outer parts of >> a CD-ROM disc. > >Interesting TIP, Guy! > >We need add this somehow to docs/tips > >What is difference? 2? 3? 10 times? It depends on the CD-ROM drive. Some drives rotate the disc at a constant speed, and some spin the disc faster when reading the inner tracks. The inner most track has a radius of about 2 cm, and the outer most track has a radius of about 5.5 cm, which means that the length of tracks are: Length of a circle = 2 * Pi * radius Inner most track length = approx 12.5 cm Outer most track length = 34.5 cm Which gives us a ratio of about 2.75 So our speed difference is a maxium of 2.75 times faster. Definitely noticeable when reading data from disc. - Guy -------------------------- Guy McLoughlin New Media Developer Toronto Ontario Canada E: theguy@sympatico.ca --------------------------- From cbpelto at pcisys.net Thu Oct 7 10:01:47 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 7 11:01:57 2004 Subject: V4RB Tutorial 1, TestAddRecords Message-ID: <3161E591-187A-11D9-8136-00306549C134@pcisys.net> I've been twiddling around with trying to get this method to work and have met with complete failure. On page 16 of the tutorial it reads.... boPerson.TestAddRecords: Sub TestAddRecords(N as integer) ... The implication being that TestAddRecords() is a method inside of the boPerson object. At the bottom of the page it says... This method will be called from the Window1 menu event handler. There is not code or guidance where to put such code to call TestAddRecords in the tutorial. The text on page 16 implies that it will be located in the source code of Window 1. So, I put in the following code into a Menu Handler of Window1 to try it. Function FileAdd() dim res as boolean dim N as integer N = 10000 res = boPerson.TestAddRecords(N) if (res = true) then MsgBox("Done adding 10000 records.") else MsgBox("Failed adding 10000 records!!!") end if End Function Note: I modified the code of TestAddRecords to include the return of a boolean if it worked or didn't work. What is happening when I compile to run is that it will not compile. It gives me an error message about the TestAddRecords not being a method or property in Window1. However, I am using the proper dot protocol for calling a method in a different object, according to communications on the RB listserv. Is there a problem using the dot protocol of RB when the method being called is in a Valentina object? Please advise.... Regards, Chuck Pelto From milan at mook.co.uk Thu Oct 7 17:56:32 2004 From: milan at mook.co.uk (Milan) Date: Thu Oct 7 11:57:09 2004 Subject: Speed Issues From CD-ROM In-Reply-To: Message-ID: <200410071843.SAA09529@moo.mook.co.uk> Hi Ruslan, Thanks for the advice. > If this is a case you need study what fields must be indexed. Is this done via the setFieldFlag() command? > You can try do XML dump/load to get the most compact db. I can *apparently* dump an XML file successfully, but loading it never works. I've tried loading the database into vStudio and exporting and XML doc too, but trying to load it back into vStudio gives the error "Kernel error: 19". -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From sunshine at public.kherson.ua Thu Oct 7 20:04:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 12:04:40 2004 Subject: Speed Issues From CD-ROM In-Reply-To: <200410071843.SAA09529@moo.mook.co.uk> Message-ID: On 10/7/04 7:56 PM, "Milan" wrote: Hi Milan, > Thanks for the advice. > >> If this is a case you need study what fields must be indexed. > > Is this done via the setFieldFlag() command? yes >> You can try do XML dump/load to get the most compact db. > > I can *apparently* dump an XML file successfully, but loading it never > works. I've tried loading the database into vStudio and exporting and XML > doc too, but trying to load it back into vStudio gives the error "Kernel > error: 19". What language is used in your texts ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 7 20:06:10 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 12:06:15 2004 Subject: V4RB Tutorial 1, TestAddRecords In-Reply-To: <3161E591-187A-11D9-8136-00306549C134@pcisys.net> Message-ID: On 10/7/04 7:01 PM, "Chuck Pelto" wrote: > I've been twiddling around with trying to get this method to work and > have met with complete failure. > > On page 16 of the tutorial it reads.... Hi Chuck, One point. I don't know have you realize, but Tutorial simply describe the Example1 Which is included into V4RB. So if something is not clear, you always can simply look into this example to see how it works there. Am I right ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From milan at mook.co.uk Thu Oct 7 18:08:37 2004 From: milan at mook.co.uk (Milan) Date: Thu Oct 7 12:09:13 2004 Subject: Speed Issues From CD-ROM In-Reply-To: Message-ID: <200410071855.SAA09837@moo.mook.co.uk> > What language is used in your texts ? The text is all in English. When I load the XML file, I get the error "error 4" in the Director message window. However, the .vdb is recreated, and it's about 50% smaller! Will this file be corrupt/incomplete? For reference, this is the code I'm using: CreateDatabase( gSchool, the moviePath & "schools.vdb" ) LoadDump( gSchool, the moviePath & "schools.xml", "schools.vdb", 2 ) Regards, Milan. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From sunshine at public.kherson.ua Thu Oct 7 21:11:54 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 13:12:03 2004 Subject: Speed Issues From CD-ROM In-Reply-To: <200410071855.SAA09837@moo.mook.co.uk> Message-ID: On 10/7/04 8:08 PM, "Milan" wrote: >> What language is used in your texts ? > > The text is all in English. > > When I load the XML file, I get the error "error 4" in the Director message > window. However, the .vdb is recreated, and it's about 50% smaller! Will > this file be corrupt/incomplete? I cannot say > For reference, this is the code I'm using: > > CreateDatabase( gSchool, the moviePath & "schools.vdb" ) > LoadDump( gSchool, the moviePath & "schools.xml", "schools.vdb", 2 ) No. You do not need the first line as I remember -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From gregkowalski at earthlink.net Thu Oct 7 17:40:05 2004 From: gregkowalski at earthlink.net (Gregory Kowalski) Date: Thu Oct 7 16:40:08 2004 Subject: query with parenthesis Message-ID: <73C61600-18A9-11D9-BF01-000393DAB46A@earthlink.net> Hello, I am having trouble making queries for strings that contain parentheses. I have a field called "form" that, among others, contains the following string: Purified F(ab')2 When I make a query using the expression: keyword = "Purified F(ab')2" "select * from TABLE where form = '"&keyword&"' " I get no results. If the keyword contains no parentheses correct results appear. i.e: keyword = "Liquid" (or whatever) The field in question is ASCII, Varchar and indexed. I changed to English, but it didn't work either. I did a sensitive test using the LIKE expression: "select * from Table where form LIKE '"&keyword&"' " If keyword = "ab" or "2" I get the right results - "Purified F(ab')2" But if I use keyword = "(ab" or "(ab')", or even just "(" all records appear as the result (!!!). The problem also appears if keyword = "'" It therefore seems as if the parenthesis and apostrophe are problematic. Any help would be greatly appreciated. Thanks Greg From sunshine at public.kherson.ua Fri Oct 8 01:04:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 7 17:04:53 2004 Subject: query with parenthesis In-Reply-To: <73C61600-18A9-11D9-BF01-000393DAB46A@earthlink.net> Message-ID: On 10/8/04 12:40 AM, "Gregory Kowalski" wrote: > Hello, > > I am having trouble making queries for strings that contain parentheses. You need escape such string, because () are used in the REGEX syntax > I have a field called "form" that, among others, contains the > following string: > > Purified F(ab')2 > > When I make a query using the expression: > > keyword = "Purified F(ab')2" > > "select * from TABLE where form = '"&keyword&"' " > > I get no results. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Fri Oct 8 09:35:21 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 8 10:35:29 2004 Subject: V4RB Tutorial 1, TestAddRecords In-Reply-To: References: Message-ID: Hi Ruslan, Are you saying there is a working version of the project(s) described in the tutorial document? Where can I find that? Thanks, Chuck On Oct 7, 2004, at 11:06 AM, Ruslan Zasukhin wrote: > On 10/7/04 7:01 PM, "Chuck Pelto" wrote: > >> I've been twiddling around with trying to get this method to work and >> have met with complete failure. >> >> On page 16 of the tutorial it reads.... > > Hi Chuck, > > One point. > > I don't know have you realize, but Tutorial simply describe the > Example1 > Which is included into V4RB. > > So if something is not clear, you always can simply look into this > example > to see how it works there. > > Am I right ? > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Fri Oct 8 19:21:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 8 11:22:04 2004 Subject: V4RB Tutorial 1, TestAddRecords In-Reply-To: Message-ID: On 10/8/04 6:35 PM, "Chuck Pelto" wrote: Hi Chuck, > Are you saying there is a working version of the project(s) described > in the tutorial document? Wow. Yes of course! It seems I have told this before: check examples! > Where can I find that? Folder Examples. Note that tutorial try to describe the example 1-5. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Fri Oct 8 11:25:22 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 8 12:25:33 2004 Subject: V4RB Tutorial 1, TestAddRecords In-Reply-To: References: Message-ID: <08E7C283-194F-11D9-8136-00306549C134@pcisys.net> Okay.... I finally figured out what you were referring to. I'd been looking for a segment on the web-site that held the examples. Did not know I had already got them with the Valentina software. May I recommend an additional change to the tutorial? Make a note, early on in the document, where the examples of all the databases described in the tutorial can be found. Searching the tutorial for mention of the examples location finds nothing. It would probably help to have those examples also found on the web-site, in the vicinity of the Valentina for REALbasic downloads area, too. Coffee breaks over. Back on my head.... Thanks, Chuck On Oct 8, 2004, at 10:21 AM, Ruslan Zasukhin wrote: > On 10/8/04 6:35 PM, "Chuck Pelto" wrote: > > Hi Chuck, > >> Are you saying there is a working version of the project(s) described >> in the tutorial document? > > Wow. Yes of course! > > It seems I have told this before: check examples! > > >> Where can I find that? > > Folder Examples. > > Note that tutorial try to describe the example 1-5. > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From cbpelto at pcisys.net Fri Oct 8 13:42:04 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 8 14:42:13 2004 Subject: ValentinaUtilities Differences Message-ID: <216D02B1-1962-11D9-8136-00306549C134@pcisys.net> Greetings Ruslan, Lots of interesting things in the Single Table example. Things that are different from the tutorial. The example does run. And that's the most important thing. I do have a question/concern about the ValentinaUtilities object in the Single Table example. I dragged it from the RB project to the desktop and compared it against the ValentinaUtilities object I have on-hand. The one from the Single Table example looks to be about twice as large as the one I have on-hand. Is that 'significant'? Regards, Chuck From sunshine at public.kherson.ua Fri Oct 8 22:52:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 8 14:52:13 2004 Subject: V4RB Tutorial 1, TestAddRecords In-Reply-To: <08E7C283-194F-11D9-8136-00306549C134@pcisys.net> Message-ID: On 10/8/04 8:25 PM, "Chuck Pelto" wrote: > Okay.... > > I finally figured out what you were referring to. > > I'd been looking for a segment on the web-site that held the examples. > Did not know I had already got them with the Valentina software. > > May I recommend an additional change to the tutorial? Make a note, > early on in the document, where the examples of all the databases > described in the tutorial can be found. Searching the tutorial for > mention of the examples location finds nothing. Ok, agree, Chuck. > It would probably help to have those examples also found on the > web-site, in the vicinity of the Valentina for REALbasic downloads > area, too. > > Coffee breaks over. Back on my head.... -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 8 22:53:37 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 8 14:53:41 2004 Subject: ValentinaUtilities Differences In-Reply-To: <216D02B1-1962-11D9-8136-00306549C134@pcisys.net> Message-ID: On 10/8/04 10:42 PM, "Chuck Pelto" wrote: > Greetings Ruslan, > > Lots of interesting things in the Single Table example. Things that are > different from the tutorial. > > The example does run. And that's the most important thing. Right! > I do have a question/concern about the ValentinaUtilities object in the > Single Table example. > I dragged it from the RB project to the desktop and compared it against > the ValentinaUtilities object I have on-hand. The one from the Single > Table example looks to be about twice as large as the one I have > on-hand. Is that 'significant'? May be that is REALbasic "miracle" ? I think you need compare not size but the properties listed in module. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 9 00:15:31 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 8 16:15:40 2004 Subject: V4RB 2.0 b18 - uploaded Message-ID: Hi All, [NEW] b18 introduce to you the new example that demonstrate 2 forms of new method Table.Sort( set, fld ) as VSet Table.Sort( set, it as VSortItem , ... ) as VSet Also added new small helper class VSortItem. This feature continue improvement of non-SQL API of V4RB. Now you can do searches and sorting using Sets. Actually now we have practically complete API for all general db tasks. Note, that I have found way implement VSortItem to support 3 forms of field specification: by name, by index, by reference. [NEW] example/SQL/Table_Alter * demonstrate usage of SQL ALTER TABLE command. * in 2.0 this command have much more power. - you can rename table - you can rename field - you can change field type - you can change field length - add/drop constraints, ... - more ------------------ We add soon also Algorithm Global methods such as Union( t1 as Table, t2 as Table ) as Table Interset( t1 as Table, t2 as Table ) as Table Difference( t1 as Table, t2 as Table ) as Table As you see this functions mirror the basic Relational Algebra operations, But again we will be able do this without SQL. -------------- Also will be added CreateSnapShoot( inTable ) as VTable This function allow you easy create new table with the same records as inTable. The main magic will be if and when we can send as inTable some cursor. Then we can build any cursor and COPY its records into new TMP or not tmp table. ---------------- Also, if I am not mistake, we have implement in kernel such cool feature as db.Clone( newLocation ) It is similar probably to FileMaker's feature. You can clone existed db and specify if new db should have - the same structure only - or also records Robert, I think THIS FEATURE can be considered as a way to REPAIR record of corrupted db. Agree? I really hope that we already have implement it in 2.0 I will ask our developers tomorrow. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 9 11:09:14 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 9 03:09:27 2004 Subject: [V4RB TIP] collect classes into project folder(s) Message-ID: Hi All, I recommend you to create in project window of RB the folders and keep here all classes related to database. If you have 2 kinds of database in your app you should use 2 folders. Then it will be much ore clean design. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From joergp at three-2-one.com Tue Oct 12 00:56:43 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Mon Oct 11 17:56:54 2004 Subject: [V4RB] Updating Blob Fields in VCursors Message-ID: How could I update a Blob field in a cursor? I tried: cur.BlobField(4).WriteRawData s If cur.Update Then // O.K. End If Cursor is read and write and field #4 is a VBlob field ? but it does not work. Any idea what's wrong? TIA, J?rg From sunshine at public.kherson.ua Tue Oct 12 09:01:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 12 01:01:41 2004 Subject: [V4RB] Updating Blob Fields in VCursors In-Reply-To: Message-ID: On 10/12/04 1:56 AM, "Joerg Pressel" wrote: Hi Joerg, > How could I update a Blob field in a cursor? > > I tried: > cur.BlobField(4).WriteRawData s > If cur.Update Then > // O.K. > End If > > Cursor is read and write and field #4 is a VBlob field ? but it does > not work. > > Any idea what's wrong? Looks to be correct. Try on some v4rb example. May be problem only in your project. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From valentina at panamind.com Tue Oct 12 04:28:08 2004 From: valentina at panamind.com (Gary Edge) Date: Tue Oct 12 03:28:14 2004 Subject: Databases sharing baseobject classes Message-ID: I'm sure the answer is yes, but before I spend the time experimenting: Two databases objects, in their constructor methods, can share baseobject classes right? -- Best Regards, Gary -- From sunshine at public.kherson.ua Tue Oct 12 11:38:39 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 12 03:38:47 2004 Subject: Databases sharing baseobject classes In-Reply-To: Message-ID: On 10/12/04 11:28 AM, "Gary Edge" wrote: Hi Gary, > I'm sure the answer is yes, but before I spend the time experimenting: > > Two databases objects, in their constructor methods, can share > baseobject classes right? I think yes, but never try. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From joergp at three-2-one.com Tue Oct 12 11:34:55 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Tue Oct 12 04:35:04 2004 Subject: [V4RB] Updating Blob Fields in VCursors In-Reply-To: References: Message-ID: > Try on some v4rb example. > May be problem only in your project. Actually it worked just fine ? I just did not refresh my display :-( Thanks for the bandwidth though, -- J?rg ____________________________________________ three-2-one interaktive Medien GmbH joerg.pressel@three-2-one.com http://www.three-2-one.com fon: +49 2151 31945-11 From wonderfef at noos.fr Tue Oct 12 19:15:30 2004 From: wonderfef at noos.fr (Eric Ferrer) Date: Tue Oct 12 12:15:44 2004 Subject: How far can I trust Vserver ? Message-ID: Hello all, I've been developing a big client-server application for more than 3 years now using RB. Today, a dozen of users use this application all day long for their job tasks. My number of users is going to jump up to 40 (and maybe more than that) because my software is going to be used in 3 new premises. The Server I developed is not made for so many users working simultaneously. So I'm now testing Vserver, which is just fine and great according to my first tests. But when I read the documentation and some posts on the lists, I ask myself this question "How far can I trust Vserver?". Here are the points I'd like hear some feedback from: ? MaxClientTimeOut There is no reason my application would suddenly disconnect from the server. Is the user makes the choice not to quit the application, the application should not disconnect. This Vserver parameter sounds important to me, because the manual gives the example of "having a coffee", which is not very long! Is is dangerous to keep clients connected all day long? ? Threads If I understood, this is the max number of clients connected simultaneously. The manual says that the recommended values are "5 to 20", which is too few for my needs. In another hand, I think I read somewhere that Vserver licence could be bought up to "unlimited connections". There must be something I did not understand... ? Testing with many connections Does anyone has already tested Vserver with many many connections ? I'm alone and I can't simulate 20 users working on the database at the same time! ? DB corruption and Vserver stability My application would make an intensive use of quite complex SQL SELECT statements, and many times some loops would be called to insert, delete or add records using cursors... Is that dangerous? Does the database get corrupted with the time? Does Vserver happen to unexpectedly quit ? Thanks all in advance for your feedback! Eric From cbpelto at pcisys.net Tue Oct 12 11:34:33 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Tue Oct 12 12:34:46 2004 Subject: Tutorial 1 Example Question Message-ID: Greetings, A question about Example 1 of V4RB. In the example project, Single Table, running the application the database window does not open up, automatically, when the application is launched. However, in the example I developed using the tutorial instructions, it does open up. I've been trying to figure out how getting the window to not open automatically was accomplished. I've been pouring over the example project and comparing it to mine, but to no avail. How did you DO that? Regards, Chuck From sunshine at public.kherson.ua Tue Oct 12 20:38:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 12 12:39:01 2004 Subject: How far can I trust Vserver ? In-Reply-To: Message-ID: On 10/12/04 8:15 PM, "Eric Ferrer" wrote: Hi Eric, You ask existed Vserver users? Then I think you should get answers from Florian, Key Ray, others. They ship RELEASE of their apps about year already. > Hello all, > > I've been developing a big client-server application for more than 3 years > now using RB. > Today, a dozen of users use this application all day long for their job > tasks. > > My number of users is going to jump up to 40 (and maybe more than that) > because my software is going to be used in 3 new premises. > The Server I developed is not made for so many users working simultaneously. > So I'm now testing Vserver, which is just fine and great according to my > first tests. > > But when I read the documentation and some posts on the lists, I ask myself > this question "How far can I trust Vserver?". > Here are the points I'd like hear some feedback from: > > ? MaxClientTimeOut > There is no reason my application would suddenly disconnect from the server. > Is the user makes the choice not to quit the application, the application > should not disconnect. But what if hardware failure? Power of hub for example ? > This Vserver parameter sounds important to me, because the manual gives the > example of "having a coffee", which is not very long! > Is is dangerous to keep clients connected all day long? No. And note, you can set this parameter to be infinity. As result your clients will NOT be disconnected. > ? Threads > If I understood, this is the max number of clients connected simultaneously. NO. > The manual says that the recommended values are "5 to 20", which is too few > for my needs. > In another hand, I think I read somewhere that Vserver licence could be > bought up to "unlimited connections". > There must be something I did not understand... No. Valentina do NOT use simple model "one user one thread" (mySQL use it for example). Instead Valentina Server use quite complex model that was suggested by ACE network framework. In short you can have 100 connected users, but 5-10 threads inside of kernel. Reason of such model is that if you have MANY threads on single SPU or even 2-CP computer, they start work just slower. > ? Testing with many connections > Does anyone has already tested Vserver with many many connections ? > I'm alone and I can't simulate 20 users working on the database at the same > time! I believe we have own tests for this. In fact you also can create MANY connections from single application! This is not prohibited. Just write loop which establish connections. Remember although, Windows on default have limit 64 connections per process. > ? DB corruption and Vserver stability > My application would make an intensive use of quite complex SQL SELECT > statements, and many times some loops would be called to insert, delete or > add records using cursors... > Is that dangerous? > Does the database get corrupted with the time? > Does Vserver happen to unexpectedly quit ? I wonder if you have use Valentina local before? Honestly. Yes, sometimes we get crash reports from Vserver. Yes, sometimes people report about corrupted dbs. In the same time some developers report that they have never see any corrupted db for years distributing tens of thousands of copies. I will recommend you develop prototype of your task and test it intensively. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue Oct 12 20:42:00 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 12 12:42:07 2004 Subject: Tutorial 1 Example Question In-Reply-To: Message-ID: On 10/12/04 8:34 PM, "Chuck Pelto" wrote: Hi Chuck, Look at REALBasic menu Edit->Project Settings Here you will see "Default Window" Choose here > Greetings, > > A question about Example 1 of V4RB. > > In the example project, Single Table, running the application the > database window does not open up, automatically, when the application > is launched. However, in the example I developed using the tutorial > instructions, it does open up. > > I've been trying to figure out how getting the window to not open > automatically was accomplished. I've been pouring over the example > project and comparing it to mine, but to no avail. > > How did you DO that? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Tue Oct 12 11:59:25 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Tue Oct 12 12:59:31 2004 Subject: Tutorial 1 Example Question In-Reply-To: References: Message-ID: <7408CF70-1C78-11D9-ACBE-00306549C134@pcisys.net> Thanks.... By the way....not to worry too much about my sending these sorts of questions. I've ordered REALbasic for Dummies. It SHOULD arrive later this week. Regards, Chuck(le) P.S. I'd have ordered the Missing Manuals item for REALbasic, but it's not available just yet. On Oct 12, 2004, at 11:42 AM, Ruslan Zasukhin wrote: > On 10/12/04 8:34 PM, "Chuck Pelto" wrote: > > Hi Chuck, > > Look at REALBasic menu Edit->Project Settings > > Here you will see "Default Window" > > Choose here > > >> Greetings, >> >> A question about Example 1 of V4RB. >> >> In the example project, Single Table, running the application the >> database window does not open up, automatically, when the application >> is launched. However, in the example I developed using the tutorial >> instructions, it does open up. >> >> I've been trying to figure out how getting the window to not open >> automatically was accomplished. I've been pouring over the example >> project and comparing it to mine, but to no avail. >> >> How did you DO that? > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From fb at memedia.de Tue Oct 12 20:45:29 2004 From: fb at memedia.de (Florian Bogeschdorfer) Date: Tue Oct 12 13:45:43 2004 Subject: How far can I trust Vserver ? In-Reply-To: Message-ID: Hi there! > > > Hello all, > > I've been developing a big client-server application for more > than 3 years now using RB. > Today, a dozen of users use this application all day long for > their job tasks. > > My number of users is going to jump up to 40 (and maybe more > than that) because my software is going to be used in 3 new premises. > The Server I developed is not made for so many users working > simultaneously. > > So I'm now testing Vserver, which is just fine and great > according to my first tests. > > But when I read the documentation and some posts on the > lists, I ask myself this question "How far can I trust Vserver?". > > Here are the points I'd like hear some feedback from: > > ? MaxClientTimeOut > There is no reason my application would suddenly disconnect > from the server. > Is the user makes the choice not to quit the application, the > application should not disconnect. > This Vserver parameter sounds important to me, because the > manual gives the example of "having a coffee", which is not very long! > Is is dangerous to keep clients connected all day long? You can set this to zero, so no timeout occurs. My application is running sometimes for days, no problems so far. > > ? Threads > If I understood, this is the max number of clients connected > simultaneously. > The manual says that the recommended values are "5 to 20", > which is too few for my needs. > In another hand, I think I read somewhere that Vserver > licence could be bought up to "unlimited connections". > There must be something I did not understand... I can not tell you about that, I have been using up to 5 users each in about 15 installations. It has never failed me. > > ? Testing with many connections > Does anyone has already tested Vserver with many many connections ? > I'm alone and I can't simulate 20 users working on the > database at the same time! > > ? DB corruption and Vserver stability > My application would make an intensive use of quite complex > SQL SELECT statements, and many times some loops would be > called to insert, delete or add records using cursors... > Is that dangerous? > Does the database get corrupted with the time? > Does Vserver happen to unexpectedly quit ? Well Eric, I am running my installations for exactly a year now. Yes there were bugs and I bet there are still some undiscovered. But I always got very fast help from Ruslan and Igor. In my applications there may be a 100.000 reads/writes in one night, if it is busy and you may bet that in 20% of all cases, the computer is not shut down properly but simply turned off. I NEVER EVER had one single damaged database, infact I did not even have a backup function till the current update. I was never forced to. Of course I may not use vServer the way you do. And yes, I took precautions like flushing my database actually after every record change (since some data is pretty critical). But these are precautions against instable PCs or stupid users. I work with Director, but I can't see a reason why the server itself should be less stable. I am running vServer as a service by the way. I have another app where I use V12 database and I have a corrupted database almost every day in one of 50 installations! So I personally was very much surprised HOW stable Valentina really is. Best regards, Florian From sunshine at public.kherson.ua Tue Oct 12 23:12:19 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 12 15:12:26 2004 Subject: How far can I trust Vserver ? In-Reply-To: Message-ID: On 10/12/04 9:45 PM, "Florian Bogeschdorfer" wrote: > I have another app where I use V12 database and I have a corrupted database > almost every day in one of 50 installations! So I personally was very much > surprised HOW stable Valentina really is. Clarification. V12 -- is one of the oldest db Xtra plugins for Macromedia Director. It is still probably major player on Director Market. May be no any more. Don't know. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From fb at memedia.de Tue Oct 12 23:38:10 2004 From: fb at memedia.de (Florian Bogeschdorfer) Date: Tue Oct 12 16:38:24 2004 Subject: How far can I trust Vserver ? In-Reply-To: Message-ID: Yes of course: V12 is not related to Valentina, it is a total different product. Florian > > On 10/12/04 9:45 PM, "Florian Bogeschdorfer" wrote: > > > I have another app where I use V12 database and I have a corrupted > > database almost every day in one of 50 installations! So I > personally > > was very much surprised HOW stable Valentina really is. > > Clarification. > > V12 -- is one of the oldest db Xtra plugins for Macromedia Director. > > It is still probably major player on Director Market. > May be no any more. Don't know. > From wonderfef at noos.fr Wed Oct 13 09:13:00 2004 From: wonderfef at noos.fr (Eric Ferrer) Date: Wed Oct 13 02:13:12 2004 Subject: How far can I trust Vserver ? In-Reply-To: Message-ID: >> ? MaxClientTimeOut > But what if hardware failure? > Power of hub for example ? This doesn't automatically disconnect? Anyway, if it's not "dangerous" to keep clients connected in normal conditions for hours, then it's OK for me. > I wonder if you have use Valentina local before? Yes, and though a DIAGNOSE command tells me that some minor problems are found, my Valentina db runs just fine. I think that REPAIR or REBUILD funtions would be highly appreciated anyway. > In fact you also can create MANY connections from single application! Yes, I tried to launch me application from different PCs, but I can't simulate all by myself: - one user updating the schedule and updating, inserting, deleting 10s of records in a go - while another user is creating a new quotation containing 10's of records - while another user is performing some complex SQL Statement to build a report - while another user is browsing the Customers table - while another user is deleting 10s of records... This is what I meant. I think this is where the threads are useful, depending on the number of connected clients. Thanks all for your feedback. Eric From sunshine at public.kherson.ua Wed Oct 13 22:32:06 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 13 14:32:21 2004 Subject: slow Vserver In-Reply-To: <000e01c4b14c$24fc6c90$67fea8c0@chris1> Message-ID: On 10/13/04 8:43 PM, "Chris Sheffield" wrote: Hi Chris, > I've got one more question for you. We've discovered that some IP addresses > either don't work or work slowly with VServer, but I'm still not totally > sure if it's because of network settings or what. > > Right now the testing network in our office is using 192.168.x.x IP > addresses and everything is working fine. We can connect to VServer from > the clients. But if we change the IP addresses to either 10.x.x.x or > 172.16.x.x, then we experience slow downs or can't connect at all. When we > can't connect we usually get error 10060. Please explain me. What are these addresses 10.x.x.x and 172.16.x.x ? Why you or your clients try to use them ? > So my question is does VServer require a certain form of IP address? We have no such info. We need stud this issue. > Everything seems to work great when using 192.168.x.x, but the other two > forms of internal IPs seem to choke, and we've found that our customers who > have problems are using one of these two forms. And they either cannot > connect at all to VServer or they are experiencing horribly long delays when > connecting. I see now. Okay, so we need setup our local network with these addresses and test. May be we have something wrong in our code. > I don't know if this offers any new ideas or not. Of course yes! > But if there is anything else you can think of, I'd be willing to try. I'm > still not totally convinced that it's hardware related. There are just too > many variables and too many different configurations involved. Yes, now it seems we have other issue. