From sunshine at public.kherson.ua Wed Dec 1 17:29:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 1 09:29:42 2004 Subject: V4RB b29 uploaded Message-ID: kernel - 2.0 b29 ================ - 0000143: [Planer] There is a wrong result for like on empty value. - 0000140: [Values] Value_text_null does not work correctly. - 0000139: [Tests] Nulable flag and methods. - 0000138: [SQL Parser] Add such operator as Shift left and right SQL Parser - 0000127: [Fields] ChangeType() to TEXT field crashes - 0000137: [NEW] Interface I_String added - 0000136: [Fields] Last record of VarChar field not updated correctly - 0000135: [Fields] Empty values in text fields. - 0000134: [API] + - * / % nodes not correctly work with NULL values - 0000133: [SQL Parser] Added support of syntax != - 0000132: [Methods (calculated fields)] fld.SetMethod() do not works V4RB - 2.0 b29 ============== - 0000119: [SQL] NOT search fails with method - 0000131: [Change] Field.Find() functions and Table funcs return now VArraySet or VBitSet - 0000130: [NEW] VBitSet class - 0000129: [NEW] VArraySet class - 0000124: [API] Wrong enum EVFieldType constant 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 sunshine at public.kherson.ua Wed Dec 1 18:15:18 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 1 10:15:22 2004 Subject: V4RB b29 uploaded In-Reply-To: Message-ID: On 12/1/04 5:58 PM, "Frank Schima" wrote: > Hi Ruslan, > > > Is the Tests project in 2b29 modified to include both of our recent > changes? Hi Frank, I want to inform, that I have check all tests up to SQL part. * now works practically all Methods tests * at least 2 crashes yet NOT fixed in this non-SQL part. -- 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 Dec 1 18:53:52 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 1 10:53:56 2004 Subject: [V4RB 0000144]: Ability of SqlExecute to return array of RecID's affected In-Reply-To: <0bc11a7af832035e453120116eadfcf1@paradigma.ukrcom.kheron.ua> Message-ID: On 12/1/04 6:03 PM, "vbt@paradigma.ukrcom.kherson.ua" wrote: Hi All, I think this have no many sense. IF you want to control the affected records, then you can do this using Cursor API: curs = db.SqlSelect().... // so you have now records which will be affected. curs.UpdateAllRecords() The same effect as UPDATE WHERE, but you have on hands the set of affected records. Works ? > The following issue has been SUBMITTED. > ====================================================================== > > ====================================================================== > Reported By: John Sims > Assigned To: > ====================================================================== > Project: V4RB > Issue ID: 144 > Category: Feature Request > Reproducibility: N/A > Severity: feature > Priority: normal > Status: new > ====================================================================== > Date Submitted: 12-01-2004 18:03 EET > Last Modified: 12-01-2004 18:03 EET > ====================================================================== > Summary: Ability of SqlExecute to return array of RecID's > affected > Description: > If at all possible, I would like the ability for SqlExecute to return an > array of RecID's affected by the Insert/Update statement that was passed. > So, if I send Valentina a statement like "UPDATE Person SET bRetired=TRUE > WHERE Age>65" then I would get an array of all of the RecID's for the > people I just marked as retired. > ====================================================================== > > Issue History > Date Modified Username Field Change > ====================================================================== > 12-01-04 18:03 John Sims New 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 Dec 1 19:14:06 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 1 11:14:08 2004 Subject: V4RB b29, ArraySet class Message-ID: Hi All, Erik, please pay attention that I have made new class ArraySet and BitSet. And Find() methods now return one or other. ArraySet class do have methods to access RecID at any position. -- 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 AYU8 at CDC.GOV Wed Dec 1 12:30:45 2004 From: AYU8 at CDC.GOV (Sims, John) Date: Wed Dec 1 11:30:52 2004 Subject: [V4RB 0000144]: Ability of SqlExecute to return array of RecID's affected Message-ID: > -----Original Message----- > From: valentina-beta-bounces@lists.macserve.net > [mailto:valentina-beta-bounces@lists.macserve.net] On Behalf > Of Ruslan Zasukhin > Sent: Wednesday, December 01, 2004 11:54 AM > To: valentina-beta@lists.macserve.net > Subject: Re: [V4RB 0000144]: Ability of SqlExecute to return > array of RecID's affected > > > On 12/1/04 6:03 PM, "vbt@paradigma.ukrcom.kherson.ua" > wrote: > > Hi All, > > I think this have no many sense. > > IF you want to control the affected records, then you can do > this using Cursor API: > > curs = db.SqlSelect().... > // so you have now records which will be affected. > > curs.UpdateAllRecords() > > > The same effect as UPDATE WHERE, but you have on hands the > set of affected records. > > Works ? > > Hi Ruslan, Yes, using the cursor API would work fine if I was looking at a specific task. However, what I am looking at trying to build is a replication engine for Valentina that I can use for several projects (and make available to others if it comes out well). To do this, I'll need to wrap all of the functionality that Valentina provides in a set of classes that also know how to maintain the replication information. I could make the replication engine work for only the classes way of Valentina but I would like to make it work with the SQL way also. To do this, I would need to know what records are affected by Insert/Update/Delete statements so that I can make appropriate updates to the replication information. Otherwise, I will have to parse the Insert/Update/Delete statements enough to rebuild them with my replication information included in the statement and to determine what other replication information needs to be added/modified. It occurred to me that if Valentina could tell me what records were affected, it would make many of my tasks much simpler --- schema changes will be more interesting :-) I hope I have made sense. As always, thank you so much for you hard work and your willingness to work with your users. Also, as many others have stated, my thoughts are with you and your country as your country tries to resolve its presidential election dispute for the betterment of its people. -John From valentina-list at vermontsoftworks.com Wed Dec 1 16:18:03 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Wed Dec 1 15:18:20 2004 Subject: V4RB b29, ArraySet class In-Reply-To: References: Message-ID: <7C8CF33E-43DE-11D9-BD71-000393DC4766@vermontsoftworks.com> Hi, Ruslan -- On Dec 1, 2004, at 12:14, Ruslan Zasukhin wrote: > Erik, please pay attention that I have made new class ArraySet and > BitSet. > > And Find() methods now return one or other. Yes! I noticed this in the release notes. Thank you!! > ArraySet class do have methods to access RecID at any position. I will examine this in detail first thing tomorrow. I'll let you know how it goes. Many thanks, -- Erik From sunshine at public.kherson.ua Thu Dec 2 00:29:03 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 1 16:29:12 2004 Subject: [V4RB 0000144]: Ability of SqlExecute to return array of RecID's affected In-Reply-To: Message-ID: On 12/1/04 7:30 PM, "Sims, John" wrote: Hi John, > Yes, using the cursor API would work fine if I was looking at a specific > task. However, what I am looking at trying to build is a replication > engine for Valentina that I can use for several projects (and make > available to others if it comes out well). To do this, I'll need to > wrap all of the functionality that Valentina provides in a set of > classes that also know how to maintain the replication information. I > could make the replication engine work for only the classes way of > Valentina but I would like to make it work with the SQL way also. To do > this, I would need to know what records are affected by > Insert/Update/Delete statements Do you realize that this can be big set of data for big db ? Well, on the other hand cursor itself is not smaller. > so that I can make appropriate updates to the replication information. >Otherwise, I will have to parse the > Insert/Update/Delete statements enough to rebuild them with my > replication information Hard task > included in the statement and to determine what > other replication information needs to be added/modified. It occurred > to me that if Valentina could tell me what records were affected, it > would make many of my tasks much simpler --- schema changes will be more > interesting :-) I see. I cannot promise this for 2.0. May be for 2.5, you have add it to mantis, so good. IT seems this will require some new mechanism. Will see. > I hope I have made sense. As always, thank you so much for you hard > work and your willingness to work with your users. Also, as many others > have stated, my thoughts are with you and your country as your country > tries to resolve its presidential election dispute for the betterment of > its people. Thank 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 AYU8 at CDC.GOV Thu Dec 2 07:47:57 2004 From: AYU8 at CDC.GOV (Sims, John) Date: Thu Dec 2 06:48:05 2004 Subject: [V4RB 0000144]: Ability of SqlExecute to return array of RecID's affected Message-ID: Hi Ruslan, Thanks for at least giving this some consideration. It is what make Paradigma Software, and yourself, so special! Take care. -John > -----Original Message----- > From: valentina-beta-bounces@lists.macserve.net > [mailto:valentina-beta-bounces@lists.macserve.net] On Behalf > Of Ruslan Zasukhin > Sent: Wednesday, December 01, 2004 5:29 PM > To: valentina-beta@lists.macserve.net > Subject: Re: [V4RB 0000144]: Ability of SqlExecute to return > array of RecID's affected > > > On 12/1/04 7:30 PM, "Sims, John" wrote: > > Hi John, > > > Yes, using the cursor API would work fine if I was looking at a > > specific task. However, what I am looking at trying to build is a > > replication engine for Valentina that I can use for several > projects > > (and make available to others if it comes out well). To do > this, I'll > > need to wrap all of the functionality that Valentina > provides in a set > > of classes that also know how to maintain the replication > information. > > I could make the replication engine work for only the > classes way of > > Valentina but I would like to make it work with the SQL way > also. To > > do this, I would need to know what records are affected by > > Insert/Update/Delete statements > > Do you realize that this can be big set of data for big db ? > Well, on the other hand cursor itself is not smaller. > > > so that I can make appropriate updates to the replication > information. > > > >Otherwise, I will have to parse the > > Insert/Update/Delete statements enough to rebuild them with my > > replication information > > Hard task > > > included in the statement and to determine what > > other replication information needs to be added/modified. > It occurred > > to me that if Valentina could tell me what records were > affected, it > > would make many of my tasks much simpler --- schema changes will be > > more interesting :-) > > I see. I cannot promise this for 2.0. > May be for 2.5, you have add it to mantis, so good. > > IT seems this will require some new mechanism. > Will see. > > > I hope I have made sense. As always, thank you so much for > you hard > > work and your willingness to work with your users. Also, as many > > others have stated, my thoughts are with you and your > country as your > > country tries to resolve its presidential election dispute for the > > betterment of its people. > > Thank you. > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- From sunshine at public.kherson.ua Fri Dec 3 00:17:35 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 2 16:17:45 2004 Subject: V4RB Reference for 2.0 uploaded Message-ID: Hi All, I have upload V4RBReference.pdf. The first 50 pages are quite polished according to 2.0 API. -- 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 Thu Dec 2 17:23:57 2004 From: jda at his.com (jda) Date: Thu Dec 2 16:24:05 2004 Subject: V4RB Reference for 2.0 uploaded In-Reply-To: References: Message-ID: >Hi All, > >I have upload V4RBReference.pdf. > >The first 50 pages are quite polished according to 2.0 API. > I see PathToICU and PathToVComponents are documented. Are they implemented yet? Jon From sunshine at public.kherson.ua Fri Dec 3 00:24:59 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 2 16:25:05 2004 Subject: V4RB Reference for 2.0 uploaded In-Reply-To: Message-ID: On 12/3/04 12:23 AM, "jda" wrote: >> Hi All, >> >> I have upload V4RBReference.pdf. >> >> The first 50 pages are quite polished according to 2.0 API. >> > > I see PathToICU and PathToVComponents are documented. Are they implemented > yet? They exists, but I have not test them 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 robert at formworkdesign.com Thu Dec 2 21:17:45 2004 From: robert at formworkdesign.com (Robert Nichols) Date: Thu Dec 2 20:17:54 2004 Subject: V4RB Reference for 2.0 uploaded In-Reply-To: References: Message-ID: <85038F81-44D1-11D9-9995-000A95D01D92@formworkdesign.com> I can't find this doc. What's the link? On Dec 2, 2004, at 5:17 PM, Ruslan Zasukhin wrote: > Hi All, > > I have upload V4RBReference.pdf. > > The first 50 pages are quite polished according to 2.0 API. > > > -- > 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-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > From sunshine at public.kherson.ua Fri Dec 3 11:00:49 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 3 03:01:02 2004 Subject: V4RB Reference for 2.0 uploaded In-Reply-To: <85038F81-44D1-11D9-9995-000A95D01D92@formworkdesign.com> Message-ID: On 12/3/04 4:17 AM, "Robert Nichols" wrote: > I can't find this doc. What's the link? http://paradigmasoft.com/download/v2_beta/v4rb_beta.html -- 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 Dec 3 11:08:04 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 3 03:08:43 2004 Subject: [V4RB 0000148]: Crash if icudt30b.dat isn't in VComponents folder In-Reply-To: <0a9f2e725f7ab031e567cf7b5a4e1e0d@paradigma.ukrcom.kheron.ua> Message-ID: On 12/3/04 1:42 AM, "vbt@paradigma.ukrcom.kherson.ua" wrote: > > The following issue has been SUBMITTED. > ====================================================================== > > ====================================================================== > Reported By: Jon Ashwell > Assigned To: > ====================================================================== > Project: V4RB > Issue ID: 148 > Category: API > Reproducibility: always > Severity: crash > Priority: normal > Status: new > ====================================================================== > Date Submitted: 12-03-2004 01:42 EET > Last Modified: 12-03-2004 01:42 EET > ====================================================================== > Summary: Crash if icudt30b.dat isn't in VComponents folder > Description: > I'm trying to set the pathtoicu like this > > f = > volume(0).child("usr").child("share").child("icu").child("icudt26b.dat") > Valentina.pathToICU = f > > But if I remove icudt30b from VComponents, it crashes on launch. The crash > occurs *before* Valentina.init. Changing the icu path isn't of much use if > we still have to ship the entire .dat file. Frankly saying this will be not so easy task, John. Deal is that we ALREADY use ICU 3.0, But MacOS Panther still use 2.8. I believe they will not be compatible. Only OS X Tiger will get ICU 3.0 In ideal, I think we will try future redo to use only C API of ICU, In this case, as point Apple engineer, we will get big chances to use SYSTEM ICU libs. ALL libs I mean. He say that Apple do work on this with IBM to polish ICU to allow such usage -- 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 Fri Dec 3 07:56:57 2004 From: jda at his.com (jda) Date: Fri Dec 3 06:57:02 2004 Subject: [V4RB 0000148]: Crash if icudt30b.dat isn't in VComponents folder In-Reply-To: References: Message-ID: >Frankly saying this will be not so easy task, John. > >Deal is that we ALREADY use ICU 3.0, >But MacOS Panther still use 2.8. >I believe they will not be compatible. > >Only OS X Tiger will get ICU 3.0 > > >In ideal, I think we will try future redo to use only C API of ICU, >In this case, as point Apple engineer, we will get big chances to use SYSTEM >ICU libs. ALL libs I mean. He say that Apple do work on this with IBM to >polish ICU to allow such usage > I see. Obviously, using the system ICU library is the best way to go -- it will greatly reduce the size of our downloads. But the problem now shifts to PathToVComponents, no? We can't change the location of the VComponents folder without having our apps crash. Did you hard-link to libraries in this file? Is that why my project crashes even before Valentina.Init (without a crash log)? Jon From sunshine at public.kherson.ua Fri Dec 3 15:02:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 3 07:02:58 2004 Subject: [V4RB 0000148]: Crash if icudt30b.dat isn't in VComponents folder In-Reply-To: Message-ID: On 12/3/04 2:56 PM, "jda" wrote: > But the problem now shifts to PathToVComponents, no? We can't change > the location of the VComponents folder without having our apps crash. > Did you hard-link to libraries in this file? Is that why my project > crashes even before Valentina.Init (without a crash log)? Yes this is important task. We will make it working -- 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 AYU8 at CDC.GOV Fri Dec 3 08:35:46 2004 From: AYU8 at CDC.GOV (Sims, John) Date: Fri Dec 3 07:41:28 2004 Subject: V4RB Reference for 2.0 uploaded Message-ID: Hi Ruslan, > Hi All, > > I have upload V4RBReference.pdf. > > The first 50 pages are quite polished according to 2.0 API. > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] I saw this in the new manual: SQLSelect( inQuery as String, inCursorLocation as EVCursorLocation = kClientSide, inLockType as EVLockType = kReadOnly, inDirection as EVCursorDirection = kForwardOnly, inBinds() as String = nil ) As Vcursor ... And I noticed that you have added "inBinds() as String". 