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed Oct 13 23:16:18 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 13 15:16:29 2004 Subject: slow Vserver In-Reply-To: <000101c4b15f$d86c6b50$67fea8c0@chris1> Message-ID: On 10/13/04 11:04 PM, "Chris Sheffield" wrote: > Thanks, Ruslan. I am by no means a networking specialist, but from what I > understand these are all the possible forms for internal IP addresses, for > use on a private network. I believe the most common form is 192.168.x.x, > which I'm guessing is what you're using on your network, and that's what our > office network is using as well. But for some reason, a lot of the schools > that we've been dealing with have been using the 10.x.x.x from, so IP > addresses range from 10.0.0.0 to 10.255.255.255. I think they have complex structure of network with sub-networks. > A couple have even been > using the other 172.16.x.x form, but that doesn't seem to be as common. > If you need further information we can probably gather together some web > sites that explain further. > > Chris Sheffield > Software Development > Read Naturally > csheffield@readnaturally.com -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From support at monkeybreadsoftware.de Wed Oct 13 23:16:24 2004 From: support at monkeybreadsoftware.de (Christian Schmitz) Date: Wed Oct 13 16:16:38 2004 Subject: [V4RB] Error 348 In-Reply-To: <3AA55870-6AA6-11D8-96D6-000A958D2BA4@beatrixwillius.de> Message-ID: <1glmbag.1pny68m1ivu2ydM%support@monkeybreadsoftware.de> Beatrix Willius wrote: > Hello Ruslan, > > I somehow managed to corrupt a Valentina database. Opening in Valentina > Carbon gives the error message 348 "Error of inventory page of varchar > field." Was a solution found? I've got a damaged file with this error, here. Mfg Christian -- Visiting Rome from 16th to 24th October 2004. Eight thousand functions in one REALbasic plug-in. The MBS Plugin. From cbpelto at pcisys.net Thu Oct 14 13:15:33 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 14 14:15:40 2004 Subject: Linking a VDB to an RB App Message-ID: <6BD27EAE-1E15-11D9-ACBE-00306549C134@pcisys.net> Greetings, In the tutorial I've worked, the Valentina DB is linked to the windows. However, if I want the db to be linked to the App, at the App level in RB, I supposed I should put the property in the App's source code. Is this correct? And, if I make it public or protected (to the sub-classes of the app) I'll be able to reference it from any window in the app, correct? If so, how would I call it from a window's source code, e.g., mDataBase.mPerson.ByteFld.Value = Val(EditField2.Text) Regards, Chuck From joergp at three-2-one.com Thu Oct 14 22:09:47 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Thu Oct 14 15:10:02 2004 Subject: SQL Problem Message-ID: I need some advice with a SQL statement (using V4RB 1.11): I have three tables: - "Albums" with Artist_Ptr pointing to "Artists" and field "Title" - "Tracks" with Album_Ptr pointing to "Albums" and Artist_Ptr pointing to "Artists" - "Artists" containing artist names (field "Artist") I want to get all Albums which contain Tracks from a certain artist. So far so easy: SELECT DISTINCT Albums.Title FROM Tracks,Albums WHERE Artist_Ptr=1606 But ? I need one additional information from the Albums which were found: the artist name. I tried: SELECT DISTINCT Albums.Title,Artist FROM Tracks,Albums,Artists WHERE Artist_Ptr=1606 This does not work, because Valentina does not know if she should take "Artist" using the Artist_Ptr from "Albums" or from "Tracks". Is there any way I can get around this problem? Any suggestion what would be the most elegant way? Many thanks, J?rg From joergp at three-2-one.com Thu Oct 14 22:16:47 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Thu Oct 14 15:16:57 2004 Subject: SQL Problem(2) Message-ID: Problem not solved, but of course I don't need the "Distinct"... From sunshine at public.kherson.ua Fri Oct 15 09:26:55 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 15 01:27:03 2004 Subject: Test, ignore Message-ID: -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From valentina at panamind.com Fri Oct 15 13:04:06 2004 From: valentina at panamind.com (Gary Edge) Date: Fri Oct 15 12:04:12 2004 Subject: Determining segment size Message-ID: <39253D04-1ECC-11D9-8E69-000A95A9A898@panamind.com> When creating a database, what do you consider when setting the segment size? The example uses 32 * 1024 which I also use, but why. I assume this number has some relation to file size. -- Best Regards, Gary -- From sunshine at public.kherson.ua Fri Oct 15 20:18:57 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 15 12:19:04 2004 Subject: Determining segment size In-Reply-To: <39253D04-1ECC-11D9-8E69-000A95A9A898@panamind.com> Message-ID: On 10/15/04 8:04 PM, "Gary Edge" wrote: > When creating a database, what do you consider when setting the segment > size? The example uses 32 * 1024 which I also use, but why. I assume > this number has some relation to file size. Hi Gary, It have relation for the size of empty db. When you will add a lots of data, this parameter practically not affects anything. You may wish it say to 2KB if you make small prefs db. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From valentina at panamind.com Fri Oct 15 13:34:14 2004 From: valentina at panamind.com (Gary Edge) Date: Fri Oct 15 12:34:22 2004 Subject: Determining segment size In-Reply-To: References: Message-ID: <6F0BB0E5-1ED0-11D9-8E69-000A95A9A898@panamind.com> Hi Ruslan, So out of curiosity, for databases that will contain from 500 to 5000 records average in the largest table, would you use the 32KB or a smaller number? -- Best Regards, Gary -- On Oct 15, 2004, at 1:18 PM, Ruslan Zasukhin wrote: > On 10/15/04 8:04 PM, "Gary Edge" wrote: > >> When creating a database, what do you consider when setting the >> segment >> size? The example uses 32 * 1024 which I also use, but why. I assume >> this number has some relation to file size. > > Hi Gary, > > It have relation for the size of empty db. > > When you will add a lots of data, this parameter practically not > affects > anything. > > You may wish it say to 2KB if you make small prefs db. > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] From sunshine at public.kherson.ua Fri Oct 15 20:45:12 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 15 12:45:21 2004 Subject: Determining segment size In-Reply-To: <6F0BB0E5-1ED0-11D9-8E69-000A95A9A898@panamind.com> Message-ID: On 10/15/04 8:34 PM, "Gary Edge" wrote: > Hi Ruslan, > > So out of curiosity, for databases that will contain from 500 to 5000 > records average in the largest table, would you use the 32KB or a > smaller number? 4-8KB will be ok -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From valentina at panamind.com Fri Oct 15 13:55:25 2004 From: valentina at panamind.com (Gary Edge) Date: Fri Oct 15 12:55:31 2004 Subject: Determining segment size - Thanks *NM In-Reply-To: References: Message-ID: <64141E16-1ED3-11D9-8E69-000A95A9A898@panamind.com> From srt at bigpond.net.au Sun Oct 17 14:05:42 2004 From: srt at bigpond.net.au (Steven) Date: Sun Oct 17 09:05:44 2004 Subject: V4WS 1.11 cannot enter serial number Message-ID: Ruslan, it is no longer possible to enter the V4WS serial number using 1.11 (or adjust the cache size). In WebSiphon.acgi go to the Preferences menu -> Builtin Preferences ->ValentinaLib Preferences and try to select it. Instead of a window opening where you can enter the serial number and cache size nothing happens. I checked and this was the case in 1.10 too. I had been using 1.9.7 where this wasn't a problem. Best Regards, Steven From srt at bigpond.net.au Sun Oct 17 14:19:26 2004 From: srt at bigpond.net.au (Steven) Date: Sun Oct 17 09:19:30 2004 Subject: VAPP 1.11 is actually 1.10 Message-ID: Ruslan, I just downloaded valentina application which is listed as version 1.11, but the splash screen and Get Info for it say 1.10? Best Regards, Steven From sunshine at public.kherson.ua Sun Oct 17 17:27:15 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 17 09:27:22 2004 Subject: VAPP 1.11 is actually 1.10 In-Reply-To: Message-ID: On 10/17/04 5:19 PM, "Steven" wrote: > Ruslan, > > I just downloaded valentina application which is listed as version 1.11, > but the splash screen and Get Info for it say 1.10? VAPP, yes, 1.10. Its page says 1.11? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sun Oct 17 17:27:50 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 17 09:27:56 2004 Subject: V4WS 1.11 cannot enter serial number In-Reply-To: Message-ID: On 10/17/04 5:05 PM, "Steven" wrote: > Ruslan, > > it is no longer possible to enter the V4WS serial number using 1.11 (or > adjust the cache size). Why? > In WebSiphon.acgi go to the Preferences menu -> Builtin Preferences > ->ValentinaLib Preferences and try to select it. Instead of a window > opening where you can enter the serial number and cache size nothing > happens. I checked and this was the case in 1.10 too. > > I had been using 1.9.7 where this wasn't a problem. Okay, I will check this Steven. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From chuck at mediamacros.com Tue Oct 19 11:46:32 2004 From: chuck at mediamacros.com (Chuck Neal) Date: Tue Oct 19 10:44:17 2004 Subject: 301 Errors Message-ID: <023e01c4b5f2$cec1ffb0$b900a8c0@batcomputer> I have a setup where a user has a master database, and then they can choose to export a portion of this to a new database. The code is set up to spider through the selections to pull in all the options needed from each table, and relink with the new id numbers. The first pass worked fine, but as I started exporting larger sets of data it began to show problems. I was getting 301 errors every time in the same place. I found the function that appeared to be the problem and reworked it so that it was not making as many database calls, then ran it again. This time it gave the same error somewhere else. It seems that its not the SQL so much as the number of sequential calls. I have tried flushing the database, etc, but no luck. It still regularly corrupts the file when I export a lot of data this way. Anyone had similar problems? I found a post about setting the size to 32 bytes instead of 1-4 but that had no effect. Thanks. -Chuck -------------------------- Chuck Neal CEO, MediaMacros, Inc. chuck@mediamacros.com http://www.mediamacros.com -------------------------- Check out the Developers Mall Your one stop shop for all your Director Xtra Needs http://www.mediamacros.net/customer From sunshine at public.kherson.ua Tue Oct 19 19:51:11 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 19 11:51:27 2004 Subject: 301 Errors In-Reply-To: <023e01c4b5f2$cec1ffb0$b900a8c0@batcomputer> Message-ID: On 10/19/04 6:46 PM, "Chuck Neal" wrote: > I have a setup where a user has a master database, and then they can choose > to export a portion of this to a new database. The code is set up to spider > through the selections to pull in all the options needed from each table, > and relink with the new id numbers. > > The first pass worked fine, but as I started exporting larger sets of data > it began to show problems. I was getting 301 errors every time in the same > place. I found the function that appeared to be the problem and reworked it > so that it was not making as many database calls, then ran it again. This > time it gave the same error somewhere else. It seems that its not the SQL > so much as the number of sequential calls. I have tried flushing the > database, etc, but no luck. It still regularly corrupts the file when I > export a lot of data this way. > > Anyone had similar problems? I found a post about setting the size to 32 > bytes instead of 1-4 but that had no effect. Hi Chuck, kFBL_DataFileError = 301, This means you have corruption. You need XML dump / load -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From chuck at mediamacros.com Tue Oct 19 13:12:11 2004 From: chuck at mediamacros.com (Chuck Neal) Date: Tue Oct 19 12:09:17 2004 Subject: 301 Errors In-Reply-To: Message-ID: <027801c4b5fe$c5886ef0$b900a8c0@batcomputer> Well the problem is that this is a new database. I create the new database and add the paired data to it but reading records, getting the info via a list, then writing them to the new database. This works fine for smaller chunks of data but the larger ones seem to corrupt it every single time. The original database is not effected, its just the new one that seems to give me problems. What causes this in the first place? Is there a command, or common event that makes this happen that I should avoid? I was able to export a small group just now with no problems, yet others fail every single time with a 301 error. I can't always predict the size or content of the data the user will have so how can I safeguard against this happening at runtime? -Chuck -------------------------- Chuck Neal CEO, MediaMacros, Inc. chuck@mediamacros.com http://www.mediamacros.com -------------------------- Check out the Developers Mall Your one stop shop for all your Director Xtra Needs http://www.mediamacros.net/customer -----Original Message----- From: valentina-bounces@lists.macserve.net [mailto:valentina-bounces@lists.macserve.net] On Behalf Of Ruslan Zasukhin Sent: Tuesday, October 19, 2004 12:51 PM To: valentina@lists.macserve.net Subject: Re: 301 Errors On 10/19/04 6:46 PM, "Chuck Neal" wrote: > I have a setup where a user has a master database, and then they can > choose to export a portion of this to a new database. The code is set > up to spider through the selections to pull in all the options needed > from each table, and relink with the new id numbers. > > The first pass worked fine, but as I started exporting larger sets of > data it began to show problems. I was getting 301 errors every time > in the same place. I found the function that appeared to be the > problem and reworked it so that it was not making as many database > calls, then ran it again. This time it gave the same error somewhere > else. It seems that its not the SQL so much as the number of > sequential calls. I have tried flushing the database, etc, but no > luck. It still regularly corrupts the file when I export a lot of > data this way. > > Anyone had similar problems? I found a post about setting the size to > 32 bytes instead of 1-4 but that had no effect. Hi Chuck, kFBL_DataFileError = 301, This means you have corruption. You need XML dump / load -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From sunshine at public.kherson.ua Tue Oct 19 20:53:40 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 19 12:53:49 2004 Subject: 301 Errors In-Reply-To: <027801c4b5fe$c5886ef0$b900a8c0@batcomputer> Message-ID: On 10/19/04 8:12 PM, "Chuck Neal" wrote: > Well the problem is that this is a new database. I create the new database > and add the paired data to it but reading records, getting the info via a > list, then writing them to the new database. This works fine for smaller > chunks of data but the larger ones seem to corrupt it every single time. > The original database is not effected, its just the new one that seems to > give me problems. What causes this in the first place? Is there a command, > or common event that makes this happen that I should avoid? There is no such info. 301 -- this is something very deep. > I was able to export a small group just now with no problems, yet others > fail every single time with a 301 error. I can't always predict the size or > content of the data the user will have so how can I safeguard against this > happening at runtime? So you do in loop read of record and you self produce some export to file? If comment export and leave only iteration by records then problem still here? What size of you cache ? Make sure it is big enough -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From chuck at mediamacros.com Tue Oct 19 15:45:31 2004 From: chuck at mediamacros.com (Chuck Neal) Date: Tue Oct 19 14:42:55 2004 Subject: 301 Errors In-Reply-To: Message-ID: <02ea01c4b614$34e52490$b900a8c0@batcomputer> Basically I have a relational database with multiple tables joined together. Here is a simple example... Box - ID, Name Item - ID, Name Item_box - Box_id, Item_id The ids are created in order they are added, so when I import/export I have to be aware that the ids must change so... First I go through the boxes. I create a new database (empty) and get all the current boxes from the original database. I then add each and set up a list of the old ID and new ID.... This way I know what the old and new numbers are. I do the same for item, then I rebuild the item_box table by checking the old numbers against the new and create new entries. I am adding images and text, but all of these were added via the same interface. It seems that it may be getting overloaded when doing this in one loop though. Is there any way to force the cache to empty out? I tried flush() at key points but it had no effect. No single item is more than a few KB, and if I do a smaller subset of the data it exports fine. I tried upping the cache from 4 MB to 8 MB and it still errors. The same commands that were used to create the initial data are also being used to copy the DB. I have a parent script that is the DB controller, so I just spawn a second one and send add() commands to this so it is not building any differently than the base application, except that its doing them quickly in a loop. Any other ideas? These are testes so clients may have much larger data sets they want to export so I can't even count on a small size being enough to get around the problem. Why is it always corrupting at the same point? Its all plain text and image data and it comes from existing records so it should be formatted correctly. -Chuck -------------------------- Chuck Neal CEO, MediaMacros, Inc. chuck@mediamacros.com http://www.mediamacros.com -------------------------- Check out the Developers Mall Your one stop shop for all your Director Xtra Needs http://www.mediamacros.net/customer -----Original Message----- From: valentina-bounces@lists.macserve.net [mailto:valentina-bounces@lists.macserve.net] On Behalf Of Ruslan Zasukhin Sent: Tuesday, October 19, 2004 1:54 PM To: valentina@lists.macserve.net Subject: Re: 301 Errors On 10/19/04 8:12 PM, "Chuck Neal" wrote: > Well the problem is that this is a new database. I create the new > database and add the paired data to it but reading records, getting > the info via a list, then writing them to the new database. This > works fine for smaller chunks of data but the larger ones seem to > corrupt it every single time. The original database is not effected, > its just the new one that seems to give me problems. What causes this > in the first place? Is there a command, or common event that makes > this happen that I should avoid? There is no such info. 301 -- this is something very deep. > I was able to export a small group just now with no problems, yet > others fail every single time with a 301 error. I can't always > predict the size or content of the data the user will have so how can > I safeguard against this happening at runtime? So you do in loop read of record and you self produce some export to file? If comment export and leave only iteration by records then problem still here? What size of you cache ? Make sure it is big enough -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From chuck at mediamacros.com Tue Oct 19 15:55:50 2004 From: chuck at mediamacros.com (Chuck Neal) Date: Tue Oct 19 14:53:09 2004 Subject: 301 Errors In-Reply-To: Message-ID: <02f201c4b615$a298d530$b900a8c0@batcomputer> A little more info. I just set debug to 2 to see everything that's happening. (Lots of DB operations there.) What is odd is I only get one 301 error and it corrupts the file, but then more commands run successfully afterwards. Here is a small portion of the dump (the SQL query is my own "put" code) * V4MD *: new VCursor...done * V4MD *: Cursor_SetBlank...done * V4MD *: Cursor_SetField...done * V4MD *: SetPicture...done * V4MD *: Cursor_AddRecord...done * V4MD *: Cursor_Destroy...done * V4MD *: GetRef...done * V4MD *: new VCursor...done * V4MD *: ValentinaError...done * V4MD *: Cursor_GetRecordsAsPropList...done * V4MD *: Cursor_Destroy...done * V4MD *: ValentinaEscapeString...done * V4MD *: DB_SqlExecute...done * V4MD *: ValentinaError...done * V4MD *: GetRef...done * V4MD *: new VCursor...error 301 * V4MD *: ValentinaError...error 301 * V4MD *: ValentinaErrorString...error 301 -- " error SELECT image FROM technique_image WHERE technique = 15 ORDER BY sort [15] DataFile error " * V4MD *: Cursor_Destroy...done * V4MD *: DB_SqlExecute...done * V4MD *: ValentinaError...done * V4MD *: GetRef...done * V4MD *: new VCursor...done * V4MD *: ValentinaError...done * V4MD *: Cursor_FirstRecord...done * V4MD *: GetPicture...done * V4MD *: Cursor_Destroy...done * V4MD *: ValentinaEscapeString...done * V4MD *: GetRef...done * V4MD *: new VCursor...done * V4MD *: ValentinaError...done * V4MD *: Cursor_GetRecordsAsPropList...done * V4MD *: Cursor_Destroy...done * V4MD *: GetRef...done * V4MD *: new VCursor...done * V4MD *: ValentinaError...done * V4MD *: Cursor_GetRecordsAsPropList...done * V4MD *: Cursor_Destroy...done * V4MD *: ValentinaEscapeString...done How is that possible? Once it errors once shouldn't it error every time after that? -Chuck -------------------------- Chuck Neal CEO, MediaMacros, Inc. chuck@mediamacros.com http://www.mediamacros.com -------------------------- Check out the Developers Mall Your one stop shop for all your Director Xtra Needs http://www.mediamacros.net/customer -----Original Message----- From: valentina-bounces@lists.macserve.net [mailto:valentina-bounces@lists.macserve.net] On Behalf Of Ruslan Zasukhin Sent: Tuesday, October 19, 2004 1:54 PM To: valentina@lists.macserve.net Subject: Re: 301 Errors On 10/19/04 8:12 PM, "Chuck Neal" wrote: > Well the problem is that this is a new database. I create the new > database and add the paired data to it but reading records, getting > the info via a list, then writing them to the new database. This > works fine for smaller chunks of data but the larger ones seem to > corrupt it every single time. The original database is not effected, > its just the new one that seems to give me problems. What causes this > in the first place? Is there a command, or common event that makes > this happen that I should avoid? There is no such info. 301 -- this is something very deep. > I was able to export a small group just now with no problems, yet > others fail every single time with a 301 error. I can't always > predict the size or content of the data the user will have so how can > I safeguard against this happening at runtime? So you do in loop read of record and you self produce some export to file? If comment export and leave only iteration by records then problem still here? What size of you cache ? Make sure it is big enough -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From sunshine at public.kherson.ua Tue Oct 19 23:30:10 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Oct 19 15:30:17 2004 Subject: 301 Errors In-Reply-To: <02f201c4b615$a298d530$b900a8c0@batcomputer> Message-ID: On 10/19/04 10:55 PM, "Chuck Neal" wrote: > How is that possible? Once it errors once shouldn't it error every time > after that? Chuck, Do you use 1.11 ? If you have 100% reproducible error then please send me project. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From chuck at mediamacros.com Wed Oct 20 08:22:48 2004 From: chuck at mediamacros.com (Chuck Neal) Date: Wed Oct 20 07:20:40 2004 Subject: 301 Errors In-Reply-To: Message-ID: <048c01c4b69f$82cf5e60$b900a8c0@batcomputer> I am using 1.11 and also rearranged some of my SQL statements. The change in order seems to have helped, but no reason why it was a problem in the first place. I will keep testing and see if the problem arises again. Thanks. -Chuck -------------------------- Chuck Neal CEO, MediaMacros, Inc. chuck@mediamacros.com http://www.mediamacros.com -------------------------- Check out the Developers Mall Your one stop shop for all your Director Xtra Needs http://www.mediamacros.net/customer -----Original Message----- From: valentina-bounces@lists.macserve.net [mailto:valentina-bounces@lists.macserve.net] On Behalf Of Ruslan Zasukhin Sent: Tuesday, October 19, 2004 4:30 PM To: valentina@lists.macserve.net Subject: Re: 301 Errors On 10/19/04 10:55 PM, "Chuck Neal" wrote: > How is that possible? Once it errors once shouldn't it error every > time after that? Chuck, Do you use 1.11 ? If you have 100% reproducible error then please send me project. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Valentina mailing list Valentina@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina From gregkowalski at earthlink.net Wed Oct 20 16:02:37 2004 From: gregkowalski at earthlink.net (Gregory Kowalski) Date: Wed Oct 20 15:02:54 2004 Subject: ascii 174 question Message-ID: Hello, I have a db that contains a ? (registered trademark) symbol (ASCII 174). When I import it into a valentina db it shows erroneously as (r). I use ASCII varchar. What can I do to correct this problem? Thanks Greg From sunshine at public.kherson.ua Wed Oct 20 23:11:39 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 20 15:11:52 2004 Subject: ascii 174 question In-Reply-To: Message-ID: On 10/20/04 11:02 PM, "Gregory Kowalski" wrote: > Hello, > > I have a db that contains a ? (registered trademark) symbol (ASCII > 174). When I import it into a valentina db it shows erroneously as > (r). > > I use ASCII varchar. > > What can I do to correct this problem? Hi Greg, If one read from db this symbol comes also as 174, Then this is not problem of db but I think of font that you use -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From gregkowalski at earthlink.net Wed Oct 20 16:13:37 2004 From: gregkowalski at earthlink.net (Gregory Kowalski) Date: Wed Oct 20 15:13:50 2004 Subject: please ignore ascii 174 question Message-ID: <86CC49BD-22D4-11D9-BDB5-000393DAB46A@earthlink.net> Please ignore ascii 174 question. Problem solved. G. From cbpelto at pcisys.net Wed Oct 20 15:29:55 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Wed Oct 20 16:30:05 2004 Subject: Bad Thing? Message-ID: <2F689BFA-22DF-11D9-ACBE-00306549C134@pcisys.net> Is it a 'bad thing' to rename the fields in a VBaseObject inside of an RB app? Example: Originally the fields were titled "this_fld", "that_fld", "another_fld". I populated the database. Then I thought better of my naming convention and changed the fields to be "fld_this", "fld_that", and "fld_another". Now I can't access the database anymore. Or was it something else I may have done? Regards, Chuck From sunshine at public.kherson.ua Thu Oct 21 00:45:37 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 21 01:12:21 2004 Subject: Bad Thing? In-Reply-To: <2F689BFA-22DF-11D9-ACBE-00306549C134@pcisys.net> Message-ID: On 10/21/04 12:29 AM, "Chuck Pelto" wrote: > Is it a 'bad thing' to rename the fields in a VBaseObject inside of an > RB app? > > Example: > > Originally the fields were titled "this_fld", "that_fld", "another_fld". > > I populated the database. > > Then I thought better of my naming convention and changed the fields to > be "fld_this", "fld_that", and "fld_another". > > Now I can't access the database anymore. > > Or was it something else I may have done? Yes you should not do this. If you just develop and you need this then you must start with fresh db If you have dbs and you need change structure then you need look at example Upgrade of schema -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Thu Oct 21 08:15:20 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Thu Oct 21 01:15:38 2004 Subject: Bad Thing? In-Reply-To: <2F689BFA-22DF-11D9-ACBE-00306549C134@pcisys.net> References: <2F689BFA-22DF-11D9-ACBE-00306549C134@pcisys.net> Message-ID: <95F86CA2-2328-11D9-867A-000393DB2B60@pandora.be> On 20 Oct 2004, at 23:29, Chuck Pelto wrote: > Is it a 'bad thing' to rename the fields in a VBaseObject inside of an > RB app? > > Example: > > Originally the fields were titled "this_fld", "that_fld", > "another_fld". > > I populated the database. > > Then I thought better of my naming convention and changed the fields > to be "fld_this", "fld_that", and "fld_another". > > Now I can't access the database anymore. > > Or was it something else I may have done? > Yep, you should have opened your database as a generic VDatabase first, then alter the table structure (i.e. rename the fields), then open it using your REALbasic VDatabase subclass. Best regards Peter De Berdt From cbpelto at pcisys.net Thu Oct 21 10:00:42 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 21 11:00:56 2004 Subject: Bad Thing? In-Reply-To: <95F86CA2-2328-11D9-867A-000393DB2B60@pandora.be> References: <2F689BFA-22DF-11D9-ACBE-00306549C134@pcisys.net> <95F86CA2-2328-11D9-867A-000393DB2B60@pandora.be> Message-ID: <5C1D0034-237A-11D9-A092-00306549C134@pcisys.net> Greetings, On Oct 21, 2004, at 12:15 AM, Peter De Berdt wrote: > Yep, you should have opened your database as a generic VDatabase > first, then alter the table structure (i.e. rename the fields), then > open it using your REALbasic VDatabase subclass. Okay...where are instructions on how to go about doing THAT business; opening the db as a generic VDB and altering the table structure? It's good that I'm doing this, learning all the pit/prat-falls, on something simple. By the way, is there a nice tool to assist in looking at and modifying a database system and its structure? Something like ANALYZER and VISUALIZER by Waves In Motion for FileMaker Pro? Those were particularly useful tools. I'd pay good money to have something like that for Valentina databases. Regards, Chuck From sunshine at public.kherson.ua Thu Oct 21 19:56:30 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 21 11:56:50 2004 Subject: Bad Thing? In-Reply-To: <5C1D0034-237A-11D9-A092-00306549C134@pcisys.net> Message-ID: On 10/21/04 7:00 PM, "Chuck Pelto" wrote: >> Yep, you should have opened your database as a generic VDatabase >> first, then alter the table structure (i.e. rename the fields), then >> open it using your REALbasic VDatabase subclass. > > Okay...where are instructions on how to go about doing THAT business; > opening the db as a generic VDB and altering the table structure? Yes, check example db = new Vdatabase alter stucture db.close db= nil > It's good that I'm doing this, learning all the pit/prat-falls, on > something simple. > By the way, is there a nice tool to assist in looking at and modifying > a database system and its structure? Something like ANALYZER and > VISUALIZER by Waves In Motion for FileMaker Pro? > Those were particularly useful tools. I'd pay good money to have > something like that for Valentina databases. Not sure what you ask about but look at VAPP and Valentina Studio -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From j.peters at valentina-db.de Thu Oct 21 19:30:37 2004 From: j.peters at valentina-db.de (Jochen Peters) Date: Thu Oct 21 12:31:10 2004 Subject: Bad Thing? In-Reply-To: References: Message-ID: Hi Chuck, >> By the way, is there a nice tool to assist in looking at and modifying >> a database system and its structure? Something like ANALYZER and >> VISUALIZER by Waves In Motion for FileMaker Pro? > >> Those were particularly useful tools. I'd pay good money to have >> something like that for Valentina databases. > > Not sure what you ask about but look at VAPP and Valentina Studio > Chuck - go to http://www.valentina-db.de/en/downloads/downvstd.shtml there you can find Valentina Studio - it can browse and change Valentina database structure and it is also able to import data from any ODBC datasource into Valentina. I have just uploaded a new alpha version 1.0a68. I think tomorrow i will upload the next build - a69 -- Best regards, Jochen Peters PIIT GmbH ------------------------------------ http://www.valentina-db.de From softil at onlinehome.de Thu Oct 21 23:29:19 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 21 16:29:52 2004 Subject: [VNET] "NO_CASE" dosent work in select statement Message-ID: <003a01c4b7b5$12e2bef0$cef4e6d9@IPSBASIS> Hi Ruslan, there is a new BUG in Valentina VNET. The "NO_CASE" clause doesnt work! If I write this: "select ohrnummer from tiere where (ohrnummercountry like 'DE00')" I get an resultset of 10 records. If I write this: "select ohrnummer from tiere where (ohrnummercountry like 'de00') NO_CASE" I get an resultset of 0 records! I need urgent an update. Best regards Carsten From softil at onlinehome.de Thu Oct 21 23:32:33 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 21 16:32:51 2004 Subject: [VNET] "NO_CASE" dosent work in select statement Message-ID: <004201c4b7b5$7bfe8d60$cef4e6d9@IPSBASIS> Hi, I'm again. I think this isn't a Bug in VNet, because I get the same bad result if I try this in ViSQL. Where can be the mistake. Best regards Carsten From sunshine at public.kherson.ua Fri Oct 22 00:34:38 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 21 16:34:56 2004 Subject: [VNET] "NO_CASE" dosent work in select statement In-Reply-To: <003a01c4b7b5$12e2bef0$cef4e6d9@IPSBASIS> Message-ID: On 10/22/04 12:29 AM, "SoftIl" wrote: > Hi Ruslan, > there is a new BUG in Valentina VNET. > The "NO_CASE" clause doesnt work! > > If I write this: > > "select ohrnummer from tiere where (ohrnummercountry like 'DE00')" > > I get an resultset of 10 records. > > If I write this: > "select ohrnummer from tiere where (ohrnummercountry like 'de00') NO_CASE" > > I get an resultset of 0 records! > > I need urgent an update. Hi, I think problem in () Try this where ohrnummercountry like 'de00' NO_CASE -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From ernestogiannotta at tiscalinet.it Thu Oct 21 23:43:26 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Oct 21 16:44:07 2004 Subject: V4RB2 Message-ID: <3D7B2990-23AA-11D9-BDAB-003065514D2E@tiscalinet.it> Hello Ruslan, I'm trying the V4RB examples and have some questions and bug reports: 1) How generation of Log Files (FBL_Asserts.txt, FBL_Asserts.txt) can be disabled? 2) MacOS_Creator example: Creator of .vdb file remains VALA even if set as MYAP (btw why there's no icon for Vind files?) 3) Locales and Collation properties what is affected by Locales? maybe date and decimal separators? there's a list of acceptable values? 4) Locale_Records example strings are returned always as UTF-8 even from UTF-16 database can't store UTF-16 strings to UTF-16 database Cool Runnings, Erne. |er| musical box |ne| a media store From sunshine at public.kherson.ua Fri Oct 22 00:52:25 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 21 16:52:43 2004 Subject: V4RB2 In-Reply-To: <3D7B2990-23AA-11D9-BDAB-003065514D2E@tiscalinet.it> Message-ID: On 10/22/04 12:43 AM, "erne" wrote: Hi Erne, > I'm trying the V4RB examples and have some questions and bug reports: > 1) How generation of Log Files (FBL_Asserts.txt, FBL_Asserts.txt) can > be disabled? We will remove them on release > 2) MacOS_Creator example: > Creator of .vdb file remains VALA even if set as MYAP > (btw why there's no icon for Vind files?) In fact for any file must not be icons. Icons is responsibility of YOUR app. > 3) Locales and Collation properties > what is affected by Locales? Strings, indexes of strings, > maybe date and decimal separators? no > there's a list of acceptable values? ISO standard. en_EN, en_GB, ru_RU, de_DE, ... > 4) Locale_Records example > strings are returned always as UTF-8 even from UTF-16 database > can't store UTF-16 strings to UTF-16 database They ARE stored in UTF16. Just with RB we communicate via UTF8 Yes, for now V4RB 2.0 beta expect from you UTF8 strings. This is default of RB, right ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From softil at onlinehome.de Fri Oct 22 01:56:31 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 21 18:56:50 2004 Subject: [VNET] "NO_CASE" dosent work in select statement References: Message-ID: <001801c4b7c9$99bbd5b0$1cefe6d9@IPSBASIS> Hi, > where ohrnummercountry like 'de00' NO_CASE Ok. Now it works fine. :-) Thanks for your fast answer. Best regards Carsten BTW: Can you say something about the "old" known problems in VNET and a new update with bugfixes? From cbpelto at pcisys.net Thu Oct 21 21:25:22 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 21 22:25:41 2004 Subject: VAPP and V Studio In-Reply-To: References: Message-ID: <01AEFD7F-23DA-11D9-A092-00306549C134@pcisys.net> I've got copies of Valentina App and Valentina Studio I can run VAPP. It looks like it is comperable to FMP's VISUALIZER with a good bit of ANALYZER thrown in. This will be very useful. Thanks. A problem with Studio. It will not run on my Mac G4 running OS X 10.3.5. It starts to open, but nothing happens. Regards, Chuck On Oct 21, 2004, at 10:56 AM, Ruslan Zasukhin wrote: > On 10/21/04 7:00 PM, "Chuck Pelto" wrote: > >>> Yep, you should have opened your database as a generic VDatabase >>> first, then alter the table structure (i.e. rename the fields), then >>> open it using your REALbasic VDatabase subclass. >> >> Okay...where are instructions on how to go about doing THAT business; >> opening the db as a generic VDB and altering the table structure? > > Yes, check example > > db = new Vdatabase > > alter stucture > > db.close > db= nil > >> It's good that I'm doing this, learning all the pit/prat-falls, on >> something simple. > > >> By the way, is there a nice tool to assist in looking at and modifying >> a database system and its structure? Something like ANALYZER and >> VISUALIZER by Waves In Motion for FileMaker Pro? > >> Those were particularly useful tools. I'd pay good money to have >> something like that for Valentina databases. > > Not sure what you ask about but look at VAPP and Valentina Studio > > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From kevo at gatorgraphics.com Thu Oct 21 23:52:56 2004 From: kevo at gatorgraphics.com (Kevin Windham) Date: Thu Oct 21 23:53:21 2004 Subject: VAPP and V Studio In-Reply-To: <01AEFD7F-23DA-11D9-A092-00306549C134@pcisys.net> References: <01AEFD7F-23DA-11D9-A092-00306549C134@pcisys.net> Message-ID: <3DBB1276-23E6-11D9-BA50-000393B29272@gatorgraphics.com> On Oct 21, 2004, at 10:25 PM, Chuck Pelto wrote: > I've got copies of Valentina App and Valentina Studio > > I can run VAPP. It looks like it is comperable to FMP's VISUALIZER > with a good bit of ANALYZER thrown in. This will be very useful. > Thanks. > > A problem with Studio. It will not run on my Mac G4 running OS X > 10.3.5. It starts to open, but nothing happens. Yes, this is a know issue with stuffit we think. You need to go into the app bundle and add execute permission to the application. Hopefully the next version will be distributed as a disk image and should take care of this issue. HTH, Kevin From sunshine at public.kherson.ua Fri Oct 22 09:23:48 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 01:24:06 2004 Subject: [VNET] "NO_CASE" dosent work in select statement In-Reply-To: <001801c4b7c9$99bbd5b0$1cefe6d9@IPSBASIS> Message-ID: On 10/22/04 2:56 AM, "SoftIl" wrote: > BTW: > Can you say something about the "old" known problems in VNET and a new > update with bugfixes? Sorry not, yet Carsten. Something very weird -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 22 09:27:09 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 01:27:54 2004 Subject: How to code a date ? In-Reply-To: Message-ID: On 10/22/04 6:01 AM, "Thierry Nauze" wrote: Hi Thierry, > How to to code a date when I kown only the year and possibily the month > > July 25, 2004 -> 2004/07/25 > > July, 2004 -> ???? : 2004/07/.. or 2004/07/00 or other string ? > > 2004 -> 2004/../.. or 2004/00/00 or other string ? > > The three dates are stored in the same field. Code for storing on disk or for searching ? For storing you can do INSERT ... ( '2004/00/00' ) For search you can use LIKE or REGEX WHERE fld LIKE '2004/??/??' WHERE fld REGEX '2004/../..' -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From ohmitou at wanadoo.fr Fri Oct 22 12:44:17 2004 From: ohmitou at wanadoo.fr (Thierry Nauze) Date: Fri Oct 22 03:44:47 2004 Subject: How to code a date ? In-Reply-To: References: Message-ID: <8F2E7282-2406-11D9-86E0-000A95BA5A32@wanadoo.fr> Le 22 oct. 04, ? 10:27, Ruslan Zasukhin a ?crit : > > Code for storing on disk or for searching ? > > For storing you can do > > INSERT ... ( '2004/00/00' ) > > > For search you can use LIKE or REGEX > > WHERE fld LIKE '2004/??/??' > > > WHERE fld REGEX '2004/../..' Thank you, it is exactly that I wanted to know. -- Thierry Nauze Saint-Denis de la R?union From sunshine at public.kherson.ua Fri Oct 22 14:15:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 06:16:09 2004 Subject: RealBasic In-Reply-To: <00d201c4b818$0a01ea00$0200a8c0@medion> Message-ID: On 10/22/04 12:18 PM, "Gunnar Jacobsen" wrote: Hi Gunnar, Excellent! thank you for info. > Hi Ruslan. > > Both C++Builder and Delphi have a action called "Import Type Library" and > now it is working fine. > > > Thanks > > Gunnar Jacobsen. > > ----- Original Message ----- > From: "Ruslan Zasukhin" > To: "Gunnar Jacobsen" > Sent: Wednesday, October 20, 2004 4:28 PM > Subject: Re: RealBasic > > >> On 10/20/04 4:51 PM, "Gunnar Jacobsen" wrote: >> >>> Hi Ruslan. >>> >>> Here are code from C++Builder how to open ole object, is there simulary > for >>> your VCOM. >>> >>> >>> *********** code begin here ********************* >>> Variant V; >>> >>> V = CreateOleObject("Word.Basic"); >>> V.Exec(Procedure("AppShow")); >>> V.Exec(Procedure("FileNew") << "Normal"); >>> V.Exec(Procedure("Insert") << "Hello from Borland"); >>> >>> *********** code end here ********************** >>> >>> Do you have some example for VCOM call for C++. >>> >>> Could you write few line how to start to use VCOM. in Delphi or > C++Builder. >> >> I have no on hands such code. >> >> To get it we will need time. >> >> May be you will try self to do that ? >> If big problems then let me know and we will make C++ example. >> >> May be there is sense for you at first play with Visual BASIC? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Fri Oct 22 12:45:37 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 22 13:45:59 2004 Subject: VAPP and V Studio In-Reply-To: <3DBB1276-23E6-11D9-BA50-000393B29272@gatorgraphics.com> References: <01AEFD7F-23DA-11D9-A092-00306549C134@pcisys.net> <3DBB1276-23E6-11D9-BA50-000393B29272@gatorgraphics.com> Message-ID: <90741636-245A-11D9-8C54-00306549C134@pcisys.net> What is the procedure to do this? I've tried the permissions repair, but no joy. Regards, Chuck On Oct 21, 2004, at 10:52 PM, Kevin Windham wrote: > > On Oct 21, 2004, at 10:25 PM, Chuck Pelto wrote: > >> I've got copies of Valentina App and Valentina Studio >> >> I can run VAPP. It looks like it is comperable to FMP's VISUALIZER >> with a good bit of ANALYZER thrown in. This will be very useful. >> Thanks. >> >> A problem with Studio. It will not run on my Mac G4 running OS X >> 10.3.5. It starts to open, but nothing happens. > > Yes, this is a know issue with stuffit we think. You need to go into > the app bundle and add execute permission to the application. > Hopefully the next version will be distributed as a disk image and > should take care of this issue. > > HTH, > Kevin > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Fri Oct 22 21:55:29 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 13:55:58 2004 Subject: VAPP and V Studio In-Reply-To: <90741636-245A-11D9-8C54-00306549C134@pcisys.net> Message-ID: On 10/22/04 9:45 PM, "Chuck Pelto" wrote: > What is the procedure to do this? > > I've tried the permissions repair, but no joy. This is way > Looking at the permissions on Valentina > Studio.app/Contents/MacOS/Valentina Studio > I see that I have read/write but no execute. > So I did a chmod u+x and it works now. You need open Terminal, And execute command chmod u+x for file Studio.app/Contents/MacOS/Valentina Studio Also you can use some OS X tools for this. For example "Path Finder" -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Fri Oct 22 14:35:29 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 22 15:36:01 2004 Subject: VAPP and V Studio In-Reply-To: References: Message-ID: Hi Ruslan, You REALLY should consider fixing that little 'problem' immediately. As an interim, I HIGHLY recommend you provide a step-by-step example of what must be done to work through this. Smart people do not like going into TERMINAL mode. And assuming people are familiar with how to navigate to a destination in Terminal without fouling things up royally, is not a good modis operandi. Regards, Chuck On Oct 22, 2004, at 12:55 PM, Ruslan Zasukhin wrote: > On 10/22/04 9:45 PM, "Chuck Pelto" wrote: > >> What is the procedure to do this? >> >> I've tried the permissions repair, but no joy. > > This is way > >> Looking at the permissions on Valentina >> Studio.app/Contents/MacOS/Valentina Studio > >> I see that I have read/write but no execute. >> So I did a chmod u+x and it works now. > > You need open Terminal, > > And execute command chmod u+x for file > Studio.app/Contents/MacOS/Valentina Studio > > > Also you can use some OS X tools for this. > For example "Path Finder" > > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Fri Oct 22 23:43:41 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 15:44:24 2004 Subject: VAPP and V Studio In-Reply-To: Message-ID: On 10/22/04 11:35 PM, "Chuck Pelto" wrote: > Hi Ruslan, > > You REALLY should consider fixing that little 'problem' immediately. > > As an interim, I HIGHLY recommend you provide a step-by-step example of > what must be done to work through this. Smart people do not like going > into TERMINAL mode. And assuming people are familiar with how to > navigate to a destination in Terminal without fouling things up > royally, is not a good modis operandi. Jochen will fix this as I know. BTW, in terminal very easy navigate if drag & drop folder/file to terminal window. So have you fix problem ? Again, you can download Path Finder utility. It can do this easy. Another solution upgrade your StuffIt to avoid this bug. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Fri Oct 22 14:50:18 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 22 15:50:42 2004 Subject: VAPP and V Studio In-Reply-To: References: Message-ID: Ahem.... ...as I was saying...... ...I attempted to do this and here is what has happened.... [Pelto-Computer:~/Desktop] cbpelto% chmod u+x Valentina Studio.app/Contents/MacOS/Valentina Studio chmod: Valentina: No such file or directory chmod: Studio.app/Contents/MacOS/Valentina: No such file or directory chmod: Studio: No such file or directory [Pelto-Computer:~/Desktop] cbpelto% This was from the proper directory, where the application is currently located, the Desktop. I even tried it with a leading "/". But to no avail. What is the proper syntax, please? Thanks, Chuck On Oct 22, 2004, at 2:35 PM, Chuck Pelto wrote: > Hi Ruslan, > > You REALLY should consider fixing that little 'problem' immediately. > > As an interim, I HIGHLY recommend you provide a step-by-step example > of what must be done to work through this. Smart people do not like > going into TERMINAL mode. And assuming people are familiar with how to > navigate to a destination in Terminal without fouling things up > royally, is not a good modis operandi. > > Regards, > > Chuck > > On Oct 22, 2004, at 12:55 PM, Ruslan Zasukhin wrote: > >> On 10/22/04 9:45 PM, "Chuck Pelto" wrote: >> >>> What is the procedure to do this? >>> >>> I've tried the permissions repair, but no joy. >> >> This is way >> >>> Looking at the permissions on Valentina >>> Studio.app/Contents/MacOS/Valentina Studio >> >>> I see that I have read/write but no execute. >>> So I did a chmod u+x and it works now. >> >> You need open Terminal, >> >> And execute command chmod u+x for file >> Studio.app/Contents/MacOS/Valentina Studio >> >> >> Also you can use some OS X tools for this. >> For example "Path Finder" >> >> >> >> -- >> Best regards, >> Ruslan Zasukhin [ I feel the need...the need for speed ] >> ------------------------------------------------------------- >> e-mail: ruslan@paradigmasoft.com >> web: http://www.paradigmasoft.com >> >> To subscribe to the Valentina mail list go to: >> http://lists.macserve.net/mailman/listinfo/valentina >> ------------------------------------------------------------- >> >> _______________________________________________ >> Valentina mailing list >> Valentina@lists.macserve.net >> http://lists.macserve.net/mailman/listinfo/valentina >> > From cbpelto at pcisys.net Fri Oct 22 14:51:13 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Fri Oct 22 15:51:35 2004 Subject: VAPP and V Studio In-Reply-To: References: Message-ID: <1C9E4438-246C-11D9-8C54-00306549C134@pcisys.net> Pathfinder sounds nice. Where can I find it? Regards, Chuck On Oct 22, 2004, at 2:43 PM, Ruslan Zasukhin wrote: > On 10/22/04 11:35 PM, "Chuck Pelto" wrote: > >> Hi Ruslan, >> >> You REALLY should consider fixing that little 'problem' immediately. >> >> As an interim, I HIGHLY recommend you provide a step-by-step example >> of >> what must be done to work through this. Smart people do not like going >> into TERMINAL mode. And assuming people are familiar with how to >> navigate to a destination in Terminal without fouling things up >> royally, is not a good modis operandi. > > Jochen will fix this as I know. > > BTW, in terminal very easy navigate if drag & drop folder/file to > terminal > window. > > So have you fix problem ? > > > Again, you can download Path Finder utility. > It can do this easy. > > > Another solution upgrade your StuffIt to avoid this bug. > > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Fri Oct 22 23:58:09 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 15:58:28 2004 Subject: VAPP and V Studio In-Reply-To: Message-ID: On 10/22/04 11:50 PM, "Chuck Pelto" wrote: > Ahem.... > > ...as I was saying...... > > ...I attempted to do this and here is what has happened.... > > [Pelto-Computer:~/Desktop] cbpelto% chmod u+x Valentina > Studio.app/Contents/MacOS/Valentina Studio > chmod: Valentina: No such file or directory > chmod: Studio.app/Contents/MacOS/Valentina: No such file or directory > chmod: Studio: No such file or directory > [Pelto-Computer:~/Desktop] cbpelto% > > This was from the proper directory, where the application is currently > located, the Desktop. I even tried it with a leading "/". But to no > avail. > > What is the proper syntax, please? I do next 3 simple steps * open terminal * type chmod u+x * drag and drop file Valentina Studio.app/Contents/MacOS/Valentina Studio Into terminal window. As result I get the next string CHMOD u+x /Users/rz/Desktop/Valentina\ Studio.app/Contents/MacOS/Valentina\ Studio You can see that in my string OS X automatically have escape the SPACE symbols. Press ENTER. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 22 23:58:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 22 15:59:55 2004 Subject: VAPP and V Studio In-Reply-To: <1C9E4438-246C-11D9-8C54-00306549C134@pcisys.net> Message-ID: On 10/22/04 11:51 PM, "Chuck Pelto" wrote: > Pathfinder sounds nice. Where can I find it? As any MAC application : www.versiontracker.com -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From softil at onlinehome.de Fri Oct 22 13:58:13 2004 From: softil at onlinehome.de (SoftIl) Date: Fri Oct 22 17:00:06 2004 Subject: [VNET] "NO_CASE" dosent work in select statement References: Message-ID: <001201c4b882$752ab950$06f3e6d9@IPSBASIS> > > Can you say something about the "old" known problems in VNET and a new > > update with bugfixes? > Sorry not, yet Carsten. > Something very weird Ok. ;-) Best regards Carsten Ilwig From totoole at rocko.lab.csuchico.edu Fri Oct 22 20:17:39 2004 From: totoole at rocko.lab.csuchico.edu (Thomas) Date: Fri Oct 22 22:17:46 2004 Subject: Base Object Method Prevents Adding Records Message-ID: <187D5060-24A2-11D9-A057-000393C799A0@rocko.lab.csuchico.edu> Hi all, The following Base Object Method is causing trouble in VAPP. if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd > ptr_child_rcd,concat(ptr_parent_rcd,'- ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) It applies to a table with the following fields: ptr_parent_rcd (an ObjectPtr to one side of a many to many join) ptr_child_rcd (an ObjectPtr to the other side of the many to many join) This is a recursive many to many so the target of both ObjectPtrs is the same, but they point to different RecIDs in the target table. The Base Object Method's Unique property is set to true. The purpose of the Base Object Method is to prevent duplicate records from being added to the table. If there is a record relating parent RecID 10 to child RecID 11, I need to prevent another record relating RecID 10 to RecID 11 from being created, regardless of which RecID is in the parent column and which is in the child column. So if we have a record: ptr_parent_rcd ptr_child_rcd 10 11 We need to prevent the following record: ptr_parent_rcd ptr_child_rcd 11 10 Furthermore, records where parent and child pointers are equal need to be prevented. So I create one unused record where this is the case, and the Base Object's Unique property prevents any others. The problem is, the Base Object Method will not allow valid records to be added using the import method or the VAPP interface. No error is returned. The import method runs, but the records are not added. The Base Object Method does not stop the make-new-record-with-data method from adding valid records. Any ideas much appreciated. Thomas From sunshine at public.kherson.ua Sat Oct 23 10:04:48 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 02:05:07 2004 Subject: Base Object Method Prevents Adding Records In-Reply-To: <187D5060-24A2-11D9-A057-000393C799A0@rocko.lab.csuchico.edu> Message-ID: On 10/23/04 6:17 AM, "Thomas" wrote: > Hi all, > > The following Base Object Method is causing trouble in VAPP. > > if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd > > ptr_child_rcd,concat(ptr_parent_rcd,'- > ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) > > It applies to a table with the following fields: > ptr_parent_rcd (an ObjectPtr to one side of a many to many join) > ptr_child_rcd (an ObjectPtr to the other side of the many to many join) > > This is a recursive many to many so the target of both ObjectPtrs is > the same, but they point to different RecIDs in the target table. The first point. I see that you compare values of ObjectPtrs, I.e. Values of RecID they point to. This is mistake. Because if you will delete a record, then its RecID is reused. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 23 10:05:35 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 02:06:01 2004 Subject: Base Object Method Prevents Adding Records In-Reply-To: <187D5060-24A2-11D9-A057-000393C799A0@rocko.lab.csuchico.edu> Message-ID: On 10/23/04 6:17 AM, "Thomas" wrote: > Hi all, > > The following Base Object Method is causing trouble in VAPP. > > if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd > > ptr_child_rcd,concat(ptr_parent_rcd,'- > ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) > > It applies to a table with the following fields: > ptr_parent_rcd (an ObjectPtr to one side of a many to many join) > ptr_child_rcd (an ObjectPtr to the other side of the many to many join) > > This is a recursive many to many so the target of both ObjectPtrs is > the same, but they point to different RecIDs in the target table. > > The Base Object Method's Unique property is set to true. As I have think. Then you have 2 records with the same value. Check this > The purpose of the Base Object Method is to prevent duplicate records > from being added to the table. > If there is a record relating parent RecID 10 to child RecID 11, I need > to prevent another record relating > RecID 10 to RecID 11 from being created, regardless of which RecID is > in the parent column and which is in the child column. > > So if we have a record: > ptr_parent_rcd ptr_child_rcd > 10 11 > > We need to prevent the following record: > ptr_parent_rcd ptr_child_rcd > 11 10 > > Furthermore, records where parent and child pointers are equal need to > be prevented. So I create one unused > record where this is the case, and the Base Object's Unique property > prevents any others. > > The problem is, the Base Object Method will not allow valid records to > be added using the import method or the VAPP interface. > No error is returned. The import method runs, but the records are not > added. > The Base Object Method does not stop the make-new-record-with-data > method from adding valid records. > > Any ideas much appreciated. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Sat Oct 23 02:58:42 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Sat Oct 23 03:59:10 2004 Subject: VAPP and V Studio In-Reply-To: References: Message-ID: Morning Ruslan, Did the 3 steps, indicated below. Here's what I got.... [Pelto-Computer:~] cbpelto% chmod u+x usage: chmod [-fv] [-R [-H | -L | -P]] mode file ... [Pelto-Computer:~] cbpelto% /Users/cbpelto/Desktop/Valentina\ Studio.app tcsh: /Users/cbpelto/Desktop/Valentina Studio.app: Permission denied. [Pelto-Computer:~] cbpelto% The term "Permission denied" is confusing, as now I can run Valentina Studio. Thanks for your patience with me on this. I'd not have figured this out for quite some time and, probably, not without doing some harm, in some fashion, to the computer system. Regards, Chuck On Oct 22, 2004, at 2:58 PM, Ruslan Zasukhin wrote: > On 10/22/04 11:50 PM, "Chuck Pelto" wrote: > >> Ahem.... >> >> ...as I was saying...... >> >> ...I attempted to do this and here is what has happened.... >> >> [Pelto-Computer:~/Desktop] cbpelto% chmod u+x Valentina >> Studio.app/Contents/MacOS/Valentina Studio >> chmod: Valentina: No such file or directory >> chmod: Studio.app/Contents/MacOS/Valentina: No such file or directory >> chmod: Studio: No such file or directory >> [Pelto-Computer:~/Desktop] cbpelto% >> >> This was from the proper directory, where the application is currently >> located, the Desktop. I even tried it with a leading "/". But to no >> avail. >> >> What is the proper syntax, please? > > I do next 3 simple steps > > * open terminal > > * type chmod u+x > > * drag and drop file > Valentina Studio.app/Contents/MacOS/Valentina Studio > > Into terminal window. > > As result I get the next string > > CHMOD u+x /Users/rz/Desktop/Valentina\ > Studio.app/Contents/MacOS/Valentina\ > Studio > > > You can see that in my string OS X automatically have escape the SPACE > symbols. > > Press ENTER. > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From cbpelto at pcisys.net Sat Oct 23 03:00:03 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Sat Oct 23 04:00:28 2004 Subject: VAPP and V Studio In-Reply-To: References: Message-ID: Actually, Pathfinder isn't there. But I did find it at TUCOWS. Regards, Chuck On Oct 22, 2004, at 2:58 PM, Ruslan Zasukhin wrote: > On 10/22/04 11:51 PM, "Chuck Pelto" wrote: > >> Pathfinder sounds nice. Where can I find it? > > As any MAC application : > > www.versiontracker.com > > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From cbpelto at pcisys.net Sat Oct 23 03:06:41 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Sat Oct 23 04:06:58 2004 Subject: Valentina Studio Registraion Message-ID: Solving the problem, of not getting into VStudio, changes the problem..... Is the registration code supposed to be on of the codes I received when I bought into Valentina? I've tried all four codes, Mac and Windows, from the two copies and I'm getting a "Wrong registration number" notice for all of them. Regards, Chuck From sunshine at public.kherson.ua Sat Oct 23 12:12:43 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 04:13:03 2004 Subject: VAPP and V Studio In-Reply-To: Message-ID: On 10/23/04 11:58 AM, "Chuck Pelto" wrote: > Did the 3 steps, indicated below. Here's what I got.... > > [Pelto-Computer:~] cbpelto% chmod u+x > usage: chmod [-fv] [-R [-H | -L | -P]] mode file ... You have press enter ? No! Simply type this words And drag file to this line OS X will self add text Now press ENTER > [Pelto-Computer:~] cbpelto% /Users/cbpelto/Desktop/Valentina\ Studio.app > tcsh: /Users/cbpelto/Desktop/Valentina Studio.app: Permission denied. > [Pelto-Computer:~] cbpelto% > > The term "Permission denied" is confusing, as now I can run Valentina > Studio. > > Thanks for your patience with me on this. I'd not have figured this out > for quite some time and, probably, not without doing some harm, in some > fashion, to the computer system. >> I do next 3 simple steps >> >> * open terminal >> >> * type chmod u+x >> >> * drag and drop file >> Valentina Studio.app/Contents/MacOS/Valentina Studio >> >> Into terminal window. >> >> As result I get the next string >> >> CHMOD u+x /Users/rz/Desktop/Valentina\ >> Studio.app/Contents/MacOS/Valentina\ >> Studio >> >> >> You can see that in my string OS X automatically have escape the SPACE >> symbols. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 23 12:36:38 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 04:38:01 2004 Subject: VAPP and V Studio In-Reply-To: <643EE49E-24D5-11D9-8D99-000A959F6B0C@valentina-db.de> Message-ID: On 10/23/04 12:24 PM, "Jochen Peters" wrote: > Hi all, > >>> You REALLY should consider fixing that little 'problem' immediately. >>> >>> As an interim, I HIGHLY recommend you provide a step-by-step example >>> of >>> what must be done to work through this. Smart people do not like going >>> into TERMINAL mode. And assuming people are familiar with how to >>> navigate to a destination in Terminal without fouling things up >>> royally, is not a good modis operandi. >> >> Jochen will fix this as I know. > This is already fixed! The actual download on > http://www.valentina-db.de7en7downloads7downvstd.shtml > is a disk image - no stuffit archive any more! Thank you Jochen, -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 23 12:36:34 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 04:38:48 2004 Subject: Valentina Studio Registraion In-Reply-To: Message-ID: On 10/23/04 12:06 PM, "Chuck Pelto" wrote: > Solving the problem, of not getting into VStudio, changes the > problem..... You mean you have RUN Vstudio now ? > Is the registration code supposed to be on of the codes I received when > I bought into Valentina? I have CC to Jochen. You need talk with him on this. > I've tried all four codes, Mac and Windows, from the two copies and I'm > getting a "Wrong registration number" notice for all of them. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 23 12:59:12 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 04:59:32 2004 Subject: VAPP and V Studio In-Reply-To: Message-ID: On 10/23/04 12:36 PM, "Ruslan Zasukhin" wrote: >>> Jochen will fix this as I know. >> This is already fixed! The actual download on >> http://www.valentina-db.de7en7downloads7downvstd.shtml >> is a disk image - no stuffit archive any more! Chuck, You can download a69 Vstudio http://www.valentina-db.de/en/downloads/downvstd.shtml -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Sat Oct 23 10:10:23 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Sat Oct 23 11:10:47 2004 Subject: Valentina Studio Registraion In-Reply-To: References: Message-ID: <0B407DBE-250E-11D9-8C54-00306549C134@pcisys.net> Morning Ruslan, YES. I was able to run VStudio after doing the step-by-step process you kindly provided. Without that step-by-step, I'd still be floundering about. Probably causing other problems in my machine. The odd message telling me PERMISSION DENIED at the end of the processing in Terminal was apparently in error, as afterwards I could run VStudio. Why it was there, I have no idea. When I ran VStudio, I noticed the requirement for registration. Something about a 10-minute limitation on each run until registration was accomplished. So, I went to my documents and got my registration numbers for Valentina and tried them. Every one of them, Mac and Windows, but none of them would satisfy the application's registration requirement. Regards, Chuck On Oct 23, 2004, at 3:36 AM, Ruslan Zasukhin wrote: > On 10/23/04 12:06 PM, "Chuck Pelto" wrote: > >> Solving the problem, of not getting into VStudio, changes the >> problem..... > > You mean you have RUN Vstudio now ? > > >> Is the registration code supposed to be on of the codes I received >> when >> I bought into Valentina? > > I have CC to Jochen. > > You need talk with him on this. > >> I've tried all four codes, Mac and Windows, from the two copies and >> I'm >> getting a "Wrong registration number" notice for all of them. > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Sat Oct 23 19:39:16 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 11:39:39 2004 Subject: Valentina Studio Registraion In-Reply-To: <0B407DBE-250E-11D9-8C54-00306549C134@pcisys.net> Message-ID: On 10/23/04 7:10 PM, "Chuck Pelto" wrote: > So, I went to my documents and got my registration numbers for > Valentina and tried them. Every one of them, Mac and Windows, but none > of them would satisfy the application's registration requirement. Chuck, Valentina Studio is separate project. If you like you can need contact Jochen Peters about purchase See CC. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From totoole at rocko.lab.csuchico.edu Sat Oct 23 09:58:58 2004 From: totoole at rocko.lab.csuchico.edu (Thomas) Date: Sat Oct 23 11:59:00 2004 Subject: Base Object Method Prevents Adding Records In-Reply-To: <20041023085928.6A9C2249E14@edison.macserve.net> References: <20041023085928.6A9C2249E14@edison.macserve.net> Message-ID: >> Hi all, >> >> The following Base Object Method is causing trouble in VAPP. >> >> if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd >> ptr_child_rcd,concat(ptr_parent_rcd,'- >> ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) >> >> It applies to a table with the following fields: >> ptr_parent_rcd (an ObjectPtr to one side of a many to many >> join) >> ptr_child_rcd (an ObjectPtr to the other side of the many to >> many join) >> >> This is a recursive many to many so the target of both ObjectPtrs is >> the same, but they point to different RecIDs in the target table. > > The first point. > > I see that you compare values of ObjectPtrs, > I.e. Values of RecID they point to. > > This is mistake. > > Because if you will delete a record, then its RecID is reused. The "On delete" property for both the parent and child ObjectPtrs in the lookup table is set to "Cascade". So the ObjectPtrs are deleted when the RecIDs they point to are deleted, no? Does this not prevent re-use mistakes? Is there a better way to manage these comparisons? >> The Base Object Method's Unique property is set to true. > > As I have think. > > Then you have 2 records with the same value. > Check this. I checked this by setting the Base Object Method's Unique property to false, then importing the records, then setting the Unique property back to true. If there were duplicate records, wouldn't VAPP prevent the Unique property from being set back to true? Thanks for your help. > The purpose of the Base Object Method is to prevent duplicate records > from being added to the table. If there is a record relating parent > RecID 10 to child RecID 11, I need to prevent another record relating > RecID 10 to RecID 11 from being created, regardless of which RecID is > in the parent column and which is in the child column. > > So if we have a record: > ptr_parent_rcd ptr_child_rcd > 10 11 > > We need to prevent the following record: > ptr_parent_rcd ptr_child_rcd > 11 10 > > Furthermore, records where parent and child pointers are equal need > to be prevented. So I create one unused record where this is the case, > and the Base Object's Unique property prevents any others. > > The problem is, the Base Object Method will not allow valid records to > be added using the import method or the VAPP interface. No error is > returned. The import method runs, but the records are not > added. The Base Object Method does not stop the > make-new-record-with-datamethod from adding valid records. > > Any ideas much appreciated. From sunshine at public.kherson.ua Sat Oct 23 20:13:20 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 12:13:40 2004 Subject: Base Object Method Prevents Adding Records In-Reply-To: Message-ID: On 10/23/04 7:58 PM, "Thomas" wrote: >>> Hi all, >>> >>> The following Base Object Method is causing trouble in VAPP. >>> >>> if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd >>> ptr_child_rcd,concat(ptr_parent_rcd,'- >>> ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) >>> >>> It applies to a table with the following fields: >>> ptr_parent_rcd (an ObjectPtr to one side of a many to many >>> join) >>> ptr_child_rcd (an ObjectPtr to the other side of the many to >>> many join) >>> >>> This is a recursive many to many so the target of both ObjectPtrs is >>> the same, but they point to different RecIDs in the target table. >> >> The first point. >> >> I see that you compare values of ObjectPtrs, >> I.e. Values of RecID they point to. >> >> This is mistake. >> >> Because if you will delete a record, then its RecID is reused. > > The "On delete" property for both the parent and child ObjectPtrs in > the lookup table is set to "Cascade". So the ObjectPtrs are deleted > when the RecIDs they point to are deleted, no? Does this not prevent > re-use mistakes? Is there a better way to manage these comparisons? But you compare RecIds ! Why ? Explain your idea please You think that PARENT always have less RecID then child ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 23 20:14:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 12:15:14 2004 Subject: Base Object Method Prevents Adding Records In-Reply-To: Message-ID: On 10/23/04 7:58 PM, "Thomas" wrote: >>> The Base Object Method's Unique property is set to true. >> >> As I have think. >> >> Then you have 2 records with the same value. >> Check this. > > I checked this by setting the Base Object Method's Unique property to > false, then importing the records, then setting the Unique property > back to true. If there were duplicate records, wouldn't VAPP prevent > the Unique property from being set back to true? > > Thanks for your help. Yes, it must refuse build unique index. Also you can do such SQL after import SELECT f, count(f) FROM T GROUP BY f ORDER BY f DESC In this case if some value has duplicate it will be on top -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cbpelto at pcisys.net Sat Oct 23 11:53:29 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Sat Oct 23 12:53:58 2004 Subject: V Studio Crash Report Message-ID: <72B2C9FA-251C-11D9-8C54-00306549C134@pcisys.net> Greetings, Enclosed with this document, please find an Apple OS X report about a crash that occurred while I was starting to use Valentina Studio. I had just hit the CREATE button in the New Database dialog. The database had been Mode 1 type and set for 64K size. Hope this helps.... -------------- next part -------------- Date/Time: 2004-10-23 11:49:31 -0600 OS Version: 10.3.5 (Build 7M34) Report Version: 2 Command: Valentina Studio Path: /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Valentina Studio Version: 1.0a68 (1.0a68) PID: 1439 Thread: 0 Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000000 Thread 0 Crashed: 0 libSystem.B.dylib 0x90010844 strrchr + 0x4 1 com.paradigmasoft.vsdk 0xb00d1f84 _Z13FBL_Path2SpecPKcP6FSSpec + 0x20 2 com.paradigmasoft.vsdk 0xb0077d90 _ZN12VDK_DataBase14CreateDataBaseEPKcm10FBL_DbModeh + 0x34 3 vf.dylib 0x0980ad50 _ZN11vsVDatabase6CreateERK8wxStringlll + 0x8c 4 vf.dylib 0x09812c00 _ZN21vsVEmbeddedConnection14CreateDatabaseERK8wxStringlll + 0x120 5 vf.dylib 0x09812fb0 _ZN21vsVEmbeddedConnection20CreateDatabaseDialogEv + 0xe8 6 admin.dylib 0x01497854 _ZN14vsAdminCommand13DoNewDatabaseEv + 0x38 7 admin.dylib 0x01497704 _ZN14vsAdminCommand2DoEv + 0x70 8 libwx_mac.dylib 0x01635548 _ZN18wxCommandProcessor6SubmitEP9wxCommandb + 0x34 9 ...digmasoft.valentina.vstudio 0x00024320 _ZN7vsFrame5OnNewER14wxCommandEvent + 0xb0 10 libwx_mac.dylib 0x0164b1ec _ZN12wxEvtHandler21ProcessEventIfMatchesERK21wxEventTableEntryBasePS_R7wxEvent + 0x9c 11 libwx_mac.dylib 0x0164a98c _ZN16wxEventHashTable11HandleEventER7wxEventP12wxEvtHandler + 0xa0 12 libwx_mac.dylib 0x0164b398 _ZN12wxEvtHandler12ProcessEventER7wxEvent + 0xcc 13 libwx_mac.dylib 0x0164b3c4 _ZN12wxEvtHandler12ProcessEventER7wxEvent + 0xf8 14 libwx_mac.dylib 0x0164b3c4 _ZN12wxEvtHandler12ProcessEventER7wxEvent + 0xf8 15 libwx_mac.dylib 0x0164b3c4 _ZN12wxEvtHandler12ProcessEventER7wxEvent + 0xf8 16 libwx_mac.dylib 0x0169add8 _ZN10wxMenuBase9SendEventEii + 0x90 17 libwx_mac.dylib 0x0182c300 _Z27wxMacAppCommandEventHandlerP25OpaqueEventHandlerCallRefP14OpaqueEventRefPv + 0x174 18 libwx_mac.dylib 0x0182a44c _Z20wxMacAppEventHandlerP25OpaqueEventHandlerCallRefP14OpaqueEventRefPv + 0xbc 19 com.apple.HIToolbox 0x927d1fc8 DispatchEventToHandlers + 0x150 20 com.apple.HIToolbox 0x927d223c SendEventToEventTargetInternal + 0x174 21 com.apple.HIToolbox 0x927e46e4 SendEventToEventTarget + 0x28 22 com.apple.HIToolbox 0x927f96dc _Z18SendHICommandEventmPK9HICommandmmhP20OpaqueEventTargetRefS3_PP14OpaqueEventRef + 0x170 23 com.apple.HIToolbox 0x9289189c SendMenuItemSelectedEvent + 0x88 24 com.apple.HIToolbox 0x92893478 _Z19FinishMenuSelectionP8MenuDataP10MenuResultS2_mmm + 0x80 25 com.apple.HIToolbox 0x9285d78c _Z14MenuSelectCore5PointdmPP13OpaqueMenuRefPt + 0x164 26 com.apple.HIToolbox 0x92883b8c MenuSelect + 0x60 27 libwx_mac.dylib 0x017e9928 _Z30wxMacTopLevelMouseEventHandlerP25OpaqueEventHandlerCallRefP14OpaqueEventRefPv + 0x388 28 libwx_mac.dylib 0x0182a4cc _Z20wxMacAppEventHandlerP25OpaqueEventHandlerCallRefP14OpaqueEventRefPv + 0x13c 29 com.apple.HIToolbox 0x927d1fc8 DispatchEventToHandlers + 0x150 30 com.apple.HIToolbox 0x927d223c SendEventToEventTargetInternal + 0x174 31 com.apple.HIToolbox 0x927e46e4 SendEventToEventTarget + 0x28 32 com.apple.HIToolbox 0x92913560 _Z16HandleMouseEventP14OpaqueEventRef + 0x184 33 com.apple.HIToolbox 0x927e2c84 _Z29ToolboxEventDispatcherHandlerP25OpaqueEventHandlerCallRefP14OpaqueEventRefPv + 0x1e8 34 com.apple.HIToolbox 0x927d2084 DispatchEventToHandlers + 0x20c 35 com.apple.HIToolbox 0x927d223c SendEventToEventTargetInternal + 0x174 36 com.apple.HIToolbox 0x927e46e4 SendEventToEventTarget + 0x28 37 libwx_mac.dylib 0x0182b1d0 _ZN5wxApp17MacHandleOneEventEPv + 0x2c 38 libwx_mac.dylib 0x0182b168 _ZN5wxApp13MacDoOneEventEv + 0x9c 39 libwx_mac.dylib 0x0182ad6c _ZN5wxApp8MainLoopEv + 0x24 40 libwx_mac.dylib 0x01835498 _Z7wxEntryRiPPw + 0x60 41 libwx_mac.dylib 0x01835608 _Z7wxEntryRiPPc + 0xc4 42 ...digmasoft.valentina.vstudio 0x0001657c main + 0x18 43 ...digmasoft.valentina.vstudio 0x00002094 _start + 0x188 (crt.c:267) 44 dyld 0x8fe1a558 _dyld_start + 0x64 Thread 1: 0 libSystem.B.dylib 0x900074c8 mach_msg_trap + 0x8 1 libSystem.B.dylib 0x90007018 mach_msg + 0x38 2 com.unsanity.ape 0xc000a954 __ape_internal + 0x90b8 3 com.unsanity.ape 0xc0001328 __ape_agent + 0x40 4 libSystem.B.dylib 0x900246e8 _pthread_body + 0x28 Thread 2: 0 libSystem.B.dylib 0x90018be8 semaphore_timedwait_signal_trap + 0x8 1 libSystem.B.dylib 0x9000e788 _pthread_cond_wait + 0x268 2 ...ple.CoreServices.CarbonCore 0x90295cc4 MPWaitOnQueue + 0xe0 3 com.apple.DesktopServices 0x90918f44 _ZN13TNodeSyncTask12SyncTaskProcEPv + 0x70 4 ...ple.CoreServices.CarbonCore 0x902c6da4 PrivateMPEntryPoint + 0x4c 5 libSystem.B.dylib 0x900246e8 _pthread_body + 0x28 PPC Thread State: srr0: 0x90010844 srr1: 0x0200f030 vrsave: 0x00000000 cr: 0x24822442 xer: 0x00000002 lr: 0xb00d1f84 ctr: 0x90010840 r0: 0xb00d1f84 r1: 0xbfffe7a0 r2: 0x00000000 r3: 0x00000000 r4: 0x0000002f r5: 0x00010000 r6: 0x00000001 r7: 0x00000001 r8: 0x0980ad2c r9: 0xa0001204 r10: 0x00000000 r11: 0xb011eb34 r12: 0x90010840 r13: 0x00000000 r14: 0x00000000 r15: 0x00000000 r16: 0x00000000 r17: 0x00000000 r18: 0x00000000 r19: 0x00000000 r20: 0x00000003 r21: 0x00000000 r22: 0x09e72c00 r23: 0x0054bfe0 r24: 0x00000001 r25: 0x00000000 r26: 0x0a05ed40 r27: 0xbfffea00 r28: 0x00000001 r29: 0x00000000 r30: 0x00010000 r31: 0xb0077d6c Binary Images Description: 0x1000 - 0x127fff com.paradigmasoft.valentina.vstudio 1.0a68 /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Valentina Studio 0x3df000 - 0x46bfff libwx_stc.