1) I'm assuming this means we can now use array binding when selecting data from the database? 2) What part(s) of the SQL statement understand array binding? Is it just in the WHERE clause? 3) Is this already implemented in 2.0 betas and I have just overlooked it? --- very, very possible :-) Thanks -John From sunshine at public.kherson.ua Fri Dec 3 15:53:54 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 3 07:54:04 2004 Subject: V4RB Reference for 2.0 uploaded In-Reply-To: Message-ID: On 12/3/04 3:35 PM, "Sims, John" wrote: Hi John, > I saw this in the new manual: > > SQLSelect( inQuery as String, inCursorLocation as EVCursorLocation = > kClientSide, inLockType as EVLockType = kReadOnly, inDirection as > EVCursorDirection = kForwardOnly, inBinds() as String = nil ) As Vcursor > > ... And I noticed that you have added "inBinds() as String". > > 1) I'm assuming this means we can now use array binding when selecting > data from the database? Yes. Example must be in "SQL way" folder > 2) What part(s) of the SQL statement understand array binding? Is it > just in the WHERE clause? You can bind ANY constant expr at any place of query. Just as I understand, constants can be in WHERE, HAVING, > 3) Is this already implemented in 2.0 betas and I have just overlooked > it? --- very, very possible :-) Yes it works in beta far far ago. Please take a look at SQL grammar of 2.0. Link is in the same place where docs. Everything you see in grammar already works. P.S. NOT IMPLEMENTED only sorting of JOIN TABLE. -- 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 AYU8 at CDC.GOV Fri Dec 3 10:33:53 2004 From: AYU8 at CDC.GOV (Sims, John) Date: Fri Dec 3 09:48:21 2004 Subject: V4RB Reference for 2.0 uploaded Message-ID: Thanks, Ruslan! I'll try to do a better job of researching the new features in 2.0 :-) Keep up the good work. -John > -----Original Message----- > From: valentina-beta-bounces@lists.macserve.net > [mailto:valentina-beta-bounces@lists.macserve.net] On Behalf > Of Ruslan Zasukhin > Sent: Friday, December 03, 2004 8:54 AM > To: valentina-beta@lists.macserve.net > Subject: Re: V4RB Reference for 2.0 uploaded > > > On 12/3/04 3:35 PM, "Sims, John" wrote: > > Hi John, > > > I saw this in the new manual: > > > > SQLSelect( inQuery as String, inCursorLocation as > EVCursorLocation = > > kClientSide, inLockType as EVLockType = kReadOnly, inDirection as > > EVCursorDirection = kForwardOnly, inBinds() as String = nil ) As > > Vcursor > > > > ... And I noticed that you have added "inBinds() as String". > > > > 1) I'm assuming this means we can now use array binding > when selecting > > data from the database? > > Yes. Example must be in "SQL way" folder > > > > 2) What part(s) of the SQL statement understand array > binding? Is it > > just in the WHERE clause? > > You can bind ANY constant expr at any place of query. > > Just as I understand, constants can be in WHERE, HAVING, > > > > 3) Is this already implemented in 2.0 betas and I have just > overlooked > > it? --- very, very possible :-) > > Yes it works in beta far far ago. > > Please take a look at SQL grammar of 2.0. > Link is in the same place where docs. > > Everything you see in grammar already works. > > > > P.S. NOT IMPLEMENTED only sorting of JOIN TABLE. > > > -- > 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-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > From sunshine at public.kherson.ua Wed Dec 8 01:08:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 7 17:08:55 2004 Subject: [V4RB 2.0 b30] uploaded Message-ID: Hi All, You can download new beta. kernel - 2.0 b30 ================ - 0000169: [Error codes/classes] ERROR: 0x59502 File "%S" does not exist. (Igor Gomon) - 0000180: [SQL Parser] Now works syntaax CREATE TABLE T (fld type NULL) (Sergey Duminskiy) - 0000179: [SQL nodes] Command REINDEX now works (Ruslan Zasukhin) - 0000178: [Error codes/classes] Error: "Table %S Not found" not show name of table (Ruslan Zasukhin) - 0000175: [Tests] Leaks in String_ICU class. (Igor Gomon) - 0000173: [Cursors] Wrong (unfilled) table name in error message for ERR_TABLE_RECORD_LOCKED. (Ivan Smahin) - 0000168: [Import/Export] Export() does not work correctly. (Igor Gomon) - 0000163: [SQL Parser] INSERT INTO ... VALUES( NULL ) parser error near '('. (Sergey Duminskiy) - 0000162: [Indexes] Byte Index cause crash (Ruslan Zasukhin) - 0000128: [Fields] ChangeType() to VarChar produce empty strings (Igor Gomon) - 0000153: [XML dump] XML dump do not load Method text. (Igor Gomon) - 0000161: [SQL] WHERE fld = 'date str' not works (Ivan Smahin) - 0000159: [SQL Parser] Parser do not throw ANTLR exceptions outside (Ruslan Zasukhin) - 0000118: [Planer] NOT( something) always scan (Sergey Duminskiy) - 0000117: [SQL] WHERE fld NOT REGEX 'str' not works (Sergey Duminskiy) - 0000158: [Fields] Updating of VarChar field records with randomly sized value fails. (Igor Gomon) - 0000157: [Fields] Append of records with random sizes to VarChar field fails. (Igor Gomon) - 0000154: [SQL nodes] SELECT * FROM T select methods also. (Ivan Smahin) - 0000152: [XML dump] XML load not correctly set segment for Text, Picture fields. (Ruslan Zasukhin) - 0000151: [Cursors] Crash if cursor is not destroyed before database close. (Ivan Smahin) - 0000150: [FBL Nodes] Empty error-info in Date-time functions. (Ivan Smahin) - 0000147: [Fields] Date and DateTime fields - put_String method problems. (Ivan Smahin) - 0000146: [Error codes/classes] Error message is not returned. (Ruslan Zasukhin) V4RB - 2.0 b30 ============== - 0000177: [API] Corrected VSet.Include() as Boolean method (Ruslan Zasukhin) - 0000176: [Examples] Crash in Do_VarChar_Test() test. (Igor Gomon) - 0000170: [Examples] Week SQL function. (Ivan Smahin) - 0000112: [Examples] Changing Field Type from Short to Text causes Crash (Ivan Smahin) - 0000123: [Examples] fld.IsIndexed = true causes crash (Ruslan Zasukhin) - 0000114: [Examples] Table.AddRecord Crashes in _Test project (Ruslan Zasukhin) - 0000155: [Examples] Do_SQLQueries_Test(), Byte field loose Nullable flag (Ivan Smahin) - 0000160: [Change] Constructors of fields get one more parameter -- inMethod (Ruslan Zasukhin) - 0000156: [API] .droprecord -> delete record from baseobject (Ivan Smahin) -- 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 ------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20041208/ded6a5a1/attachment.html From frank-list3 at mindstarprods.com Wed Dec 8 13:05:55 2004 From: frank-list3 at mindstarprods.com (Frank Schima) Date: Wed Dec 8 14:06:44 2004 Subject: [V4RB] VVarchar and Language Message-ID: <91B19C4A-4954-11D9-843B-000D932CA350@mindstarprods.com> Hi Ruslan, I'm testing b30 and I'm seeing a new behavior in the VVarchar Class. When defining a Varchar field using the API method, I cannot specify a language anymore: So this used to work: vDescription = New VVarChar("Description", 504, "ASCII", kV_Nullable) But now it only compiles when I remove the language parameter. vDescription = New VVarChar("Description", 504, kV_Nullable) The new docs say it should be an optional parameter like it always used to be. How can I specify the language in the constructor? Workaround is to set it afterwards, i.e: vDescription.Language = "ASCII" Related question: How would I specify the language if using SQL to create the table? CREATE TABLE myTable (Description Varchar(504)) Final question: What is the default language when none is specified? Best regards, Frank Schima From sunshine at public.kherson.ua Wed Dec 8 22:16:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 8 14:16:38 2004 Subject: [V4RB] VVarchar and Language In-Reply-To: <91B19C4A-4954-11D9-843B-000D932CA350@mindstarprods.com> Message-ID: On 12/8/04 10:05 PM, "Frank Schima" wrote: Hi Frank, > I'm testing b30 and I'm seeing a new behavior in the VVarchar Class. > When defining a Varchar field using the API method, I cannot specify a > language anymore: > > So this used to work: > vDescription = New VVarChar("Description", 504, "ASCII", kV_Nullable) > > But now it only compiles when I remove the language parameter. > vDescription = New VVarChar("Description", 504, kV_Nullable) It is removed > The new docs say it should be an optional parameter like it always used > to be. I have point that only 50 pages are polished. I think now I can upload document with 120 pages polished. > How can I specify the language in the constructor? Forget about it. Instead of language for each field, now you you define once db.Locale = "en_US" db.StorageEncoding = "UTF-16" // on default. I still self do not see where Locale make affect. Deal is that it seems Valentina do indexing, sorting and searching correctly nevertheless of this parameter. It looks that it affect only some Date/Time/Money operations in the ICU library (which we not use at least now). > Workaround is to set it afterwards, i.e: > vDescription.Language = "ASCII" > > Related question: > How would I specify the language if using SQL to create the table? > > CREATE TABLE myTable (Description Varchar(504)) > > Final question: > What is the default language when none is specified? For UTF16 it practically does not mater. Try just do nothing :-) and see if this works 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 jda at his.com Wed Dec 8 15:55:31 2004 From: jda at his.com (jda) Date: Wed Dec 8 14:55:39 2004 Subject: [V4RB] VVarchar and Language In-Reply-To: References: Message-ID: > How would I specify the language if using SQL to create the table? > > >> CREATE TABLE myTable (Description Varchar(504)) >> >> Final question: >> What is the default language when none is specified? > >For UTF16 it practically does not mater. > >Try just do nothing :-) >and see if this works for you. > One caveat -- I think we agreed that the length parameter is in bytes, not characters, regardless of whether UTF-16 or UTF-8 is used as the storageEncoding. This is to avoid confusion as to how much memory is actually allocated when we use different encodings. Is this correct? Jon From sunshine at public.kherson.ua Wed Dec 8 23:14:44 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 8 15:14:51 2004 Subject: [V4RB] VVarchar and Language In-Reply-To: Message-ID: On 12/8/04 10:55 PM, "jda" wrote: >> How would I specify the language if using SQL to create the table? >>> >>> CREATE TABLE myTable (Description Varchar(504)) >>> >>> Final question: >>> What is the default language when none is specified? >> >> For UTF16 it practically does not mater. >> >> Try just do nothing :-) >> and see if this works for you. >> > > One caveat -- I think we agreed that the length parameter is in > bytes, not characters, regardless of whether UTF-16 or UTF-8 is used > as the storageEncoding. This is to avoid confusion as to how much > memory is actually allocated when we use different encodings. Is this > correct? No Jon. After discussion on list we have think here more on this. It is better to use chars. This is more natural way. This follow to SQL92 standard also. We have check other respected DBMS vendors, we have read their docs. Nobody use bytes. And who use unicode also get NOT easy rules. We offer next rules: UTF16 N chars. Each char is 2 bytes N chars = 2*N bytes single byte encodings N chars. Each char east 1 byte such as N chars = 1*N bytes Latin1, Cyrilic 1251 ... EXCEPTION: UTF8 N bytes = N chars for Western N bytes = N/2 chars for others -- 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 IvanSmahin at public.kherson.ua Thu Dec 9 02:15:51 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Wed Dec 8 18:15:59 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. Message-ID: <1659270105.20041209021551@public.kherson.ua> Hello All, This is short info about usage of DateTime features in Valentina 2. 1. There are three datetime types in Valentina 2: Date, Time and DateTime. As far as I know, storing milliseconds in time and datetime fields is new to version 1. Important feature is "year auto correction". It means you can set such database params as CenturyBound and Century for getting proper dates, putting ones in "short-year" forms. Also you can set own date and time separators and choose date format for comfortable usage. It is possible to change those params at any time. ( Please note - you should use current datetime format in any queries, expected values and methods' text. ) 2. "date", "time", "datetime" keywords. What is it for. Where it must be used and so on. The first two keywords are from SQL-92 standard, the third is Valentina SQL-syntax extension. You HAVE TO use these keywords in methods predicates in case you need to define datetime-literal. Example: There are samples of method's text: "fld1 <> date'2004-01-01'" "fld2 = time'16:00:00'" "fld3 < datetime'2004-01-01 01:00:00'" Also, you CAN use these keywords in any other places. For example: "YEAR(date'2004-01-01')" "... where fld1 > date'2004-01-01' " but "YEAR('2004-01-01')" "... where fld1 > '2004-01-01' " are also allowable. Strict usage of such keywords must be only in methods' predicates based on datetime-literals. 3. There is powerful set of datetime SQL-functions in Valentina 2. I would say this is a most powerful set in DBMS world. -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From Claudius at sailer-online.de Thu Dec 9 06:31:14 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Wed Dec 8 23:31:26 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <1659270105.20041209021551@public.kherson.ua> References: <1659270105.20041209021551@public.kherson.ua> Message-ID: <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> Hi, Am 09. Dez 2004 um 01:15 Uhr schrieb Ivan Smahin: > Strict usage of such keywords must be only in methods' > predicates based on > datetime-literals. > > > 3. There is powerful set of datetime SQL-functions in Valentina 2. > I would say this is a most powerful set in DBMS world. is it also planed to install a function/method weekoftheyear? thanks for help Claudius -- G4/733 QS / MacOS X 10.3.5de / RB 5.5.3de/ Valentina 1.10.0 Homepage http://www.ClaSai.de From sunshine at public.kherson.ua Thu Dec 9 09:52:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 01:52:10 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> Message-ID: On 12/9/04 7:31 AM, "Claudius Sailer" wrote: > Hi, > > Am 09. Dez 2004 um 01:15 Uhr schrieb Ivan Smahin: > >> Strict usage of such keywords must be only in methods' >> predicates based on >> datetime-literals. >> >> >> 3. There is powerful set of datetime SQL-functions in Valentina 2. >> I would say this is a most powerful set in DBMS world. > > is it also planed to install a function/method weekoftheyear? Hi Claudius, I am sure this function was in Valentina 1.x also. Right ? And yes it exactly exists in 2.0. I need upload PDF with all functions from 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 sunshine at public.kherson.ua Thu Dec 9 09:53:17 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 01:53:22 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <1659270105.20041209021551@public.kherson.ua> Message-ID: On 12/9/04 2:15 AM, "Ivan Smahin" wrote: > Hello All, > > This is short info about usage of DateTime features in Valentina 2. > > 1. There are three datetime types in Valentina 2: > Date, Time and DateTime. > > As far as I know, storing milliseconds in time and datetime fields > is new to version 1. > > Important feature is "year auto correction". It means you can set > such database params as CenturyBound and Century for getting proper > dates, putting ones in "short-year" forms. > > Also you can set own date and time separators and choose date format > for comfortable usage. It is possible to change those params at any > time. > ( Please note - you should use current datetime format in any queries, > expected values and methods' text. ) > > > 2. "date", "time", "datetime" keywords. What is it for. > Where it must be used and so on. > > The first two keywords are from SQL-92 standard, the third is > Valentina > SQL-syntax extension. > > You HAVE TO use these keywords in methods predicates > in case you need to define datetime-literal. > > Example: > There are samples of method's text: > "fld1 <> date'2004-01-01'" > "fld2 = time'16:00:00'" > "fld3 < datetime'2004-01-01 01:00:00'" Question: how in literal we specify the milliseconds? It seems after DOT? "fld3 < datetime'2004-01-01 01:00:00.123'" > Also, you CAN use these keywords in any other places. For example: > > "YEAR(date'2004-01-01')" > "... where fld1 > date'2004-01-01' " > > but > > "YEAR('2004-01-01')" > "... where fld1 > '2004-01-01' " > > are also allowable. > > Strict usage of such keywords must be only in methods' predicates > based on > datetime-literals. > > > 3. There is powerful set of datetime SQL-functions in Valentina 2. > I would say this is a most powerful set in DBMS world. > > -- 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 IvanSmahin at public.kherson.ua Thu Dec 9 12:58:37 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 9 04:58:42 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> References: <1659270105.20041209021551@public.kherson.ua> <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> Message-ID: <1181592915.20041209125837@public.kherson.ua> Hello Claudius, Thursday, December 9, 2004, 7:31:14 AM, you wrote: CS> Hi, CS> Am 09. Dez 2004 um 01:15 Uhr schrieb Ivan Smahin: >> Strict usage of such keywords must be only in methods' >> predicates based on >> datetime-literals. >> >> >> 3. There is powerful set of datetime