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/../Frameworks/libwx_stc.dylib 0x6d4000 - 0x6f0fff com.unsanity.windowshadex WindowShade X version 3.1 (3.1) /Users/cbpelto/Library/Application Enhancers/WindowShade X.ape/Contents/MacOS/WindowShade X 0x1496000 - 0x14a1fff admin.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Plugins/admin.dylib 0x162b000 - 0x1983fff libwx_mac.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/../Frameworks/libwx_mac.dylib 0x1f1e000 - 0x1f85fff js.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Plugins/js.dylib 0x1f9d000 - 0x1fcafff otl_odbc.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Plugins/otl_odbc.dylib 0x967b000 - 0x96b3fff sql.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Plugins/sql.dylib 0x9807000 - 0x9898fff vf.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Plugins/vf.dylib 0x9a98000 - 0x9b5ffff vs.dylib /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/Plugins/vs.dylib 0x780c0000 - 0x780c9fff libz.1.1.3.dylib /usr/lib/libz.1.1.3.dylib 0x82920000 - 0x82936fff libiodbc.2.1.6.dylib /usr/lib/libiodbc.2.1.6.dylib 0x8fe00000 - 0x8fe4ffff dyld /usr/lib/dyld 0x90000000 - 0x90122fff libSystem.B.dylib /usr/lib/libSystem.B.dylib 0x90190000 - 0x9023dfff com.apple.CoreFoundation 6.3.5 (299.32) /System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation 0x90280000 - 0x904f9fff com.apple.CoreServices.CarbonCore 10.3.5 /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/CarbonCore.framework/Versions/A/CarbonCore 0x90570000 - 0x905dffff com.apple.framework.IOKit 1.3.5 (???) /System/Library/Frameworks/IOKit.framework/Versions/A/IOKit 0x90610000 - 0x9069afff com.apple.CoreServices.OSServices 3.0.1 /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/OSServices.framework/Versions/A/OSServices 0x90700000 - 0x90700fff com.apple.CoreServices 10.3 (???) /System/Library/Frameworks/CoreServices.framework/Versions/A/CoreServices 0x90720000 - 0x90787fff com.apple.audio.CoreAudio 2.1.2 /System/Library/Frameworks/CoreAudio.framework/Versions/A/CoreAudio 0x907f0000 - 0x907f9fff com.apple.DiskArbitration 2.0.3 /System/Library/PrivateFrameworks/DiskArbitration.framework/Versions/A/DiskArbitration 0x90810000 - 0x90810fff com.apple.ApplicationServices 1.0 (???) /System/Library/Frameworks/ApplicationServices.framework/Versions/A/ApplicationServices 0x90910000 - 0x90983fff com.apple.DesktopServices 1.2.2 /System/Library/PrivateFrameworks/DesktopServicesPriv.framework/Versions/A/DesktopServicesPriv 0x90d00000 - 0x90d1bfff com.apple.SystemConfiguration 1.7.1 (???) /System/Library/Frameworks/SystemConfiguration.framework/Versions/A/SystemConfiguration 0x90d40000 - 0x90d40fff com.apple.Carbon 10.3 (???) /System/Library/Frameworks/Carbon.framework/Versions/A/Carbon 0x910b0000 - 0x91101fff com.apple.bom 1.2.5 (63.2) /System/Library/PrivateFrameworks/Bom.framework/Versions/A/Bom 0x912a0000 - 0x912bdfff com.apple.audio.SoundManager 3.8 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/CarbonSound.framework/Versions/A/CarbonSound 0x912e0000 - 0x912f7fff com.apple.LangAnalysis 1.5.4 /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/LangAnalysis.framework/Versions/A/LangAnalysis 0x91320000 - 0x913defff ColorSync /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ColorSync.framework/Versions/A/ColorSync 0x91460000 - 0x91473fff com.apple.speech.synthesis.framework 3.2 /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/SpeechSynthesis.framework/Versions/A/SpeechSynthesis 0x914a0000 - 0x91509fff com.apple.htmlrendering 1.1.2 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/HTMLRendering.framework/Versions/A/HTMLRendering 0x91560000 - 0x91619fff com.apple.QD 3.4.65 (???) /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/QD.framework/Versions/A/QD 0x91670000 - 0x916a8fff com.apple.AE 1.4 /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/AE.framework/Versions/A/AE 0x916e0000 - 0x91773fff com.apple.print.framework.PrintCore 3.3 /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/PrintCore.framework/Versions/A/PrintCore 0x917e0000 - 0x917f0fff com.apple.speech.recognition.framework 3.3 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/SpeechRecognition.framework/Versions/A/SpeechRecognition 0x91810000 - 0x9182afff com.apple.openscripting 1.2.1 (???) /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/OpenScripting.framework/Versions/A/OpenScripting 0x91850000 - 0x91860fff com.apple.ImageCapture 2.1.5 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/ImageCapture.framework/Versions/A/ImageCapture 0x91890000 - 0x9189cfff com.apple.help 1.0.1 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/Help.framework/Versions/A/Help 0x918c0000 - 0x918cdfff com.apple.CommonPanels 1.2.1 (1.0) /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/CommonPanels.framework/Versions/A/CommonPanels 0x918f0000 - 0x9193efff com.apple.print.framework.Print 3.3 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/Print.framework/Versions/A/Print 0x91990000 - 0x9199bfff com.apple.securityhi 1.2 (90) /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/SecurityHI.framework/Versions/A/SecurityHI 0x919c0000 - 0x91a33fff com.apple.NavigationServices 3.3.2 /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/NavigationServices.framework/Versions/A/NavigationServices 0x91ab0000 - 0x91ac4fff libCGATS.A.dylib /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/CoreGraphics.framework/Versions/A/Resources/libCGATS.A.dylib 0x91ae0000 - 0x91aebfff libCSync.A.dylib /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/CoreGraphics.framework/Versions/A/Resources/libCSync.A.dylib 0x91b10000 - 0x91b2afff libPDFRIP.A.dylib /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/CoreGraphics.framework/Versions/A/Resources/libPDFRIP.A.dylib 0x91b50000 - 0x91b5ffff libPSRIP.A.dylib /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/CoreGraphics.framework/Versions/A/Resources/libPSRIP.A.dylib 0x91b80000 - 0x91b93fff libRIP.A.dylib /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/CoreGraphics.framework/Versions/A/Resources/libRIP.A.dylib 0x91bb0000 - 0x91d45fff com.apple.QuickTime 6.5.1 /System/Library/Frameworks/QuickTime.framework/Versions/A/QuickTime 0x92070000 - 0x92096fff com.apple.FindByContent 1.4 (1.2) /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/FindByContent.framework/Versions/A/FindByContent 0x920c0000 - 0x922a7fff com.apple.security 2.3 (176) /System/Library/Frameworks/Security.framework/Versions/A/Security 0x92430000 - 0x92468fff com.apple.LaunchServices 10.3.5 (98.4) /System/Library/Frameworks/ApplicationServices.framework/Versions/A/Frameworks/LaunchServices.framework/Versions/A/LaunchServices 0x92740000 - 0x92777fff com.apple.CFNetwork 1.2.1 (7) /System/Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/CFNetwork.framework/Versions/A/CFNetwork 0x927d0000 - 0x92b54fff com.apple.HIToolbox 1.3.4 (???) /System/Library/Frameworks/Carbon.framework/Versions/A/Frameworks/HIToolbox.framework/Versions/A/HIToolbox 0xb0000000 - 0xb011cfff com.paradigmasoft.vsdk kernel version 1.9.10b11 (1.9.8) /Users/cbpelto/Desktop/Valentina Studio.app/Contents/MacOS/../Frameworks/VSDK.framework/Versions/A/VSDK 0xc0000000 - 0xc000efff com.unsanity.ape 1.4.1 /Library/Frameworks/ApplicationEnhancer.framework/Versions/A/ApplicationEnhancer -------------- next part -------------- Regards, Chuck From sunshine at public.kherson.ua Sat Oct 23 21:03:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 23 13:04:30 2004 Subject: V Studio Crash Report In-Reply-To: <72B2C9FA-251C-11D9-8C54-00306549C134@pcisys.net> Message-ID: On 10/23/04 8:53 PM, "Chuck Pelto" wrote: > > Greetings, > > Enclosed with this document, please find an Apple OS X report about a > crash that occurred while I was starting to use Valentina Studio. > > I had just hit the CREATE button in the New Database dialog. The > database had been Mode 1 type and set for 64K size. > > Hope this helps.... Hi Chuck Please note, VStudio has own separate mail list. Please subscribe to it and send any Vstudio issues there. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From totoole at rocko.lab.csuchico.edu Sat Oct 23 18:20:48 2004 From: totoole at rocko.lab.csuchico.edu (Thomas) Date: Sat Oct 23 20:20:51 2004 Subject: Valentina Digest, Vol 22, Issue 17 In-Reply-To: <20041023171558.53FE324B8A8@edison.macserve.net> References: <20041023171558.53FE324B8A8@edison.macserve.net> Message-ID: >>> The following Base Object Method is causing trouble in VAPP. >>> >>> if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd >>> ptr_child_rcd,concat(ptr_parent_rcd,'- >>> ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) >> The "On delete" property for both the parent and child ObjectPtrs in >> the lookup table is set to "Cascade". So the ObjectPtrs are deleted >> when the RecIDs they point to are deleted, no? Does this not prevent >> re-use mistakes? Is there a better way to manage these comparisons? > > But you compare RecIds ! > > Why ? > > Explain your idea please > > You think that PARENT always have less RecID then child ? A little hard to explain. Maybe this will help. Fact Table_______________________ Fct_Tbl_RecID Fct_Tbl.Data_Field 10 Thomas 20 Mary 30 Jane 40 Lena 50 Lily Lookup Table_(2 Ptrs to 1 Fact_Table)__ Ptr_Parent_Rcd Ptr_Child_Rcd 10 20 10 30 40 30 Recursive many to many on the one Fact table. Any fact record can be related to any other fact record. Lookup table tells which records are related (linked) to which other records. Thomas is related to Mary and Jane. Lena is related to Jane. Lookup table records have to be unique. We have Thomas related to Mary (Lookup_Table first record 10 to 20). We don't need, and so try to prevent, Mary related to Thomas (20 to 10). Base Object Method won't allow import of new Lookup_Table records. Would like to import 50 to 40, 50 to 10, etc. But Base Object Method thinks these are not unique. Hope that helps, and thanks again. From kim at torchpublishing.com.au Sun Oct 24 13:37:11 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Sat Oct 23 22:37:37 2004 Subject: Primary key Message-ID: All, I've just this weekend started looking at Valentina for use with RB. I'm surprised there is no field of type Integer and am wondering how I go about creating a Primary key/auto-incrementing ID field? Cheers and thanks Kim From sunshine at public.kherson.ua Sun Oct 24 09:27:44 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 24 01:28:06 2004 Subject: Primary key In-Reply-To: Message-ID: On 10/24/04 6:37 AM, "Kim Kohen" wrote: Hi Kim, > I've just this weekend started looking at Valentina for use with RB. I'm > surprised there is no field of type Integer and am wondering how I go about > creating a Primary key/auto-incrementing ID field? 1) Valentina 1.x do have SEVERAL integer types, Just they have other names BYTE - unsigned 1 byte SHORT - signed 2 bytes USHORT - unsigned 2 bytes MEDIUM - signed 3 bytes UMEDIUM - unsigned 3 bytes LONG - signed 4 bytes ULONG - unsigned 4 bytes LLONG - signed 8 bytes ULLONG - unsigned 8 bytes Enough Kim? :-) The INTGER type which you need, is analog LONG. 2) Valentina 1.x do not have PRIMARY KEY, Valentina 2.0 do have. but Valentina developers prefer to use RecID field of table. please read docs. Valentina 1.x have mechanism RecID - OjectPtr fields, Which allow you do the same as Primary Key concept, But more easy and in more effective way. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sun Oct 24 09:39:08 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 24 01:39:42 2004 Subject: Valentina Digest, Vol 22, Issue 17 In-Reply-To: Message-ID: On 10/24/04 4:20 AM, "Thomas" wrote: >>>> The following Base Object Method is causing trouble in VAPP. >>>> >>>> if(ptr_parent_rcd = ptr_child_rcd,'False',if(ptr_parent_rcd >>>> ptr_child_rcd,concat(ptr_parent_rcd,'- >>>> ',ptr_child_rcd),concat(ptr_child_rcd,'-',ptr_parent_rcd))) >>> The "On delete" property for both the parent and child ObjectPtrs in >>> the lookup table is set to "Cascade". So the ObjectPtrs are deleted >>> when the RecIDs they point to are deleted, no? Does this not prevent >>> re-use mistakes? Is there a better way to manage these comparisons? >> >> But you compare RecIds ! >> >> Why ? >> >> Explain your idea please >> >> You think that PARENT always have less RecID then child ? > > A little hard to explain. Maybe this will help. > > Fact Table_______________________ > Fct_Tbl_RecID Fct_Tbl.Data_Field > 10 Thomas > 20 Mary > 30 Jane > 40 Lena > 50 Lily > > Lookup Table_(2 Ptrs to 1 Fact_Table)__ > Ptr_Parent_Rcd Ptr_Child_Rcd > 10 20 > 10 30 > 40 30 > > Recursive many to many on the one Fact table. Any fact record can be > related to any other fact record. Lookup table tells which records are > related (linked) to which other records. This is clear. You try to build self-recursive M : M link on table. > Thomas is related to Mary and Jane. Lena is related to Jane. Clear. > Lookup table records have to be unique. We have Thomas related to Mary > (Lookup_Table first record 10 to 20). We don't need, and so try to > prevent, Mary related to Thomas (20 to 10). Okay I see your task. if( ptr_parent_rcd = ptr_child_rcd,'False', if( ptr_parent_rcd > ptr_child_rcd, concat(ptr_parent_rcd,'-',ptr_child_rcd), concat(ptr_child_rcd,'-',ptr_parent_rcd))) Aha. So you try to build string that contains 2 numbers sorted in ASC order. I see. Correct. I think you get problems because for cases when ptrs are equal you produce string 'False'. This means that you can have 2 and 3 and more time the same string 'False'. And this is why you get NOT UNIQUE error. Right ? > Base Object Method won't allow import of new Lookup_Table records. > Would like to import 50 to 40, 50 to 10, etc. But Base Object Method > thinks these are not unique. Wait. How you are going import values into RecID field ????? This is READ ONLY field. Thomas, I think you need create in the table FactTable own "ID" as Long field Then you will be able import into it. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From softil at onlinehome.de Sun Oct 24 18:49:37 2004 From: softil at onlinehome.de (SoftIl) Date: Sun Oct 24 11:50:49 2004 Subject: [VNET] Question about the dataBase.SqlExecute() Method Message-ID: <000f01c4b9e9$93ca6550$09fee6d9@IPSBASIS> Hi, I've got following problem: If I do this in my application: uint _returnValue = myDataBase.SqlExecute("SELECT COUNT(ohrNummer) FROM tiere WHERE (ohrnummer = '0577105079') AND (ohrnummerCountry LIKE 'DE00' NO_CASE)") I get '0' as result. That's wrong. If I run the same command in Valentina Studio (or in visql) I get as result '1'. That's right. Where is my mistake? Thanks for your help Carsten From sunshine at public.kherson.ua Sun Oct 24 20:28:09 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 24 12:28:32 2004 Subject: [VNET] Question about the dataBase.SqlExecute() Method In-Reply-To: <000f01c4b9e9$93ca6550$09fee6d9@IPSBASIS> Message-ID: On 10/24/04 7:49 PM, "SoftIl" wrote: > Hi, > I've got following problem: > > If I do this in my application: > uint _returnValue = myDataBase.SqlExecute("SELECT COUNT(ohrNummer) FROM > tiere WHERE (ohrnummer = '0577105079') AND (ohrnummerCountry LIKE 'DE00' > NO_CASE)") > I get '0' as result. That's wrong. Wrong is to use SqlExecute() with SELECT You need SqlSelect() > If I run the same command in Valentina Studio (or in visql) I get as result > '1'. That's right. > > Where is my mistake? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From totoole at rocko.lab.csuchico.edu Sun Oct 24 14:02:19 2004 From: totoole at rocko.lab.csuchico.edu (Thomas) Date: Sun Oct 24 16:02:24 2004 Subject: Valentina Digest, Vol 22, Issue 18 In-Reply-To: <20041024063947.5E59D24E077@edison.macserve.net> References: <20041024063947.5E59D24E077@edison.macserve.net> Message-ID: Hi Ruslan, >> Fact Table_______________________ >> Fct_Tbl_RecID Fct_Tbl.Data_Field >> 10 Thomas >> 20 Mary >> 30 Jane >> 40 Lena >> 50 Lily >> >> Lookup Table_(2 Ptrs to 1 Fact_Table)__ >> Ptr_Parent_Rcd Ptr_Child_Rcd >> 10 20 >> 10 30 >> 40 30 > This is clear. > You try to build self-recursive M : M link on table. > >> Lookup table records have to be unique. We have Thomas related to Mary >> (Lookup_Table first record 10 to 20). We don't need, and so try to >> prevent, Mary related to Thomas (20 to 10). > > Okay I see your task. > > if( ptr_parent_rcd = ptr_child_rcd,'False', > if( ptr_parent_rcd > ptr_child_rcd, > concat(ptr_parent_rcd,'-',ptr_child_rcd), > concat(ptr_child_rcd,'-',ptr_parent_rcd))) > > Aha. So you try to build string that contains 2 numbers sorted in ASC > order. > I see. Correct. > > I think you get problems because for cases when ptrs are equal you > produce > string 'False'. > > This means that you can have 2 and 3 and more time the same string > 'False'. > And this is why you get NOT UNIQUE error. > > Right ? No. This is not the problem. >> Base Object Method won't allow import of new Lookup_Table records. >> Would like to import 50 to 40, 50 to 10, etc. But Base Object Method >> thinks these are not unique. > > Wait. How you are going import values into RecID field ????? > This is READ ONLY field. No. This is not the problem either. I import to Lookup_Table ObjectPtrs. Lookup_Table has Base Object Method that is causing problem. It is Lookup_Table that has to have unique records. I put data in fact table and collect Fact_Table RecIDs first. I tested with simplest case: new database, one table, two fields (type = short), one base object method (unique set to true). No importing, no ObjectPtrs, no RecIDs involved. Still, not able to add records with VAPP interface. So I moved 'False' to other end of Base Object Method: if(ptr_parent_rcd>ptr_child_rcd,concat(ptr_parent_rcd,'- ',ptr_child_rcd),if(ptr_child_rcd>ptr_parent_rcd,concat(ptr_child_rcd,'- ',ptr_parent_rcd),?False?)) Still not able to add records. Removed 'False': if(ptr_parent_rcd>ptr_child_rcd,concat(ptr_parent_rcd,'- ',ptr_child_rcd),if(ptr_child_rcd>ptr_parent_rcd,concat(ptr_child_rcd,'- ',ptr_parent_rcd))) Not able to add records after first record. Removed nested If statements: if(num_one>num_two,concat(num_one,'-',num_two),concat(num_two,'- ',num_one)) Now able to add records. But have to delete existing records first. I can send simplest case files if you like, or you can create the same simple test very easy. Thanks again, Thomas From sunshine at public.kherson.ua Mon Oct 25 00:08:25 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 24 16:08:49 2004 Subject: Valentina Digest, Vol 22, Issue 18 In-Reply-To: Message-ID: On 10/25/04 12:02 AM, "Thomas" wrote: >> Okay I see your task. >> >> if( ptr_parent_rcd = ptr_child_rcd,'False', >> if( ptr_parent_rcd > ptr_child_rcd, >> concat(ptr_parent_rcd,'-',ptr_child_rcd), >> concat(ptr_child_rcd,'-',ptr_parent_rcd))) >> >> Aha. So you try to build string that contains 2 numbers sorted in ASC >> order. >> I see. Correct. >> >> I think you get problems because for cases when ptrs are equal you >> produce >> string 'False'. >> >> This means that you can have 2 and 3 and more time the same string >> 'False'. >> And this is why you get NOT UNIQUE error. >> >> Right ? > > No. This is not the problem. I still think it is >>> Base Object Method won't allow import of new Lookup_Table records. >>> Would like to import 50 to 40, 50 to 10, etc. But Base Object Method >>> thinks these are not unique. >> >> Wait. How you are going import values into RecID field ????? >> This is READ ONLY field. > > No. This is not the problem either. > I import to Lookup_Table ObjectPtrs. Lookup_Table has Base Object > Method that is causing problem. It is Lookup_Table that has to have > unique records. I put data in fact table and collect Fact_Table RecIDs > first. > > I tested with simplest case: new database, one table, two fields (type > = short), one base object method (unique set to true). No importing, no > ObjectPtrs, no RecIDs involved. > Still, not able to add records with VAPP interface. > > So I moved 'False' to other end of Base Object Method: > if(ptr_parent_rcd>ptr_child_rcd,concat(ptr_parent_rcd,'- > ',ptr_child_rcd),if(ptr_child_rcd>ptr_parent_rcd,concat(ptr_child_rcd,'- > ',ptr_parent_rcd),?False?)) > Still not able to add records. But this is the samm. No difference. > Removed 'False': > if(ptr_parent_rcd>ptr_child_rcd,concat(ptr_parent_rcd,'- > ',ptr_child_rcd),if(ptr_child_rcd>ptr_parent_rcd,concat(ptr_child_rcd,'- > ',ptr_parent_rcd))) > Not able to add records after first record. Here it needs to look deeply. > Removed nested If statements: > if(num_one>num_two,concat(num_one,'-',num_two),concat(num_two,'- > ',num_one)) > Now able to add records. But have to delete existing records first. This one looks to be correct and simple. > I can send simplest case files if you like, or you can create the same > simple test very easy. I think you have resolve you task -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From softil at onlinehome.de Sun Oct 24 23:24:30 2004 From: softil at onlinehome.de (SoftIl) Date: Sun Oct 24 16:26:29 2004 Subject: [VNET] Question about the dataBase.SqlExecute() Method References: Message-ID: <000001c4ba10$1bf7bc40$29fbe6d9@IPSBASIS> Hi Ruslan, > > uint _returnValue = myDataBase.SqlExecute("SELECT COUNT(ohrNummer) FROM > > tiere WHERE (ohrnummer = '0577105079') AND (ohrnummerCountry LIKE 'DE00' > > NO_CASE)") > Wrong is to use SqlExecute() with SELECT > You need SqlSelect() But SqlSelect returns an cursor object. What I need is only the number of records for them the condition is true. Best regards Carsten From sunshine at public.kherson.ua Mon Oct 25 00:43:26 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 24 16:43:49 2004 Subject: [VNET] Question about the dataBase.SqlExecute() Method In-Reply-To: <000001c4ba10$1bf7bc40$29fbe6d9@IPSBASIS> Message-ID: On 10/25/04 12:24 AM, "SoftIl" wrote: Hi Carsten, >>> uint _returnValue = myDataBase.SqlExecute("SELECT COUNT(ohrNummer) FROM >>> tiere WHERE (ohrnummer = '0577105079') AND (ohrnummerCountry LIKE 'DE00' >>> NO_CASE)") >> Wrong is to use SqlExecute() with SELECT >> You need SqlSelect() > But SqlSelect returns an cursor object. > What I need is only the number of records for them the condition is true. You MUST use it. And later count = curs.Field(1).value -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From softil at onlinehome.de Mon Oct 25 01:29:44 2004 From: softil at onlinehome.de (SoftIl) Date: Sun Oct 24 18:29:56 2004 Subject: [VNET] Question about the dataBase.SqlExecute() Method References: Message-ID: <001001c4ba21$5b013770$7be7e6d9@IPSBASIS> > You MUST use it. > And later > count = curs.Field(1).value Ok. I will do that. Best regards and have a good sleep, Carsten From bibiko at eva.mpg.de Mon Oct 25 09:35:31 2004 From: bibiko at eva.mpg.de (Hans-Joerg Bibiko) Date: Mon Oct 25 02:41:09 2004 Subject: [VJDK] supports Unicode? In-Reply-To: <000001c4ba10$1bf7bc40$29fbe6d9@IPSBASIS> References: <000001c4ba10$1bf7bc40$29fbe6d9@IPSBASIS> Message-ID: <417CACC3.3060509@eva.mpg.de> Hi, maybe this is a very stupid question, but it is a short one. Does VJDK support full range Unicode? Thanks in advance Hans From sunshine at public.kherson.ua Mon Oct 25 12:31:37 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Oct 25 04:32:05 2004 Subject: [VJDK] supports Unicode? In-Reply-To: <417CACC3.3060509@eva.mpg.de> Message-ID: On 10/25/04 10:35 AM, "Hans-Joerg Bibiko" wrote: > Hi, > > maybe this is a very stupid question, but it is a short one. > > Does VJDK support full range Unicode? No, Hans, Full support of unicode will have Valentina 2.0 -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From james.milne at mac.com Mon Oct 25 20:53:39 2004 From: james.milne at mac.com (James Milne) Date: Mon Oct 25 14:54:16 2004 Subject: Full Text Index Search Message-ID: <9099F20D-26BF-11D9-B575-000393DB941A@mac.com> Hi guys & gals, I use V4RB in a product I'm developing. I'm looking to implement a full-text search on a field in one of my tables. This field contains a fairly short comment string. Usually it will have a maximum of around 256 to 512 characters in it. There will likely also be around a maximum of 50,000 rows in this particular table. At the moment, I have an interface which allows the users to type in search terms, ala Google, into a text field. A listbox is populated with records in the table whose comment field matches the search terms entered by the user. The terms are very simple. I don't require support for boolean operands. I simply look for any rows in the table which contain words starting with, or equal to, the search terms. All the search terms must be found in the comment for it to be a valid result. As far as I can tell, Valentina doesn't support Full Text Indexing per-se. You can index on a text field, but the closest you can get to a full text index is to use the LIKE operator on the field. I was wondering whether anyone knew of any ways to implement a full text index using Valentina? Alternatively, whether there was a good way to build ones' own full-text index? This is something I haven't spent a lot of time researching. As far as rolling one's own solution goes: I'd been considering maintaining a table of words extracted from the comments. Each unique word, that passed a 'noise filter' test to remove small words like 'it', 'be', etc., would be added to this table. I would then have another table which linked entries in the word table to entries in the table containing the comments. To perform the search, I would have to find the primary keys of each valid word in the search terms, then find all the record IDs which referred to those words. My result set would be the intersection of the sets of comment record IDs for each word. Whenever one changed the value of the comment, one would have to drop all the rows in the word->comment map table referring to the row containing the comment, then rebuild the word->comment mappings. This is fair enough, as the comment fields do not change frequently in this application. I was just wondering whether anyone else has had to solve a similar problem using Valentina, and whether this is an efficient approach. -- Kind Regards, James Milne From ernestogiannotta at tiscalinet.it Mon Oct 25 22:17:00 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Mon Oct 25 15:17:30 2004 Subject: Full Text Index Search In-Reply-To: <9099F20D-26BF-11D9-B575-000393DB941A@mac.com> References: <9099F20D-26BF-11D9-B575-000393DB941A@mac.com> Message-ID: Hello James, On 25 ott 2004, at 21:53, James Milne wrote: > At the moment, I have an interface which allows the users to type in > search terms, ala Google, into a text field. A listbox is populated > with records in the table whose comment field matches the search terms > entered by the user. > > The terms are very simple. I don't require support for boolean > operands. I simply look for any rows in the table which contain words > starting with, or equal to, the search terms. All the search terms > must be found in the comment for it to be a valid result. > > As far as I can tell, Valentina doesn't support Full Text Indexing > per-se. You can index on a text field, but the closest you can get to > a full text index is to use the LIKE operator on the field. > why don't you try the IndexByWords option of Valentina for this? then you split search field contents in a group of And queries e.g. the search field contains "nice girls" the query will be like "Where (Left(Comment,4) = 'nice' and Left(Comment,5) = 'girls')" you can also filter out unrelevant words like "the", "in" etc. HTH Cool Runnings, Erne. |er| musical box |ne| a media store From valentina-list at vermontsoftworks.com Mon Oct 25 16:36:45 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Mon Oct 25 15:37:18 2004 Subject: Full Text Index Search In-Reply-To: <9099F20D-26BF-11D9-B575-000393DB941A@mac.com> References: <9099F20D-26BF-11D9-B575-000393DB941A@mac.com> Message-ID: <967B99F0-26C5-11D9-852D-000393DC4766@vermontsoftworks.com> Hi, James -- If I'm understanding your need correctly, Valentina already supports exactly what you're looking for. I have an internal application which archives old e-mail messages, for example. I've simply defined the message-body field as Indexed set the IndexByWords flag. I simply split up the user's search field up into words and set up the SQL inquiry with an "OR" for each word. RB fragment: sqlWhereClause = "WHERE messageBody IS '" + userWord(1) + "' OR messageBody IS '" + userWord(2) + "'" etc. Actually, what I do is a little fancier: I build a dictionary out of the words in the user's search field (thus avoiding duplicates), and make sure that the *last* term that the user types matches against word-beginnings (using "Left" in the SQL and matching only the number of characters in the last term). Depending on the size of your database, this can actually be done real-time, including the refreshing of the listbox. Or, if that's too slow, it can be initiated from a timer firing every 0.5 seconds or so. Adjust to taste. Good luck! -- Erik Mueller-Harder On Oct 25, 2004, at 15:53, James Milne wrote: > Hi guys & gals, > > I use V4RB in a product I'm developing. I'm looking to implement a > full-text search on a field in one of my tables. > > This field contains a fairly short comment string. Usually it will > have a maximum of around 256 to 512 characters in it. There will > likely also be around a maximum of 50,000 rows in this particular > table. > > At the moment, I have an interface which allows the users to type in > search terms, ala Google, into a text field. A listbox is populated > with records in the table whose comment field matches the search terms > entered by the user. > > The terms are very simple. I don't require support for boolean > operands. I simply look for any rows in the table which contain words > starting with, or equal to, the search terms. All the search terms > must be found in the comment for it to be a valid result. > > As far as I can tell, Valentina doesn't support Full Text Indexing > per-se. You can index on a text field, but the closest you can get to > a full text index is to use the LIKE operator on the field. > > I was wondering whether anyone knew of any ways to implement a full > text index using Valentina? Alternatively, whether there was a good > way to build ones' own full-text index? This is something I haven't > spent a lot of time researching. > > As far as rolling one's own solution goes: I'd been considering > maintaining a table of words extracted from the comments. Each unique > word, that passed a 'noise filter' test to remove small words like > 'it', 'be', etc., would be added to this table. I would then have > another table which linked entries in the word table to entries in the > table containing the comments. To perform the search, I would have to > find the primary keys of each valid word in the search terms, then > find all the record IDs which referred to those words. My result set > would be the intersection of the sets of comment record IDs for each > word. > > Whenever one changed the value of the comment, one would have to drop > all the rows in the word->comment map table referring to the row > containing the comment, then rebuild the word->comment mappings. This > is fair enough, as the comment fields do not change frequently in this > application. > > I was just wondering whether anyone else has had to solve a similar > problem using Valentina, and whether this is an efficient approach. From james.milne at mac.com Mon Oct 25 22:02:53 2004 From: james.milne at mac.com (James Milne) Date: Mon Oct 25 16:03:18 2004 Subject: Full Text Index Search In-Reply-To: References: <9099F20D-26BF-11D9-B575-000393DB941A@mac.com> Message-ID: <3C9F6B45-26C9-11D9-B575-000393DB941A@mac.com> On 25 Oct, 2004, at 21:17, erne wrote: > Hello James, > > why don't you try the IndexByWords option of Valentina for this? > then you split search field contents in a group of And queries > > e.g. > the search field contains "nice girls" > the query will be like > "Where (Left(Comment,4) = 'nice' and Left(Comment,5) = 'girls')" > > you can also filter out unrelevant words like "the", "in" etc. On 25 Oct, 2004, at 21:36, Erik Mueller-Harder wrote: > Hi, James -- > > If I'm understanding your need correctly, Valentina already supports > exactly what you're looking for. > > I have an internal application which archives old e-mail messages, for > example. I've simply defined the message-body field as Indexed set > the IndexByWords flag. Hi Erne & Erik, Thanks guys. I think I'll investigate the IndexByWords property further. -- Kind Regards, James Milne From lists at pedro.Net.au Tue Oct 26 12:45:04 2004 From: lists at pedro.Net.au (Pedro fp) Date: Mon Oct 25 21:45:21 2004 Subject: [V4RB] difference between string & varchar Message-ID: <0A6D244C-26F9-11D9-933D-000D9366C144@pedro.Net.au> G'day Folks This may be a bit of a daft question & even been asked before, it also probably applies to more than just V4RB, so please forgive & bear with me. I have never quite figured what, for practical purposes, the difference between a string field & a varchar field is. Even from what I read in the V4Rb docs they appear to be pretty much the same except in name. So could someone please explain the difference for me & offer some guide as to when I should choose one over the other? Cheers, Pedro :-) Web: PGP Key ID: 387CD96F Instant messaging... AIM: bandidoOfOz ICQ: 27671678 Jabber: pedrofp MSN: mail@pedro.net.au Yahoo: pedro_fp ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~ "She [Margaret Mead] said: 'We are all immigrants into a new time.' This inevitably requires that the patterns we have inherited from the past must be tested against a new set of realities that nobody sees clearly. Immigrants into a new culture can learn from current inhabitants. Immigrants into a new time face a more difficult challenge." Robert Theobald in 'The Healing Century' From peter.de.berdt at pandora.be Tue Oct 26 08:53:01 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Tue Oct 26 01:53:24 2004 Subject: [V4RB] difference between string & varchar In-Reply-To: <0A6D244C-26F9-11D9-933D-000D9366C144@pedro.Net.au> References: <0A6D244C-26F9-11D9-933D-000D9366C144@pedro.Net.au> Message-ID: On 26 Oct 2004, at 04:45, Pedro fp wrote: > G'day Folks > > This may be a bit of a daft question & even been asked before, it also > probably applies to more than just V4RB, so please forgive & bear with > me. > > I have never quite figured what, for practical purposes, the > difference between a string field & a varchar field is. Even from what > I read in the V4Rb docs they appear to be pretty much the same except > in name. So could someone please explain the difference for me & offer > some guide as to when I should choose one over the other? > A VString field takes up the space of the max length you define, no matter how many characters are actually used. Varchar only uses the space it needs (variable character length), hence the name. Ruslan will probably go more technical on this, but this is the easy explanation ;-) Best regards Peter De Berdt From lists at pedro.Net.au Tue Oct 26 19:05:51 2004 From: lists at pedro.Net.au (Pedro fp) Date: Tue Oct 26 04:06:00 2004 Subject: [V4RB] difference between string & varchar In-Reply-To: References: <0A6D244C-26F9-11D9-933D-000D9366C144@pedro.Net.au> Message-ID: <3C303360-272E-11D9-933D-000D9366C144@pedro.Net.au> On 26/10/2004, at 4:53 PM, Peter De Berdt wrote: >> I have never quite figured what, for practical purposes, the >> difference between a string field & a varchar field is. Even from >> what I read in the V4Rb docs they appear to be pretty much the same >> except in name. So could someone please explain the difference for me >> & offer some guide as to when I should choose one over the other? > > A VString field takes up the space of the max length you define, no > matter how many characters are actually used. Varchar only uses the > space it needs (variable character length), hence the name. Ruslan > will probably go more technical on this, but this is the easy > explanation ;-) That is as I thought but it still leaves me unsure about why I'd choose one over the other. I take it VVarChar would result in a smaller data file, particularly where the string data is highly variable in length with a lot of it being a lot smaller than the maximum length, is a VString faster? Cheers, Pedro :-) Web: PGP Key ID: 387CD96F Instant messaging... AIM: bandidoOfOz ICQ: 27671678 Jabber: pedrofp MSN: mail@pedro.net.au Yahoo: pedro_fp ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~ "I think we need more of the wordless in our lives. We need more stillness, more of a sense of wonder, a feeling for the mystery of life. We need more love, more silence, more deep listening, more deep giving." Ben Okri in 'A Way of Being Free', 1997, p90 From Kurt.Klamp at abdnb.bayern.de Tue Oct 26 14:08:17 2004 From: Kurt.Klamp at abdnb.bayern.de (Klamp, Kurt) Date: Tue Oct 26 07:08:45 2004 Subject: [SQL/VAPP] Problems with field recid Message-ID: <21674C363854D511B526006008F6B9CD01312A6D@ABDN-6.dir.abdnb.bayern.de> Hi Ruslan, since migration to OSX i have trouble with some sql statements which worked before. Don't know wether it is a VAPP issue (always 1.10) or an Mac OS one. This sql's lead to an error: 'Not uniqe field name' ??? SELECT Investments.**, Investments.Recid,count(InvBuchungen.InvPtr) FROM Investments,InvBuchungen GROUP BY Investments.InvDepotNr SELECT Investments.recid, Investments.InvName, Investments.InvInhaber, count(InvBuchungen.InvPtr) FROM Investments,InvBuchungen GROUP BY InvBuchungen.InvPtr This similar statement works: SELECT Investments.InvName, Investments.InvInhaber, count(InvBuchungen.InvPtr) FROM Investments,InvBuchungen GROUP BY InvBuchungen.InvPtr If i replace the GROUP BY part with 'Investments.recid' which must leads to the same result i get the error again. The 2 tables (Investments := One Table ; InvBuchungen := Many Table) are linked with objPtr where Investments.recid=InvBuchungen.InvPtr. I want to get some basic info from One Table together with the number of related records in Many Table. In general i assume that there are problems with the field 'recid' eiter as part of the selected fields or as direct or indirect part of the group by statement. Kurt ########################################### This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.F-Secure.com/ From milan at mook.co.uk Tue Oct 26 15:18:29 2004 From: milan at mook.co.uk (Milan) Date: Tue Oct 26 09:19:14 2004 Subject: Valentina on Mac OSX In-Reply-To: <21674C363854D511B526006008F6B9CD01312A6D@ABDN-6.dir.abdnb.bayern.de> Message-ID: <200410261518.PAA02886@moo.mook.co.uk> We are developing a cdrom using the valentina xtra which will need to run on PC, Mac OS9 and OSX. As per normal for Director-authored content, we have created a 'stub' movie which sets up the projector environment before calling in the main Director movie from a sub-directory on the cdrom. This works perfectly on PC, but we've noticed an issue with an OSX projector crashing as soon as it launches. This *only* happens if the classic xtra is also included on the cd - without this, the OSX projector runs and performs as expected. Are there any known issues surrounding the inclusion of both xtras on a cdrom? If so, how do we get round this? From mo.ritz at gmx.de Tue Oct 26 17:10:35 2004 From: mo.ritz at gmx.de (mo) Date: Tue Oct 26 10:10:54 2004 Subject: Commit cursor changes to DB Message-ID: <3033012E-2761-11D9-B216-000A95BC6F40@gmx.de> Hi, I recently started to sort out Valentina & Diretcor, so this is a very basic question. What is the general modus operandi to store cursor changes in the actual db? What I do right now is: saveConfiguration = function(o){ var RecID=o.RecID; delete o.RecID; trace("Model.saveConfiguration "); var c= _global.gDBManager.getDBCursor( "SELECT * FROM configurations WHERE RecID="+RecID); var r=c.updateAllRecords(o.toPList()); _global.gDBManager.DBInstance.flush(); } but the data is not stored in the database. If I trace the cursor after the updateAllRecords command, the values are changed, however only in the cursor as it seems. Could someone be so kind and shed some light on this? Much appreciated, cheers, mo . . Moritz Stefaner 0179 - 525 21 26 http://der-mo.net . . From cbpelto at pcisys.net Tue Oct 26 09:50:01 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Tue Oct 26 10:50:19 2004 Subject: Automatic Data Entry Message-ID: Greetings, Looking over the documentation and the archived e-mail, I get the impression that there is NO automatic population of fields with data in Valentina 1.0. This includes no automatic incrementation of a serial number field. Is this correct? I also note that Valentina 2.0 is supposed to have the automatic incrementing of serial numbers, according to one e-mail in the archive. Will V2.0 have automatic data input where a field is supposed to be a constant value? When is V2.0 supposed to be available in final form? Regards, Chuck From lists at pedro.Net.au Wed Oct 27 02:04:24 2004 From: lists at pedro.Net.au (Pedro fp) Date: Tue Oct 26 11:04:31 2004 Subject: [V4RB] SQL again ... arrrgh! Message-ID: G'day Folks I'm trying to construct a set of canned searched in my current project & while most are working fine I'm stumbling on one pair. They are to find records where one VString field is empty and not empty respectively. As a precaution I've tried to construct my searches to cover both possibilities, that an empty field is an empty string and that an empty field is null. The SQL strings I've tried so far are (in pairs for the empty field then the not empty field) ... SELECT ... WHERE ( NOT extention LIKE '' and NOT extention = NULL ) ORDER BY name ASC SELECT ... WHERE ( extention LIKE '' or extention = NULL ) ORDER BY name ASC SELECT ... WHERE ( NOT extention = '' and NOT extention = NULL ) ORDER BY name ASC SELECT ... WHERE ( extention = '' or extention = NULL ) ORDER BY name ASC SELECT ... WHERE ( NOT extention = '' and NOT extention IS null ) ORDER BY name ASC SELECT ... WHERE ( extention = '' or extention IS null ) ORDER BY name ASC SELECT ... WHERE NOT ( extention = '' or extention IS null ) ORDER BY name ASC SELECT ... WHERE ( extention = '' or extention IS null ) ORDER BY name ASC Any tips on where I'm going wrong? If possible I'd prefer to negate outside the parentheses as in my last attempt but it's not a big issue if that's not possible. Cheers, Pedro :-) Web: PGP Key ID: 387CD96F Instant messaging... AIM: bandidoOfOz ICQ: 27671678 Jabber: pedrofp MSN: mail@pedro.net.au Yahoo: pedro_fp ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ "How would we think if we saw the twenty-first century as 'The Healing Century' and left the economic emphasis behind in the twentieth?" Robert Theobald, 1998 From IvanSmahin at public.kherson.ua Tue Oct 26 20:08:17 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Tue Oct 26 12:08:16 2004 Subject: [V4RB] SQL again ... arrrgh! In-Reply-To: References: Message-ID: <82679920.20041026200817@public.kherson.ua> Hello Pedro, Tuesday, October 26, 2004, 7:04:24 PM, you wrote: Pf> G'day Folks Pf> I'm trying to construct a set of canned searched in my current project Pf> & while most are working fine I'm stumbling on one pair. They are to Pf> find records where one VString field is empty and not empty Pf> respectively. As a precaution I've tried to construct my searches to Pf> cover both possibilities, that an empty field is an empty string and Pf> that an empty field is null. Pf> The SQL strings I've tried so far are (in pairs for the empty field Pf> then the not empty field) ... Pf> SELECT ... WHERE ( NOT extention LIKE '' and NOT extention = NULL ) Pf> ORDER BY name ASC Pf> SELECT ... WHERE ( extention LIKE '' or extention = NULL ) ORDER BY Pf> name ASC Pf> SELECT ... WHERE ( NOT extention = '' and NOT extention = NULL ) ORDER Pf> BY name ASC Pf> SELECT ... WHERE ( extention = '' or extention = NULL ) ORDER BY name Pf> ASC Pf> SELECT ... WHERE ( NOT extention = '' and NOT extention IS null ) ORDER Pf> BY name ASC Pf> SELECT ... WHERE ( extention = '' or extention IS null ) ORDER BY name Pf> ASC Pf> SELECT ... WHERE NOT ( extention = '' or extention IS null ) ORDER BY Pf> name ASC Pf> SELECT ... WHERE ( extention = '' or extention IS null ) ORDER BY name Pf> ASC Pf> Any tips on where I'm going wrong? If possible I'd prefer to negate Pf> outside the parentheses as in my last attempt but it's not a big issue Pf> if that's not possible. Pf> Cheers, Pedro :-) Pf> Web: PGP Key ID: 387CD96F Pf> Instant messaging... AIM: bandidoOfOz ICQ: 27671678 Jabber: pedrofp Pf> MSN: mail@pedro.net.au Yahoo: pedro_fp Pf> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Pf> "How would we think if we saw the twenty-first century as Pf> 'The Healing Century' and left the economic emphasis behind Pf> in the twentieth?" Pf> Robert Theobald, 1998 Pf> _______________________________________________ Pf> Valentina mailing list Pf> Valentina@lists.macserve.net Pf> http://lists.macserve.net/mailman/listinfo/valentina I'm not familiar with 1.0 but anyway you should not think that it is possible to compare NULL-value with any another value. Strictly speaking the result of such comparison is undefined. So you will get a zero result. Some DBMS allow to set such mode, but it's allowed for "equal" only. (I mean - mode when NULL is equal to NULL and not-NULL is NOT equal to NULL). But I believe this is not good thing. You, or another person might be confused with such query. As far as I know valentina does not have such mode. -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From frank-list3 at mindstarprods.com Tue Oct 26 11:12:55 2004 From: frank-list3 at mindstarprods.com (Frank Schima) Date: Tue Oct 26 12:13:22 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <472715B5-2772-11D9-8DAE-0003939246BC@mindstarprods.com> Hi Chuck, On Oct 26, 2004, at 9:50 AM, Chuck Pelto wrote: > Looking over the documentation and the archived e-mail, I get the > impression that there is NO automatic population of fields with data > in Valentina 1.0. This includes no automatic incrementation of a > serial number field. > > Is this correct? Correct. > I also note that Valentina 2.0 is supposed to have the automatic > incrementing of serial numbers, according to one e-mail in the > archive. > > Will V2.0 have automatic data input where a field is supposed to be a > constant value? Ruslan stated that default values will be in V2 as I recall. But I don't understand how that is not what you want? Also, why do you need to fill a field with the same value every record? That is a poor design. Constant values should not be in a field. That wastes space and slows down your queries. In any event, a field Method would be a solution for this situation in 1.0. Just have a field Method that always returns 2, or pi, or whatever. That has the advantage of not wasting disk space. What would be useful, IMHO, is a default function. So, for instance, you could automatically populate a DateTime field with the current DateTime so you know when the record was added. > When is V2.0 supposed to be available in final form? When it's done! :^) Best regards, Frank From IvanSmahin at public.kherson.ua Tue Oct 26 20:19:57 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Tue Oct 26 12:19:54 2004 Subject: [V4RB] SQL again ... arrrgh! In-Reply-To: References: Message-ID: <1988663353.20041026201957@public.kherson.ua> Hello Pedro, Tuesday, October 26, 2004, 7:04:24 PM, you wrote: Pf> G'day Folks Pf> I'm trying to construct a set of canned searched in my current project Pf> & while most are working fine I'm stumbling on one pair. They are to Pf> find records where one VString field is empty and not empty Pf> respectively. As a precaution I've tried to construct my searches to Pf> cover both possibilities, that an empty field is an empty string and Pf> that an empty field is null. Pf> The SQL strings I've tried so far are (in pairs for the empty field Pf> then the not empty field) ... Pf> SELECT ... WHERE ( NOT extention LIKE '' and NOT extention = NULL ) Pf> ORDER BY name ASC Pf> SELECT ... WHERE ( extention LIKE '' or extention = NULL ) ORDER BY Pf> name ASC Pf> SELECT ... WHERE ( NOT extention = '' and NOT extention = NULL ) ORDER Pf> BY name ASC Pf> SELECT ... WHERE ( extention = '' or extention = NULL ) ORDER BY name Pf> ASC Pf> SELECT ... WHERE ( NOT extention = '' and NOT extention IS null ) ORDER Pf> BY name ASC Pf> SELECT ... WHERE ( extention = '' or extention IS null ) ORDER BY name Pf> ASC Pf> SELECT ... WHERE NOT ( extention = '' or extention IS null ) ORDER BY Pf> name ASC Pf> SELECT ... WHERE ( extention = '' or extention IS null ) ORDER BY name Pf> ASC Pf> Any tips on where I'm going wrong? If possible I'd prefer to negate Pf> outside the parentheses as in my last attempt but it's not a big issue Pf> if that's not possible. Pf> Cheers, Pedro :-) Pf> Web: PGP Key ID: 387CD96F Pf> Instant messaging... AIM: bandidoOfOz ICQ: 27671678 Jabber: pedrofp Pf> MSN: mail@pedro.net.au Yahoo: pedro_fp Pf> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Pf> "How would we think if we saw the twenty-first century as Pf> 'The Healing Century' and left the economic emphasis behind Pf> in the twentieth?" Pf> Robert Theobald, 1998 Pf> _______________________________________________ Pf> Valentina mailing list Pf> Valentina@lists.macserve.net Pf> http://lists.macserve.net/mailman/listinfo/valentina Another tip: You say " the string field is empty" - so I think that NULL value is the best description. "Empty string" approach is possible but not so clear. I would say that empty string is value with zero length but it is DEFINED value - you see? -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From valentina-list at vermontsoftworks.com Tue Oct 26 14:02:35 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Oct 26 13:02:47 2004 Subject: [V4RB] difference between string & varchar In-Reply-To: <3C303360-272E-11D9-933D-000D9366C144@pedro.Net.au> References: <0A6D244C-26F9-11D9-933D-000D9366C144@pedro.Net.au> <3C303360-272E-11D9-933D-000D9366C144@pedro.Net.au> Message-ID: <37088614-2779-11D9-B069-000393DC4766@vermontsoftworks.com> On Oct 26, 2004, at 05:05, Pedro fp wrote: > That is as I thought but it still leaves me unsure about why I'd > choose one over the other. I take it VVarChar would result in a > smaller data file, particularly where the string data is highly > variable in length with a lot of it being a lot smaller than the > maximum length, is a VString faster? Speed: My suspicion would be that Strings are faster than VarChars when they're not being used wastefully. [Though after looking through my Valentina message archive, I see that Ruslan recently said, "That was surprise for mew few years ago, but VarChar fields work much faster of String." So I guess my suspicion was wrong!] Size: VarChars will pretty much always end up saving you space when you're not sure of the size of the data you need to store, though there is 8 bytes of overhead per VarChar field per record. FYI, there is no point ever in defining a VarChar's maximum size as less than 504 bytes -- the most efficient maximum sizes can be represented as (x / 2) - 8, where x is a multiple of 1024; so 504, 1016, 1528, etc. Reliability: Ruslan acknowledges that VarChars are slightly more likely to become corrupted than Strings "in hard cases" because of their complex structure. In the real world, I've ended up leaving Strings for VarChars in most situations. Exceptions are few: things like 2-letter U.S. state abbreviations and internal codes or keys whose length I'm certain of. HTH, -- Erik From valentina-list at vermontsoftworks.com Tue Oct 26 14:10:09 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Oct 26 13:10:21 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> On Oct 26, 2004, at 11:50, Chuck Pelto wrote: > Looking over the documentation and the archived e-mail, I get the > impression that there is NO automatic population of fields with data > in Valentina 1.0. This includes no automatic incrementation of a > serial number field. > > Is this correct? I believe that is correct, yes, though the RecID field in every table does contain a unique serial number that seems in fact to be automatically incremented. If records are deleted, however, there is no guarantee that previous RecID values won't be re-used. > I also note that Valentina 2.0 is supposed to have the automatic > incrementing of serial numbers, according to one e-mail in the > archive. > > Will V2.0 have automatic data input where a field is supposed to be a > constant value? Just curious: why would you want to store a constant value? > When is V2.0 supposed to be available in final form? Alpha releases seem to be progressing nicely, but I think Ruslan is not making release dates, if he knows them, public. HTH, -- Erik From nfriesen at dyton.com Tue Oct 26 13:21:25 2004 From: nfriesen at dyton.com (Nathan Friesen) Date: Tue Oct 26 13:21:43 2004 Subject: Valentina on Mac OSX In-Reply-To: <200410261518.PAA02886@moo.mook.co.uk> References: <200410261518.PAA02886@moo.mook.co.uk> Message-ID: I've had this same problem. The only way I've found to get around it is to have the two stub movies reside in separate folders with their own Xtras folders, so they don't see each other. I then have the stub movie look up one level to call the main Director movie. Finally, each time I burn a CD I make an alias of each of the stub movies at the top level of the CD so the user doesn't have to dig down into a folder to start the application. Not the most elegant solution, but it does seem to work. If anyone else has a better suggestion, I'd be glad to hear it. Nathan On Oct 26, 2004, at 9:18 AM, Milan wrote: > We are developing a cdrom using the valentina xtra which will need to > run on > PC, Mac OS9 and OSX. > > As per normal for Director-authored content, we have created a 'stub' > movie > which sets up the projector environment before calling in the main > Director > movie from a sub-directory on the cdrom. > > This works perfectly on PC, but we've noticed an issue with an OSX > projector > crashing as soon as it launches. This *only* happens if the classic > xtra is > also included on the cd - without this, the OSX projector runs and > performs > as expected. > > Are there any known issues surrounding the inclusion of both xtras on a > cdrom? If so, how do we get round this? > From mo.ritz at gmx.de Tue Oct 26 22:15:52 2004 From: mo.ritz at gmx.de (mo) Date: Tue Oct 26 15:16:14 2004 Subject: [V4MD] Commit cursor changes to DB In-Reply-To: <3033012E-2761-11D9-B216-000A95BC6F40@gmx.de> References: <3033012E-2761-11D9-B216-000A95BC6F40@gmx.de> Message-ID: Hi, update on this: I was so stupid and forgot to create the cursor with the additional flag to set it to read and write. Now it works... partially ;) The next problem I encounter is that when I call the saveConfiguration function for the second time, it won't work. If I trace the cursor after the update, I see the changes. If I reselect the same query after that, all field values are aas they used to be before the update. Any Ideas? Valentina gives no error message, btw., neither when creating the cursor nor when updating. It would be great if someone could help me out here.. Cheers! mo . . Moritz Stefaner 0179 - 525 21 26 http://der-mo.net . . On 26.10.2004, at 17:10, mo wrote: > Hi, > > I recently started to sort out Valentina & Diretcor, so this is a > very basic question. What is the general modus operandi to store > cursor changes in the actual db? What I do right now is: > > saveConfiguration = function(o){ > var RecID=o.RecID; delete o.RecID; > trace("Model.saveConfiguration "); > var c= _global.gDBManager.getDBCursor( "SELECT * FROM configurations > WHERE RecID="+RecID); > var r=c.updateAllRecords(o.toPList()); > _global.gDBManager.DBInstance.flush(); > } > > but the data is not stored in the database. If I trace the cursor > after the updateAllRecords command, the values are changed, however > only in the cursor as it seems. > > Could someone be so kind and shed some light on this? > Much appreciated, > > cheers, > mo . . Moritz Stefaner 0179 - 525 21 26 http://der-mo.net . . > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Wed Oct 27 09:23:16 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 01:23:32 2004 Subject: [V4MD] Commit cursor changes to DB In-Reply-To: Message-ID: On 10/26/04 11:15 PM, "mo" wrote: > Hi, > > update on this: I was so stupid and forgot to create the cursor with > the additional flag to set it to read and write. > > Now it works... partially ;) The next problem I encounter is that when > I call the saveConfiguration function for the second time, it won't > work. If I trace the cursor after the update, I see the changes. If I > reselect the same query after that, all field values are aas they used > to be before the update. > > Any Ideas? > > Valentina gives no error message, btw., neither when creating the > cursor nor when updating. > > It would be great if someone could help me out here.. Why you use UpdateAll() ? As far as I see you SELECT exactly one record, Because you do WHERE RecID="+RecID So you should use just curs.updateRecord() > Cheers! > mo > > . . > > Moritz Stefaner > 0179 - 525 21 26 > http://der-mo.net > > . . > On 26.10.2004, at 17:10, mo wrote: > >> Hi, >> >> I recently started to sort out Valentina & Diretcor, so this is a >> very basic question. What is the general modus operandi to store >> cursor changes in the actual db? What I do right now is: >> >> saveConfiguration = function(o){ >> var RecID=o.RecID; delete o.RecID; >> trace("Model.saveConfiguration "); >> var c= _global.gDBManager.getDBCursor( "SELECT * FROM configurations >> WHERE RecID="+RecID); >> var r=c.updateAllRecords(o.toPList()); >> _global.gDBManager.DBInstance.flush(); >> } >> >> but the data is not stored in the database. If I trace the cursor >> after the updateAllRecords command, the values are changed, however >> only in the cursor as it seems. >> >> Could someone be so kind and shed some light on this? >> Much appreciated, -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed Oct 27 09:24:10 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 01:24:26 2004 Subject: Full Text Index Search In-Reply-To: <20041026110951.D9725507AFE@lists.realsoftware.com> Message-ID: On 10/26/04 2:09 PM, "realbasic-nug-request@lists.realsoftware.com" wrote: >>> I use V4RB in a product I'm developing. I'm looking to implement a >>> full-text search on a field in one of my tables. >> >> Now I'm confused - I thought from stuff that Ruslan had said here >> before that Valentina has full text searching and also fragment >> searching (as well as making the coffee whilst you're out taking a >> stroll on the lake :-). >> >> I'd do some speed trials with LIKE before making any heavier decisions. >> > LIKE in Valentina is extremely fast (even though it's an unindexed > search). I did some testing a while ago and unindexed searches in > Valentina are faster than an indexed search (FULLTEXT) in other DBMS. Hello guys! I was absent one day, so sorry for delay. Let me give answers. > From: Andy Dent > Date: Tue, 26 Oct 2004 13:18:03 +0800 >> I use V4RB in a product I'm developing. I'm looking to implement a >> full-text search on a field in one of my tables. > > Now I'm confused - I thought from stuff that Ruslan had said here > before that Valentina has full text searching and also fragment > searching (as well as making the coffee whilst you're out taking a > stroll on the lake :-). Andy, I think James understand under term FULL TEXT indexing, the feature similar to MS SQL Server or may be some special search text engines. Such feature can give you ability for example do search as word1 NEAR TO word1 in 2-3 steps. Valentina cannot do this. Valentina do have special INDEX BY WORDS for String, VarChar and TEXT fields. Up to now most developers are quite satisfied wit this. 2) about speed of LIKE in in Valentina. Yes James, it is much faster than in mySQL or others. Because of internal Valentina structure. Let me to remind: If you have in mySQL table with 30 VarChar fields, then one record can be in average say 30 * (15-20) = 600 bytes or more. To scan such table for LIKE DBMS need load into RAM all this. Valentina instead use for each column one file. So we need scan only 20 bytes * N of records. For this example Valentina is expected to be 30 times faster. 3) in Valentina 2.0 we will use SOMETIMES the index file to be even more faster. Example. Assume you have table in million of records. Assume that you have only 10,000 different words in column. So index will contains only 10,000 columns. Then using INDEX SCAN instead of COLUMN SCAN we get speed even 100 times faster. 4) Next, in Valentina 2.0 (you can see this in beta) we have introduce new special STRING SEACH functions in V4RB API. Similar as you have on OS X or emails app... 5) Next, in 2.0 we have implement that not only left() function use index, But also right() and substr(). So James, I think you can look on Valentina 2.0 beta. I believe you will be impressed :-) -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed Oct 27 09:57:35 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 01:57:48 2004 Subject: [SQL/VAPP] Problems with field recid In-Reply-To: <21674C363854D511B526006008F6B9CD01312A6D@ABDN-6.dir.abdnb.bayern.de> Message-ID: On 10/26/04 3:08 PM, "Klamp, Kurt" wrote: Hi Kurt, > since migration to OSX i have trouble with some sql statements which worked > before. Don't know wether it is a VAPP issue (always 1.10) or an Mac OS > one. > > This sql's lead to an error: > 'Not uniqe field name' ??? > > SELECT Investments.**, Investments.Recid,count(InvBuchungen.InvPtr) FROM > Investments,InvBuchungen GROUP BY Investments.InvDepotNr > SELECT Investments.recid, Investments.InvName, Investments.InvInhaber, > count(InvBuchungen.InvPtr) FROM Investments,InvBuchungen GROUP BY > InvBuchungen.InvPtr > > This similar statement works: > > SELECT Investments.InvName, Investments.InvInhaber, > count(InvBuchungen.InvPtr) FROM Investments,InvBuchungen GROUP BY > InvBuchungen.InvPtr > > If i replace the GROUP BY part with 'Investments.recid' which must leads to > the same result i get the error again. The 2 tables (Investments := One > Table ; InvBuchungen := Many Table) are linked with objPtr where > Investments.recid=InvBuchungen.InvPtr. I want to get some basic info from > One Table together with the number of related records in Many Table. > > In general i assume that there are problems with the field 'recid' eiter as > part of the selected fields or as direct or indirect part of the group by > statement. I believe that Valentina 1.x still is not able to sort on RecID field. (workaround is to use ULONG method = "RecID") GROUP BY use sorting, so I think this is the same problem. Yes you can use ObjectPtr field. Is this solution for you ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From Peter.Lindgren at fdt.se Wed Oct 27 08:57:38 2004 From: Peter.Lindgren at fdt.se (Peter Lindgren) Date: Wed Oct 27 01:58:29 2004 Subject: [V4RB] SQL again ... arrrgh! Message-ID: Hej! valentina-bounces@lists.macserve.net <> skrev Tuesday, October 26, 2004 7:20 PM: > Tuesday, October 26, 2004, 7:04:24 PM, you wrote: > Pf> I'm trying to construct a set of canned searched in my current > Pf> project & while most are working fine I'm stumbling on one pair. > Pf> They are to find records where one VString field is empty and not > Pf> empty respectively. As a precaution I've tried to construct my > Pf> searches to cover both possibilities, that an empty field is an > Pf> empty string and that an empty field is null. > You say " the string field is empty" - so I think that NULL > value is the best description. "Empty string" approach is > possible but not so clear. > I would say that empty string is value with zero length but > it is DEFINED value - you see? Would it be possible to prepare the data first before running the query? First set all empty strings to NULL, then query for IS NULL? UPDATE table SET column=NULL WHERE column='' Your application handles NULL as empty strings, right? /Peter -- Peter Lindgren, dataingenj?r FDT System AB http://www.fdt.se/ 0920-24 33 00 From peter.de.berdt at pandora.be Wed Oct 27 10:10:32 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Wed Oct 27 03:10:48 2004 Subject: Automatic Data Entry In-Reply-To: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> References: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> Message-ID: On 26 Oct 2004, at 20:10, Erik Mueller-Harder wrote: > On Oct 26, 2004, at 11:50, Chuck Pelto wrote: > >> Looking over the documentation and the archived e-mail, I get the >> impression that there is NO automatic population of fields with data >> in Valentina 1.0. This includes no automatic incrementation of a >> serial number field. >> >> Is this correct? > > I believe that is correct, yes, though the RecID field in every table > does contain a unique serial number that seems in fact to be > automatically incremented. If records are deleted, however, there is > no guarantee that previous RecID values won't be re-used. > >> I also note that Valentina 2.0 is supposed to have the automatic >> incrementing of serial numbers, according to one e-mail in the >> archive. >> >> Will V2.0 have automatic data input where a field is supposed to be a >> constant value? > > Just curious: why would you want to store a constant value? > My guess would be Filemaker background (globals from Filemaker). I've seen a lot of Filemaker developers trying to do this when switching to another database system. It's not a good practice, that's for sure. Best regards Peter De Berdt From mo.ritz at gmx.de Wed Oct 27 11:01:05 2004 From: mo.ritz at gmx.de (mo) Date: Wed Oct 27 04:01:37 2004 Subject: [V4MD] Commit cursor changes to DB In-Reply-To: References: Message-ID: Hi, > Why you use UpdateAll() ? > > As far as I see you SELECT exactly one record, > Because you do WHERE RecID="+RecID > > So you should use just curs.updateRecord() I also tried it with curs.updateRecord(), the problem remains... hmmm... further ideas? Cheers! mo . . Moritz Stefaner 0179 - 525 21 26 http://der-mo.net . . On 27.10.2004, at 08:23, Ruslan Zasukhin wrote: > On 10/26/04 11:15 PM, "mo" wrote: > >> Hi, >> >> update on this: I was so stupid and forgot to create the cursor with >> the additional flag to set it to read and write. >> >> Now it works... partially ;) The next problem I encounter is that when >> I call the saveConfiguration function for the second time, it won't >> work. If I trace the cursor after the update, I see the changes. If I >> reselect the same query after that, all field values are aas they used >> to be before the update. >> >> Any Ideas? >> >> Valentina gives no error message, btw., neither when creating the >> cursor nor when updating. >> >> It would be great if someone could help me out here.. > > Why you use UpdateAll() ? > > As far as I see you SELECT exactly one record, > Because you do WHERE RecID="+RecID > > So you should use just curs.updateRecord() > >> Cheers! >> mo >> >> . . >> >> Moritz Stefaner >> 0179 - 525 21 26 >> http://der-mo.net >> >> . . >> On 26.10.2004, at 17:10, mo wrote: >> >>> Hi, >>> >>> I recently started to sort out Valentina & Diretcor, so this is a >>> very basic question. What is the general modus operandi to store >>> cursor changes in the actual db? What I do right now is: >>> >>> saveConfiguration = function(o){ >>> var RecID=o.RecID; delete o.RecID; >>> trace("Model.saveConfiguration "); >>> var c= _global.gDBManager.getDBCursor( "SELECT * FROM >>> configurations >>> WHERE RecID="+RecID); >>> var r=c.updateAllRecords(o.toPList()); >>> _global.gDBManager.DBInstance.flush(); >>> } >>> >>> but the data is not stored in the database. If I trace the cursor >>> after the updateAllRecords command, the values are changed, however >>> only in the cursor as it seems. >>> >>> Could someone be so kind and shed some light on this? >>> Much appreciated, > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From cbpelto at pcisys.net Wed Oct 27 03:23:22 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Wed Oct 27 04:23:37 2004 Subject: Automatic Data Entry In-Reply-To: References: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> Message-ID: Morning Guys, On Oct 27, 2004, at 2:10 AM, Peter De Berdt wrote: > My guess would be Filemaker background (globals from Filemaker). I've > seen a lot of Filemaker developers trying to do this when switching to > another database system. It's not a good practice, that's for sure. In part correct, Peter. But not quite. A technique I use for key fields consists of a alpa-char code AND a serial number in the key field to help me keep track of which table is being linked to. So I'd like to have an automatic alpha-char populated into a field and a serial number automatically incremented to go along with it to make a key that is unique to the record and the table. E.g. "Datafile" & "0000000123" to make Datafile0000000123 as the key field for a record in a datafile in a database system. The other aspect is automatically populating a given field with a given value. That value may be changed at a later date, e.g., all ItemQuantity values for records in a given data file are populated with 1, when a new record is created in a sales-related datafile. This can be changed by the user at a latter date, but since 1 is the most common occurrence, then that reduces their work for entry. Regards, Chuck From kim at torchpublishing.com.au Wed Oct 27 19:53:34 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Wed Oct 27 04:53:53 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Peter > My guess would be Filemaker background (globals from Filemaker). I've > seen a lot of Filemaker developers trying to do this when switching to > another database system. It's not a good practice, that's for sure. ???? It's really common within the FileMaker community to use global fields to store regularly used values or to create interface tricks that are not otherwise available. One of the most common uses is to create a relationship to a global to allow portals to appear or disappear depending on the value of the non-constant field. They're also often used as 'placeholders' in a calc field eg. Have some HTML tags in a global and then calc it into some DB data for use on a web page. I know it appears strange to people who haven't done it, but then I find it really strange that I can't create my own Primary Key. Cheers Kim From james.milne at mac.com Wed Oct 27 10:55:11 2004 From: james.milne at mac.com (James Milne) Date: Wed Oct 27 04:55:27 2004 Subject: Full Text Index Search In-Reply-To: References: Message-ID: <4B23865D-27FE-11D9-96FF-000393DB941A@mac.com> On 27 Oct, 2004, at 07:24, Ruslan Zasukhin wrote: > Valentina cannot do this. > Valentina do have special INDEX BY WORDS for String, VarChar and TEXT > fields. Up to now most developers are quite satisfied wit this. From what I can understand from other developers using IndexByWords, it will probably be sufficient for my current needs. > 2) about speed of LIKE in in Valentina. > Yes James, it is much faster than in mySQL or others. Because of > internal > Valentina structure. Let me to remind: > > If you have in mySQL table with 30 VarChar fields, then one record can > be in > average say 30 * (15-20) = 600 bytes or more. To scan such table for > LIKE > DBMS need load into RAM all this. > > Valentina instead use for each column one file. > So we need scan only 20 bytes * N of records. > For this example Valentina is expected to be 30 times faster. OK- This makes sense! > 3) in Valentina 2.0 we will use SOMETIMES the index file to be even > more > faster. Example. Assume you have table in million of records. Assume > that > you have only 10,000 different words in column. So index will contains > only > 10,000 columns. Then using INDEX SCAN instead of COLUMN SCAN we get > speed > even 100 times faster. If IndexByWords is enabled in Valentina 1.0, does this not generate an index for the column? If so, does this index get used when performing a LIKE search? > 4) Next, in Valentina 2.0 (you can see this in beta) we have introduce > new > special STRING SEACH functions in V4RB API. Similar as you have on OS > X or > emails app... Cool; that sounds like it will be a nice upgrade when it becomes available. > 5) Next, in 2.0 we have implement that not only left() function use > index, > But also right() and substr(). This is also very nice! > So James, I think you can look on Valentina 2.0 beta. > I believe you will be impressed :-) Beta sounds great! However, I need to ship a product within a couple of months. Will Valentina 2.0 be available for me to integrate into a shipping application by January? For the time being, I'm intending using Valentina 1 and the IndexByWords property, with a LIKE search. That might be sufficient for the time-being. Thanks for your feedback, Ruslan. -- Kind Regards, James Milne From sunshine at public.kherson.ua Wed Oct 27 13:45:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 05:45:19 2004 Subject: Full Text Index Search In-Reply-To: <20041027102338.21E6D50DD77@lists.realsoftware.com> Message-ID: > From: James Milne > Date: Wed, 27 Oct 2004 11:01:21 +0100 > > On 27 Oct, 2004, at 07:24, Ruslan Zasukhin wrote: > >> Valentina cannot do this. >> Valentina do have special INDEX BY WORDS for String, VarChar and TEXT >> fields. Up to now most developers are quite satisfied wit this. > > From what I can understand from other developers using IndexByWords, it > will probably be sufficient for my current needs. > >> 2) about speed of LIKE in in Valentina. Yes James, it is much faster than in >> mySQL or others. Because of internal Valentina structure. Let me to remind: >> >> If you have in mySQL table with 30 VarChar fields, then one record can be in >> average say 30 * (15-20) = 600 bytes or more. To scan such table for LIKE >> DBMS need load into RAM all this. >> >> Valentina instead use for each column one file. So we need scan only 20 bytes >> * N of records. For this example Valentina is expected to be 30 times faster. > > OK- This makes sense! > >> 3) in Valentina 2.0 we will use SOMETIMES the index file to be even more >> faster. Example. Assume you have table in million of records. Assume that you >> have only 10,000 different words in column. So index will contains only >> 10,000 columns. Then using INDEX SCAN instead of COLUMN SCAN we get speed >> even 100 times faster. > > If IndexByWords is enabled in Valentina 1.0, does this not generate an > index for the column? Yes of course! This feature can be done ONLY if index exists > If so, does this index get used when performing a LIKE search? No, LIKE do not use index in 1.x >> 4) Next, in Valentina 2.0 (you can see this in beta) we have introduce new >> special STRING SEACH functions in V4RB API. Similar as you have on OS X or >> emails app... > > Cool; that sounds like it will be a nice upgrade when it becomes > available. > >> 5) Next, in 2.0 we have implement that not only left() function use index, >> But also right() and substr(). > > This is also very nice! Yes, to add to 4) and 5) I mean that all this allow you do fld.FindStartsWith() fld.FindEndsWith() fld.Findcontains() Kinds of search, and Valentina will use INDEX SCAN if it can. >> So James, I think you can look on Valentina 2.0 beta. >> I believe you will be impressed :-) > > Beta sounds great! > > However, I need to ship a product within a couple of months. > > Will Valentina 2.0 be available for me to integrate into a shipping > application by January? I believe yes. We practically have done all new features and we go into mode of bug fixing and docs writing. > For the time being, I'm intending using Valentina 1 and the > IndexByWords property, with a LIKE search. That might be sufficient for the time-being. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed Oct 27 13:56:57 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 05:57:11 2004 Subject: [V4MD] Commit cursor changes to DB In-Reply-To: Message-ID: On 10/27/04 12:01 PM, "mo" wrote: > Hi, > >> Why you use UpdateAll() ? >> >> As far as I see you SELECT exactly one record, >> Because you do WHERE RecID="+RecID >> >> So you should use just curs.updateRecord() > > I also tried it with curs.updateRecord(), the problem remains... > hmmm... > > further ideas? I'd say that problem with record locks, but you say you have set ReadWrite. What cursor you use? ServerSide or ClientSide ? ClientSide always is READ ONLY. Check this point > Cheers! > mo > > > . . > > Moritz Stefaner > 0179 - 525 21 26 > http://der-mo.net > > . . > On 27.10.2004, at 08:23, Ruslan Zasukhin wrote: > >> On 10/26/04 11:15 PM, "mo" wrote: >> >>> Hi, >>> >>> update on this: I was so stupid and forgot to create the cursor with >>> the additional flag to set it to read and write. >>> >>> Now it works... partially ;) The next problem I encounter is that when >>> I call the saveConfiguration function for the second time, it won't >>> work. If I trace the cursor after the update, I see the changes. If I >>> reselect the same query after that, all field values are aas they used >>> to be before the update. >>> >>> Any Ideas? >>> >>> Valentina gives no error message, btw., neither when creating the >>> cursor nor when updating. >>> >>> It would be great if someone could help me out here.. >> >> Why you use UpdateAll() ? >> >> As far as I see you SELECT exactly one record, >> Because you do WHERE RecID="+RecID >> >> So you should use just curs.updateRecord() >> >>> Cheers! >>> mo >>> >>> . . >>> >>> Moritz Stefaner >>> 0179 - 525 21 26 >>> http://der-mo.net >>> >>> . . >>> On 26.10.2004, at 17:10, mo wrote: >>> >>>> Hi, >>>> >>>> I recently started to sort out Valentina & Diretcor, so this is a >>>> very basic question. What is the general modus operandi to store >>>> cursor changes in the actual db? What I do right now is: >>>> >>>> saveConfiguration = function(o){ >>>> var RecID=o.RecID; delete o.RecID; >>>> trace("Model.saveConfiguration "); >>>> var c= _global.gDBManager.getDBCursor( "SELECT * FROM >>>> configurations >>>> WHERE RecID="+RecID); >>>> var r=c.updateAllRecords(o.toPList()); >>>> _global.gDBManager.DBInstance.flush(); >>>> } >>>> >>>> but the data is not stored in the database. If I trace the cursor >>>> after the updateAllRecords command, the values are changed, however >>>> only in the cursor as it seems. >>>> >>>> Could someone be so kind and shed some light on this? >>>> Much appreciated, -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Wed Oct 27 13:01:36 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Wed Oct 27 06:01:49 2004 Subject: Full Text Index Search In-Reply-To: References: Message-ID: <91E61E18-2807-11D9-A6CC-000393DB2B60@pandora.be> > >>> So James, I think you can look on Valentina 2.0 beta. >>> I believe you will be impressed :-) >> >> Beta sounds great! >> >> However, I need to ship a product within a couple of months. >> >> Will Valentina 2.0 be available for me to integrate into a shipping >> application by January? > > I believe yes. > > We practically have done all new features and we go into mode of bug > fixing > and docs writing. > The docs are a very important aspect. The docs from Valentina 1 caused some of the people I know to go for another database just because the docs were very limited and some of the example code just wasn't correct. Valentina 1 was very easy to learn (once you understood the object-relational approach), but knowing the plethora of new features in Valentina 2, the docs should be clear and educating. Something similar to the REALbasic Guide (QuickStart and Developer's Guide) would be very nice. Best regards Peter De Berdt From peter.de.berdt at pandora.be Wed Oct 27 13:07:52 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Wed Oct 27 06:08:05 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <72559C8D-2808-11D9-A6CC-000393DB2B60@pandora.be> On 27 Oct 2004, at 11:53, Kim Kohen wrote: > G'day Peter > >> My guess would be Filemaker background (globals from Filemaker). I've >> seen a lot of Filemaker developers trying to do this when switching to >> another database system. It's not a good practice, that's for sure. > ???? > It's really common within the FileMaker community to use global fields > to > store regularly used values or to create interface tricks that are not > otherwise available. > True, filemaker uses globals to make up for the lack of object properties and in Filemaker 6 the lack of local variables. > One of the most common uses is to create a relationship to a global to > allow > portals to appear or disappear depending on the value of the > non-constant > field. They're also often used as 'placeholders' in a calc field eg. > Have > some HTML tags in a global and then calc it into some DB data for use > on a > web page. > You can include the HTML tags in your compiled code or put it in true Valentina records. > I know it appears strange to people who haven't done it, but then I > find it > really strange that I can't create my own Primary Key. > Then Valentina 2 will rock your head :-) Best regards Peter De Berdt From peter.de.berdt at pandora.be Wed Oct 27 13:20:03 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Wed Oct 27 06:20:13 2004 Subject: Automatic Data Entry In-Reply-To: References: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> Message-ID: <25E33DBA-280A-11D9-A6CC-000393DB2B60@pandora.be> On 27 Oct 2004, at 11:23, Chuck Pelto wrote: > Morning Guys, > > On Oct 27, 2004, at 2:10 AM, Peter De Berdt wrote: > >> My guess would be Filemaker background (globals from Filemaker). I've >> seen a lot of Filemaker developers trying to do this when switching >> to another database system. It's not a good practice, that's for >> sure. > > In part correct, Peter. But not quite. A technique I use for key > fields consists of a alpa-char code AND a serial number in the key > field to help me keep track of which table is being linked to. So I'd > like to have an automatic alpha-char populated into a field and a > serial number automatically incremented to go along with it to make a > key that is unique to the record and the table. > > E.g. "Datafile" & "0000000123" to make Datafile0000000123 as the key > field for a record in a datafile in a database system. > > The other aspect is automatically populating a given field with a > given value. That value may be changed at a later date, e.g., all > ItemQuantity values for records in a given data file are populated > with 1, when a new record is created in a sales-related datafile. This > can be changed by the user at a latter date, but since 1 is the most > common occurrence, then that reduces their work for entry. > This is something you have to code in REALbasic (or whatever environment you're programming in), not from Valentina (you could use a calculation/method field, but this can't be changed). If your application is object-oriented, this should be very easy (because your object will automatically generate the field). Once you get the grasp of object-oriented programming and your framework classes are well written, you'll soon discover all the repeating and tedious work you need to do in Filemaker simply isn't necessary. If you try to maintaining the same workflow of Filemaker, i.e. throwing in another field whenever you like, coding in the controls themselves instead of thinking out a decent structure, your application will soon become unmaintainable. Filemaker is like Lego, you just puzzle together some pieces, and you can make very beautiful (and productive) solutions with it, but it isn't smooth (the lego blocks have edges). There will be many times where you have to use tricks to get what you want (every filemaker developer knows of the hundreds of workarounds and need for plug-ins). When something isn't well implemented, you just break out a piece of the puzzle, then put some new lego blocks back in, if there's some junk or holes in the middle and no one notices, you can get away with it. Object-oriented development is just like real sculpting, everything needs to be well thought out before you start chopping of pieces. If you start chopping away, you'll soon make some irrepairable damage and have to start over. But the end result will be so much smoother and realistic. Best regards Peter De Berdt From sunshine at public.kherson.ua Wed Oct 27 14:30:42 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 06:30:56 2004 Subject: Full Text Index Search In-Reply-To: <91E61E18-2807-11D9-A6CC-000393DB2B60@pandora.be> Message-ID: On 10/27/04 2:01 PM, "Peter De Berdt" wrote: >>>> So James, I think you can look on Valentina 2.0 beta. >>>> I believe you will be impressed :-) >>> >>> Beta sounds great! >>> >>> However, I need to ship a product within a couple of months. >>> >>> Will Valentina 2.0 be available for me to integrate into a shipping >>> application by January? >> >> I believe yes. >> >> We practically have done all new features and we go into mode of bug >> fixing >> and docs writing. >> > The docs are a very important aspect. The docs from Valentina 1 caused > some of the people I know to go for another database just because the > docs were very limited and some of the example code just wasn't > correct. Valentina 1 was very easy to learn (once you understood the > object-relational approach), but knowing the plethora of new features > in Valentina 2, the docs should be clear and educating. > > Something similar to the REALbasic Guide (QuickStart and Developer's > Guide) would be very nice. Yes, we move in this way. We have draw about 50 pictures to explain different features, And I think we will make more. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From kim at torchpublishing.com.au Wed Oct 27 22:07:08 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Wed Oct 27 07:07:24 2004 Subject: Automatic Data Entry In-Reply-To: <25E33DBA-280A-11D9-A6CC-000393DB2B60@pandora.be> Message-ID: G'day Peter > you'll soon discover all the repeating and tedious work you > need to do in Filemaker simply isn't necessary. ???? Get ready for a lot more of this type of question given that RS are offering Realbasic professional edition for half price to FMP users. I just ordered it today because it was cheaper than getting the upgrade from standard to professional. Maybe paradigma could offer a similar deal for valentina:) Cheers kim From Kurt.Klamp at abdnb.bayern.de Wed Oct 27 14:32:28 2004 From: Kurt.Klamp at abdnb.bayern.de (Klamp, Kurt) Date: Wed Oct 27 07:33:06 2004 Subject: [SQL/VAPP] Problems with field recid Message-ID: <21674C363854D511B526006008F6B9CD01312A73@ABDN-6.dir.abdnb.bayern.de> Hi Ruslan, >I believe that Valentina 1.x still is not able to sort on RecID field. >(workaround is to use ULONG method = "RecID") >GROUP BY use sorting, so I think this is the same problem. >Yes you can use ObjectPtr field. >Is this solution for you ? Yes it can be.Knowing a workaround solves the problem. But i'm wondering why it works formerly in OS 9. Are there differences between Vapp 1.10 PPC and Vapp 1.10 carbon ? Best Kurt ########################################### This message has been scanned by F-Secure Anti-Virus for Microsoft Exchange. For more information, connect to http://www.F-Secure.com/ From peter.de.berdt at pandora.be Wed Oct 27 15:44:59 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Wed Oct 27 08:45:12 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <65487584-281E-11D9-85BD-000393DB2B60@pandora.be> On 27 Oct 2004, at 14:07, Kim Kohen wrote: > G'day Peter > >> you'll soon discover all the repeating and tedious work you >> need to do in Filemaker simply isn't necessary. > ???? > Get ready for a lot more of this type of question given that RS are > offering > Realbasic professional edition for half price to FMP users. I just > ordered > it today because it was cheaper than getting the upgrade from standard > to > professional. > No problem, I'm a full time Filemaker developer, but my former job I was a fulltime REALbasic developer. I still prefer RB over FM, but one's gotta do what the boss wants you to :-) > Maybe paradigma could offer a similar deal for valentina:) > I don't think this is necessary: Filemaker Server is more expensive than Valentina Server, less performant. REALbasic's reduction deal makes it cheaper than one Filemaker license, and you don't have an extra per-user license cost. Even if you would opt for e.g. Openbase and have ? 5 users, you are making profit over buying FM Server + 5 licenses. You just have to take a bit of extra time into account to design a decent framework. Once you have your framework, development time is even less than Filemaker. Even the plug-ins for REALbasic, if you would need them, are so cheap (Einhugur + Monkeybread for example). Ruslan, is the database now automatically adding fields if the object-structure is changed (instead of having to open the valentina db as a generic database, altering the structure and then initializing the subclass)? Best regards Peter De Berdt From sunshine at public.kherson.ua Wed Oct 27 18:20:39 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 10:21:51 2004 Subject: [SQL/VAPP] Problems with field recid In-Reply-To: <21674C363854D511B526006008F6B9CD01312A73@ABDN-6.dir.abdnb.bayern.de> Message-ID: On 10/27/04 3:32 PM, "Klamp, Kurt" wrote: > Yes it can be.Knowing a workaround solves the problem. But i'm wondering why > it works formerly in OS 9. Are there differences between Vapp 1.10 PPC and > Vapp 1.10 carbon ? OS X is more strict. It can crash in place where OS 9 continue work. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed Oct 27 18:23:11 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Oct 27 10:24:21 2004 Subject: Automatic Data Entry In-Reply-To: <65487584-281E-11D9-85BD-000393DB2B60@pandora.be> Message-ID: On 10/27/04 4:44 PM, "Peter De Berdt" wrote: > Ruslan, is the database now automatically adding fields if the > object-structure is changed (instead of having to open the valentina db > as a generic database, altering the structure and then initializing the > subclass)? You mean in 2.0 ? No Peter. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From softil at onlinehome.de Wed Oct 27 19:37:05 2004 From: softil at onlinehome.de (SoftIl) Date: Wed Oct 27 12:38:05 2004 Subject: DELETE Statement won't work correct Message-ID: <000f01c4bc4b$b4885860$29e4e6d9@IPSBASIS> Hi, what's wrong with this statement: DELETE FROM behandlungsliste BHL, basisliste BL WHERE ((BL.betriebs_id = '15427782-c133-11d8-a802-0010b53f93f1') AND (BHL.masterrec_id = BL.record_id)) If I call this statement, all records in table "behandlungsliste" are deleted. I would like to delete in table "behandlungsliste" only the records for which the condition is true? How is the correct statement? Thanks for your help Carsten Ilwig From cbpelto at pcisys.net Wed Oct 27 15:34:33 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Wed Oct 27 16:34:55 2004 Subject: Keywords Question Message-ID: Greetings, Getting my mind around SQL. Want to know if the following keywords I'm reading about in SQL are available for queries in Valentina: BETWEEN IN NOT IN LIKE NOT LIKE SIMILAR ALL SOME ANY DISTINCT OVERLAPS Regards, Chuck From cbpelto at pcisys.net Wed Oct 27 15:59:02 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Wed Oct 27 16:59:13 2004 Subject: Automatic Data Entry In-Reply-To: <25E33DBA-280A-11D9-A6CC-000393DB2B60@pandora.be> References: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> <25E33DBA-280A-11D9-A6CC-000393DB2B60@pandora.be> Message-ID: <6A092E0E-2863-11D9-A4A0-00306549C134@pcisys.net> On Oct 27, 2004, at 5:20 AM, Peter De Berdt wrote: > This is something you have to code in REALbasic (or whatever > environment you're programming in), not from Valentina (you could use > a calculation/method field, but this can't be changed). If your > application is object-oriented, this should be very easy (because your > object will automatically generate the field). Once you get the grasp > of object-oriented programming and your framework classes are well > written, you'll soon discover all the repeating and tedious work you > need to do in Filemaker simply isn't necessary. If you try to > maintaining the same workflow of Filemaker, i.e. throwing in another > field whenever you like, coding in the controls themselves instead of > thinking out a decent structure, your application will soon become > unmaintainable. Are you suggesting that each record in a database system designed with Valentina is an object in and of itself? > Once you get the grasp of object-oriented programming and your > framework classes are well written, you'll soon discover all the > repeating and tedious work you need to do in Filemaker simply isn't > necessary. If you try to maintaining the same workflow of Filemaker, > i.e. throwing in another field whenever you like, coding in the > controls themselves instead of thinking out a decent structure, your > application will soon become unmaintainable. One of the great aspects of FMP is that you have the flexibility to modify the data file/table, database structure at will at any time. As the old adage goes, "NOTHING endures like 'change'." We are constantly changing things. Have a structure that is cast in concrete makes it very difficult to cope with such a concept as 'change'. But change it must. Especially if it is going to survive. Most especially in a competitive environment. And maybe that's what V Studio is supposed to help with; changing a Valentina database structure without breaking the application. I certainly hope so. Regards, Chuck From joergp at three-2-one.com Thu Oct 28 01:27:36 2004 From: joergp at three-2-one.com (Joerg Pressel) Date: Wed Oct 27 18:27:51 2004 Subject: SQL Problem Message-ID: I'm looking for some advice with a SQL statement (using V4RB 1.11): I have three tables: - "Albums" with Artist_Ptr pointing to "Artists" and field "Title" - "Tracks" with Album_Ptr pointing to "Albums" and Artist_Ptr pointing to "Artists" - "Artists" containing artist names (field "Artist") I want to get all Albums which contain Tracks from a certain artist. So far so easy: SELECT DISTINCT Albums.Title FROM Tracks,Albums WHERE Artist_Ptr=1606 But ? I need one additional information from the Albums which were found: the artist name. I tried: SELECT DISTINCT Albums.Title,Artist FROM Tracks,Albums,Artists WHERE Artist_Ptr=1606 This does not work, because Valentina does not know if she should take "Artist" using the Artist_Ptr from "Albums" or from "Tracks". Is there any way I can get around this problem? Any suggestion what would be the most elegant way? Many thanks, J?rg From peter.de.berdt at pandora.be Thu Oct 28 08:30:00 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Thu Oct 28 01:30:12 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: On 27 Oct 2004, at 17:23, Ruslan Zasukhin wrote: > On 10/27/04 4:44 PM, "Peter De Berdt" > wrote: > >> Ruslan, is the database now automatically adding fields if the >> object-structure is changed (instead of having to open the valentina >> db >> as a generic database, altering the structure and then initializing >> the >> subclass)? > > You mean in 2.0 ? > > No Peter. > That's a pity. Seeing so many Filemaker developers switching to other databases, this in one of the features that could have made Valentina stand out (because now it just crashes if the class structure doesn't match the database structure and you try to access a non-available field). Best regards Peter De Berdt From peter.de.berdt at pandora.be Thu Oct 28 08:46:56 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Thu Oct 28 01:47:04 2004 Subject: Automatic Data Entry In-Reply-To: <6A092E0E-2863-11D9-A4A0-00306549C134@pcisys.net> References: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> <25E33DBA-280A-11D9-A6CC-000393DB2B60@pandora.be> <6A092E0E-2863-11D9-A4A0-00306549C134@pcisys.net> Message-ID: <28BAF53C-28AD-11D9-9042-000393DB2B60@pandora.be> On 27 Oct 2004, at 23:59, Chuck Pelto wrote: > > On Oct 27, 2004, at 5:20 AM, Peter De Berdt wrote: > >> This is something you have to code in REALbasic (or whatever >> environment you're programming in), not from Valentina (you could use >> a calculation/method field, but this can't be changed). If your >> application is object-oriented, this should be very easy (because >> your object will automatically generate the field). Once you get the >> grasp of object-oriented programming and your framework classes are >> well written, you'll soon discover all the repeating and tedious work >> you need to do in Filemaker simply isn't necessary. If you try to >> maintaining the same workflow of Filemaker, i.e. throwing in another >> field whenever you like, coding in the controls themselves instead of >> thinking out a decent structure, your application will soon become >> unmaintainable. > > Are you suggesting that each record in a database system designed with > Valentina is an object in and of itself? > No, you need to have a class to which you pass your record and that class then automatically adds your merged constant-field to the insert. Let me give you an example. Using my framework, I can just drop some control subclasses on a window, give them a name, type, maxlength, ... If the SchemaVersion of the database doesn't match a constant "CurrentVersion" in my REALbasic project, I instantiate all windows and loop the controls. If they are of a certain ClassInterface, my REALbasic app checks if the field exists in the database and if the type is correct, if not, it updates the database structure. My Window subclass automatically handles editing, saving, ... records. I could easily add a new class "ConstantDatabaseField" using the same ClassInterface "DatabaseObject", which would then also be included in the chain of fields to be checked. I still have my early versions of the framework lying around, I'm prepared to send them over to you directly if you want to have a look, mail me offlist. It won't be bug free, and is not optimized, but you'll probably find some parts of it very useful. A grasp of the features: instant multilanguage (able to change application language at runtime, not using REALbasic's constants way), automatic adding of new fields to the db, automatic saving of records, handling of changes and asking the user wether or not he wants to save a record, discard it or cancel the action (i.e. the commit feature added in Filemaker 7) and a lot more, I don't remember. I think I'm using Einhugur plug-ins (DataGrid mainly) in my testing, so you may need to download those. >> Once you get the grasp of object-oriented programming and your >> framework classes are well written, you'll soon discover all the >> repeating and tedious work you need to do in Filemaker simply isn't >> necessary. If you try to maintaining the same workflow of Filemaker, >> i.e. throwing in another field whenever you like, coding in the >> controls themselves instead of thinking out a decent structure, your >> application will soon become unmaintainable. > > One of the great aspects of FMP is that you have the flexibility to > modify the data file/table, database structure at will at any time. > > As the old adage goes, "NOTHING endures like 'change'." We are > constantly changing things. Have a structure that is cast in concrete > makes it very difficult to cope with such a concept as 'change'. But > change it must. Especially if it is going to survive. Most especially > in a competitive environment. > > And maybe that's what V Studio is supposed to help with; changing a > Valentina database structure without breaking the application. I > certainly hope so. > You can alter your database structure using SQL too, but it's not the same as just opening up Field Definitions and changing a field's name or type. But generally speaking, changing a field's type should not be necessary (in Filemaker you have to change it sometimes if you want to get things to work, in Valentina a field that holds an number, will always be a number-type field, it shouldn't be text). Best regards Peter De Berdt From sunshine at public.kherson.ua Thu Oct 28 09:53:37 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 01:54:51 2004 Subject: Keywords Question In-Reply-To: Message-ID: On 10/28/04 12:34 AM, "Chuck Pelto" wrote: > Greetings, > > Getting my mind around SQL. > > Want to know if the following keywords I'm reading about in SQL are > available for queries in Valentina: > > BETWEEN > IN > NOT IN > LIKE > NOT LIKE > SIMILAR > ALL > SOME > ANY > DISTINCT > OVERLAPS THIS words are not in Valentina SQL > SIMILAR > SOME > OVERLAPS -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 09:54:58 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 01:56:06 2004 Subject: Automatic Data Entry In-Reply-To: <6A092E0E-2863-11D9-A4A0-00306549C134@pcisys.net> Message-ID: On 10/28/04 12:59 AM, "Chuck Pelto" wrote: >> Once you get the grasp of object-oriented programming and your >> framework classes are well written, you'll soon discover all the >> repeating and tedious work you need to do in Filemaker simply isn't >> necessary. If you try to maintaining the same workflow of Filemaker, >> i.e. throwing in another field whenever you like, coding in the >> controls themselves instead of thinking out a decent structure, your >> application will soon become unmaintainable. > > One of the great aspects of FMP is that you have the flexibility to > modify the data file/table, database structure at will at any time. Actually this is not aspect of FMP This MUST be able do any db -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 09:59:27 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 02:00:38 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/28/04 9:30 AM, "Peter De Berdt" wrote: > On 27 Oct 2004, at 17:23, Ruslan Zasukhin wrote: > >> On 10/27/04 4:44 PM, "Peter De Berdt" >> wrote: >> >>> Ruslan, is the database now automatically adding fields if the >>> object-structure is changed (instead of having to open the valentina >>> db >>> as a generic database, altering the structure and then initializing >>> the >>> subclass)? >> >> You mean in 2.0 ? >> >> No Peter. >> > That's a pity. Seeing so many Filemaker developers switching to other > databases, this in one of the features that could have made Valentina > stand out (because now it just crashes if the class structure doesn't > match the database structure and you try to access a non-available > field). In 2.0 we have add special error codes as FieldNotInTable FieldNotInClass Deal is that REALbasic code is REALbasic code, And Valentina's kernel is RB code. I do not see way EXACTLY and explicitly map changes in RB code to STEPS of modification in db structure. If you think this is easy then let's try formulate this rules together. May be we did not think deeply. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 10:00:55 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 02:02:14 2004 Subject: Automatic Data Entry In-Reply-To: <28BAF53C-28AD-11D9-9042-000393DB2B60@pandora.be> Message-ID: On 10/28/04 9:46 AM, "Peter De Berdt" wrote: > You can alter your database structure using SQL too, but it's not the > same as just opening up Field Definitions and changing a field's name > or type. But generally speaking, changing a field's type should not be > necessary (in Filemaker you have to change it sometimes if you want to > get things to work, in Valentina a field that holds an number, will > always be a number-type field, it shouldn't be text). Peter, Actually in Valentina developer also can easy change e.g. ushort to long type I agree that change from number to TEXT has not many sense -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 10:03:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 02:04:42 2004 Subject: SQL Problem In-Reply-To: Message-ID: On 10/28/04 2:27 AM, "Joerg Pressel" wrote: Hi Joerg, > I'm looking for some advice with a SQL statement (using V4RB 1.11): > > I have three tables: > - "Albums" with Artist_Ptr pointing to "Artists" and field "Title" > - "Tracks" with Album_Ptr pointing to "Albums" and Artist_Ptr pointing > to "Artists" > - "Artists" containing artist names (field "Artist") > > I want to get all Albums which contain Tracks from a certain artist. > > So far so easy: > SELECT DISTINCT Albums.Title FROM Tracks,Albums WHERE Artist_Ptr=1606 > > But ? I need one additional information from the Albums which were > found: the artist name. > > I tried: > SELECT DISTINCT Albums.Title,Artist FROM Tracks,Albums,Artists WHERE > Artist_Ptr=1606 > This does not work, because Valentina does not know if she should take > "Artist" using the Artist_Ptr from "Albums" or from "Tracks". > > Is there any way I can get around this problem? Any suggestion what > would be the most elegant way? You need self specify the path SELECT DISTINCT Albums.Title,Artist FROM Tracks,Albums,Artists WHERE Artist_Ptr=1606 And albums.artist_ptr = artist.recid -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 10:05:12 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 02:06:25 2004 Subject: DELETE Statement won't work correct In-Reply-To: <000f01c4bc4b$b4885860$29e4e6d9@IPSBASIS> Message-ID: On 10/27/04 8:37 PM, "SoftIl" wrote: Hi Carsten, > Hi, > what's wrong with this statement: > > DELETE FROM behandlungsliste BHL, basisliste BL WHERE ((BL.betriebs_id = > '15427782-c133-11d8-a802-0010b53f93f1') AND (BHL.masterrec_id = > BL.record_id)) I am not sure that 1.x do support ALIAS OF TABLE in the DELETE command. I am also not sure that this is by standard, DELETE FROM behandlungsliste BHL Why you use it ?! Aha! You specify 2 tables This is wrong. DELETE can work only on 1 table. > If I call this statement, all records in table "behandlungsliste" are > deleted. > I would like to delete in table "behandlungsliste" only the records for > which the condition is true? > How is the correct statement? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 10:12:01 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 02:13:06 2004 Subject: Automatic Data Entry In-Reply-To: <6A092E0E-2863-11D9-A4A0-00306549C134@pcisys.net> Message-ID: On 10/28/04 12:59 AM, "Chuck Pelto" wrote: > And maybe that's what V Studio is supposed to help with; changing a > Valentina database structure without breaking the application. I > certainly hope so. Chuck, I do not think that FMP can allow me REMOVE some field, And that it will remove this field from layouts CORRECTLY. I still will need to go and polish layouts. Right? How you can CHANGE structure and not break code? If you rename field in Vstudio then your code will be broken Because you use that name in CODE. If you delete field then your code will be broken because you use that field. If you add field, then MAY BE your code will work. But may happens that will be some glitches also. Example, if you use SELECT * then you will start to get MORE columns for cursor than before -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Thu Oct 28 09:22:58 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Thu Oct 28 02:23:08 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <317D0934-28B2-11D9-9042-000393DB2B60@pandora.be> On 28 Oct 2004, at 08:59, Ruslan Zasukhin wrote: > On 10/28/04 9:30 AM, "Peter De Berdt" > wrote: > >> On 27 Oct 2004, at 17:23, Ruslan Zasukhin wrote: >> >>> On 10/27/04 4:44 PM, "Peter De Berdt" >>> wrote: >>> >>>> Ruslan, is the database now automatically adding fields if the >>>> object-structure is changed (instead of having to open the valentina >>>> db >>>> as a generic database, altering the structure and then initializing >>>> the >>>> subclass)? >>> >>> You mean in 2.0 ? >>> >>> No Peter. >>> >> That's a pity. Seeing so many Filemaker developers switching to other >> databases, this in one of the features that could have made Valentina >> stand out (because now it just crashes if the class structure doesn't >> match the database structure and you try to access a non-available >> field). > > In 2.0 we have add special error codes as > > FieldNotInTable > FieldNotInClass > > Deal is that REALbasic code is REALbasic code, > And Valentina's kernel is RB code. > > I do not see way EXACTLY and explicitly map changes in RB code > to STEPS of modification in db structure. > > If you think this is easy then let's try formulate this rules together. > May be we did not think deeply. > I don't know how the Valentina plug-in handles things (and I don't have a problem with maintaining a correct database structure myself, my classes handle this), but let me give you an example: SchemaVersion 1 MyVDatabase -> MyBaseObject1 ? MyField1 as VVarchar ? MyField2 as VULong ? MyField3 as VDouble SchemaVersion 2 MyVDatabase -> MyBaseObject1 ? MyField1 as VVarchar ? MyField2 as VLong ? MyField3 as VDouble ? MyField4 as VString What you have to do now, is before creating an instance of MyVDatabase, open up that database as a generic VDatabase, then create the new field, alter the type of MyField2, then close the database again, instantiate a MyVDatabase and then start working with your database. If the Valentina plug-in would somehow be able to reflect the changes in the class structure to the database structure, it would save a lot of people time and would stay closer to the Filemaker paradigm of being able to change the database structure instantly. Best regards Peter De Berdt From sunshine at public.kherson.ua Thu Oct 28 11:00:55 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 03:02:07 2004 Subject: Automatic Data Entry In-Reply-To: <317D0934-28B2-11D9-9042-000393DB2B60@pandora.be> Message-ID: On 10/28/04 10:22 AM, "Peter De Berdt" wrote: >> > I don't know how the Valentina plug-in handles things (and I don't have > a problem with maintaining a correct database structure myself, my > classes handle this), but let me give you an example: > > SchemaVersion 1 > MyVDatabase > -> MyBaseObject1 > ? MyField1 as VVarchar > ? MyField2 as VULong > ? MyField3 as VDouble > > SchemaVersion 2 > MyVDatabase > -> MyBaseObject1 > ? MyField1 as VVarchar > ? MyField2 as VLong > ? MyField3 as VDouble > ? MyField4 as VString > > What you have to do now, is before creating an instance of MyVDatabase, > open up that database as a generic VDatabase, then create the new > field, alter the type of MyField2, then close the database again, > instantiate a MyVDatabase and then start working with your database. But instance we create BEFORE do open db = new myDatabase > If the Valentina plug-in would somehow be able to reflect the changes > in the class structure to the database structure, it would save a lot > of people time and would stay closer to the Filemaker paradigm of being > able to change the database structure instantly. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Thu Oct 28 10:26:07 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Thu Oct 28 03:26:14 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <043058E6-28BB-11D9-9042-000393DB2B60@pandora.be> On 28 Oct 2004, at 10:00, Ruslan Zasukhin wrote: > On 10/28/04 10:22 AM, "Peter De Berdt" > wrote: > >>> >> I don't know how the Valentina plug-in handles things (and I don't >> have >> a problem with maintaining a correct database structure myself, my >> classes handle this), but let me give you an example: >> >> SchemaVersion 1 >> MyVDatabase >> -> MyBaseObject1 >> ? MyField1 as VVarchar >> ? MyField2 as VULong >> ? MyField3 as VDouble >> >> SchemaVersion 2 >> MyVDatabase >> -> MyBaseObject1 >> ? MyField1 as VVarchar >> ? MyField2 as VLong >> ? MyField3 as VDouble >> ? MyField4 as VString >> >> What you have to do now, is before creating an instance of >> MyVDatabase, >> open up that database as a generic VDatabase, then create the new >> field, alter the type of MyField2, then close the database again, >> instantiate a MyVDatabase and then start working with your database. > > But instance we create BEFORE do open > > db = new myDatabase > Wouldn't it be possible to check the database file on opening then? Best regards Peter De Berdt From sunshine at public.kherson.ua Thu Oct 28 11:27:52 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 03:28:59 2004 Subject: Automatic Data Entry In-Reply-To: <043058E6-28BB-11D9-9042-000393DB2B60@pandora.be> Message-ID: On 10/28/04 11:26 AM, "Peter De Berdt" wrote: >>> I don't know how the Valentina plug-in handles things (and I don't >>> have >>> a problem with maintaining a correct database structure myself, my >>> classes handle this), but let me give you an example: >>> >>> SchemaVersion 1 >>> MyVDatabase >>> -> MyBaseObject1 >>> ? MyField1 as VVarchar >>> ? MyField2 as VULong >>> ? MyField3 as VDouble >>> >>> SchemaVersion 2 >>> MyVDatabase >>> -> MyBaseObject1 >>> ? MyField1 as VVarchar >>> ? MyField2 as VLong >>> ? MyField3 as VDouble >>> ? MyField4 as VString >>> >>> What you have to do now, is before creating an instance of >>> MyVDatabase, >>> open up that database as a generic VDatabase, then create the new >>> field, alter the type of MyField2, then close the database again, >>> instantiate a MyVDatabase and then start working with your database. >> >> But instance we create BEFORE do open >> >> db = new myDatabase >> > Wouldn't it be possible to check the database file on opening then? Let's' think. You have made example only of ONE simple change. * adding of new field. --- YES I think such change we can catch. What about other changes: * remove of field * rename of field * change of type * change of flags * change of constraints * add of table * remove of table * change table name -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Thu Oct 28 10:38:17 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Thu Oct 28 03:38:25 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: > Let's' think. > > You have made example only of ONE simple change. > * adding of new field. > > --- YES I think such change we can catch. > > What about other changes: > > * remove of field This should be easy enough to catch > * rename of field Very hard to do, I agree, you could use the order in which the properties are instantiated (the field order in the database is also determined by the order in which you create them using classes). > * change of type > * change of flags If fieldname is equal, changing type and flags should also be possible, no? > * add of table > * remove of table > * change table name > Same remarks as for the fields I'm not saying it's a trivial task, not at all. I'm using a dynamic database structure (using a generic VDatabase class) and let my own REALbasic classes handle the update of the db structure by the way and am very happy with the results. I was just trying to provide Valentina (which I still think is one of the most promising databases around) a way of getting more revenues from Filemaker switchers. Best regards Peter De Berdt From fb at memedia.de Thu Oct 28 10:43:39 2004 From: fb at memedia.de (Florian Bogeschdorfer) Date: Thu Oct 28 03:43:47 2004 Subject: Synchronising dbs from several servers Message-ID: Hi! I need to synchronize adresses and payback cards between different outlets, each running it's own vServer and DB in my current application. I thought about sending changes (change,new,delete) in the two important tables to an online Db/php,mySQL where each outlet can then download the changes. Do you see a better way? Since I made a single handler doing all record changes like On VALchangerecord cursor,data --change data here .. .. .. End I am now in trouble, because I would have to catch if the changes are in the 2 mentioned tables, something like: On VALchangerecord cursor,data --change data here .. .. .. If table of cursor is "Adress" or "card" then --synchronize with internet end if End But I see no way to find out the table(s), that a cursor refers to. Is there a way? Otherwise I would have to change every call to the changerecord sub... Best regards, Florian From softil at onlinehome.de Thu Oct 28 13:21:50 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 28 07:43:35 2004 Subject: DELETE Statement won't work correct References: Message-ID: <004b01c4bceb$c3454610$a0e7e6d9@IPSBASIS> Hi Ruslan, > This is wrong. DELETE can work only on 1 table. Oh, I haven't know this. How can I solve this problem? At first: Select all records from mastertable (basisliste). SELECT record_id FROM basisiliste WHERE betriebs_id = '15427782-c133-11d8-a802-0010b53f93f' and second: In an loop I delete all records in childTable: DELETE FROM behandlungsliste WHERE masterrec_id = myVar //this is the variable which include the record_id from the masterrecord Is there a faster way to do this? Best regards Carsten ----- Original Message ----- > > DELETE FROM behandlungsliste BHL, basisliste BL WHERE ((BL.betriebs_id = > > '15427782-c133-11d8-a802-0010b53f93f1') AND (BHL.masterrec_id = > > BL.record_id)) > > You specify 2 tables > From softil at onlinehome.de Thu Oct 28 14:38:15 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 28 07:43:53 2004 Subject: [VNET] using length in where clause Message-ID: <004c01c4bceb$c37e7e80$a0e7e6d9@IPSBASIS> Hi all, is it possible to use the length function in an where clause like this: SELECT * FROM myTable WHERE length('myFieldName') < 15 If I try this, I get an error "wrong expression". In case of not, is there some other possibility to check the length of an field in a where clause? Thanks for your help Carsten From mo.ritz at gmx.de Thu Oct 28 14:57:26 2004 From: mo.ritz at gmx.de (mo) Date: Thu Oct 28 08:03:57 2004 Subject: Fwd: [V4MD] Commit cursor changes to DB Message-ID: Hi, I found a solution for my problem. By isolating the script from the larger application background, I discovered that the problem was not the update itself, but a subsequent new creation of a server side read-only cursor to read out the modified table. What I did was a desperate guess: I created the second cursor with the same properties as the first one (readAndWrite, server, randomAccess), and voila it worked. Looks like a bug of the Xtra to me - is this a known issue? Cheers! mo . . Moritz Stefaner 0179 - 525 21 26 http://der-mo.net . . On 27.10.2004, at 12:56, Ruslan Zasukhin wrote: > On 10/27/04 12:01 PM, "mo" wrote: > >> Hi, >> >>> Why you use UpdateAll() ? >>> >>> As far as I see you SELECT exactly one record, >>> Because you do WHERE RecID="+RecID >>> >>> So you should use just curs.updateRecord() >> >> I also tried it with curs.updateRecord(), the problem remains... >> hmmm... >> >> further ideas? > > I'd say that problem with record locks, but you say you have set > ReadWrite. > > What cursor you use? ServerSide or ClientSide ? > ClientSide always is READ ONLY. > Check this point > > >> Cheers! >> mo >> >> >> . . >> >> Moritz Stefaner >> 0179 - 525 21 26 >> http://der-mo.net >> >> . . >> On 27.10.2004, at 08:23, Ruslan Zasukhin wrote: >> >>> On 10/26/04 11:15 PM, "mo" wrote: >>> >>>> Hi, >>>> >>>> update on this: I was so stupid and forgot to create the cursor with >>>> the additional flag to set it to read and write. >>>> >>>> Now it works... partially ;) The next problem I encounter is that >>>> when >>>> I call the saveConfiguration function for the second time, it won't >>>> work. If I trace the cursor after the update, I see the changes. If >>>> I >>>> reselect the same query after that, all field values are aas they >>>> used >>>> to be before the update. >>>> >>>> Any Ideas? >>>> >>>> Valentina gives no error message, btw., neither when creating the >>>> cursor nor when updating. >>>> >>>> It would be great if someone could help me out here.. >>> >>> Why you use UpdateAll() ? >>> >>> As far as I see you SELECT exactly one record, >>> Because you do WHERE RecID="+RecID >>> >>> So you should use just curs.updateRecord() >>> >>>> Cheers! >>>> mo >>>> >>>> . . >>>> >>>> Moritz Stefaner >>>> 0179 - 525 21 26 >>>> http://der-mo.net >>>> >>>> . . >>>> On 26.10.2004, at 17:10, mo wrote: >>>> >>>>> Hi, >>>>> >>>>> I recently started to sort out Valentina & Diretcor, so this is a >>>>> very basic question. What is the general modus operandi to store >>>>> cursor changes in the actual db? What I do right now is: >>>>> >>>>> saveConfiguration = function(o){ >>>>> var RecID=o.RecID; delete o.RecID; >>>>> trace("Model.saveConfiguration "); >>>>> var c= _global.gDBManager.getDBCursor( "SELECT * FROM >>>>> configurations >>>>> WHERE RecID="+RecID); >>>>> var r=c.updateAllRecords(o.toPList()); >>>>> _global.gDBManager.DBInstance.flush(); >>>>> } >>>>> >>>>> but the data is not stored in the database. If I trace the cursor >>>>> after the updateAllRecords command, the values are changed, however >>>>> only in the cursor as it seems. >>>>> >>>>> Could someone be so kind and shed some light on this? >>>>> Much appreciated, > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina mailing list > Valentina@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina > From sunshine at public.kherson.ua Thu Oct 28 16:11:38 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 08:12:47 2004 Subject: DELETE Statement won't work correct In-Reply-To: <004b01c4bceb$c3454610$a0e7e6d9@IPSBASIS> Message-ID: On 10/28/04 2:21 PM, "SoftIl" wrote: > Hi Ruslan, >> This is wrong. DELETE can work only on 1 table. > Oh, I haven't know this. > How can I solve this problem? > > At first: > Select all records from mastertable (basisliste). > SELECT record_id FROM basisiliste WHERE betriebs_id = > '15427782-c133-11d8-a802-0010b53f93f' This select many records ? > and second: > In an loop I delete all records in childTable: > DELETE FROM behandlungsliste WHERE masterrec_id = myVar //this is the > variable which include the record_id from the masterrecord > > Is there a faster way to do this? You can try WHERE ... IN (SELECT) -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 16:13:37 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 08:14:47 2004 Subject: [VNET] using length in where clause In-Reply-To: <004c01c4bceb$c37e7e80$a0e7e6d9@IPSBASIS> Message-ID: On 10/28/04 3:38 PM, "SoftIl" wrote: > Hi all, > is it possible to use the length function in an where clause like this: > > SELECT * FROM myTable WHERE length('myFieldName') < 15 no > If I try this, I get an error "wrong expression". > > In case of not, is there some other possibility to check the length of an > field in a where clause? Using BaseObject, method -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From kevo at gatorgraphics.com Thu Oct 28 08:38:00 2004 From: kevo at gatorgraphics.com (Kevin Windham) Date: Thu Oct 28 08:38:14 2004 Subject: Automatic Data Entry In-Reply-To: <28BAF53C-28AD-11D9-9042-000393DB2B60@pandora.be> References: <46070628-277A-11D9-B069-000393DC4766@vermontsoftworks.com> <25E33DBA-280A-11D9-A6CC-000393DB2B60@pandora.be> <6A092E0E-2863-11D9-A4A0-00306549C134@pcisys.net> <28BAF53C-28AD-11D9-9042-000393DB2B60@pandora.be> Message-ID: <95BB361A-28E6-11D9-88BF-000393B29272@gatorgraphics.com> On Oct 28, 2004, at 1:46 AM, Peter De Berdt wrote: > No, you need to have a class to which you pass your record and that > class then automatically adds your merged constant-field to the > insert. > > Let me give you an example. Using my framework, I can just drop some > control subclasses on a window, give them a name, type, maxlength, ... > If the SchemaVersion of the database doesn't match a constant > "CurrentVersion" in my REALbasic project, I instantiate all windows > and loop the controls. If they are of a certain ClassInterface, my > REALbasic app checks if the field exists in the database and if the > type is correct, if not, it updates the database structure. My Window > subclass automatically handles editing, saving, ... records. I could > easily add a new class "ConstantDatabaseField" using the same > ClassInterface "DatabaseObject", which would then also be included in > the chain of fields to be checked. > > I still have my early versions of the framework lying around, I'm > prepared to send them over to you directly if you want to have a look, > mail me offlist. It won't be bug free, and is not optimized, but > you'll probably find some parts of it very useful. A grasp of the > features: instant multilanguage (able to change application language > at runtime, not using REALbasic's constants way), automatic adding of > new fields to the db, automatic saving of records, handling of changes > and asking the user wether or not he wants to save a record, discard > it or cancel the action (i.e. the commit feature added in Filemaker 7) > and a lot more, I don't remember. I think I'm using Einhugur plug-ins > (DataGrid mainly) in my testing, so you may need to download those. This sounds very interesting. I would like to see them if you don't mind. Thanks, Kevin From sunshine at public.kherson.ua Thu Oct 28 17:02:15 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 09:03:23 2004 Subject: Synchronising dbs from several servers In-Reply-To: Message-ID: On 10/28/04 11:43 AM, "Florian Bogeschdorfer" wrote: Hi Florian, > I need to synchronize adresses and payback cards between different outlets, > each running it's own vServer and DB in my current application. > > I thought about sending changes (change,new,delete) in the two important > tables to an online Db/php,mySQL where each outlet can then download the > changes. Do you see a better way? First of all, you MUST to have some online center. Right? > Since I made a single handler doing all record changes like > > On VALchangerecord cursor,data > --change data here > .. > .. > .. > End > > I am now in trouble, because I would have to catch if the changes are in the > 2 mentioned tables, something like: > > On VALchangerecord cursor,data > --change data here > .. > .. > .. > If table of cursor is "Adress" or "card" then > --synchronize with internet > end if > End > > But I see no way to find out the table(s), that a cursor refers to. Is there > a way? No such way. > Otherwise I would have to change every call to the changerecord sub... This is hard task. You can read about replication. One server publish some tables, other servers subscribe to them. Only publisher can write to that tables, rest read. You can think about usage in these 2 tables the COLUMN 'datetime'. And use its for synchronization. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From fb at memedia.de Thu Oct 28 16:30:43 2004 From: fb at memedia.de (Florian Bogeschdorfer) Date: Thu Oct 28 09:30:56 2004 Subject: Synchronising dbs from several servers In-Reply-To: Message-ID: Hi Ruslan! > > On 10/28/04 11:43 AM, "Florian Bogeschdorfer" wrote: > > Hi Florian, > > > I need to synchronize adresses and payback cards between different > > outlets, each running it's own vServer and DB in my current > application. > > > > I thought about sending changes (change,new,delete) in the two > > important tables to an online Db/php,mySQL where each > outlet can then > > download the changes. Do you see a better way? > > First of all, you MUST to have some online center. Right? Yes, I am planning to store the changes until the have been downloaded from the other outlets, then delete them. > > > > > > But I see no way to find out the table(s), that a cursor > refers to. Is > > there a way? > > No such way. Too bad. Could it be implemented? > > > Otherwise I would have to change every call to the > changerecord sub... > > This is hard task. Yes, especially because the project is very large. Hundreds of routines are calling this sub. Well maybe less, since some are parent/child > > You can read about replication. > One server publish some tables, other servers subscribe to them. > Only publisher can write to that tables, rest read. > > You can think about usage in these 2 tables the COLUMN 'datetime'. > And use its for synchronization. Each outlet can be a publisher, eg customer goes to outlet 10, buys something, then 15 minutes later he goes to outlet 15 and there must be the changes already downloaded... I know that the common way is to store the relevant data on the card, but this is no option here. So I upload the data to a server, adding to more fields FROM_OUTLET and FROM_ADRESS_ID. This is actually no big problem. A script is running constantly to synchronize non-stop. But having to rewrite all parts that change is a hard way since I have to filter out 2 tables from 50, only changes in these tables are to be synchronized. I need something like "Hey, table 1 is changed, we must synchronize..." No idea? Best regards, Florian From softil at onlinehome.de Thu Oct 28 16:08:50 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 28 09:39:59 2004 Subject: [VNET] using length in where clause References: Message-ID: <000001c4bcfc$04a26ab0$feefe6d9@IPSBASIS> Hi Ruslan, > Using BaseObject, method Ok. Thanks. Best regards, Carsten From softil at onlinehome.de Thu Oct 28 16:39:11 2004 From: softil at onlinehome.de (SoftIl) Date: Thu Oct 28 09:40:16 2004 Subject: DELETE Statement won't work correct References: Message-ID: <000101c4bcfc$04d93220$feefe6d9@IPSBASIS> Hi, > > SELECT record_id FROM basisiliste WHERE betriebs_id = > > '15427782-c133-11d8-a802-0010b53f93f' > This select many records ? That's right. > You can try WHERE ... IN (SELECT) Great, this works. Many thanks. :-) Best regards, Carsten From fvanlerberghe at freegates.be Thu Oct 28 16:54:51 2004 From: fvanlerberghe at freegates.be (Francois Van Lerberghe) Date: Thu Oct 28 09:55:48 2004 Subject: [V4RB] Must I close the database before copy the database files Message-ID: Hello Ruslan, I want to develop an automatic backup procedure in my project. I wonder if is it safe to do a VDatabase.Flush and next copy the database files, letting the database opened ? Or must I close the database first, before copy the files ? I ask it because the open command takes several seconds and I would avoid it if it's possible. Fran?ois Van Lerberghe Rue Thier Monty, 15 A 4570 Marchin Belgique From sunshine at public.kherson.ua Thu Oct 28 18:19:42 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 10:20:48 2004 Subject: [V4RB] Must I close the database before copy the database files In-Reply-To: Message-ID: On 10/28/04 5:54 PM, "Francois Van Lerberghe" wrote: > Hello Ruslan, > > I want to develop an automatic backup procedure in my project. > I wonder if is it safe to do a VDatabase.Flush and next copy the database > files, letting the database opened ? > Or must I close the database first, before copy the files ? Yes of course > I ask it because the open command takes several seconds and I would avoid it > if it's possible. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 21:25:15 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 13:26:24 2004 Subject: Synchronising dbs from several servers In-Reply-To: Message-ID: On 10/28/04 5:30 PM, "Florian Bogeschdorfer" wrote: >>> But I see no way to find out the table(s), that a cursor >> refers to. Is >>> there a way? >> >> No such way. > > Too bad. Could it be implemented? I afraid no. Cursor use RESULT TABLE. This can be NEW TMP table, or sometimes even SECOND TMP table. It have no info about original tables. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 21:25:59 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 13:27:04 2004 Subject: Synchronising dbs from several servers In-Reply-To: Message-ID: On 10/28/04 5:30 PM, "Florian Bogeschdorfer" wrote: >> >>> Otherwise I would have to change every call to the >> changerecord sub... >> >> This is hard task. > > Yes, especially because the project is very large. Hundreds of routines are > calling this sub. Well maybe less, since some are parent/child You can refactor it to be router to several new functions -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu Oct 28 21:28:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Oct 28 13:29:43 2004 Subject: Synchronising dbs from several servers In-Reply-To: Message-ID: On 10/28/04 5:30 PM, "Florian Bogeschdorfer" wrote: >> You can read about replication. >> One server publish some tables, other servers subscribe to them. >> Only publisher can write to that tables, rest read. >> >> You can think about usage in these 2 tables the COLUMN 'datetime'. >> And use its for synchronization. > > Each outlet can be a publisher, eg customer goes to outlet 10, buys > something, then 15 minutes later he goes to outlet 15 and there must be the > changes already downloaded... So you need 2-ways replication... Even more harder. > I know that the common way is to store the relevant data on the card, but > this is no option here. > So I upload the data to a server, adding to more fields FROM_OUTLET and > FROM_ADRESS_ID. This is actually no big problem. A script is running > constantly to synchronize non-stop. > But having to rewrite all parts that change is a hard way since I have to > filter out 2 tables from 50, only changes in these tables are to be > synchronized. > I need something like "Hey, table 1 is changed, we must synchronize..." > > No idea? In Lingo ? I think in Lingo you have procedural way of programming. So you do not have polymorphism, And therefore you have only one way: to use if() or switch() -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From cindy at kowhaiprogramming.com Fri Oct 29 08:10:42 2004 From: cindy at kowhaiprogramming.com (Cindy Brown) Date: Thu Oct 28 14:13:06 2004 Subject: Synchronising dbs from several servers In-Reply-To: Message-ID: Hi, I do this at the moment but with TCP/IP across a LAN. What I did was set up an additional baseobject which is unique to each machine and keeps track of all changes. This database in that baseobject has four fields - the name of the database where the change occurred, a unique recid for the record that has been changed, the type of change (add, edit, delete) and lastly the name of the machine (or in your case server) where the change was made. I then use a timer to check every 30 seconds (you could make it quicker) whether there have been any changes on the workstation. If there are any new additions to the sync list, these changes are sent across to a main server and added to the main server's sync list. Then if there are any changes on the main server, these are sent back to the workstation. It took a bit of work to get this to go right but I have managed it with REALbasic and Valentina 1.10. I now have it working with up to 50 machines on a network talking to a server and syncing whenever they connect. Hope that helps. Cindy >> Each outlet can be a publisher, eg customer goes to outlet 10, buys >> something, then 15 minutes later he goes to outlet 15 and there must be the >> changes already downloaded... > -- From kim at torchpublishing.com.au Fri Oct 29 07:54:42 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Thu Oct 28 16:55:00 2004 Subject: Automatic Data Entry In-Reply-To: <72559C8D-2808-11D9-A6CC-000393DB2B60@pandora.be> Message-ID: G'day Peter > True, filemaker uses globals to make up for the lack of object > properties and in Filemaker 6 the lack of local variables. > You can include the HTML tags in your compiled code or put it in true > Valentina records. ???? Using RB as the front end I can't see where I'd need to use global fields any more. The fact that I can create an admin layout somewhere in the app (which I always did with FMP) and then put either an editfield or label there to hold constant data makes it unecessary. I know constants could probably be used for this but I haven't gotten that far in the user guide yet:) > Then Valentina 2 will rock your head :-) ???? Hmmm, I'm afraid I'm struggling with Valentina. It appears from the tutorials and list discussion that it's so OO it seems I'll need to learn a hell of a lot more RB/OOP before I can use it effectively. I'm surprised because I had a MySQL app running in no time using RB and simple SQL statements. I'd really like to use Valentina because there are great merits in being able to use the same code for a single or multi user application but I'm now resigning myself to RBD for single user and MySQL (where I feel pretty comfortable) for multi user - at least until I learn more RB. Regards Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From cbpelto at pcisys.net Thu Oct 28 17:49:21 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 28 18:49:34 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: On Oct 28, 2004, at 3:54 PM, Kim Kohen wrote: >> Then Valentina 2 will rock your head :-) > ???? > Hmmm, I'm afraid I'm struggling with Valentina. It appears from the > tutorials and list discussion that it's so OO it seems I'll need to > learn a > hell of a lot more RB/OOP before I can use it effectively. I'm > surprised > because I had a MySQL app running in no time using RB and simple SQL > statements. Unfortunately, the tutorials, all of them, are not very good. Especially if one is new to SQL. I've addressed this with Ruslan. Hopefully, he'll have time to sort that out, as a good tutorial is a good selling point for a product like this. > I'd really like to use Valentina because there are great merits in > being > able to use the same code for a single or multi user application but > I'm now > resigning myself to RBD for single user and MySQL (where I feel pretty > comfortable) for multi user - at least until I learn more RB. I need a mult-user SQL database system. The product we're moving into the 21st Century was of that sort. That's why we went with Valentina. However, the learning curve has been something of a burden. And, I'm sure, it will continue to be something of that for a bit longer. But at least the slope is not quite so steep. I can come up with work-arounds on the matters of automatic data entry. That's not too much of a problem. But it would be a nice feature to see available. Especially the serial number aspect. I've not read all of the e-mail on this subject, yet. Been away most of the day. I'll likely have more comments come tomorrow. Regards, Chuck From cbpelto at pcisys.net Thu Oct 28 17:51:25 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 28 18:51:31 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <47932458-293C-11D9-8BD2-00306549C134@pcisys.net> P.S. I'm going to use my venerated FMP to help me model the database system structure I'll implement in REALbasic/Valentina. On Oct 28, 2004, at 5:49 PM, Chuck Pelto wrote: > > On Oct 28, 2004, at 3:54 PM, Kim Kohen wrote: > >>> Then Valentina 2 will rock your head :-) >> ???? >> Hmmm, I'm afraid I'm struggling with Valentina. It appears from the >> tutorials and list discussion that it's so OO it seems I'll need to >> learn a >> hell of a lot more RB/OOP before I can use it effectively. I'm >> surprised >> because I had a MySQL app running in no time using RB and simple SQL >> statements. > > Unfortunately, the tutorials, all of them, are not very good. > Especially if one is new to SQL. I've addressed this with Ruslan. > Hopefully, he'll have time to sort that out, as a good tutorial is a > good selling point for a product like this. > >> I'd really like to use Valentina because there are great merits in >> being >> able to use the same code for a single or multi user application but >> I'm now >> resigning myself to RBD for single user and MySQL (where I feel pretty >> comfortable) for multi user - at least until I learn more RB. > > I need a mult-user SQL database system. The product we're moving into > the 21st Century was of that sort. That's why we went with Valentina. > However, the learning curve has been something of a burden. And, I'm > sure, it will continue to be something of that for a bit longer. But > at least the slope is not quite so steep. > > I can come up with work-arounds on the matters of automatic data > entry. That's not too much of a problem. But it would be a nice > feature to see available. Especially the serial number aspect. > > I've not read all of the e-mail on this subject, yet. Been away most > of the day. I'll likely have more comments come tomorrow. > > Regards, > > Chuck > From kim at torchpublishing.com.au Fri Oct 29 10:03:14 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Thu Oct 28 19:03:32 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Chuck > Unfortunately, the tutorials, all of them, are not very good. > Especially if one is new to SQL. I've addressed this with Ruslan. ???? If I could get my head around it I'd write a tutorial myself:) > I can come up with work-arounds on the matters of automatic data entry. > That's not too much of a problem. But it would be a nice feature to see > available. Especially the serial number aspect. ???? I know people will refer to the RecID but I'm really struggling to come to grips with not having my own ID field. It's been the first field in every db I've created in the last 10 years. The auto-entry thing can be overcome pretty easily. In RB I have a 'sheet' for a new record where people enter basic information. It would be easy to just hard-code a constant into the SQL insert statement to give a default value to a particular field every time a new record is created. Cheers Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From cbpelto at pcisys.net Thu Oct 28 19:49:55 2004 From: cbpelto at pcisys.net (Chuck Pelto) Date: Thu Oct 28 20:50:05 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: On Oct 28, 2004, at 6:03 PM, Kim Kohen wrote: > G'day Chuck > >> Unfortunately, the tutorials, all of them, are not very good. >> Especially if one is new to SQL. I've addressed this with Ruslan. > ???? > If I could get my head around it I'd write a tutorial myself:) I was of the same mind, but considering that other third-party tutorials didn't seem to be keeping theirs up to speed, I suspect that the only one with the real motivation to do a good tutorial and keep it current is the manufacturer. >> I can come up with work-arounds on the matters of automatic data >> entry. >> That's not too much of a problem. But it would be a nice feature to >> see >> available. Especially the serial number aspect. > ???? > I know people will refer to the RecID but I'm really struggling to > come to > grips with not having my own ID field. It's been the first field in > every db > I've created in the last 10 years. You still need a key/ID field in SQL, as far as I can tell. It's just that its not going to be as automated as we are used to in FMP. We'll need to build a routine to generated that data to go into that field. > The auto-entry thing can be overcome pretty easily. In RB I have a > 'sheet' > for a new record where people enter basic information. It would be > easy to > just hard-code a constant into the SQL insert statement to give a > default > value to a particular field every time a new record is created. Yes, the routine can be accomplished inside of RB and the data passed to the key field in the SQL tables, as necessary. More drudge, but I suspect a module can be developed and passed from RB app to RB app so we don't have to recreate the wheel in each new project. Later, Chuck From peter.de.berdt at pandora.be Fri Oct 29 08:26:16 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Fri Oct 29 01:26:27 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <706D8AB0-2973-11D9-A976-000393DB2B60@pandora.be> On 29 Oct 2004, at 03:49, Chuck Pelto wrote: > > On Oct 28, 2004, at 6:03 PM, Kim Kohen wrote: > >> G'day Chuck >> >>> Unfortunately, the tutorials, all of them, are not very good. >>> Especially if one is new to SQL. I've addressed this with Ruslan. >> ???? >> If I could get my head around it I'd write a tutorial myself:) > > I was of the same mind, but considering that other third-party > tutorials didn't seem to be keeping theirs up to speed, I suspect that > the only one with the real motivation to do a good tutorial and keep > it current is the manufacturer. > To defend Ruslan, most other databases don't have the complete SQL explanation in their manual either, SQL is a standard, so you can just look up the specs. However, Valentina 1 only supports a subset of SQL, so I agree it's sometimes hard to know what you can and what you can't use. But I agree to the statement that in order for Valentina to become one of the serious players on the market, it needs a clear and educating manual, if it were my company, I'd even outsource it. Programmers always tend to assume the end user is smarter than he/she really is :-) >>> I can come up with work-arounds on the matters of automatic data >>> entry. >>> That's not too much of a problem. But it would be a nice feature to >>> see >>> available. Especially the serial number aspect. >> ???? >> I know people will refer to the RecID but I'm really struggling to >> come to >> grips with not having my own ID field. It's been the first field in >> every db >> I've created in the last 10 years. > > You still need a key/ID field in SQL, as far as I can tell. It's just > that its not going to be as automated as we are used to in FMP. We'll > need to build a routine to generated that data to go into that field. You can use the RecID (and in Valentina 2, you have primary keys and foreign keys too, those are the ones you're looking for), but if a record is deleted, the RecID of that record will be reused when you create a new one. I believe this has also changed in Valentina 2. > >> The auto-entry thing can be overcome pretty easily. In RB I have a >> 'sheet' >> for a new record where people enter basic information. It would be >> easy to >> just hard-code a constant into the SQL insert statement to give a >> default >> value to a particular field every time a new record is created. > > Yes, the routine can be accomplished inside of RB and the data passed > to the key field in the SQL tables, as necessary. More drudge, but I > suspect a module can be developed and passed from RB app to RB app so > we don't have to recreate the wheel in each new project. > Exactly, that's object-oriented programming. You just let the object handle your automatic data entry. I wouldn't put it in a module, but in a class (modules are like scripts, they're global, but it's not a good practice to use them if you can use a class for it). This class would then do record updating, record inserting, record deleting and all other database operations for you. If your class is well written and you don't use direct database operations from within the rest of your application, you could even decouple the database backend from your frontend application. Switching databases then becomes a breeze, as you only have to rewrite the class that interfaces with the database (look at a lot of PHP solutions, they have database abstraction, which allows you to choose your database backend). I want to emphasize switching from Filemaker to REALbasic requires you to turn a switch and start thinking differently. If you don't, you'll have the same problems you have with Filemaker in no time: speed issues, unmaintainable code, broken code, ... If you are really serious about object-oriented programming, get yourself some books on the subject, don't forget to get "Design Patterns" once you really know OO, it's like the programmer's bible. Good luck! Best regards Peter De Berdt From sunshine at public.kherson.ua Fri Oct 29 09:47:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 01:48:14 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 12:54 AM, "Kim Kohen" wrote: >> Then Valentina 2 will rock your head :-) > ???? > Hmmm, I'm afraid I'm struggling with Valentina. It appears from the > tutorials and list discussion that it's so OO it seems I'll need to learn a > hell of a lot more RB/OOP before I can use it effectively. I'm surprised > because I had a MySQL app running in no time using RB and simple SQL > statements. Kim, Valentina is very flexable. It support at least 3 different ways of programming on choice of develoepr. You can program Valentina in EXACTLY THE SAME way as you did with mySQL! No problems. Not remember if you have ask about this few days ago or somebody else? And I did show example how Valentina c be similar to mySQL. So if you do not like OO or if you cannot use it right now then just not use it. And go by familiar to you way. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 29 09:47:57 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 01:49:04 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 12:54 AM, "Kim Kohen" wrote: > I'd really like to use Valentina because there are great merits in being > able to use the same code for a single or multi user application but I'm now > resigning myself to RBD for single user and MySQL (where I feel pretty > comfortable) for multi user - at least until I learn more RB. Well. Show on list your example of RB + mySQL Few lines of code. We will show you analog for Valentina -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 29 09:54:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 01:55:47 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 3:03 AM, "Kim Kohen" wrote: Hi Kim, > I know people will refer to the RecID but I'm really struggling to come to > grips with not having my own ID field. It's been the first field in every db > I've created in the last 10 years. :-) Right. And Valentina introduce you a NEW PARADIGM RecID fields and ObjectPtr. And I believe they are better :-) If you don't know why I will explain Look, we want to be EXTENTION to Relational model. Valentina 2.0 do this perfectly. We have made great jon in 2.0. And very soon you will see that. We have implement full SQL92 in 2.0 So you will have EXACTLY ALL as in Oracle, mySQL,. ... If you want Primary key then you have it, if you want FOREIGN KEY then you have it... But we also have improve object model of Valentina in 2.0. We have add not only RecID but also OID field. We have add new abstraction LINK into Valentina Database model. Any other db in the world do not have it. > The auto-entry thing can be overcome pretty easily. In RB I have a 'sheet' > for a new record where people enter basic information. It would be easy to > just hard-code a constant into the SQL insert statement to give a default > value to a particular field every time a new record is created. Right. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 29 09:57:07 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 01:58:13 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 4:49 AM, "Chuck Pelto" wrote: >> I know people will refer to the RecID but I'm really struggling to >> come to >> grips with not having my own ID field. It's been the first field in >> every db >> I've created in the last 10 years. > > You still need a key/ID field in SQL, as far as I can tell. It's just > that its not going to be as automated as we are used to in FMP. We'll > need to build a routine to generated that data to go into that field. But RecID field 1) exists always. You need do ZERO steps to get it. 2) it provide referential integrity automatically 3) it is 4 times faster on joins that ID of ULONG type. So tell me. WHAT REASONS to use own ID field ??? Only because all other DBMS do this ? Only because you are used to do this during 10 years ? :-) -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 29 10:00:05 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 02:01:13 2004 Subject: Automatic Data Entry In-Reply-To: <706D8AB0-2973-11D9-A976-000393DB2B60@pandora.be> Message-ID: On 10/29/04 9:26 AM, "Peter De Berdt" wrote: >> You still need a key/ID field in SQL, as far as I can tell. It's just >> that its not going to be as automated as we are used to in FMP. We'll >> need to build a routine to generated that data to go into that field. > > You can use the RecID (and in Valentina 2, you have primary keys and > foreign keys too, those are the ones you're looking for), yes > but if a > record is deleted, the RecID of that record will be reused when you > create a new one. I believe this has also changed in Valentina 2. No Peter! This is feature of RecID field. There is nothing bad in this! This is good that it is reused :-) There is no problem here because Valentina with its ObjectPtr field do automatical referential integrity. IF this feature for some task is not god, then and only then developer should prefer to use own ID field. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Oct 29 10:04:51 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 02:06:00 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/28/04 11:38 AM, "Peter De Berdt" wrote: >> Let's' think. >> >> You have made example only of ONE simple change. >> * adding of new field. >> >> --- YES I think such change we can catch. >> >> What about other changes: >> >> * remove of field > > This should be easy enough to catch > >> * rename of field > > Very hard to do, I agree, you could use the order in which the > properties are instantiated (the field order in the database is also > determined by the order in which you create them using classes). > >> * change of type >> * change of flags > > If fieldname is equal, changing type and flags should also be possible, > no? > >> * add of table >> * remove of table >> * change table name >> > Same remarks as for the fields Well, I agree. And I know that 4D have such feature. But 4D have advantage. They do this in the own IDE. We need invent way to do this from DIFFERENT program languages to control engine. I know that 4D use for this ID property of Fields, tables. In Valentina 2.0 Tables, Links, Fields all have now ID field which always grow. So yes we will try this but I think not for 2.0 > I'm not saying it's a trivial task, not at all. I'm using a dynamic > database structure (using a generic VDatabase class) and let my own > REALbasic classes handle the update of the db structure by the way and > am very happy with the results. I was just trying to provide Valentina > (which I still think is one of the most promising databases around) a > way of getting more revenues from Filemaker switchers. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From bibiko at eva.mpg.de Fri Oct 29 09:51:22 2004 From: bibiko at eva.mpg.de (Hans-Joerg Bibiko) Date: Fri Oct 29 02:56:50 2004 Subject: An administrative request Message-ID: <4181F67A.1060302@eva.mpg.de> Dear adminstrators of this list, I receive all the interesting mails from this list and not only from this list. I use a mail programme with a filter function in order store these lists in different folders. But for the Valentina list I can't declare an easy filter. Would it be possible to program the list server to add an unique identifier like e.g. '[VDB]' to the subject field automatically? I guess this problem have some other subscribers too. Thank you in advance Hans-Joerg From sunshine at public.kherson.ua Fri Oct 29 11:10:56 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 03:12:05 2004 Subject: An administrative request In-Reply-To: <4181F67A.1060302@eva.mpg.de> Message-ID: On 10/29/04 10:51 AM, "Hans-Joerg Bibiko" wrote: > > Dear adminstrators of this list, > > I receive all the interesting mails from this list and not only from this > list. I use a mail > programme with a filter function in order store these lists in different > folders. > > But for the Valentina list I can't declare an easy filter. Would it be > possible to program the list > server to add an unique identifier like e.g. '[VDB]' to the subject field > automatically? I guess > this problem have some other subscribers too. I can do this easy on my side. You can note that field Reply-To contains email Valentina@lists.macserve.net This allow you filter easy -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From bibiko at eva.mpg.de Fri Oct 29 10:24:30 2004 From: bibiko at eva.mpg.de (Hans-Joerg Bibiko) Date: Fri Oct 29 03:29:56 2004 Subject: An administrative request In-Reply-To: References: Message-ID: <4181FE3E.6020508@eva.mpg.de> Hi Ruslan, > You can note that field Reply-To contains email > Valentina@lists.macserve.net Yes, I know but I have to switch my user environment from Farsi to Chinese to English to Japanese etc. The tricky thing is (and I don't know why) if I declare this as a filter it works only for English and Japanese. I figured out that this 'only' works with the subject field correctly. I know it sounds a bit ridiculous but .... ? Either I'm so stupid in my mind or it is a question of the OS. For Linux and MAC there is no problem, but now I have to use a few windows programmes and while using them I want to read my mails too. Hans-Joerg From sunshine at public.kherson.ua Fri Oct 29 11:38:17 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 03:39:25 2004 Subject: An administrative request In-Reply-To: <4181FE3E.6020508@eva.mpg.de> Message-ID: On 10/29/04 11:24 AM, "Hans-Joerg Bibiko" wrote: Hi Hans, >> You can note that field Reply-To contains email >> Valentina@lists.macserve.net > > Yes, I know but I have to switch my user environment from Farsi to Chinese to > English to Japanese > etc. The tricky thing is (and I don't know why) if I declare this as a filter > it works only for > English and Japanese. I figured out that this 'only' works with the subject > field correctly. I know > it sounds a bit ridiculous but .... ? > Either I'm so stupid in my mind or it is a question of the OS. > For Linux and MAC there is no problem, but now I have to use a few windows > programmes and while > using them I want to read my mails too. Please try to find solution on your side. I do not want add prefix because we often use self prefixes on this list [V4RB] [V4MD], ... To differ products. Use several prefixes will be confusing. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From kim at torchpublishing.com.au Fri Oct 29 22:49:52 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 07:50:05 2004 Subject: Automatic Data Entry In-Reply-To: <706D8AB0-2973-11D9-A976-000393DB2B60@pandora.be> Message-ID: G'day Peter > To defend Ruslan, most other databases don't have the complete SQL > explanation in their manual either, ???? I'm not looking for an explanation of SQL - in fact I feel my level of SQL knowledge is well suited to the tasks I need to achieve. My problem is I'm having a lot of trouble following the tutorials because they are so OOP based and my head is so procedural. > I wouldn't put it in a module, but in > a class (modules are like scripts, they're global, but it's not a good > practice to use them if you can use a class for it). ???? I know this is probably off topic but I'm interested to know why this is? Cheers Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From kim at torchpublishing.com.au Fri Oct 29 23:23:44 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 08:23:54 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Ruslan > Right. And Valentina introduce you a NEW PARADIGM RecID fields and > ObjectPtr. And I believe they are better :-) > If you don't know why I will explain ???? I'm listening:) Is the Objectptr just Valentina's equivalent of a foreign key? Some of Valentina's terminology makes comparing it to another database quite confusing. > But we also have improve object model of Valentina in 2.0. > We have add not only RecID but also OID field. > We have add new abstraction LINK into Valentina Database model. ???? Well, it remains to be seen if this will make things easier to understand or not. Of course the problems with Valentina are exacerbated by the fact I'm learning RB at the same time - and still trying to update existing FMP projects, write very complex applescripts for InDesign and updating our PHP/MySQL based web site at the same time - no wonder my head is spinning. Cheers Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From kim at torchpublishing.com.au Fri Oct 29 23:23:52 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 08:24:16 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Ruslan > Well. Show on list your example of RB + mySQL > Few lines of code. ???? OK, display cursor in listbox without binding: TextWindow.ListBox2.deleteAllRows While not rs.eof TextWindow.ListBox2.addrow("") For i = 1 to rs.fieldcount TextWindow.ListBox2.cell(TextWindow.ListBox2.lastIndex,i -1)=rs.IdxField(i).StringValue next rs.MoveNext wend Also, a basic query using primary key - I don't understand how to directly query the RecID when it's a 'hidden' field. if mydb.connect() then s = "select ID, story_name, heading, sub_heading, caption, body, story_writer, section, publication, depth, page, comment from stories where ID='"+str(storyID)+"'" there's others but these come to mind immediately. Cheers Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From kim at torchpublishing.com.au Fri Oct 29 23:23:55 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 08:24:33 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Ruslan > But RecID field > 1) exists always. You need do ZERO steps to get it. ???? I still don't see an easy way to get it via an SQL select but I'm sure if there is you'll demonstrate:) > 2) it provide referential integrity automatically ???? This is good > 3) it is 4 times faster on joins that ID of ULONG type. ???? This is good too > So tell me. WHAT REASONS to use own ID field ??? ???? Easily accessible through SQL select and customisable > Only because all other DBMS do this ? ???? Not necessarily but it would seem to make the code more portable if the developer needed to move to another DBMS that relied on user defined primary keys > Only because you are used to do this during 10 years ? ???? 10 years is a long time:) Cheers Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From sunshine at public.kherson.ua Fri Oct 29 16:29:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 08:30:57 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 4:23 PM, "Kim Kohen" wrote: > G'day Ruslan > >> Right. And Valentina introduce you a NEW PARADIGM RecID fields and >> ObjectPtr. And I believe they are better :-) >> If you don't know why I will explain > ???? > I'm listening:) > > Is the Objectptr just Valentina's equivalent of a foreign key? Analog but this is not EXACTLY the same. Foreign key is CONSTRAINT. ObjectPtr is a real field (actually you can think it is ULONG type) of table. ObjectPtr keep integer values of RecID field. > Some of Valentina's terminology makes comparing it to another database quite > confusing. Kim, I wonder, if you have read ValentinaKernel.pdf ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Fri Oct 29 15:39:42 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Fri Oct 29 08:39:54 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: >> I wouldn't put it in a module, but in >> a class (modules are like scripts, they're global, but it's not a good >> practice to use them if you can use a class for it). > ???? > I know this is probably off topic but I'm interested to know why this > is? > Because one of the basic principles of OO-programming is encapsulation, a module doesn't encapsulate anything. It's from OO point of view bad practice. That's why other OO languages don't have them. You should ask this same question on the REALbasic NUG itself and see the comments you'll get, probably better than mine and more in depth :-) (I don't have the time right now to write pages on good OO programming ;-)) Best regards Peter De Berdt From sunshine at public.kherson.ua Fri Oct 29 16:39:53 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 08:41:01 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 4:23 PM, "Kim Kohen" wrote: > G'day Ruslan > >> Well. Show on list your example of RB + mySQL >> Few lines of code. > ???? > OK, display cursor in listbox without binding: > > TextWindow.ListBox2.deleteAllRows > While not rs.eof > TextWindow.ListBox2.addrow("") > For i = 1 to rs.fieldcount > TextWindow.ListBox2.cell(TextWindow.ListBox2.lastIndex,i > -1)=rs.IdxField(i).StringValue > next > rs.MoveNext > wend dim curs as VCursor Mydb.open( location ) curs = mydb.SqlSelect( s ) TextWindow.ListBox2.deleteAllRows curs.FirstRecord do TextWindow.ListBox2.addrow("") For i = 1 to curs.fieldcount TextWindow.ListBox2.cell(TextWindow.ListBox2.lastIndex,I ?1 ) = curs.Field(i).GetString next until curs.NextRecord() = false > Also, a basic query using primary key - I don't understand how to directly > query the RecID when it's a 'hidden' field. > > if mydb.connect() then > s = "select ID, story_name, heading, sub_heading, caption, body, > story_writer, section, publication, depth, page, comment from stories where > ID='"+str(storyID)+"'" > > there's others but these come to mind immediately. Simply Kim: s = "select RecID, story_name, heading, sub_heading, caption, body, story_writer, section, publication, depth, page, comment from stories where If you need join tables then SELECT .. FROM T1, T2 WHERE T1.RecID = T2.ObjectPtr You can use RecID by name. You can read and compare its values. Just you can be lazy and do not set its values :-) -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From kim at torchpublishing.com.au Fri Oct 29 23:41:37 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 08:41:49 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Ruslan > Kim, I wonder, if you have read ValentinaKernel.pdf ? ???? To be honest I can't remember. You wouldn't believe how much I've read in the past two weeks. Literally thousands of pages on a variety of subjects - MySQL, SQLite. SQLite server, valentina, valentina server and several tutorials and about a million things on RB - including two books and everything on the mailing lists. Now if only I could store all that data in a database:)) Cheers Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From sunshine at public.kherson.ua Fri Oct 29 16:44:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 08:45:47 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: On 10/29/04 4:23 PM, "Kim Kohen" wrote: Hi Kim, >> But RecID field >> 1) exists always. You need do ZERO steps to get it. > ???? > I still don't see an easy way to get it via an SQL select but I'm sure if > there is you'll demonstrate:) SELECT RecID, ... SELECT T1.Recid, T2.Recid FROM T1, T2 WHERE T1.Recid = T2.Ptr Btw, Valentina allow you to be lazy and do NOT specify link conditions for join! SELECT T1.Recid, T2.Recid FROM T1, T2 >> 2) it provide referential integrity automatically > ???? > This is good > >> 3) it is 4 times faster on joins that ID of ULONG type. > ???? > This is good too > >> So tell me. WHAT REASONS to use own ID field ??? > ???? > Easily accessible through SQL select and customisable Ok, more detailed example: dim r1 as integer curs = db.SqlSelect( "SEELCT Recid, f1, f2 FROM T" ) r1 = curs.ULongField("Recid").value // read found value q2 = "INSERT INTO T2 (f1, ptr) VALUES (5," + str(r1) + ")" db.SqlExecute( q2 ) This will create record in T2 which is linked to r1 of T1 table. >> Only because all other DBMS do this ? > ???? > Not necessarily but it would seem to make the code more portable if the > developer needed to move to another DBMS that relied on user defined primary > keys This is why in 2.0, yes, we have implement standard things also. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From peter.de.berdt at pandora.be Fri Oct 29 15:48:01 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Fri Oct 29 08:48:05 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <26A7EE15-29B1-11D9-89AC-000393DB2B60@pandora.be> > Simply Kim: > > s = "select RecID, story_name, heading, sub_heading, caption, body, > story_writer, section, publication, depth, page, comment from stories > where > > > If you need join tables then > > SELECT .. > FROM T1, T2 > WHERE T1.RecID = T2.ObjectPtr > You can even just say "SELECT ... FROM T1,T2" because the ObjectPtr already knows how to join the two tables. > > You can use RecID by name. > And if you need the RecID in the SELECT query, you need to include it in the query, even if you are using SELECT * (it's all in the docs if you have not read them yet): SELECT ..., RecID FROM T1 Best regards Peter De Berdt From kim at torchpublishing.com.au Fri Oct 29 23:53:04 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 08:53:15 2004 Subject: Automatic Data Entry In-Reply-To: <26A7EE15-29B1-11D9-89AC-000393DB2B60@pandora.be> Message-ID: G'day Peter > And if you need the RecID in the SELECT query, you need to include it > in the query, even if you are using SELECT * (it's all in the docs if > you have not read them yet): > SELECT ..., RecID FROM T1 ???? Would I use Select RecID, * from t1 or does it need to be Select *, RecID from t1 or does it not matter which order as long as it's specifically requested? Cheers Kim From peter.de.berdt at pandora.be Fri Oct 29 15:57:20 2004 From: peter.de.berdt at pandora.be (Peter De Berdt) Date: Fri Oct 29 08:57:26 2004 Subject: Automatic Data Entry In-Reply-To: References: Message-ID: <7397D32C-29B2-11D9-89AC-000393DB2B60@pandora.be> On 29 Oct 2004, at 15:53, Kim Kohen wrote: > G'day Peter > >> And if you need the RecID in the SELECT query, you need to include it >> in the query, even if you are using SELECT * (it's all in the docs if >> you have not read them yet): >> SELECT ..., RecID FROM T1 > ???? > Would I use Select RecID, * from t1 or does it need to be > Select *, RecID from t1 or does it not matter which order as long as > it's > specifically requested? > No specific order, it just needs to be specifically requested. Best regards Peter De Berdt From kim at torchpublishing.com.au Sat Oct 30 00:02:59 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Fri Oct 29 09:03:08 2004 Subject: Automatic Data Entry In-Reply-To: Message-ID: G'day Ruslan > SELECT RecID, ... ???? Thanks for all your posts. It's just gone midnight so I won't be doing anything now but I will check them out as best I can tomorrow. Thanks again - I very much appreciate your (and everyone else's) efforts. Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From sunshine at public.kherson.ua Fri Oct 29 18:38:25 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 10:39:38 2004 Subject: Automatic Data Entry In-Reply-To: <7397D32C-29B2-11D9-89AC-000393DB2B60@pandora.be> Message-ID: On 10/29/04 4:57 PM, "Peter De Berdt" wrote: > > On 29 Oct 2004, at 15:53, Kim Kohen wrote: > >> G'day Peter >> >>> And if you need the RecID in the SELECT query, you need to include it >>> in the query, even if you are using SELECT * (it's all in the docs if >>> you have not read them yet): >>> SELECT ..., RecID FROM T1 >> ???? >> Would I use Select RecID, * from t1 or does it need to be >> Select *, RecID from t1 or does it not matter which order as long as >> it's >> specifically requested? >> > No specific order, it just needs to be specifically requested. Yes, as any other virtual field. Kim, you know that in FMP exists virtual fields? Valentina also have them. You can have 10 normal fields, and 10 virtual fields in table. Select * -- returns only normal fields. To be compatible to other RDBMS. Select ** -- return 10 + 10 fields Select RecID, f1, method1, f2, method2 -- you have tell EXACTLY what you way to see. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From seanmancuso at yahoo.com Fri Oct 29 11:26:40 2004 From: seanmancuso at yahoo.com (Sean Mancuso) Date: Fri Oct 29 13:26:47 2004 Subject: I need speed Message-ID: <20041029182640.59614.qmail@web10901.mail.yahoo.com> Hello list, I´m having a bit of a speed problem working on a project where the computer has to generate the starting 11 players of a football team: repeat with I = 1 to 16 giverole tt = 0 fx = "select * from Table WHERE club = "&"'"&avversario&"'"&" AND Titstatus = "&0&" AND Ruolo > "&d1&" AND Ruolo < "&d2&" AND Side LIKE "&drs[I]&" ORDER BY Ability Desc" gMyCursorP = new( xtra "VCursor", GetRef(gMyDBP), fx) recordsT1 = GetRecords(gMyCursorP) end repeat I need to use the * in the selection, but the cursor has to be created 16 times for team and there are 40 teams, on a Pentium4 2.4 MHz it takes about 75 seconds, would you have any tips to do the same thing a bit faster ? __________________________________ Do you Yahoo!? Take Yahoo! Mail with you! Get it on your mobile phone. http://mobile.yahoo.com/maildemo From sunshine at public.kherson.ua Fri Oct 29 22:06:30 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Oct 29 14:07:38 2004 Subject: I need speed In-Reply-To: <20041029182640.59614.qmail@web10901.mail.yahoo.com> Message-ID: On 10/29/04 9:26 PM, "Sean Mancuso" wrote: > Hello list, > I?m having a bit of a speed problem working on a > project where the computer has to generate the > starting 11 players of a football team: > > repeat with I = 1 to 16 > > giverole > tt = 0 > fx = "select * from Table WHERE club > = "&"'"&avversario&"'"&" AND Titstatus = "&0&" AND > Ruolo > "&d1&" AND > Ruolo < "&d2&" AND Side LIKE "&drs[I]&" ORDER BY > Ability Desc" > gMyCursorP = new( xtra "VCursor", GetRef(gMyDBP), fx) > > > recordsT1 = GetRecords(gMyCursorP) > > end repeat > > I need to use the * in the selection, but the cursor > has to be created 16 times for team and there are 40 > teams, on a Pentium4 2.4 MHz it takes about 75 > seconds, would you have any tips to do the same thing > a bit faster ? Hi Sean, Too long. * what cache size ? * how many records in table? * how many records are found by cursors? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From forgete at cafederic.com Fri Oct 29 16:30:11 2004 From: forgete at cafederic.com (Eric Forget) Date: Fri Oct 29 15:30:22 2004 Subject: [VSDK] SetCompareFunction for VDK_Text Message-ID: Hi Ruslan, If you recall a few months ago you added a parameter to SetCompareFunction() of VDK_String to prohibit an automatic rebuild of the index each time an application is launched. Unfortunately, you forgot to add the same to the VDK_Text. Would you mind adding it? ?ric ___________________________________________________________________ Eric Forget Cafederic ForgetE@cafederic.com From sunshine at public.kherson.ua Sat Oct 30 08:36:31 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 30 00:37:39 2004 Subject: [VSDK] SetCompareFunction for VDK_Text In-Reply-To: Message-ID: On 10/29/04 11:30 PM, "Eric Forget" wrote: > Hi Ruslan, > > If you recall a few months ago you added a parameter to SetCompareFunction() > of VDK_String to prohibit an automatic rebuild of the index each time an > application is launched. Unfortunately, you forgot to add the same to the > VDK_Text. > > Would you mind adding it? It is not hard. I will try catch time Eric. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat Oct 30 08:42:40 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 30 00:43:49 2004 Subject: VCSDK compilation problem with Project Bilder In-Reply-To: Message-ID: On 10/30/04 2:05 AM, "joe.farage@dplanet.ch" wrote: > Name: Joe Farage > > Question regarding: vsdk > > Hello, > > I am trying your Valentina database but I cant even compile the given example > file in C language with Project Builder 2.01. > I get a link error during compilation but I add all the header files and .dll > and .lib to the project. > > I get the following error: > > /usr/bin/jam -d1 > JAMBASE=/Developer/Makefiles/pbx_jamfiles/ProjectBuilderJambase JAMFILE=- > build ACTION=build _DEFAULT_GCC_VERSION=3.1 BUILD_STYLE=Development > CPP_HEADERMAP_FILE=/Users/joe/Desktop/valentina/build/valentina.build/valentin > a.build/valentina.hmap SRCROOT=/Users/joe/Desktop/valentina > OBJROOT=/Users/joe/Desktop/valentina/build > SYMROOT=/Users/joe/Desktop/valentina/build DSTROOT=/tmp/valentina.dst > ..updating 6 target(s)... > StandaloneExecutable /Users/joe/Desktop/valentina/build/valentina > ld: can't locate file for: -lVCSDK_Carbon > > /usr/bin/gcc3 -o "/Users/joe/Desktop/valentina/build/valentina" > "-L/Users/joe/Desktop/valentina/build" > "-L/Users/joe/Desktop/VCSDK_Carbon/VCSDK" > "-F/Users/joe/Desktop/valentina/build" > "-F/Users/joe/Desktop/VCSDK_Carbon/VCSDK" -filelist > "/Users/joe/Desktop/valentina/build/valentina.build/valentina.build/Object > s-normal/LinkFileList" "-arch" "ppc" "-prebind" > "-lVCSDK_Carbon" "-lVCSDK_WIN_cw" > > ..failed StandaloneExecutable.LinkUsingFileList > /Users/joe/Desktop/valentina/build/valentina ... > > > So it would be so nice if you can give me a tip to resolve that problem. I > found nothing in your documentation and in forums. > > many thanks and best regards, Hi Joe, Well, I think we even did not try do that this. I mean VCSDK was made with CodeWarrior. In theory it must work with ProjectBuilder and xCode. As I see, problem that it cannot find library. Am I right that you have create PB project self ? I think you need or add LIBRARY PATH to location where is libVCSDK_Carbon Or you need copy this library to path /usr/lib where all unix libs are. Am I right that you have download and use VCSDK_macho archive ? Let me check if it have such library with such name... Aha, note that archive contains VCSDK_Macho.lib. GCC expect to see library with UNIX style name libVCSDK_Macho.lib I think you can rename it self. And I wonder, you have made CARBON or MACHO project ? You need macho for ProjectBuilder. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From seanmancuso at yahoo.com Sat Oct 30 02:19:05 2004 From: seanmancuso at yahoo.com (Sean Mancuso) Date: Sat Oct 30 04:19:09 2004 Subject: I need speed Message-ID: <20041030091905.15301.qmail@web10904.mail.yahoo.com> * what cache size ? ValentinaInit( 4 * 1024 * 1024 * how many records in table? about 8000 * how many records are found by cursors? between 1 to 5 first it selects only the players in the database of the opposing team, then if they are available, then it selects the goalkeeper or attacker depending on the repeat loop state and then the side (left, right) could it be the Side LIKE that slows it down? fx = "select * from Table WHERE club > = "&"'"&avversario&"'"&" AND Titstatus = "&0&" AND > Ruolo > "&d1&" AND > Ruolo < "&d2&" AND Side LIKE "&drs[I]&" ORDER BY > Ability Desc" __________________________________ Do you Yahoo!? Yahoo! Mail Address AutoComplete - You start. We finish. http://promotions.yahoo.com/new_mail From sunshine at public.kherson.ua Sat Oct 30 20:24:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Oct 30 12:25:43 2004 Subject: I need speed In-Reply-To: <20041030091905.15301.qmail@web10904.mail.yahoo.com> Message-ID: On 10/30/04 12:19 PM, "Sean Mancuso" wrote: > * what cache size ? > ValentinaInit( 4 * 1024 * 1024 First of all try 20 MB -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From kim at torchpublishing.com.au Sun Oct 31 18:59:48 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Sun Oct 31 02:00:08 2004 Subject: DB corruption In-Reply-To: Message-ID: G'day Ruslan I've been fiddling around a bit today and have had everything working OK when suddenly I've struck the following error (only visible in valentina Carbon): error 351:VarChar field corrupted Can this be fixed or is it lost? Regards Kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au From sunshine at public.kherson.ua Sun Oct 31 11:01:38 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Oct 31 03:02:45 2004 Subject: DB corruption In-Reply-To: Message-ID: On 10/31/04 10:59 AM, "Kim Kohen" wrote: > G'day Ruslan > > I've been fiddling around a bit today and have had everything working OK > when suddenly I've struck the following error (only visible in valentina > Carbon): > > error 351:VarChar field corrupted > > Can this be fixed or is it lost? Hi Kim, You can try do XML dump then XML load into fresh new db. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From kim at torchpublishing.com.au Sun Oct 31 20:36:00 2004 From: kim at torchpublishing.com.au (Kim Kohen) Date: Sun Oct 31 03:36:12 2004 Subject: DB corruption In-Reply-To: Message-ID: G'day Ruslan > You can try do XML dump then XML load into fresh new db. ???? OK, I exported the data OK (it didn't matter much as it was only a test) but it has me a little concerned that it got corrupt in the first place. Cheers kim Kim Kohen Group Pre-Press and Systems Manager Torch Publishing Company P/L 47 Allingham St Condell Park NSW 2200 Australia 612 9795 0000 http://www.torchpublishing.com.au