SQL-functions in Valentina 2. >> I would say this is a most powerful set in DBMS world. CS> is it also planed to install a function/method weekoftheyear? CS> thanks for help CS> Claudius It is the same to WEEK function. Week( date, [firstWeekDay] ) Returns the week of year in the range of 0 - 53 firstWeekDay == 0 - is for Sunday 1 is for Monday and so on... up to 6 (Sat) -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From IvanSmahin at public.kherson.ua Thu Dec 9 13:00:44 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 9 05:00:47 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: References: <1659270105.20041209021551@public.kherson.ua> Message-ID: <491284132.20041209130044@public.kherson.ua> Hello Ruslan, Thursday, December 9, 2004, 9:53:17 AM, you wrote: RZ> On 12/9/04 2:15 AM, "Ivan Smahin" RZ> wrote: RZ> Question: how in literal we specify the milliseconds? RZ> It seems after DOT? RZ> "fld3 < datetime'2004-01-01 01:00:00.123'" Yes. '01:00:00.123' - the last three digits is milliseconds. You can omit ms in literals. -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From IvanSmahin at public.kherson.ua Thu Dec 9 13:09:23 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 9 05:09:25 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <491284132.20041209130044@public.kherson.ua> References: <1659270105.20041209021551@public.kherson.ua> <491284132.20041209130044@public.kherson.ua> Message-ID: <1221281056.20041209130923@public.kherson.ua> Hello Ivan, Thursday, December 9, 2004, 1:00:44 PM, you wrote: IS> Hello Ruslan, IS> Thursday, December 9, 2004, 9:53:17 AM, you wrote: RZ>> On 12/9/04 2:15 AM, "Ivan Smahin" RZ>> wrote: RZ>> Question: how in literal we specify the milliseconds? RZ>> It seems after DOT? RZ>> "fld3 < datetime'2004-01-01 01:00:00.123'" IS> Yes. IS> '01:00:00.123' - the last three digits is milliseconds. IS> You can omit ms in literals. Sorry - it must be ':' as delimiter. '01:00:00:123' -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From ernestogiannotta at tiscalinet.it Thu Dec 9 16:53:29 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 9 09:53:38 2004 Subject: test In-Reply-To: Message-ID: test -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From Claudius at sailer-online.de Thu Dec 9 17:59:06 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Thu Dec 9 10:59:16 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: References: Message-ID: Hi Ruslan, Am 09. Dez 2004 um 08:52 Uhr schrieb Ruslan Zasukhin: >> Am 09. Dez 2004 um 01:15 Uhr schrieb Ivan Smahin: >> >>> Strict usage of such keywords must be only in methods' >>> predicates based on >>> datetime-literals. >>> >>> >>> 3. There is powerful set of datetime SQL-functions in Valentina >>> 2. >>> I would say this is a most powerful set in DBMS world. >> >> is it also planed to install a function/method weekoftheyear? > > I am sure this function was in Valentina 1.x also. > Right ? > > And yes it exactly exists in 2.0. > > I need upload PDF with all functions from 2.0 sorry that I tried to asked this. You can see other eMail from me that problem is not so trivial. At the moment I don't use methods, because I am waiting for functions in SQL-Statement like SELECT WEKKOFYEAR(date) I have no time to test week results at the moment in methods, I want to wait until I can use functions. I am not in a hurry, but it would be unusable for me to get an US-Week as result. Because than I have same result as in RealBasic and have to write my own function, where I didn't have the speed I would sugest from valentina. bye Claudius -- G4/733 QS / MacOS X 10.3.5de / RB 5.5.3de/ Valentina 1.10.0 Homepage http://www.ClaSai.de Download finale Versionen, Betaversionen auf der Homepage RealBasic ListBoxes: [ I feel the need...the need for speed!!! ] From Claudius at sailer-online.de Thu Dec 9 17:59:09 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Thu Dec 9 10:59:18 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <1181592915.20041209125837@public.kherson.ua> References: <1659270105.20041209021551@public.kherson.ua> <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> <1181592915.20041209125837@public.kherson.ua> Message-ID: HI Ivan, Am 09. Dez 2004 um 11:58 Uhr schrieb Ivan Smahin: > CS> Hi, > > CS> Am 09. Dez 2004 um 01:15 Uhr schrieb Ivan Smahin: > >>> Strict usage of such keywords must be only in methods' >>> predicates based on >>> datetime-literals. >>> >>> >>> 3. There is powerful set of datetime SQL-functions in Valentina >>> 2. >>> I would say this is a most powerful set in DBMS world. > > CS> is it also planed to install a function/method weekoftheyear? > > It is the same to WEEK function. > Week( date, [firstWeekDay] ) > Returns the week of year in the range of 0 - 53 > firstWeekDay == 0 - is for Sunday > 1 is for Monday and so on... up to 6 (Sat) Thanks, but this is my problem. IMHO do we have several calculation methods. The US-Method says 1.1.YYYY ist week 1 using DIN week 1 ist where the first thursday is in. So you can have 01.01.XXXX in week 52 or week 53 by DIN but week 1 in US Method. The information when is a new week starting is a good thing, because US-week starts on sunday and by DIN on Monday. But after that, what is calculationMethod in Valentina? Thanks for help Claudius -- G4/733 QS / MacOS X 10.3.5de / RB 5.5.3de/ Valentina 1.10.0 Homepage http://www.ClaSai.de Download finale Versionen, Betaversionen auf der Homepage RealBasic ListBoxes: [ I feel the need...the need for speed!!! ] From sunshine at public.kherson.ua Thu Dec 9 19:17:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 11:17:39 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: Message-ID: On 12/9/04 6:59 PM, "Claudius Sailer" wrote: >>>> Strict usage of such keywords must be only in methods' >>>> predicates based on >>>> datetime-literals. >>>> >>>> >>>> 3. There is powerful set of datetime SQL-functions in Valentina >>>> 2. >>>> I would say this is a most powerful set in DBMS world. >>> >>> is it also planed to install a function/method weekoftheyear? >> >> I am sure this function was in Valentina 1.x also. >> Right ? >> >> And yes it exactly exists in 2.0. >> >> I need upload PDF with all functions from 2.0 > > sorry that I tried to asked this. > You can see other eMail from me that problem is not so trivial. At the > moment I don't use methods, because I am waiting for functions in > SQL-Statement like > > SELECT WEKKOFYEAR(date) Aha, yes this works in 2.0 > I have no time to test week results at the moment in methods, I want to > wait until I can use functions. You can in 2.0 beta. Right now > I am not in a hurry, but it would be unusable for me to get an US-Week > as result. Ivan have point that you can specify which day to consider as FIRST DAY of week. > Because than I have same result as in RealBasic and have to > write my own function, where I didn't have the speed I would sugest > from 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 Thu Dec 9 19:19:16 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 11:19:21 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: Message-ID: On 12/9/04 6:59 PM, "Claudius Sailer" wrote: >> It is the same to WEEK function. >> Week( date, [firstWeekDay] ) >> Returns the week of year in the range of 0 - 53 >> firstWeekDay == 0 - is for Sunday >> 1 is for Monday and so on... up to 6 (Sat) > > Thanks, but this is my problem. IMHO do we have several calculation > methods. > The US-Method says 1.1.YYYY ist week 1 using DIN week 1 ist > where the first thursday is in. So you can have 01.01.XXXX in week 52 > or week 53 by DIN but week 1 in US Method. The information when is a > new week starting is a good thing, because US-week starts on sunday and > by DIN on Monday. > But after that, what is calculationMethod in Valentina? Claudius, Please give us exact example. Ivan will check. Or you can check self right now with latest V4RB. Give us some date, what you get in RB what you get in 1.x what YOU WANT to get. -- 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 Claudius at sailer-online.de Thu Dec 9 18:39:42 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Thu Dec 9 11:39:52 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: References: Message-ID: <4EE0A044-4A09-11D9-80B7-00039365848C@sailer-online.de> Am 09. Dez 2004 um 18:19 Uhr schrieb Ruslan Zasukhin: > On 12/9/04 6:59 PM, "Claudius Sailer" > wrote: > >>> It is the same to WEEK function. >>> Week( date, [firstWeekDay] ) >>> Returns the week of year in the range of 0 - 53 >>> firstWeekDay == 0 - is for Sunday >>> 1 is for Monday and so on... up to 6 (Sat) >> >> Thanks, but this is my problem. IMHO do we have several calculation >> methods. > >> The US-Method says 1.1.YYYY ist week 1 using DIN week 1 ist >> where the first thursday is in. So you can have 01.01.XXXX in week 52 >> or week 53 by DIN but week 1 in US Method. The information when is a >> new week starting is a good thing, because US-week starts on sunday >> and >> by DIN on Monday. >> But after that, what is calculationMethod in Valentina? > > Claudius, > > Please give us exact example. Ivan will check. > Or you can check self right now with latest V4RB. > > Give us some date, > > what you get in RB > what you get in 1.x > > what YOU WANT to get. Okay can I what version of viSQL works with 2.0? I can make an example in RB where you can see differences in US and DIN, would this help? It is an MacOS X executable, but I can also send code!! thanks for help Claudius From Claudius at sailer-online.de Thu Dec 9 19:24:53 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Thu Dec 9 12:25:01 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <4EE0A044-4A09-11D9-80B7-00039365848C@sailer-online.de> References: <4EE0A044-4A09-11D9-80B7-00039365848C@sailer-online.de> Message-ID: <9EF30020-4A0F-11D9-80B7-00039365848C@sailer-online.de> Am 09. Dez 2004 um 18:39 Uhr schrieb Claudius Sailer: > > Am 09. Dez 2004 um 18:19 Uhr schrieb Ruslan Zasukhin: > >> On 12/9/04 6:59 PM, "Claudius Sailer" >> wrote: >> >>>> It is the same to WEEK function. >>>> Week( date, [firstWeekDay] ) >>>> Returns the week of year in the range of 0 - 53 >>>> firstWeekDay == 0 - is for Sunday >>>> 1 is for Monday and so on... up to 6 (Sat) >>> >>> Thanks, but this is my problem. IMHO do we have several calculation >>> methods. >> >>> The US-Method says 1.1.YYYY ist week 1 using DIN week 1 ist >>> where the first thursday is in. So you can have 01.01.XXXX in week 52 >>> or week 53 by DIN but week 1 in US Method. The information when is a >>> new week starting is a good thing, because US-week starts on sunday >>> and >>> by DIN on Monday. >>> But after that, what is calculationMethod in Valentina? >> >> Claudius, >> >> Please give us exact example. Ivan will check. >> Or you can check self right now with latest V4RB. >> >> Give us some date, >> >> what you get in RB >> what you get in 1.x >> >> what YOU WANT to get. > > Okay can I what version of viSQL works with 2.0? Okay I found it and compiled it,. then I tried this new database --> Database1 was created I type create table Test(Datum date) commit Pressing SQL-Button everything is fine insert into Test(Datum) values("01.01.2004") ViSQL crashes with errormessage and was after restarting not able to open database. By trying to open database ViSQl crashes without an errormessage. *hmmm* Surely I made some mistakes ;-)) Sorry Claudius From IvanSmahin at public.kherson.ua Thu Dec 9 20:27:50 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 9 12:27:53 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: References: <1659270105.20041209021551@public.kherson.ua> <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> <1181592915.20041209125837@public.kherson.ua> Message-ID: <1632955301.20041209202750@public.kherson.ua> Hello Claudius, Thursday, December 9, 2004, 6:59:09 PM, you wrote: CS> HI Ivan, CS> Am 09. Dez 2004 um 11:58 Uhr schrieb Ivan Smahin: >> CS> Hi, >> >> CS> Am 09. Dez 2004 um 01:15 Uhr schrieb Ivan Smahin: >> >>>> Strict usage of such keywords must be only in methods' >>>> predicates based on >>>> datetime-literals. >>>> >>>> >>>> 3. There is powerful set of datetime SQL-functions in Valentina >>>> 2. >>>> I would say this is a most powerful set in DBMS world. >> >> CS> is it also planed to install a function/method weekoftheyear? >> >> It is the same to WEEK function. >> Week( date, [firstWeekDay] ) >> Returns the week of year in the range of 0 - 53 >> firstWeekDay == 0 - is for Sunday >> 1 is for Monday and so on... up to 6 (Sat) CS> Thanks, but this is my problem. IMHO do we have several calculation CS> methods. The US-Method says 1.1.YYYY ist week 1 using DIN week 1 ist CS> where the first thursday is in. So you can have 01.01.XXXX in week 52 CS> or week 53 by DIN but week 1 in US Method. The information when is a CS> new week starting is a good thing, because US-week starts on sunday and CS> by DIN on Monday. CS> But after that, what is calculationMethod in Valentina? CS> Thanks for help CS> Claudius I see. So for making week function more flexible we add third optional param to this sql-function - minimalDaysInFirstWeek. So, now it seems like: Week( date, [firstWeekDay], [minimalDaysInFirstWeek] ) where minimalDaysInFirstWeek must be in range 1-7 (1 by default - "US calendar" - XXXX-01-01 is always first week) Some examples: SELECT week('2005-01-01') SELECT week('2005-01-01', 0) SELECT week('2005-01-01', 1) SELECT week('2005-01-01', 2) SELECT week('2005-01-01', 3) SELECT week('2005-01-01', 4) SELECT week('2005-01-01', 5) SELECT week('2005-01-01', 6) SELECT week('2005-01-01', 2, 3) SELECT week('2005-01-01', 3, 3) SELECT week('2005-01-01', 4, 3) SELECT week('2005-01-01', 5, 3) SELECT week('2005-01-01', 6, 3) All these queries produce 1 as result. In other words, you get first week for 01.01,XXXX. ---------------- SELECT week('2005-01-01', 0, 3) SELECT week('2005-01-01', 1, 3) But this queries produce 53 as result. Is it ok for you? -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From sunshine at public.kherson.ua Thu Dec 9 20:30:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 12:30:49 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <9EF30020-4A0F-11D9-80B7-00039365848C@sailer-online.de> Message-ID: On 12/9/04 8:24 PM, "Claudius Sailer" wrote: >> Okay can I what version of viSQL works with 2.0? > > Okay I found it and compiled it,. > > then I tried this > > new database --> Database1 was created > > I type > > create table Test(Datum date) > commit > > Pressing SQL-Button everything is fine > > insert into Test(Datum) values("01.01.2004") Single quotes must be > ViSQL crashes with errormessage and was after restarting not able to > open database. By trying to open database ViSQl crashes without an > errormessage. > > *hmmm* > Surely I made some mistakes ;-)) -- 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 Claudius at sailer-online.de Thu Dec 9 19:46:25 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Thu Dec 9 12:46:34 2004 Subject: DateTime fields, methods over such fields, DateTime depended SQL where-clause and SQL-functions. In-Reply-To: <1632955301.20041209202750@public.kherson.ua> References: <1659270105.20041209021551@public.kherson.ua> <8AC8FE88-49A3-11D9-9CD0-00039365848C@sailer-online.de> <1181592915.20041209125837@public.kherson.ua> <1632955301.20041209202750@public.kherson.ua> Message-ID: Hi Ivan, Hi Ruslan, Am 09. Dez 2004 um 19:27 Uhr schrieb Ivan Smahin: > I see. So for making week function more flexible we add third optional > param > to this sql-function - minimalDaysInFirstWeek. > > So, now it seems like: > > Week( date, [firstWeekDay], [minimalDaysInFirstWeek] ) > > where > minimalDaysInFirstWeek must be in range 1-7 > (1 by default - "US calendar" - XXXX-01-01 is always first week) > > > Some examples: > > SELECT week('2005-01-01') > SELECT week('2005-01-01', 0) > SELECT week('2005-01-01', 1) > SELECT week('2005-01-01', 2) > SELECT week('2005-01-01', 3) > SELECT week('2005-01-01', 4) > SELECT week('2005-01-01', 5) > SELECT week('2005-01-01', 6) I checked this function against my critical few test-dates and FANTASTIC you had implemented calculation by DIN and not US-Style. For me I don't need 3rd value, but I am not sure what US-Users need!!!!!! > SELECT week('2005-01-01', 2, 3) > SELECT week('2005-01-01', 3, 3) > SELECT week('2005-01-01', 4, 3) > SELECT week('2005-01-01', 5, 3) > SELECT week('2005-01-01', 6, 3) > > All these queries produce 1 as result. > In other words, you get first week for 01.01,XXXX. > > ---------------- > > SELECT week('2005-01-01', 0, 3) > SELECT week('2005-01-01', 1, 3) > > But this queries produce 53 as result. I think your third parameter will help US-Users, but I can life with 2. It is now yours decission :-)) MANY THANKS bye Claudius From jda at his.com Thu Dec 9 14:23:33 2004 From: jda at his.com (jda) Date: Thu Dec 9 13:23:39 2004 Subject: DB size Message-ID: Hi Ruslan. Maybe it's too soon to start optimizing (but perhaps not), but I wonder about what is going to happen with db sizes. One of my beta testers reports that a db that was 26 MB in V1.x is now 55 MB in V2 (11,000 records). I have users with well over 50,000 records, which if this is linear will be > 250 MB! I've found that a db with one (empty) records has gone from about 260K (V1.11) to 1.7 MB (v2). Are we going to be able to get back to the V1.x db sizes? Thanks, Jon From sunshine at public.kherson.ua Thu Dec 9 21:27:24 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 13:27:30 2004 Subject: DB size In-Reply-To: Message-ID: On 12/9/04 9:23 PM, "jda" wrote: Hi Jon, > Maybe it's too soon to start optimizing (but perhaps not), but I > wonder about what is going to happen with db sizes. One of my beta > testers reports that a db that was 26 MB in V1.x is now 55 MB in V2 > (11,000 records). Actually this is quite reasonable. Database is in UTF16 format now, so each string use 2 time more. > I have users with well over 50,000 records, which > if this is linear will be > 250 MB! And now it is 125MB, right ? > I've found that a db with one > (empty) records has gone from about 260K (V1.11) to 1.7 MB (v2). Are > we going to be able to get back to the V1.x db sizes? Empty database is a question. But if remember that empty database also have system tables with about 200 records, and indexes on them, then may be also looks reasonable. Solution will be only in use of UTF8. It must be the same size -- 1 byte per char. In the same time SYSTEM TABLES always will be in UTF16 -- 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 Thu Dec 9 14:36:24 2004 From: jda at his.com (jda) Date: Thu Dec 9 13:36:26 2004 Subject: DB size In-Reply-To: References: Message-ID: > > >Solution will be only in use of UTF8. >It must be the same size -- 1 byte per char. That's my goal, then. When will that be usable? > >In the same time SYSTEM TABLES always will be in UTF16 > That's not a lot though, is it? Jon From sunshine at public.kherson.ua Thu Dec 9 21:40:29 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 13:40:34 2004 Subject: DB size In-Reply-To: Message-ID: On 12/9/04 9:36 PM, "jda" wrote: >> Solution will be only in use of UTF8. >> It must be the same size -- 1 byte per char. > > That's my goal, then. When will that be usable? First of all all major features and bugs must work. Task of UTF8 and other encodings, if issue of several hard days after that :-) >> In the same time SYSTEM TABLES always will be in UTF16 >> > > That's not a lot though, is it? Yes, just 200-1000 records. -- 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 Dec 9 21:46:14 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 13:46:18 2004 Subject: Tech. support inquiry In-Reply-To: <8BE2BE13-4A1A-11D9-9E9E-000A95C377AA@maxprog.com> Message-ID: On 12/9/04 9:43 PM, "MAX Programming LLC" wrote: > Hi, > > Actually I was waiting for that v1 to v2 conversion tool. I am in the > beta list but I don't know where to download the current V$RB pluggin > beta... Hi Stan, Download here http://paradigmasoft.com/download/v2_beta/download.html Conversion tool, this is good of course. But you should start port your application to 2.0 API. And test for now on NEW FRESH 2.0 format dbs. -- 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 Dec 9 23:48:20 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 9 15:48:26 2004 Subject: Tech. support inquiry In-Reply-To: Message-ID: On 12/9/04 10:27 PM, "MAX Programming LLC" wrote: > Hi, > > Just a small question. I hope "Step 2: Drag and drop VComponents folder > into "/Library/CFMSupport" folder on your OS X." and "Step 3: Drag all > files from VEngine folder into "/Library/CFMSupport/VComponents" > folder" are for the beta cycle only. As you know Mac Application > doesn't use installers and I don't want to tell my customers to do > that... It would be a show-stopper. I already read things about that in > the list but I just want to make sure. :) Hi Stan, Yes we must be able create Packages. Just I am not sure that REALBasic 5.x or 6.x support this as built-in Feature. So In the worse case we will build package manually. And you all -- REALBasic developers should press REAL add this feature, As have xCode and CodeWarrior -- 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-list3 at mindstarprods.com Sat Dec 11 16:15:58 2004 From: frank-list3 at mindstarprods.com (Frank Schima) Date: Sat Dec 11 17:16:45 2004 Subject: [V4RB] Pointer to Self Message-ID: <9DAAE4AE-4BCA-11D9-8601-000D932CA350@mindstarprods.com> Hi Ruslan, Is a pointer to Self supposed to work in V2? Here is the code that creates the field (same as in V1 basically): ptrParent = new VObjectPtr("ParentPtr", self, EVFlag.fNullable) I'm having strange problems importing text to a table with a pointer to self. The SQL used was this: SELECT Category, ParentPtr, Code FROM Category The error message mentioned ParentPtr pointing to an ambiguous field! Best regards, Frank Schima From sunshine at public.kherson.ua Sun Dec 12 01:22:28 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Dec 11 17:22:37 2004 Subject: [V4RB] Pointer to Self In-Reply-To: <9DAAE4AE-4BCA-11D9-8601-000D932CA350@mindstarprods.com> Message-ID: On 12/12/04 1:15 AM, "Frank Schima" wrote: Hi Frank, > Is a pointer to Self supposed to work in V2? > > Here is the code that creates the field (same as in V1 basically): > > ptrParent = new VObjectPtr("ParentPtr", self, EVFlag.fNullable) > > I'm having strange problems importing text to a table with a pointer to > self. The SQL used was this: > > SELECT Category, ParentPtr, Code FROM Category SQL used for what? For import ? But why you import something into ObjectPtr ? > The error message mentioned ParentPtr pointing to an ambiguous field! Mmm, I have nothing understand. -- 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-list3 at mindstarprods.com Sat Dec 11 18:11:43 2004 From: frank-list3 at mindstarprods.com (Frank Schima) Date: Sat Dec 11 19:12:24 2004 Subject: [V4RB] Pointer to Self In-Reply-To: References: Message-ID: On Dec 11, 2004, at 4:22 PM, Ruslan Zasukhin wrote: > On 12/12/04 1:15 AM, "Frank Schima" > wrote: > > Hi Frank, > >> Is a pointer to Self supposed to work in V2? >> >> Here is the code that creates the field (same as in V1 basically): >> >> ptrParent = new VObjectPtr("ParentPtr", self, EVFlag.fNullable) >> >> I'm having strange problems importing text to a table with a pointer >> to >> self. The SQL used was this: >> >> SELECT Category, ParentPtr, Code FROM Category > > SQL used for what? For import ? Yes. But the problem occurs before any import can occur (see below). > But why you import something into ObjectPtr ? The Category table is hierarchical. I am trying to import the hierarchical structure from a text file. It all works fine in 1.11. >> The error message mentioned ParentPtr pointing to an ambiguous field! > > Mmm, I have nothing understand. I try to create the cursor for import with the SQL above. But it gives an error - an exception actually. Here is the exact text: Exception Message: Identifier "ParentPtr" is ambiguous. Found in "(null)" and "(null)". Exception Error Number: 591119 I hope that helps you understand because I don't either! -Frank Schima From sunshine at public.kherson.ua Sun Dec 12 13:56:29 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Dec 12 05:56:36 2004 Subject: [V4RB] Pointer to Self In-Reply-To: Message-ID: On 12/12/04 3:11 AM, "Frank Schima" wrote: >> But why you import something into ObjectPtr ? > > The Category table is hierarchical. I am trying to import the > hierarchical structure from a text file. It all works fine in 1.11. Frank, This is danger. This will work only if you have NO ANY deleted record in the table. -- 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-list3 at mindstarprods.com Sun Dec 12 08:19:43 2004 From: frank-list3 at mindstarprods.com (Frank Schima) Date: Sun Dec 12 09:20:23 2004 Subject: [V4RB] Pointer to Self In-Reply-To: References: Message-ID: <3FCAF0F1-4C51-11D9-B294-000D932CA350@mindstarprods.com> On Dec 12, 2004, at 4:56 AM, Ruslan Zasukhin wrote: > On 12/12/04 3:11 AM, "Frank Schima" > wrote: > >>> But why you import something into ObjectPtr ? >> >> The Category table is hierarchical. I am trying to import the >> hierarchical structure from a text file. It all works fine in 1.11. > > Frank, > > This is danger. This will work only if you have NO ANY deleted record > in the > table. Yes. This is only for initial import into an empty table. From sunshine at public.kherson.ua Mon Dec 13 19:23:39 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Dec 13 11:23:46 2004 Subject: V4RB 2.0b31 uploaded Message-ID: Hi All, kernel - 2.0 b31 ================ - 0000184: [Fields] Update of VarChar field causes damage of data in adjacent records. (Igor Gomon) V4RB - 2.0 b31 ============== - 0000202: [Examples] [NEW] API_way/Link_ForeignKey_Recursion. (Ruslan Zasukhin) - 0000203: [Examples] [NEW] API_way/Link_Binary_Recursion. (Ruslan Zasukhin) - 0000201: [Examples] [NEW] API_way/Link_ObjectPtr_Recursion. (Ruslan Zasukhin) - 0000166: [Examples] Test_BLOB() crashes (Igor Gomon) - 0000200: [Examples] ObjectPtr_Recursive, FindAll() not correct (Ruslan Zasukhin) - 0000196: [API] VLink class, some methods now have parameter inRecursionDirection. (Ruslan Zasukhin) - 0000181: [Examples] TEST: Test_BLOB crashes. (Ivan Smahin) - 0000167: [Examples] Do_Method_Date_Long() // CRASH (Ivan Smahin) -- 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 ------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20041213/129bb44d/attachment.html From sunshine at public.kherson.ua Mon Dec 13 19:24:26 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Dec 13 11:24:30 2004 Subject: Jon, VarChar bug Message-ID: Hi Jon, Please check your report on VarChar show values from different field. I think this bug is fixed in b31 Let me 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 jda at his.com Mon Dec 13 12:34:25 2004 From: jda at his.com (jda) Date: Mon Dec 13 11:34:38 2004 Subject: Jon, VarChar bug In-Reply-To: References: Message-ID: >Hi Jon, > >Please check your report on VarChar show values from different field. > >I think this bug is fixed in b31 > >Let me know > Hi Ruslan, Yes, the "data contamination" bug seems to have been fixed! Jon From sunshine at public.kherson.ua Mon Dec 13 19:37:43 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Dec 13 11:37:50 2004 Subject: Jon, VarChar bug In-Reply-To: Message-ID: On 12/13/04 7:34 PM, "jda" wrote: >> Hi Jon, >> >> Please check your report on VarChar show values from different field. >> >> I think this bug is fixed in b31 >> >> Let me know >> > > Hi Ruslan, > > Yes, the "data contamination" bug seems to have been fixed! Excellent! :-) -- 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 Mon Dec 13 15:07:36 2004 From: jda at his.com (jda) Date: Mon Dec 13 14:07:43 2004 Subject: DB size In-Reply-To: References: Message-ID: > > >>> In the same time SYSTEM TABLES always will be in UTF16 >>> >> >> That's not a lot though, is it? > >Yes, just 200-1000 records. > Hi Ruslan, Sorry to follow up so late, but the inordinate size of "small" db's is going to be a big problem for me. With my main app I distribute hundreds of dbs containing perhaps a total of 1K of actual data that is used for controlling certain functions. When they were 200 KB (V1.x) it wasn't so bad. But now an empty file is clocking in at 1.4 MB! Storing as UTF-16 has nothing to do with this enormous increase, of course (I did have to increase the segment size from 512 to 1024 in v2.x, too, or I would get crashes, and that accounts for some of it). I simply can't use hundreds of MB for these files. I certainly hope there is a way to get the size back to 1.x levels. Jon From sunshine at public.kherson.ua Mon Dec 13 22:45:03 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Dec 13 14:45:13 2004 Subject: DB size In-Reply-To: Message-ID: On 12/13/04 10:07 PM, "jda" wrote: > > Hi Ruslan, > > Sorry to follow up so late, but the inordinate size of "small" db's > is going to be a big problem for me. With my main app I distribute > hundreds of dbs containing perhaps a total of 1K of actual data that > is used for controlling certain functions. When they were 200 KB > (V1.x) it wasn't so bad. But now an empty file is clocking in at 1.4 > MB! Storing as UTF-16 has nothing to do with this enormous increase, > of course (I did have to increase the segment size from 512 to 1024 > in v2.x, too, or I would get crashes, and that accounts for some of > it). I simply can't use hundreds of MB for these files. I certainly > hope there is a way to get the size back to 1.x levels. Jon, This is not urgent task from our point of view. You should wait. We need first of all remove all known crashes. -- 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 Mon Dec 13 15:54:37 2004 From: jda at his.com (jda) Date: Mon Dec 13 14:54:47 2004 Subject: DB size In-Reply-To: References: Message-ID: >Jon, > >This is not urgent task from our point of view. >You should wait. > >We need first of all remove all known crashes. > By all means, fix the crashes first -- that's why I've been reporting them so religiously via Mantis. But I just wanted to point out that the blossoming size of databases is a problem, and at least in my case will make it impossible for me to release my app until it is dealt with. Jon From sunshine at public.kherson.ua Mon Dec 13 22:59:15 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Dec 13 14:59:26 2004 Subject: DB size In-Reply-To: Message-ID: On 12/13/04 10:54 PM, "jda" wrote: >> Jon, >> >> This is not urgent task from our point of view. >> You should wait. >> >> We need first of all remove all known crashes. >> > > By all means, fix the crashes first -- that's why I've been reporting > them so religiously via Mantis. But I just wanted to point out that > the blossoming size of databases is a problem, and at least in my > case will make it impossible for me to release my app until it is > dealt with. Jon, but in any case you should not yet make release based on beta. Only when we will see that we have stable build I will inform vdeveloeprs they if you want you can make release. Because we can spend some time more on docs and other polishing yet. -- 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 robelko.com Tue Dec 14 11:36:40 2004 From: rjb at robelko.com (Robert Brenstein) Date: Tue Dec 14 04:47:17 2004 Subject: DB size In-Reply-To: References: Message-ID: > >Sorry to follow up so late, but the inordinate size of "small" db's >is going to be a big problem for me. With my main app I distribute >hundreds of dbs containing perhaps a total of 1K of actual data that >is used for controlling certain functions. When they were 200 KB >(V1.x) it wasn't so bad. But now an empty file is clocking in at 1.4 >MB! Storing as UTF-16 has nothing to do with this enormous increase, >of course (I did have to increase the segment size from 512 to 1024 >in v2.x, too, or I would get crashes, and that accounts for some of >it). I simply can't use hundreds of MB for these files. I certainly >hope there is a way to get the size back to 1.x levels. > >Jon I am curious why you would use hundreds of dbs with tiny bits of data each. Seems that combining them into a larger db be reasonable. Or if this is just control data, use non-db files. Robert From rblists at carsten-friehe.de Tue Dec 14 13:14:01 2004 From: rblists at carsten-friehe.de (=?iso-8859-1?Q?Carsten_Friehe?=) Date: Tue Dec 14 06:14:11 2004 Subject: Install? Message-ID: <26883301$110302619041bed80e1abcb5.16940682@config6.schlund.de> Hi! I am an old user of V4RB and now I am looking for V4RB 2.0beta. I only have one question at the beginning. Is it possible to make a standalone application like it was possible in V4RB 1.x? Or do my users have to install something more? Or will this be possible in a future version? Thank you! Carsten From sunshine at public.kherson.ua Tue Dec 14 14:25:50 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 06:26:06 2004 Subject: Install? In-Reply-To: <26883301$110302619041bed80e1abcb5.16940682@config6.schlund.de> Message-ID: On 12/14/04 2:14 PM, "Carsten Friehe" wrote: > Hi! > > I am an old user of V4RB and now I am looking for V4RB 2.0beta. I only > have one question at the beginning. > Is it possible to make a standalone application like it was possible in > V4RB 1.x? Or do my users have to install something more? > Or will this be possible in a future version? Application package should be the best solution. Please go to REAL an all vote to add this feature. http://support.realsoftware.com/feedback/viewreport.php? reportid=jltucquj http://support.realsoftware.com/feedback/viewreport.php? reportid=uulmsobd -- 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 rblists at carsten-friehe.de Tue Dec 14 13:38:03 2004 From: rblists at carsten-friehe.de (=?iso-8859-1?Q?Carsten_Friehe?=) Date: Tue Dec 14 06:40:11 2004 Subject: Install? Message-ID: <26883301$110302769441beddee5bfe44.74069880@config20.schlund.de> Hi Ruslan! > Application package should be the best solution. Ok. So no direct support like in 1.x is there. My users have to install the libraries themselves. > Please go to REAL an all vote to add this feature. Done! Carsten From sunshine at public.kherson.ua Tue Dec 14 14:46:57 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 06:47:04 2004 Subject: Install? In-Reply-To: <26883301$110302769441beddee5bfe44.74069880@config20.schlund.de> Message-ID: On 12/14/04 2:38 PM, "Carsten Friehe" wrote: > Hi Ruslan! > >> Application package should be the best solution. > > Ok. So no direct support like in 1.x is there. My users have to install > the libraries themselves. NO. 1) you always can provide for your user Installer, made with Apple PackageMaker for example or any other installer 2) I ma sure that RIGHT NOW we also can create App Package, just we will do this little manually. >> Please go to REAL an all vote to add this feature. > > Done! -- 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 Dec 14 08:11:59 2004 From: jda at his.com (jda) Date: Tue Dec 14 07:12:09 2004 Subject: DB size In-Reply-To: References: Message-ID: >>Sorry to follow up so late, but the inordinate size of "small" db's >>is going to be a big problem for me. With my main app I distribute >>hundreds of dbs containing perhaps a total of 1K of actual data >>that is used for controlling certain functions. When they were 200 >>KB (V1.x) it wasn't so bad. But now an empty file is clocking in at >>1.4 MB! Storing as UTF-16 has nothing to do with this enormous >>increase, of course (I did have to increase the segment size from >>512 to 1024 in v2.x, too, or I would get crashes, and that accounts >>for some of it). I simply can't use hundreds of MB for these files. >>I certainly hope there is a way to get the size back to 1.x levels. >> >>Jon > >I am curious why you would use hundreds of dbs with tiny bits of >data each. Seems that combining them into a larger db be reasonable. >Or if this is just control data, use non-db files. > It's control data, mostly. The main problem for me is that this is the approach I took with V1.x, and it was satisfactory. Now my code is riddled with calls to these dbs, and I have thousands of users who have modified them to suit their own needs. So for me to move to another method (e.g. saving as XML) is going to require a ton of work and present compatibility problems when my users migrate to V2. If I had it to do all over again I'd use another approach, and in fact I may have to bite the bullet and do so. But it still seems to me that an empty db should not require 1.4 MB. If optimizations are the problem, that can be dealt with and we will all benefit. If this is inherent to the new db structures, then... Jon From sunshine at public.kherson.ua Tue Dec 14 15:29:34 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 07:29:43 2004 Subject: DB size In-Reply-To: Message-ID: On 12/14/04 3:11 PM, "jda" wrote: > If I had it to do all over again I'd use another approach, and in > fact I may have to bite the bullet and do so. But it still seems to > me that an empty db should not require 1.4 MB. If optimizations are > the problem, that can be dealt with and we will all benefit. If this > is inherent to the new db structures, then... Btw, have you see that viSQL now show internal files of db volumes ? You can see how many files get your db, And their size. Also you can send me one of empty 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 jda at his.com Tue Dec 14 08:42:19 2004 From: jda at his.com (jda) Date: Tue Dec 14 07:42:33 2004 Subject: DB size In-Reply-To: References: Message-ID: >Btw, have you see that viSQL now show internal files of db volumes ? > >You can see how many files get your db, >And their size. > >Also you can send me one of empty db. > Thanks Ruslan, I can see them in viSQL --132 files, no names, so I can't tell what's what, but all are reasonable (< 20K) except one that is 1.2 MB! System table? I'll send it to you. Jon From sunshine at public.kherson.ua Tue Dec 14 15:55:22 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 07:55:33 2004 Subject: DB size In-Reply-To: Message-ID: On 12/14/04 3:42 PM, "jda" wrote: > Thanks Ruslan, > > I can see them in viSQL --132 files, no names, so I can't tell what's > what, but all are reasonable (< 20K) except one that is 1.2 MB! wow > System table? I do not think so > I'll send it to 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 valentina-list at vermontsoftworks.com Tue Dec 14 10:02:42 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Dec 14 09:02:34 2004 Subject: .FindContains ignores "inSelection" VSet parameter Message-ID: <345BA346-4DE1-11D9-B9D6-000393DC4766@vermontsoftworks.com> Just submitted this, FYI: #208 Summary: 0000208: .FindContains ignores "inSelection" VSet parameter Description: myVSet = VarCharField.FindContains(inValue, inSelection) ignores the InSelection parameter, so it's impossible to iterate through multiple search terms on progressively smaller VSets. Additional Information: This may or may not also be true for other field types and other .Find methods. From sunshine at public.kherson.ua Tue Dec 14 17:12:50 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 09:13:02 2004 Subject: .FindContains ignores "inSelection" VSet parameter In-Reply-To: <345BA346-4DE1-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: On 12/14/04 5:02 PM, "Erik Mueller-Harder" wrote: > Just submitted this, FYI: > > #208 Ok Erik > Summary: > 0000208: .FindContains ignores "inSelection" VSet parameter > > Description: > myVSet = VarCharField.FindContains(inValue, inSelection) ignores the > InSelection parameter, so it's impossible to iterate through multiple > search terms on progressively smaller VSets. Please note! Smaller selection MEANS SENSE only if you do search for NON-indexed field. For indexed field, Valentina still will do search on index. Just comfort is that it will self make AND-union of sets. And you can reduce your code. > Additional Information: > This may or may not also be true for other field types and other .Find > methods. -- 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-list at vermontsoftworks.com Tue Dec 14 10:19:53 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Dec 14 09:19:37 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive Message-ID: <9B0F3FE6-4DE3-11D9-B9D6-000393DC4766@vermontsoftworks.com> Just submitted this, FYI: #209 Summary: 0000209: "collationstrength = 1" doesn't render field case-insensitive Description: VarChar fields with a collationstrength of 1 are not found with FindContains unless the search-text capitalization exactly matches. E.g., "Cabot" is not found if you search for "cabot". Additional Information: Other field types and/or collationstrengths may or may not also be broken. From valentina-list at vermontsoftworks.com Tue Dec 14 10:23:20 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Dec 14 09:23:06 2004 Subject: .FindContains ignores "inSelection" VSet parameter In-Reply-To: References: Message-ID: <160F14FA-4DE4-11D9-B9D6-000393DC4766@vermontsoftworks.com> >> myVSet = VarCharField.FindContains(inValue, inSelection) ignores the >> InSelection parameter, so it's impossible to iterate through multiple >> search terms on progressively smaller VSets. > > Please note! > > Smaller selection MEANS SENSE only if you do search for NON-indexed > field. > > For indexed field, Valentina still will do search on index. > Just comfort is that it will self make AND-union of sets. And you can > reduce > your code. Sounds good. So how can I modify the following command to search for records containing both "Ruslan" and "Erik"? myVSet = myVarChar.FindContains("Ruslan") Thanks! -- Erik From sunshine at public.kherson.ua Tue Dec 14 17:21:16 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 09:23:57 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive In-Reply-To: <9B0F3FE6-4DE3-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: On 12/14/04 5:19 PM, "Erik Mueller-Harder" wrote: > Just submitted this, FYI: > > #209 > > Summary: > 0000209: "collationstrength = 1" doesn't render field case-insensitive > > Description: > VarChar fields with a collationstrength of 1 are not found with > FindContains unless the search-text capitalization exactly matches. > E.g., "Cabot" is not found if you search for "cabot". > > Additional Information: > Other field types and/or collationstrengths may or may not also be > broken. Erik, This is correct! I will explain why. -- 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 Dec 14 17:30:29 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 09:30:41 2004 Subject: .FindContains ignores "inSelection" VSet parameter In-Reply-To: <160F14FA-4DE4-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: On 12/14/04 5:23 PM, "Erik Mueller-Harder" wrote: >>> myVSet = VarCharField.FindContains(inValue, inSelection) ignores the >>> InSelection parameter, so it's impossible to iterate through multiple >>> search terms on progressively smaller VSets. >> >> Please note! >> >> Smaller selection MEANS SENSE only if you do search for NON-indexed >> field. >> >> For indexed field, Valentina still will do search on index. >> Just comfort is that it will self make AND-union of sets. And you can >> reduce >> your code. > > Sounds good. So how can I modify the following command to search for > records containing both "Ruslan" and "Erik"? > > myVSet = myVarChar.FindContains("Ruslan") 1) General way is: s1 = myVarChar.FindContains("Ruslan") s2 = myVarChar.FindContains("Erik") s1.Intersection( s2 ) // hmm, may be we need also add // s3 = Intersection( s1, s2 ) 2) second way s1 = myVarChar.FindContains("Ruslan") s2 = myVarChar.FindContains("Erik", s1) s1 = nil // to free RAM 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 sunshine at public.kherson.ua Tue Dec 14 17:33:31 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 09:33:41 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive In-Reply-To: <9B0F3FE6-4DE3-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: On 12/14/04 5:19 PM, "Erik Mueller-Harder" wrote: > Just submitted this, FYI: > > #209 > > Summary: > 0000209: "collationstrength = 1" doesn't render field case-insensitive > > Description: > VarChar fields with a collationstrength of 1 are not found with > FindContains unless the search-text capitalization exactly matches. > E.g., "Cabot" is not found if you search for "cabot". 0 - ignore case + ignore accents 1 - ignore accents 2 - nothing ignore (i.e. The same srtings) So what you want to get ? -- 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 Dec 14 17:49:11 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 09:49:19 2004 Subject: V4RBReference_en.pdf updated Message-ID: Hi All, You can download fresh version of PDF. Practically all 145 pages was proofread. -- 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-list at vermontsoftworks.com Tue Dec 14 11:06:29 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Dec 14 10:06:14 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive In-Reply-To: References: Message-ID: <1D1CE306-4DEA-11D9-B9D6-000393DC4766@vermontsoftworks.com> >> VarChar fields with a collationstrength of 1 are not found with >> FindContains unless the search-text capitalization exactly matches. >> E.g., "Cabot" is not found if you search for "cabot". > > 0 - ignore case + ignore accents > 1 - ignore accents > 2 - nothing ignore (i.e. The same srtings) > > So what you want to get ? Sounds like "0" is what I want. Given the lack of documentation (not your fault! I know this is an alpha! :-), I'd been using for definitions and explanations of collation. There, your "0" is defined as "primary," your "1" as "secondary," etc. Hmmm.... Actually, I'd used your EVColAttributeValue.kPrimary enum. You (like Rogue Wave) have enums kPrimary, kSecondary, kTertiary, and kQuaternary. How do these correspond to your numerals above? kPrimary is *not* ignoring case, currently. Thanks, -- Erik From valentina-list at vermontsoftworks.com Tue Dec 14 11:07:34 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Dec 14 10:07:17 2004 Subject: .FindContains ignores "inSelection" VSet parameter In-Reply-To: References: Message-ID: <442D0E54-4DEA-11D9-B9D6-000393DC4766@vermontsoftworks.com> > 1) General way is: > > s1 = myVarChar.FindContains("Ruslan") > > s2 = myVarChar.FindContains("Erik") > > s1.Intersection( s2 ) > > // hmm, may be we need also add > // s3 = Intersection( s1, s2 ) So I'll use this for now, thanks! > 2) second way > > s1 = myVarChar.FindContains("Ruslan") > > s2 = myVarChar.FindContains("Erik", s1) > > s1 = nil // to free RAM ASAP Seems a little cleaner, or at least more terse. I'll switch to this when it works. Thank you! -- Erik From valentina-list at vermontsoftworks.com Tue Dec 14 11:21:59 2004 From: valentina-list at vermontsoftworks.com (Erik Mueller-Harder) Date: Tue Dec 14 10:21:40 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive In-Reply-To: <1D1CE306-4DEA-11D9-B9D6-000393DC4766@vermontsoftworks.com> References: <1D1CE306-4DEA-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: <4777889C-4DEC-11D9-B9D6-000393DC4766@vermontsoftworks.com> On Dec 14, 2004, at 11:06, I wrote: > kPrimary is *not* ignoring case, currently. Well, I've changed no code -- but I think Valentina did just rebuilt my field's index. At any rate, the field is now case-insensitive. So please consider this report withdrawn. Thanks for your help! Oh, I would still be interested in how you define the various EVColAttributeValue enums (kPrimary, kSecondary, kTertiary, and kQuaternary). Many thanks! -- Erik From sunshine at public.kherson.ua Tue Dec 14 18:27:52 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 10:34:56 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive In-Reply-To: <4777889C-4DEC-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: On 12/14/04 6:21 PM, "Erik Mueller-Harder" wrote: > On Dec 14, 2004, at 11:06, I wrote: > >> kPrimary is *not* ignoring case, currently. > > Well, I've changed no code -- but I think Valentina did just rebuilt my > field's index. > > At any rate, the field is now case-insensitive. > > So please consider this report withdrawn. Thanks for your help! > > Oh, I would still be interested in how you define the various > EVColAttributeValue enums (kPrimary, kSecondary, kTertiary, and > kQuaternary). You mean numeric values ? Look into fresh PDF on site -- 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 Dec 14 18:43:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 14 10:43:56 2004 Subject: "collationstrength = 1" doesn't render field case-insensitive In-Reply-To: <1D1CE306-4DEA-11D9-B9D6-000393DC4766@vermontsoftworks.com> Message-ID: On 12/14/04 6:06 PM, "Erik Mueller-Harder" wrote: >>> VarChar fields with a collationstrength of 1 are not found with >>> FindContains unless the search-text capitalization exactly matches. >>> E.g., "Cabot" is not found if you search for "cabot". >> >> 0 - ignore case + ignore accents >> 1 - ignore accents >> 2 - nothing ignore (i.e. The same srtings) >> >> So what you want to get ? > > Sounds like "0" is what I want. > > Given the lack of documentation (not your fault! I know this is an > alpha! :-), I'd been using > > rwucollator.html> > > for definitions and explanations of collation. There, your "0" is > defined as "primary," your "1" as "secondary," etc. > > Hmmm.... Actually, I'd used your EVColAttributeValue.kPrimary enum. > You (like Rogue Wave) have enums kPrimary, kSecondary, kTertiary, and > kQuaternary. How do these correspond to your numerals above? > > kPrimary is *not* ignoring case, currently. /*************************************************************************** *******************/ REALconstant EVOsConstants[] = { { "kDefault = 0" }, { "kMacOS = 1" }, { "kWindows = 2" }, { "kLinux = 3" }, }; /*************************************************************************** *******************/ REALconstant EVDebugLevelConstants[] = { { "kLogNothing = 0" }, { "kLogErrors = 1" }, { "kLogFunctions = 2" }, { "kLogParams = 3" }, }; /*************************************************************************** *******************/ REALconstant EVDbModeConstants[] = { { "kDscDatBlbInd = 1" }, { "kDsc_DatBlbInd = 2" }, { "kDsc_DatBlb_Ind = 3" }, { "kDsc_Dat_Blb_Ind = 4" }, { "kDscDatBlb_Ind = 5" }, { "kDscDat_Blb_Ind = 6" }, { "kDscDatInd_Blb = 7" }, { "kDsc_DatInd_Blb = 8" }, }; /*************************************************************************** *******************/ REALconstant EVFlagConstants[] = { { "fNone = 0" }, { "fNullable = 1" }, { "fIndexed = 2" }, { "fUnique = 4" }, { "fIndexByWords = 8" }, { "fCompressed = 32" }, { "fMethod = 64" }, }; /*************************************************************************** *******************/ REALconstant EVOnDeletionConstants[] = { { "kNoAction = 0" }, { "kSetNull = 1" }, { "kCascade = 2" }, { "kRestrict = 3" }, { "kDefault = 4" }, }; /*************************************************************************** *******************/ REALconstant EVOnUpdateConstants[] = { { "kNoAction = 0" }, { "kSetNull = 1" }, { "kCascade = 2" }, { "kRestrict = 3" }, { "kDefault = 4" }, }; /*************************************************************************** *******************/ REALconstant EVRecursionDirectionConstants[] = { { "kFromParentToChild = 0" }, { "kFromChildToParent = 1" }, }; /*************************************************************************** *******************/ REALconstant EVStorageTypeConstants[] = { { "kDefault = 0" }, { "kDisk = 1" }, { "kRAM = 2" }, }; /*************************************************************************** *******************/ REALconstant EVCursorLocationConstants[] = { { "kClientSide = 1" }, { "kServerSide = 2" }, { "kServerSideBulk = 3" }, }; /*************************************************************************** *******************/ REALconstant EVLockTypeConstants[] = { { "kNoLocks = 1" }, { "kReadOnly = 2" }, { "kReadWrite = 3" }, }; /*************************************************************************** *******************/ REALconstant EVCursorDirectionConstants[] = { { "kForwardOnly = 1" }, { "kRandom = 2" }, }; /*************************************************************************** *******************/ REALconstant EVLinkTypeConstants[] = { { "kMany = 0" }, { "kOne = 1" }, }; /*************************************************************************** *******************/ REALconstant EVFieldTypeConstants[] = { { "kTypeEmpty = 0" }, { "kTypeEnum = 1" }, { "kTypeBoolean = 2" }, { "kTypeByte = 3" }, { "kTypeShort = 4" }, { "kTypeUShort = 5" }, { "kTypeMedium = 6" }, { "kTypeUMedium = 7" }, { "kTypeLong = 8" }, { "kTypeULong = 9" }, { "kTypeLLong = 10" }, { "kTypeULLong = 11" }, { "kTypeFloat = 12" }, { "kTypeDouble = 13" }, { "kTypeLDouble = 14" }, { "kTypeDecimal = 15" }, { "kTypeDate = 16" }, { "kTypeTime = 17" }, { "kTypeDateTime = 18" }, { "kTypeString = 19" }, { "kTypeVarChar = 20" }, { "kTypeFixedBinary = 21" }, { "kTypeVarBinary = 22" }, { "kTypeBLOB = 23" }, { "kTypeText = 24" }, { "kTypePicture = 25" }, { "kTypeSound = 26" }, { "kTypeMovie = 27" }, { "kTypeRecID = 28" }, { "kTypeOID = 29" }, { "kTypeObjectPtr = 30" }, { "kTypeObjectsPtr = 31" }, { "kTypeTimeStamp = 32" }, }; /*************************************************************************** *******************/ REALconstant EVDumpTypeConstants[] = { { "kSQL = 1" }, { "kXML = 2" }, }; /*************************************************************************** *******************/ REALconstant EVDumpDataConstants[] = { { "kStructureOnly = 1" }, { "kStructureAndRecords = 2" }, { "kRecordsOnly = 3" }, }; /*************************************************************************** *******************/ REALconstant EVVerboseLevelConstants[] = { { "kNone = 0" }, { "kLow = 1" }, { "kNormal = 2" }, { "kHigh = 3" }, { "kVeryHigh = 4" }, }; /*************************************************************************** *******************/ REALconstant EVColAttributeConstants[] = { { "kFrenchCollation = 0" }, { "kAlternateHandling = 1" }, { "kCaseFirst = 2" }, { "kCaseLevel = 3" }, { "kNormalizationMode = 4" }, { "kStrength = 5" }, { "kHiraganaQuaternaryMode = 6" }, { "kNumericCollation = 7" }, { "kAttributeCount = 8" }, }; /*************************************************************************** *******************/ REALconstant EVColAttributeValueConstants[] = { { "kDefault = -1" }, { "kPrimary = 0" }, { "kSecondary = 1" }, { "kTertiary = 2" }, { "kDefaultStrength = 2" }, { "kQuaternary = 3" }, { "kIdentical = 15" }, { "kOFF = 16" }, { "kON = 17" }, { "kShifted = 20" }, { "kNonIgnorable = 21" }, { "kLowerFirst = 24" }, { "kUpperFirst = 25" } }; -- 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 Dec 15 14:41:11 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 15 06:41:34 2004 Subject: [V4RB 0000206]: SELECT with ObjectPtr to Self fails In-Reply-To: <7970828972e4ef8ec0b72efb3d333d07@paradigma.ukrcom.kheron.ua> Message-ID: Hi Frank, I cannot reproduce with b31 In the example Link_ObjectPtr_Recursion I have add the following line Sub Action for button STEP2 AddRecordsAndLinkThem( mDatabase ) ShowStatistic() call mDatabase.SqlSelect( "SELECT ParentPtr FROM Person" ) // Open 2 browser windows mBrowser1 = new TableBrowser mBrowser2 = new TableBrowser No problems. It works. So try again, and if problem send me something So tell me how to reproduce on example > The following issue has been SUBMITTED. > ====================================================================== > > ====================================================================== > Reported By: Frank Schima > Assigned To: > ====================================================================== > Project: V4RB > Issue ID: 206 > Category: SQL > Reproducibility: always > Severity: major > Priority: normal > Status: new > ====================================================================== > Date Submitted: 12-13-2004 21:07 EET > Last Modified: 12-13-2004 21:07 EET > ====================================================================== > Summary: SELECT with ObjectPtr to Self fails > Description: > I have a table with an ObjectPtr field pointing to the same table (a self > pointer). When I try to SELECT that ObjectPtr: > > SELECT ObjectPtrToSelfField FROM myTable > > I get the following Exception: > > Exception Message: Identifier "ParentPtr" is ambiguous. Found in "(null)" > and "(null)". > Exception Error Number: 591119 > > ====================================================================== > > Issue History > Date Modified Username Field Change > ====================================================================== > 12-13-04 21:07 Frank Schima New 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 Dec 15 15:41:12 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 15 07:41:24 2004 Subject: [V4RB 0000206]: SELECT with ObjectPtr to Self fails In-Reply-To: Message-ID: On 12/15/04 2:41 PM, "Ruslan Zasukhin" wrote: > Hi Frank, > > I cannot reproduce with b31 I have reproduce it in Classes way example. -- 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 Dec 18 00:53:44 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 17 16:53:56 2004 Subject: V4RB b32 uploaded Message-ID: kernel - 2.0 b32 ================ - 0000214: [SQL] "select 2 as 'name'" does not set cursor field name to 'name'. (Ivan Smahin) - 0000213: [FBL Nodes] DateTime functions don't work correctly in string oriented environment. (Ivan Smahin) - 0000207: [Cursors] No proper method values after update using cursor. (Ivan Smahin) V4RB - 2.0 b32 ============== - 0000206: [SQL] SELECT with ObjectPtr to Self fails (Ivan Smahin) - 0000141: [Feature Request] Smaller db sizes (Ruslan Zasukhin) - 0000208: [API] .FindContains ignores "inSelection" VSet parameter (Ruslan Zasukhin) -- 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 ------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20041218/c4991ba1/attachment.html From sunshine at public.kherson.ua Sun Dec 19 01:03:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Dec 18 17:03:39 2004 Subject: ValentinaSQL_en.pdf uploaded Message-ID: Hi Guys, http://paradigmasoft.com/download/v2_beta/download.html We have polish pages about functions (about 30 pages). In the first turn I recommend to looks at * date time functions. Changes here and improvements. * link functions. new cool things in 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 sunshine at public.kherson.ua Sun Dec 19 09:15:25 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Dec 19 01:15:33 2004 Subject: [V4RB 0000216]: Database.Open crashes for my database In-Reply-To: Message-ID: On 12/19/04 4:11 AM, "vbt@paradigma.ukrcom.kherson.ua" wrote: Hi Frank, Oh, I forget to mention. b32 have change system table field. So you need create new db with b32 > The following issue has been SUBMITTED. > ====================================================================== > > ====================================================================== > Reported By: Frank Schima > Assigned To: > ====================================================================== > Project: V4RB > Issue ID: 216 > Category: API > Reproducibility: always > Severity: crash > Priority: normal > Status: new > ====================================================================== > Date Submitted: 12-19-2004 04:11 EET > Last Modified: 12-19-2004 04:11 EET > ====================================================================== > Summary: Database.Open crashes for my database > Description: > Attempting to open my database with ViSQL 2 causes a crash. I'm sending a > project to create the database to Ruslan. > ====================================================================== > > Issue History > Date Modified Username Field Change > ====================================================================== > 12-19-04 04:11 Frank Schima New 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 frank-list3 at mindstarprods.com Sun Dec 19 08:00:17 2004 From: frank-list3 at mindstarprods.com (Frank Schima) Date: Sun Dec 19 09:00:31 2004 Subject: [V4RB 0000216]: Database.Open crashes for my database In-Reply-To: References: Message-ID: Hi Ruslan, This is a new database. The project creates a new one which then crashes ViSQL 2. Please let me know if you need to me to resend the project which reproduces the problem. Best regards, Frank On Dec 19, 2004, at 12:15 AM, Ruslan Zasukhin wrote: > On 12/19/04 4:11 AM, "vbt@paradigma.ukrcom.kherson.ua" > wrote: > > Hi Frank, > > Oh, I forget to mention. > > b32 have change system table field. > So you need create new db with b32 > > >> The following issue has been SUBMITTED. >> ====================================================================== >> >> ====================================================================== >> Reported By: Frank Schima >> Assigned To: >> ====================================================================== >> Project: V4RB >> Issue ID: 216 >> Category: API >> Reproducibility: always >> Severity: crash >> Priority: normal >> Status: new >> ====================================================================== >> Date Submitted: 12-19-2004 04:11 EET >> Last Modified: 12-19-2004 04:11 EET >> ====================================================================== >> Summary: Database.Open crashes for my database >> Description: >> Attempting to open my database with ViSQL 2 causes a crash. I'm >> sending a >> project to create the database to Ruslan. >> ====================================================================== >> >> Issue History >> Date Modified Username Field Change >> ====================================================================== >> 12-19-04 04:11 Frank Schima New 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 > ------------------------------------------------------------- > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > From sunshine at public.kherson.ua Sun Dec 19 17:22:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Dec 19 09:22:51 2004 Subject: [V4RB 0000216]: Database.Open crashes for my database In-Reply-To: Message-ID: On 12/19/04 5:00 PM, "Frank Schima" wrote: > Hi Ruslan, > > > This is a new database. The project creates a new one which then > crashes ViSQL 2. Please let me know if you need to me to resend the > project which reproduces the problem. Aha. Yes, Frank, send project again please, -- 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 Sun Dec 19 12:24:11 2004 From: jda at his.com (jda) Date: Sun Dec 19 11:24:22 2004 Subject: Trapping index errors Message-ID: I'm finding that index damage is the most common form of db corruption. I'd like to trap for that when opening a database and automatically rebuild the indexes. I have a db with an index error of 537857 ($83501) -- Inventory corrupted for index "%S". I've tried trapping for a number of different EVErrors, and even if theError.ErrorNumber >= EVError.IndexByteReclistCorrupted and theError.ErrorNumber <= EVError.IndexPageInventoryCorrupted then... but can't catch it. I can of course trap for the error number I've given, but I'd like a more general solution. Ruslan, what enumerated constants would you suggest I use to trap for repairable index errors? Thanks, Jon From sunshine at public.kherson.ua Sun Dec 19 20:01:11 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Dec 19 12:01:17 2004 Subject: Trapping index errors In-Reply-To: Message-ID: On 12/19/04 7:24 PM, "jda" wrote: > I'm finding that index damage is the most common form of db > corruption. I'd like to trap for that when opening a database and > automatically rebuild the indexes. > > I have a db with an index error of 537857 ($83501) -- Inventory > corrupted for index "%S". > > I've tried trapping for a number of different EVErrors, and even > > if theError.ErrorNumber >= EVError.IndexByteReclistCorrupted and > theError.ErrorNumber <= EVError.IndexPageInventoryCorrupted then... This should work. Look into XML files. BTW, use for this MS IE. It can show XML in nice way. Or any other good XML Viewer Errors make groups. They are sorted. So range search should work! > but can't catch it. > > I can of course trap for the error number I've given, but I'd like a > more general solution. > > Ruslan, what enumerated constants would you suggest I use to trap for > repairable index errors? And of course I need way to reproduce errors -- 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 Sun Dec 19 13:08:52 2004 From: jda at his.com (jda) Date: Sun Dec 19 12:08:58 2004 Subject: Trapping index errors In-Reply-To: References: Message-ID: > I have a db with an index error of 537857 ($83501) -- Inventory > > corrupted for index "%S". >> >> I've tried trapping for a number of different EVErrors, and even >> >> if theError.ErrorNumber >= EVError.IndexByteReclistCorrupted and >> theError.ErrorNumber <= EVError.IndexPageInventoryCorrupted then... > >This should work. > >Look into XML files. >BTW, use for this MS IE. It can show XML in nice way. >Or any other good XML Viewer > >Errors make groups. >They are sorted. >So range search should work! Good. It turns out that you can't repair this particular error by reindexing -- the db throws the same error when you try to open it again. I wonder if this is something that Valentina should deal with itself on myDatabase.open I mean, index errors should not prevent you from opening the database, and could be repaired by V itself by removing the indexes and forcing the creation of new indexes -- what do you think? > > > Ruslan, what enumerated constants would you suggest I use to trap for >> repairable index errors? > >And of course I need way to reproduce errors > It happens sporadically after lots of developing/debugging, so I'm afraid I have no way to reproduce this particular error. If I find one, of course I'll report it. Jon From sunshine at public.kherson.ua Sun Dec 19 20:14:21 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Dec 19 12:14:25 2004 Subject: Trapping index errors In-Reply-To: Message-ID: On 12/19/04 8:08 PM, "jda" wrote: >> Errors make groups. >> They are sorted. >> So range search should work! > > Good. > > It turns out that you can't repair this particular error by > reindexing -- the db throws the same error when you try to open it > again. > > I wonder if this is something that Valentina should deal with itself on > > myDatabase.open > > I mean, index errors should not prevent you from opening the > database, and could be repaired by V itself by removing the indexes > and forcing the creation of new indexes -- what do you think? In fact index error must NEVER happens. If we get this means some bugs in my algorithms. We need fix them. So you say that you have db which REINDEX produce again bad index? Send me this db please. -- 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 lists at mostrom.pp.se Tue Dec 28 19:56:45 2004 From: lists at mostrom.pp.se (Jan Erik =?iso-8859-1?Q?Mostr=F6m?=) Date: Tue Dec 28 12:56:53 2004 Subject: newbie question - getting v2 to compile Message-ID: I've just downloaded v2 for the first time and I'm now trying to compile some example application but it fails. What I've done is + Moved the VEngine folder into the VComponents folder + Moved the VComponents folder into the /Library/CFMSupport folder + Dragged the V2RB into the plugin folder When I open an example and try to compile it, I get an error message for every line that has something to do with Valentina. When I try command-completion RB doesn't recognize any valentina stuff. I assume that I've made some basic mistake??? jem -- Jan Erik Mostr?m www.mostrom.pp.se From sunshine at public.kherson.ua Tue Dec 28 22:10:49 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Dec 28 14:10:57 2004 Subject: newbie question - getting v2 to compile In-Reply-To: Message-ID: On 12/28/04 8:56 PM, "Jan Erik Mostr?m" wrote: > I've just downloaded v2 for the first time and I'm now trying to compile some > example application but it fails. What I've done is > > + Moved the VEngine folder into the VComponents folder NOT folder!!! But files in the folder > + Moved the VComponents folder into the /Library/CFMSupport folder > > + Dragged the V2RB into the plugin folder > > When I open an example and try to compile it, I get an error message for > every line that has something to do with Valentina. When I try > command-completion RB doesn't recognize any valentina stuff. > > I assume that I've made some basic 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 jda at his.com Wed Dec 29 08:50:15 2004 From: jda at his.com (jda) Date: Wed Dec 29 07:50:33 2004 Subject: locked records Message-ID: Hi Ruslan, I've run across a situation where can't make a cursor because a "record (null) is locked in the table "mytable". I've quit my app and opened the database -- I get the same error. Shouldn't locked records revert to unlocked when the db is closed? If so, either that isn't happening in this case, or the error message is incorrect. Jon From sunshine at public.kherson.ua Wed Dec 29 15:56:57 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 29 07:57:21 2004 Subject: locked records In-Reply-To: Message-ID: On 12/29/04 3:50 PM, "jda" wrote: > Hi Ruslan, > > I've run across a situation where can't make a cursor because a > "record (null) is locked in the table "mytable". > > I've quit my app and opened the database -- I get the same error. > Shouldn't locked records revert to unlocked when the db is closed? If > so, either that isn't happening in this case, or the error message is > incorrect. Ivan, Do you have notes on 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 sunshine at public.kherson.ua Wed Dec 29 18:12:27 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 29 10:12:36 2004 Subject: FIXED: V4RB2 - Method_Create example In-Reply-To: <94227E8E-416C-11D9-8B9F-003065514D2E@tiscalinet.it> Message-ID: On 11/28/04 8:37 PM, "erne" wrote: Actual this was fixed in b31 or b32 I think. > Hello Ruslan, > > runnings the Method_Create example in the Api_Way folder > I find that the Upper(VarChar) method fails to update correctly > > it can't display strings longer than the one that was first input > > e.g. > if first string input is "abra" you get "ABRA" > then you input "very cool" you get "VERY" > > Trying the method Length(VarChar) > if first string input is "abra" you get 4 > then you input "very cool" you still get 4 -- 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 Dec 29 21:22:10 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 29 13:22:18 2004 Subject: V4RB2 vLink In-Reply-To: <96E02AC6-3CC5-11D9-898A-003065514D2E@tiscalinet.it> Message-ID: On 11/22/04 10:32 PM, "erne" wrote: Hi Erne, Sorry for delay. I have not realize here present questions. So here we go. > hello Ruslan, > > looking into new vLink class and have some questions > > we create BinaryLink with DB method: > DB.CreateBinaryLink( > inName as String, > inLeftTable as VTable, > inRightTable as VTable, > [inLeftPower as Integer = 1], > [inRightPower as Integer = 0], // 1 : M > [inOnDelete as integer = 1], > [inStorageType as integer = 0], // kDefault > inTemporary as Boolean = false) As VLink > > how can we delete this if we don't need it anymore? We have method: Vdatabase.DropLink( link ) -- 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 Dec 29 21:28:20 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 29 13:28:26 2004 Subject: V4RB2 vLink In-Reply-To: <96E02AC6-3CC5-11D9-898A-003065514D2E@tiscalinet.it> Message-ID: On 11/22/04 10:32 PM, "erne" wrote: > hello Ruslan, > > looking into new vLink class and have some questions > > we create BinaryLink with DB method: > DB.CreateBinaryLink( > inName as String, > inLeftTable as VTable, > inRightTable as VTable, > [inLeftPower as Integer = 1], > [inRightPower as Integer = 0], // 1 : M > [inOnDelete as integer = 1], > [inStorageType as integer = 0], // kDefault > inTemporary as Boolean = false) As VLink > > how can we delete this if we don't need it anymore? > > so the vLink object knows the Tables it is linking > > but when we want to find linked we must call: > vLink.FindLinked( inRecID as Integer, inTableA as VTable, inTableB as > VTable ) as VSet > > why must we specify the tables here? > mmm... I see, we need specify the direction of desired link Absolutely correct! inRecID this is RecID of tableA > so it could also be implemented by index as in: > vLink.FindLinked( inRecID as Integer, inTableA as Integer, inTableB as > Integer ) as VSet But what sense, Erne? I think pointers on objects is the best choice. > after all we are required to know the Index order of linked tables if > we must link records by: > vLink.LinkRecords( inRecIDs() as Integer ) Agree. Index of table is strict thing. It correspond to the order of creation. > anyway, to specify the 2nd table seems redundant as we have: > vLink.DeleteLinkedRecords( inRecID as Integer, inTableA as VTable ) > vLink.DeleteAllLinkedRecords( inTableA as VTable ) > > so it should be enough: > vLink.FindLinked( inRecID as Integer, inTableA as [VTable or Integer] ) > as VSet I see your point...and > unless vLink will be able to link more than 2 tables in the future Exactly! Bingo! :-) > in which case the LinkRecords, DeleteLinkedRecords and > DeleteAllLinkedRecords methods must be updated NO. because when link will have 3 branches, I still can say vLink.DeleteAllLinkedRecords( inTableA as VTable ) And this means break links to ALL linked tables of TableA. -- 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 Dec 29 22:06:00 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Dec 29 14:06:08 2004 Subject: V4RB2 vLink In-Reply-To: <96E02AC6-3CC5-11D9-898A-003065514D2E@tiscalinet.it> Message-ID: On 11/22/04 10:32 PM, "erne" wrote: > What the BranchCount property indicates? Can it be more than 2? In Valentina 2.0 it is always 2. Yes in future we plan add support of N-branch links. Just I do not see yet full picture. Please download new V4RBRef.pdf It now have description of 2 more classes Vlink + ---- VLink2 +------ VBinaryLink So Vlink is general kind link. Can be N-link. Link2 adds more properties and methods for Link with 2 branches. Note that today we have 3 links of link2 ObjectPtr ForeignKey BinaryLink So we always must be able CAST TO Link2 any of this 3 classes, And work with it on general abstraction of link-2 > Why the Owner must be one of the linked tables? Can't vLink live at DB > level? Erne, please note that Links are on the db level. You ask DB to create or drop link. Owner here have sense of CASCADE deletion. If you delete record of owner then child records can be deleted. > What exactly Temporary means? The link will be deleted on DB.Close? The same as for table. Temporary Table or link. It is stored in the .tmp volume. > and finally, how can vLink be used via SQL? > Thanks for listening > (and, of course, for all these new supercool features :-) Please look at our SQL grammar. Link is on the same page where you download V4RB 2.0 By memory, we have CREATE LINK DROP LINK INSERT INTO LINK We have extend the syntax of SQL92 join to use Link instead of fields SQL92 FROM T1 join T2 on T1.id = T2.ptr Valentina also can do FROM T1 join T2 on LinkName -- 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 Dec 30 09:02:32 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 30 02:02:55 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: Hi Ruslan, on 29-12-2004 20:22, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > Sorry for delay. I have not realize here present questions. > So here we go. > not a problem at all :-) -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From ernestogiannotta at tiscalinet.it Thu Dec 30 09:17:27 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 30 02:17:37 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: on 29-12-2004 20:28, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: >> after all we are required to know the Index order of linked tables if >> we must link records by: >> vLink.LinkRecords( inRecIDs() as Integer ) > > Agree. Index of table is strict thing. > It correspond to the order of creation. > >> anyway, to specify the 2nd table seems redundant as we have: >> vLink.DeleteLinkedRecords( inRecID as Integer, inTableA as VTable ) >> vLink.DeleteAllLinkedRecords( inTableA as VTable ) >> >> so it should be enough: >> vLink.FindLinked( inRecID as Integer, inTableA as [VTable or Integer] ) >> as VSet > > I see your point...and > >> unless vLink will be able to link more than 2 tables in the future > > Exactly! Bingo! :-) > >> in which case the LinkRecords, DeleteLinkedRecords and >> DeleteAllLinkedRecords methods must be updated > > NO. because when link will have 3 branches, I still can say > > vLink.DeleteAllLinkedRecords( inTableA as VTable ) > > And this means break links to ALL linked tables of TableA. > still vLink.LinkRecords( inRecIDs() as Integer ) remains not consistent... it does not give way to specify link direction should be vLink.LinkRecords( inRecIDs() as Integer, inTableA as VTable, inTableB as VTable ) just like in the vLink.FindLinked syntax Cool Runnings, Erne. -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From sunshine at public.kherson.ua Thu Dec 30 12:05:23 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 04:05:33 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: On 12/30/04 10:17 AM, "erne" wrote: >>> after all we are required to know the Index order of linked tables if >>> we must link records by: >>> vLink.LinkRecords( inRecIDs() as Integer ) >> >> Agree. Index of table is strict thing. >> It correspond to the order of creation. >> >>> anyway, to specify the 2nd table seems redundant as we have: >>> vLink.DeleteLinkedRecords( inRecID as Integer, inTableA as VTable ) >>> vLink.DeleteAllLinkedRecords( inTableA as VTable ) >>> >>> so it should be enough: >>> vLink.FindLinked( inRecID as Integer, inTableA as [VTable or Integer] ) >>> as VSet >> >> I see your point...and >> >>> unless vLink will be able to link more than 2 tables in the future >> >> Exactly! Bingo! :-) >> >>> in which case the LinkRecords, DeleteLinkedRecords and >>> DeleteAllLinkedRecords methods must be updated >> >> NO. because when link will have 3 branches, I still can say >> >> vLink.DeleteAllLinkedRecords( inTableA as VTable ) >> >> And this means break links to ALL linked tables of TableA. >> > > still > > vLink.LinkRecords( inRecIDs() as Integer ) > > remains not consistent... it does not give way to specify link direction and > should be > > vLink.LinkRecords( inRecIDs() as Integer, inTableA as VTable, inTableB as > VTable ) > > just like in the vLink.FindLinked syntax No no no. When you create link you specify order of tables: db.CreateBinaryLink( ... T1, T2 ...) So when you want link 2 records of these tables you do link.LinkRecords( Array( 3, 5) ) And you and Valentina know that 3 -- this is RecID of T1 5 -- this is RecID of T2. Direction is not important here. Absolutely. -- 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 Dec 30 12:01:17 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 30 05:01:41 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: on 30-12-2004 11:05, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > On 12/30/04 10:17 AM, "erne" wrote: > >>>> after all we are required to know the Index order of linked tables if >>>> we must link records by: >>>> vLink.LinkRecords( inRecIDs() as Integer ) >>> >>> Agree. Index of table is strict thing. >>> It correspond to the order of creation. >>> >>>> anyway, to specify the 2nd table seems redundant as we have: >>>> vLink.DeleteLinkedRecords( inRecID as Integer, inTableA as VTable ) >>>> vLink.DeleteAllLinkedRecords( inTableA as VTable ) >>>> >>>> so it should be enough: >>>> vLink.FindLinked( inRecID as Integer, inTableA as [VTable or Integer] ) >>>> as VSet >>> >>> I see your point...and >>> >>>> unless vLink will be able to link more than 2 tables in the future >>> >>> Exactly! Bingo! :-) >>> >>>> in which case the LinkRecords, DeleteLinkedRecords and >>>> DeleteAllLinkedRecords methods must be updated >>> >>> NO. because when link will have 3 branches, I still can say >>> >>> vLink.DeleteAllLinkedRecords( inTableA as VTable ) >>> >>> And this means break links to ALL linked tables of TableA. >>> >> >> still >> >> vLink.LinkRecords( inRecIDs() as Integer ) >> >> remains not consistent... it does not give way to specify link direction > > and > >> should be >> >> vLink.LinkRecords( inRecIDs() as Integer, inTableA as VTable, inTableB as >> VTable ) >> >> just like in the vLink.FindLinked syntax > > No no no. > > When you create link you specify order of tables: > > db.CreateBinaryLink( ... T1, T2 ...) > that's why I asked option to refer them by index also > So when you want link 2 records of these tables you do > > link.LinkRecords( Array( 3, 5) ) > > And you and Valentina know that > > 3 -- this is RecID of T1 > 5 -- this is RecID of T2. > > Direction is not important here. Absolutely. still... this limits linked tables to 2 and not more and... what about a 2 dimensional array to link several record pairs at once? e.g. dim LinksArray(2,2) LinksArray(0,1) = 3 //-- this is RecID of T1 LinksArray(0,2) = 5 //-- this is RecID of T2 LinksArray(1,1) = 3 //-- this is RecID of T1 LinksArray(1,2) = 6 //-- this is RecID of T2 LinksArray(2,1) = 4 //-- this is RecID of T1 LinksArray(2,2) = 5 //-- this is RecID of T2 link.LinkRecords( LinksArray, T1, T2 ) // specify Tables for future multitables links imagine this: dim LinksArray(1,3) LinksArray(0,1) = 3 //-- this is RecID of T1 LinksArray(0,2) = 5 //-- this is RecID of T2 LinksArray(0,3) = 5 //-- this is RecID of T3 LinksArray(1,1) = 3 //-- this is RecID of T1 LinksArray(1,2) = 6 //-- this is RecID of T2 LinksArray(1,3) = 4 //-- this is RecID of T3 link.LinkRecords( LinksArray, T1, T2, T3 ) // specify Tables for future multitables links Cool Runnings, Erne. -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From IvanSmahin at public.kherson.ua Thu Dec 30 13:05:13 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 30 05:05:15 2004 Subject: locked records In-Reply-To: References: Message-ID: <1169365047.20041230130513@public.kherson.ua> Hello jda, Wednesday, December 29, 2004, 3:50:15 PM, you wrote: j> Hi Ruslan, j> I've run across a situation where can't make a cursor because a j> "record (null) is locked in the table "mytable". j> I've quit my app and opened the database -- I get the same error. j> Shouldn't locked records revert to unlocked when the db is closed? If j> so, either that isn't happening in this case, or the error message is j> incorrect. j> Jon j> _______________________________________________ j> Valentina-beta mailing list j> Valentina-beta@lists.macserve.net j> http://lists.macserve.net/mailman/listinfo/valentina-beta That is strange. Could you give me an error code and message. And the simple project with this problem? -- Best regards, Ivan mailto:IvanSmahin@public.kherson.ua From sunshine at public.kherson.ua Thu Dec 30 14:17:13 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 06:17:22 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: On 12/30/04 1:01 PM, "erne" wrote: >> When you create link you specify order of tables: >> >> db.CreateBinaryLink( ... T1, T2 ...) >> > > that's why I asked option to refer them by index also A Table ? But we have this Vlink.Table( index ) Vlink.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 sunshine at public.kherson.ua Thu Dec 30 14:20:49 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 06:20:59 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: On 12/30/04 1:01 PM, "erne" wrote: >> When you create link you specify order of tables: >> >> db.CreateBinaryLink( ... T1, T2 ...) >> So when you want link 2 records of these tables you do >> >> link.LinkRecords( Array( 3, 5) ) >> >> And you and Valentina know that >> >> 3 -- this is RecID of T1 >> 5 -- this is RecID of T2. >> >> Direction is not important here. Absolutely. > > still... this limits linked tables to 2 and not more No limits here. Why you think so? > and... > what about a 2 dimensional array to link several record pairs at once? > > e.g. > dim LinksArray(2,2) > > LinksArray(0,1) = 3 //-- this is RecID of T1 > LinksArray(0,2) = 5 //-- this is RecID of T2 > LinksArray(1,1) = 3 //-- this is RecID of T1 > LinksArray(1,2) = 6 //-- this is RecID of T2 > LinksArray(2,1) = 4 //-- this is RecID of T1 > LinksArray(2,2) = 5 //-- this is RecID of T2 > > link.LinkRecords( LinksArray, T1, T2 ) // specify Tables for future > multitables links This have sense...but it is just a cosmetic thing. And I prefer C++ style LinksArray[3][2] = { { 3, 5 }, { 3, 6 }, { 4, 5 } } > imagine this: > > dim LinksArray(1,3) > > LinksArray(0,1) = 3 //-- this is RecID of T1 > LinksArray(0,2) = 5 //-- this is RecID of T2 > LinksArray(0,3) = 5 //-- this is RecID of T3 > LinksArray(1,1) = 3 //-- this is RecID of T1 > LinksArray(1,2) = 6 //-- this is RecID of T2 > LinksArray(1,3) = 4 //-- this is RecID of T3 > > link.LinkRecords( LinksArray, T1, T2, T3 ) // specify Tables for future > multitables links -- 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 Thu Dec 30 07:49:31 2004 From: jda at his.com (jda) Date: Thu Dec 30 06:49:44 2004 Subject: locked records In-Reply-To: <1169365047.20041230130513@public.kherson.ua> References: <1169365047.20041230130513@public.kherson.ua> Message-ID: >Hello jda, > >Wednesday, December 29, 2004, 3:50:15 PM, you wrote: > > >j> Hi Ruslan, > >j> I've run across a situation where can't make a cursor because a >j> "record (null) is locked in the table "mytable". > >j> I've quit my app and opened the database -- I get the same error. >j> Shouldn't locked records revert to unlocked when the db is closed? If >j> so, either that isn't happening in this case, or the error message is >j> incorrect. > >j> Jon >j> _______________________________________________ >j> Valentina-beta mailing list >j> Valentina-beta@lists.macserve.net >j> http://lists.macserve.net/mailman/listinfo/valentina-beta > >That is strange. >Could you give me an error code and message. >And the simple project with this problem? > > The error message was "record (null) is locked in the table "mytable" I'm afraid that I tried to debug this database, and in the course of that the index became totally trashed and wouldn't open at all. I don't know the steps to reproduce this, but it happens to me regularly (when I've created a new database and worked with it for about 10 minutes). I'll try to see if I can figure out the pattern and let you know. Jon From jda at his.com Thu Dec 30 09:15:12 2004 From: jda at his.com (jda) Date: Thu Dec 30 08:15:28 2004 Subject: locked records In-Reply-To: <1169365047.20041230130513@public.kherson.ua> References: <1169365047.20041230130513@public.kherson.ua> Message-ID: > >That is strange. >Could you give me an error code and message. >And the simple project with this problem? > Ivan, I've managed to get another database to give the same error. It clearly has to do with deleting records -- possible it is related to Mantis bug 219, for which I provided an easy way to reproduce. Anyway, the error is Record "(null)" is locked in table "myTable". and the error number is 460037 Closing and reopening the database does not fix this error. I can send you the database, if you like. Unfortunately it is 1.4 MB zipped. I can email it to you or upload it to my ftp site. Please let me know if you want it. Jon From sunshine at public.kherson.ua Thu Dec 30 16:21:55 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 08:22:09 2004 Subject: locked records In-Reply-To: Message-ID: On 12/30/04 4:15 PM, "jda" wrote: >> That is strange. >> Could you give me an error code and message. >> And the simple project with this problem? >> > > Ivan, I've managed to get another database to give the same error. It > clearly has to do with deleting records -- possible it is related to > Mantis bug 219, for which I provided an easy way to reproduce. > > Anyway, the error is > > Record "(null)" is locked in table "myTable". > > and the error number is > > 460037 > > Closing and reopening the database does not fix this error. > > I can send you the database, if you like. Unfortunately it is 1.4 MB > zipped. I can email it to you or upload it to my ftp site. Please let > me know if you want it. Send to Ivan's email. Ivan please open this db in viSQL and check if any LOCK files present here. All lock files actually must go to .tmp volume. So they must die on the next open. May be we have made mistake and put them (sometimes) not to .tmp volume... -- 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 IvanSmahin at public.kherson.ua Thu Dec 30 16:42:50 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 30 08:42:51 2004 Subject: locked records In-Reply-To: References: <1169365047.20041230130513@public.kherson.ua> Message-ID: <41D413EA.1040909@public.kherson.ua> jda ?????: >> >> That is strange. >> Could you give me an error code and message. >> And the simple project with this problem? >> > > Ivan, I've managed to get another database to give the same error. It > clearly has to do with deleting records -- possible it is related to > Mantis bug 219, for which I provided an easy way to reproduce. > > Anyway, the error is > > Record "(null)" is locked in table "myTable". > > and the error number is > > 460037 > > Closing and reopening the database does not fix this error. > > I can send you the database, if you like. Unfortunately it is 1.4 MB > zipped. I can email it to you or upload it to my ftp site. Please let me > know if you want it. > > Jon > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > I'll try to resolve this problem without your db. Just one question - how much records in the cursor do you expect to get? It seems to be solved already - in 2.0 b33. There was 222 mantis bug. "There is unable to lock more than 1024 records." -- Best regards, Ivan Smahin. From sunshine at public.kherson.ua Thu Dec 30 16:47:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 08:47:43 2004 Subject: locked records In-Reply-To: <41D413EA.1040909@public.kherson.ua> Message-ID: On 12/30/04 4:42 PM, "Ivan Smahin" wrote: > > I'll try to resolve this problem without your db. > Just one question - how much records in the cursor do you expect to get? > It seems to be solved already - in 2.0 b33. > There was 222 mantis bug. > "There is unable to lock more than 1024 records." Aha, really! Jon, if you have more than 1024 records in table then this is 222 bug. It is fixed already -- 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 Thu Dec 30 09:50:26 2004 From: jda at his.com (jda) Date: Thu Dec 30 08:50:35 2004 Subject: locked records In-Reply-To: <41D413EA.1040909@public.kherson.ua> References: <1169365047.20041230130513@public.kherson.ua> <41D413EA.1040909@public.kherson.ua> Message-ID: > >I'll try to resolve this problem without your db. Too late. :-) >Just one question - how much records in the cursor do you expect to get? >It seems to be solved already - in 2.0 b33. >There was 222 mantis bug. >"There is unable to lock more than 1024 records." > There is a 1089 "recordcount", and 1090 "physicalrecordcount" when I look in viSQL. I don't see bug 222 on Mantis (must be internal to Paradigma), but yes, this may be the issue because I seem to see it happens only with rather large databases (but the problem is reported as a record being incorrectly locked, of course). I'll check it with b33 when it is released. Jon From jda at his.com Thu Dec 30 09:55:52 2004 From: jda at his.com (jda) Date: Thu Dec 30 08:55:58 2004 Subject: locked records In-Reply-To: References: Message-ID: > >Aha, really! > >Jon, if you have more than 1024 records in table then this is 222 bug. >It is fixed already > We'll see. I deleted some records to get down to 971 -- I get the same error. Jon From ernestogiannotta at tiscalinet.it Thu Dec 30 16:40:07 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 30 09:40:37 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: on 30-12-2004 13:17, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > On 12/30/04 1:01 PM, "erne" wrote: > >>> When you create link you specify order of tables: >>> >>> db.CreateBinaryLink( ... T1, T2 ...) >>> >> >> that's why I asked option to refer them by index also > > A Table ? > > But we have this > > Vlink.Table( index ) > Vlink.Table( name ) true! this is more than enough :-) -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From ernestogiannotta at tiscalinet.it Thu Dec 30 16:53:17 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 30 09:53:31 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: on 30-12-2004 13:20, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: >>> When you create link you specify order of tables: >>> >>> db.CreateBinaryLink( ... T1, T2 ...) > >>> So when you want link 2 records of these tables you do >>> >>> link.LinkRecords( Array( 3, 5) ) >>> >>> And you and Valentina know that >>> >>> 3 -- this is RecID of T1 >>> 5 -- this is RecID of T2. >>> >>> Direction is not important here. Absolutely. >> >> still... this limits linked tables to 2 and not more > > No limits here. Why you think so? > because if assume: 3 -- this is RecID of T1 5 -- this is RecID of T2. how can I specify when an array element refers to an eventual T3? as in 3 -- this is RecID of T1 5 -- this is RecID of T3. Cool Runnings, Erne. -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From sunshine at public.kherson.ua Thu Dec 30 18:38:26 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 10:38:38 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: On 12/30/04 5:53 PM, "erne" wrote: >>> still... this limits linked tables to 2 and not more >> >> No limits here. Why you think so? >> > > because if assume: > > 3 -- this is RecID of T1 > 5 -- this is RecID of T2. > > how can I specify when an array element refers to an eventual T3? > as in > > 3 -- this is RecID of T1 > 5 -- this is RecID of T3. Aha I see. IMPORTANT POINT is that N-link link all 3 records in the same time! I.e. You always will do LinkRecords( Array(3, 5, 7) ) Only in this case N-link have 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 ernestogiannotta at tiscalinet.it Thu Dec 30 17:50:14 2004 From: ernestogiannotta at tiscalinet.it (erne) Date: Thu Dec 30 10:50:25 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: on 30-12-2004 17:38, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > On 12/30/04 5:53 PM, "erne" wrote: > >>>> still... this limits linked tables to 2 and not more >>> >>> No limits here. Why you think so? >>> >> >> because if assume: >> >> 3 -- this is RecID of T1 >> 5 -- this is RecID of T2. >> >> how can I specify when an array element refers to an eventual T3? >> as in >> >> 3 -- this is RecID of T1 >> 5 -- this is RecID of T3. > > Aha I see. > > IMPORTANT POINT is that N-link link all 3 records in the same time! > > I.e. You always will do > > LinkRecords( Array(3, 5, 7) ) > > Only in this case N-link have sense. > Ok, I see so if I don't have something to link in T2 I'll do LinkRecords( Array(3, 0, 7) ) makes sense? Cool Runnings, Erne. -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From sunshine at public.kherson.ua Thu Dec 30 19:04:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 11:04:51 2004 Subject: V4RB2 vLink In-Reply-To: Message-ID: On 12/30/04 6:50 PM, "erne" wrote: >> Aha I see. >> >> IMPORTANT POINT is that N-link link all 3 records in the same time! >> >> I.e. You always will do >> >> LinkRecords( Array(3, 5, 7) ) >> >> Only in this case N-link have sense. >> > > Ok, I see > > so if I don't have something to link in T2 I'll do > > LinkRecords( Array(3, 0, 7) ) > > makes sense? I think yes. The main is that we store triplet. In any case, we discuss future %-) -- 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 IvanSmahin at public.kherson.ua Thu Dec 30 19:39:36 2004 From: IvanSmahin at public.kherson.ua (Ivan Smahin) Date: Thu Dec 30 11:39:41 2004 Subject: locked records In-Reply-To: References: Message-ID: <41D43D58.1060106@public.kherson.ua> jda ?????: >> >> Aha, really! >> >> Jon, if you have more than 1024 records in table then this is 222 bug. >> It is fixed already >> > > We'll see. I deleted some records to get down to 971 -- I get the same > error. > > Jon > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > It might be no effect from such deletions - because phisical record count remains unchanged in most cases. So you should try to delete last records only - for getting phisical records count less than 1024 - but not random records. Or you can reduce cursor records in other manner - using "where" clause. -- Best regards, Ivan Smahin. From jda at his.com Thu Dec 30 14:15:51 2004 From: jda at his.com (jda) Date: Thu Dec 30 13:16:02 2004 Subject: locked records In-Reply-To: <41D43D58.1060106@public.kherson.ua> References: <41D43D58.1060106@public.kherson.ua> Message-ID: > >It might be no effect from such deletions - because phisical record >count remains unchanged in most cases. So you should try to delete >last records only - for getting phisical records count less than >1024 - but not random records. > >Or you can reduce cursor records in other manner - using "where" clause. > IVan, More investigation has revealed that record 1030 is the culprit -- it is locked, and closing/opening the db does not unlock it. I sent you the db a few hours ago -- hopefully you will see this, too. Jon From sunshine at public.kherson.ua Thu Dec 30 22:20:59 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Dec 30 14:21:07 2004 Subject: locked records In-Reply-To: Message-ID: On 12/30/04 9:15 PM, "jda" wrote: >> >> It might be no effect from such deletions - because phisical record >> count remains unchanged in most cases. So you should try to delete >> last records only - for getting phisical records count less than >> 1024 - but not random records. >> >> Or you can reduce cursor records in other manner - using "where" clause. >> > > IVan, > > More investigation has revealed that record 1030 is the culprit -- it > is locked, and closing/opening the db does not unlock it. I sent you > the db a few hours ago -- hopefully you will see this, too. Jon, I am finish one bug, so today will be b33 -- 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 Dec 31 12:33:56 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 31 04:34:03 2004 Subject: [V4RB] b33 uploaded Message-ID: kernel - 2.0 b33 ================ - 0000217: [SQL nodes] Query " where 2-2=0 is true" does not work. (Ivan Smahin) - 0000241: [Indexes] FindContains() and FindEndsWith() ignore collator settings (Ruslan Zasukhin) - 0000235: [SQL] Crash in case "select * from... group by field". (Ivan Smahin) - 0000234: [SQL] There are wrong results in join case tables have fields with the same name. (Ivan Smahin) - 0000233: [NEW] There are two new error codes added for xLinkError exception. (Ivan Smahin) - 0000232: [Error codes/classes] Table name is not set in many places for ERR_TABLE_NOT_FOUND. (Ivan Smahin) - 0000230: [Fields] I_Field::SetBlank() (Ivan Smahin) - 0000231: [FBL Nodes] Amost no one sql-function controls params count. (Ivan Smahin) - 0000225: [FBL Nodes] String sql-functions don't check argue count. (Ivan Smahin) - 0000222: [Cursors] There is unable to lock more than 1024 records. (Ruslan Zasukhin) - 0000223: [SQL] There is crash in Database.close after exception in subquery. (Ivan Smahin) - 0000221: [SQL] Wrong error message for incorrect aliasing usage. (Ivan Smahin) - 0000220: [FBL Nodes] "HEAD" and "TAIL" methods does not work correctly. (Ivan Smahin) - 0000218: [FBL Nodes] "Replace" ENode returns empty string if no replace was made. (Ivan Smahin) V4RB - 2.0 b33 ============== - 0000240: [Examples] [NEW] API_way and SQL_way Field_Picture examples. (Ruslan Zasukhin) - 0000239: [Change] Vcursor.ReadOnly => .IsReadOnly() (Ruslan Zasukhin) - 0000238: [Change] Vtable.GoToRecId() and GetRecID() made as property .RecID (Ruslan Zasukhin) - 0000237: [Change] VCursor.CurrentPostion => .Position (Ruslan Zasukhin) - 0000236: [Indexes] IndexStyle.IgnoreLength =3 work as = 2 (Ruslan Zasukhin) - 0000226: [NEW] VTable.EOF property (Ruslan Zasukhin) - 0000227: [NEW] VTable.BOF (Ruslan Zasukhin) - 0000228: [NEW] VTable.UpdateAllRecords() method. (Ruslan Zasukhin) - 0000229: [Change] Methods VTable.GetRecID() and GoToRecID now made as property VTable.RecID (Ruslan Zasukhin) -- 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 ------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20041231/3c79ad75/attachment.html From jda at his.com Fri Dec 31 07:27:38 2004 From: jda at his.com (jda) Date: Fri Dec 31 06:27:47 2004 Subject: .deleteField In-Reply-To: References: Message-ID: Hi Ruslan, I create a temp field like this fld = myDatabase.myReferences.createVarCharField("authorEditorSortMethod", 2000, EVFlag.fNone, "IFNULL(authors, IFNULL(editors,title))") and dispose of it later with myDatabase.myReferences.deleteField(fld) With b33 I get an error that deleteField does not exist. Is .dropField the new equivalent? Thanks, Jon From sunshine at public.kherson.ua Fri Dec 31 14:52:20 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 31 06:52:27 2004 Subject: .deleteField In-Reply-To: Message-ID: On 12/31/04 2:27 PM, "jda" wrote: > Hi Ruslan, > > I create a temp field like this > > fld = > myDatabase.myReferences.createVarCharField("authorEditorSortMethod", > 2000, EVFlag.fNone, "IFNULL(authors, IFNULL(editors,title))") > > > and dispose of it later with > > myDatabase.myReferences.deleteField(fld) > > > With b33 I get an error that deleteField does not exist. > > Is .dropField the new equivalent? Yes Jon -- 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 Dec 31 14:52:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 31 06:52:51 2004 Subject: .deleteField In-Reply-To: Message-ID: On 12/31/04 2:27 PM, "jda" wrote: > Hi Ruslan, > > I create a temp field like this > > fld = > myDatabase.myReferences.createVarCharField("authorEditorSortMethod", > 2000, EVFlag.fNone, "IFNULL(authors, IFNULL(editors,title))") > > > and dispose of it later with > > myDatabase.myReferences.deleteField(fld) > > > With b33 I get an error that deleteField does not exist. > > Is .dropField the new equivalent? We correct names to get the same names in all products., And we choose names that are close to SQL commands. -- 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 Fri Dec 31 08:02:42 2004 From: jda at his.com (jda) Date: Fri Dec 31 07:02:48 2004 Subject: b33 first impressions Message-ID: Good news -- the problem with the locked record (null) in myTable seems to be fixed! Not so good news -- bug/issue #55 seems to be a bit worse (at least, I suspect it is related). Now, files with long names give an error -47 (busy) when I try to open them. Even worse, the database is then deleted from the hard drive! Jon From jda at his.com Fri Dec 31 13:36:44 2004 From: jda at his.com (jda) Date: Fri Dec 31 12:37:01 2004 Subject: SQL LIKE Message-ID: Sorry if this is received twice -- it didn't show up for me the first time I sent it: Looking at the latest SQL docs for V2 (p. 24), it seems that LIKE should be acceptable for searching (equal to REGEX). But it doesn't work for me. Should it? Or am I misreading the documentation? Jon From sunshine at public.kherson.ua Fri Dec 31 20:45:30 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 31 12:45:39 2004 Subject: SQL LIKE In-Reply-To: Message-ID: On 12/31/04 8:36 PM, "jda" wrote: > Sorry if this is received twice -- it didn't show up for me the first > time I sent it: > > Looking at the latest SQL docs for V2 (p. 24), it seems that LIKE > should be acceptable for searching (equal to REGEX). But it doesn't > work for me. Should it? Or am I misreading the documentation? It is not such powerful as REGEX, Show your examples, Jon. -- 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 Fri Dec 31 13:53:50 2004 From: jda at his.com (jda) Date: Fri Dec 31 12:54:02 2004 Subject: SQL LIKE In-Reply-To: References: Message-ID: > > Looking at the latest SQL docs for V2 (p. 24), it seems that LIKE >> should be acceptable for searching (equal to REGEX). But it doesn't >> work for me. Should it? Or am I misreading the documentation? > >It is not such powerful as REGEX, > >Show your examples, Jon. > Oh, OK. I tried to do a case-insensitive search with authors LIKE '(?i)ruslan' and got no hits (worked with REGEX '(?i)ruslan' I also tried authors LIKE 'ruslan' no_case and got an error (Table no_case not found). Maybe the V4RB pdf, which shows an example like this on p. 103, is wrong? Jon From sunshine at public.kherson.ua Fri Dec 31 21:11:40 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 31 13:11:49 2004 Subject: SQL LIKE In-Reply-To: Message-ID: On 12/31/04 8:53 PM, "jda" wrote: >>> Looking at the latest SQL docs for V2 (p. 24), it seems that LIKE >>> should be acceptable for searching (equal to REGEX). But it doesn't >>> work for me. Should it? Or am I misreading the documentation? >> >> It is not such powerful as REGEX, >> >> Show your examples, Jon. >> > > Oh, OK. I tried to do a case-insensitive search with > > authors LIKE '(?i)ruslan' ^^^^ this is regex command! If you want in 2.0 do the same you need \ authors REGEX '(?i)ruslan' > and got no hits (worked with REGEX '(?i)ruslan' > I also tried > authors LIKE 'ruslan' no_case and got an error (Table no_case not found). ^^^^^^^^^^ no_case key word from 1.x not implemented yet in 2.0 > Maybe the V4RB pdf, which shows an example like this on p. 103, is wrong? I need to check 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 sunshine at public.kherson.ua Fri Dec 31 23:22:44 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Dec 31 15:22:53 2004 Subject: [V4RB 0000242]: Database.Open crashes for my NEW database In-Reply-To: Message-ID: On 12/31/04 10:31 PM, "vbt@paradigma.ukrcom.kherson.ua" wrote: > The following issue has been SUBMITTED. > ====================================================================== > > ====================================================================== > Reported By: Frank Schima > Assigned To: > ====================================================================== > Project: V4RB > Issue ID: 242 > Category: API > Reproducibility: always > Severity: crash > Priority: normal > Status: new > ====================================================================== > Date Submitted: 12-31-2004 22:31 EET > Last Modified: 12-31-2004 22:31 EET > ====================================================================== > Summary: Database.Open crashes for my NEW database > Description: > Attempting to open my newly created database with ViSQL 2 causes a crash. > I'm sending a project that creates the database to Ruslan. Guys, Forget to mention. I have change format of system tables, So b33 cannot open dbs from b32 !!!!!!!!!!!!! You need create new db by b33 -- 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 Fri Dec 31 12:07:28 2004 From: jda at his.com (jda) Date: Fri Dec 31 17:56:58 2004 Subject: SQL LIKE Message-ID: Looking at the latest SQL docs for V2 (p. 24), it seems that LIKE should be acceptable for searching (equal to REGEX). But it doesn't work for me. Should it? Or am I misreading the documentation? Jon