From Ed at Kleban.com Wed Mar 1 16:34:41 2006 From: Ed at Kleban.com (Ed Kleban) Date: Wed Mar 1 23:17:57 2006 Subject: [ANN] ROADMAP for next builds In-Reply-To: Message-ID: I've no info on RB app sales, but I will offer a few comments from presentations at REAL World 2006: Geoff did show a graph of RB sales, which I can probably drag a copy out of the presentations CD we got, showing that that Linux sales were just starting, but growing strong and are going to be a major focus of RS sales in the coming year. He also mentioned that on one of the unix platforms, Linspire I believe, that RB was the top unavailable application asked for, not just across development tools, but across all tools. RS also announced that that they are partnering with Linspire and that RS is now available as a Linspire download. Also, Matt noted later in a Linspire session that there is no cost for the standard edition of RB for Linux. I also have to add that the Linspire presentation was one of the most impressive presentations I saw at RW. Reading the NUG list I've frequently had the question of, "Ok, so if I want to mess around with unix, which of the numerous port should I try and deploy. After seeing the Linspire presentation that questions has been completely answered as far as I'm concerned, and Linspire is certainly the Unix port of choice for my personal needs. It is oriented toward desktop and home users rather than for use on a server, and the support, ease-of-use, and very low cost are very attractive. I just checked the presentation CD. I didn't see a presentation there for Geoff's keynote unfortunately. Hope this helps. --Ed On 2/24/06 9:01 AM, "Ruslan Zasukhin" wrote: > On 2/24/06 4:56 PM, "Ken Jordan" wrote: > >> And the Linux version of V4RB? :-) > > Anybody from RB developers already have experience of sales > Of their applications on Linux ? > > Compare to other mac + win markets ? > From Ed at Kleban.com Wed Mar 1 16:35:35 2006 From: Ed at Kleban.com (Ed Kleban) Date: Wed Mar 1 23:18:01 2006 Subject: [ANN] ROADMAP for next builds In-Reply-To: Message-ID: On 2/24/06 6:33 AM, "Ruslan Zasukhin" wrote: > Hi All members of Beta list, > > I want inform you what to expect in nearest builds. > > 9) and of course Link improvements as I have discuss with Ed. > Three cheers and a green balloon! From ktekinay at mactechnologies.com Sat Mar 4 04:29:14 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sat Mar 4 03:29:23 2006 Subject: FindLinkedAsBitSet crashes with nil Message-ID: I'll file a report later today, but I discovered that the following statement will crash my RB program without an exception. link.FindLinkedAsBitSet( nil, tbl1, tbl2 ) That should just return an empty set or nil. At the very least, it should throw an exception. BTW, I discovered this because field.FindContains is returning nil if the term is not found. I'm under the impression that it wasn't like this before 2.2, but perhaps I'm wrong. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sat Mar 4 12:03:11 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 4 04:03:18 2006 Subject: FindLinkedAsBitSet crashes with nil In-Reply-To: Message-ID: On 3/4/06 11:29 AM, "Kem Tekinay" wrote: Hi Kem, > I'll file a report later today, but I discovered that the following > statement will crash my RB program without an exception. > > link.FindLinkedAsBitSet( nil, tbl1, tbl2 ) > > That should just return an empty set or nil. At the very least, it should > throw an exception. right > BTW, I discovered this because field.FindContains is returning nil if the > term is not found. I'm under the impression that it wasn't like this before > 2.2, but perhaps I'm wrong. Yes few weeks ago we have make sure that functions which return Set in case of ZERO found records return always NULL but not empty set. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sat Mar 4 10:15:29 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sat Mar 4 09:15:37 2006 Subject: FindLinkedAsBitSet crashes with nil In-Reply-To: Message-ID: On 3/4/06 5:03 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > Yes few weeks ago we have make sure that functions which return Set in case > of ZERO found records return always NULL but not empty set. Just to be sure I understand, any search that yields no records will now return nil, not a set of zero records, yes? And if that's true, I suppose any search that takes a set as a parameter should be able to accept nil, yes? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sat Mar 4 17:41:13 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 4 09:41:19 2006 Subject: FindLinkedAsBitSet crashes with nil In-Reply-To: Message-ID: On 3/4/06 5:15 PM, "Kem Tekinay" wrote: >> Yes few weeks ago we have make sure that functions which return Set in case >> of ZERO found records return always NULL but not empty set. > > Just to be sure I understand, any search that yields no records will now > return nil, not a set of zero records, yes? yes > And if that's true, I suppose any search that takes a set as a parameter > should be able to accept nil, yes? yes -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sat Mar 4 12:35:36 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sat Mar 4 11:35:43 2006 Subject: DIAGNOSE crashes Message-ID: Calling DIAGNOSE with no parameters leads to an instant crash even with a brand new copy of my database. The relevant code is: gDB.Open( f ) if not gDB.Diagnose() then MsgBox( "The database appears to be corrupted." ) Quit end if RB never gets to the MsgBox. I'll file a report later. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sun Mar 5 00:00:25 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 4 16:00:33 2006 Subject: DIAGNOSE crashes In-Reply-To: Message-ID: On 3/4/06 7:35 PM, "Kem Tekinay" wrote: > Calling DIAGNOSE with no parameters leads to an instant crash even with a > brand new copy of my database. The relevant code is: > > gDB.Open( f ) > if not gDB.Diagnose() then > MsgBox( "The database appears to be corrupted." ) > Quit > end if > > RB never gets to the MsgBox. > > I'll file a report later. This is already fixed! Will be available in next beta build You can use for now SqlExecute( "DIAGNOSE DATABASE" ) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sat Mar 4 17:40:25 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sat Mar 4 16:40:31 2006 Subject: DIAGNOSE crashes In-Reply-To: Message-ID: On 3/4/06 5:00 PM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > This is already fixed! > Will be available in next beta build That's great news. Did you get the e-mails I sent to your directly about the database corruption? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sun Mar 5 10:45:20 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Mar 5 02:45:34 2006 Subject: DIAGNOSE crashes In-Reply-To: Message-ID: On 3/5/06 12:40 AM, "Kem Tekinay" wrote: > On 3/4/06 5:00 PM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > >> This is already fixed! >> Will be available in next beta build > > That's great news. > > Did you get the e-mails I sent to your directly about the database > corruption? Yes I did -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Sun Mar 5 18:34:32 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Mar 5 10:35:03 2006 Subject: [ANN] V4RB, V4MD mac, 2.3 b2 Message-ID: Hi All, You can download MAC builds of V4RB and V4MD 2.3 b2. * V4MD and V4RB now have Examples/TestProject. This new project will have target to test C++ code of plugins itself, not code of kernel (for kernel we have own C++ tests). This project will grow with time. It should test all and each function that plugin expose. In all possible modes. For now I have add into each project only few tests which I have use to tests reported last bugs. kernel - 2.3 b1 =============== - 0001362: [SQL] Order by clause result effected by field order - 0001337: [SQL] [NEW] SHOW DATABASE - 0001339: [SQL] [NEW] SHOW COLUMNS - 0001338: [SQL] [NEW] SHOW TABLES V4RB - 2.3 b2 ============= - 0001229: [API-way] Allow For Naming Of The Link In The Constructor When Creating A VObjectPtr - 0001364: [Links] link.FindLinkedAsBitSet will cause a crash if nil is used as first parameter - 0001288: [API] vdatabase.diagnose -> crash V4RB - 2.3 b1 ============= - 0001361: [Database] Trashing index causes an Error - 0001344: [API] db= new VDatabase(VServer) does not work V4MD - 2.3 b1 ============= - 0001347: [Import/Export] [V4MD] convert_1_2 - all text is -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20060305/bd25f0f3/attachment-0001.html From sunshine at public.kherson.ua Sun Mar 5 18:37:15 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Mar 5 10:37:27 2006 Subject: INSERT not working In-Reply-To: <44086DC3.3040307@mediamacros.com> Message-ID: On 3/3/06 6:24 PM, "Chuck Neal" wrote: Chuck, This is fixed in 2.3 b2 (mac) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sun Mar 5 19:48:59 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 5 18:49:06 2006 Subject: Rebuild indexes? Message-ID: When I launched my app for the first time after installing 2.3b2, a search on a related text field took over 400 seconds with a lot of hard drive activity. After that, searches took around 3 seconds even after quitting and restarting the app and REALbasic. Did the beta rebuild the indexes? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Mon Mar 6 10:28:12 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 6 02:28:35 2006 Subject: Rebuild indexes? In-Reply-To: Message-ID: On 3/6/06 2:48 AM, "Kem Tekinay" wrote: Hi Kem, > When I launched my app for the first time after installing 2.3b2, a search > on a related text field took over 400 seconds with a lot of hard drive > activity. After that, searches took around 3 seconds even after quitting and > restarting the app and REALbasic. Did the beta rebuild the indexes? I think not beta itself but engine. Probably it have recognize something bad. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Tue Mar 7 01:56:56 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 6 17:57:05 2006 Subject: [ANN] V4MD and VCOM (win) 2.3 b2 uploaded Message-ID: Hi All, http://paradigmasoft.com/en/products/download/beta ChangeLog: http://valentina-db.com/bt/changelog_page.php -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Wed Mar 8 14:57:18 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Mar 8 06:57:24 2006 Subject: [ANN] V4RB 2.3 b3 MAC Message-ID: Hi All, V4RB - 2.3 b3 ============= - 0001364: [Links] link.FindLinkedAsBitSet will cause a crash if nil is used as first parameter - 0001229: [API-way] Allow For Naming Of The Link In The Constructor When Creating A VObjectPtr -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20060308/abc7fa9f/attachment.html From ktekinay at mactechnologies.com Wed Mar 8 08:09:45 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Wed Mar 8 07:09:49 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: Does this do anything about the bad index on boolean problem? Is that something I even need to worry about? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Wed Mar 8 15:34:04 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Mar 8 07:34:09 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/8/06 3:09 PM, "Kem Tekinay" wrote: Hi Kem, > Does this do anything about the bad index on boolean problem? Is that > something I even need to worry about? Have you made report into Mantis about this ? -------------- Please report this into mantis * Each separate issue report into separate issue report. * if you have database or project then ATTACH it compressed to report. you can put up to 250Kb archives. * Provide in the report as many detailed info as you can. * If possible provide EXACT steps to reproduce problem. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Wed Mar 8 08:47:41 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Wed Mar 8 07:47:45 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/8/06 8:34 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > Have you made report into Mantis about this ? No, as you requested, I had sent the database that demonstrated the problem directly to you. Should I also file this on Mantis? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Wed Mar 8 16:19:23 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Mar 8 08:19:30 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/8/06 3:47 PM, "Kem Tekinay" wrote: > On 3/8/06 8:34 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > >> Have you made report into Mantis about this ? > > No, as you requested, I had sent the database that demonstrated the problem > directly to you. Should I also file this on Mantis? Hmm, it seems I have not get it -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Wed Mar 8 14:22:45 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Wed Mar 8 13:22:51 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/8/06 9:19 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > Hmm, it seems I have not get it I just resent it. Please confirm. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Wed Mar 8 23:06:46 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Mar 8 15:06:53 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/8/06 9:22 PM, "Kem Tekinay" wrote: >> Hmm, it seems I have not get it > > I just resent it. Please confirm. Still nothing from you. Try sunshine@public.kherson.ua ruslan@paradigmasoft.com ruslan@valentina-db.com -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Wed Mar 8 20:10:39 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Wed Mar 8 19:10:45 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/8/06 4:06 PM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > Still nothing from you. > > Try I just resent to all three addresses. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Fri Mar 10 11:14:40 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 10 03:14:48 2006 Subject: [ANN] VCOM 2.3 b3 uploaded. Message-ID: Hi All, You can download b3 of VCOM. http://paradigmasoft.com/en/products/download/beta -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Fri Mar 10 11:50:43 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 10 03:50:54 2006 Subject: [ANN] Valentina Studio 2.3 fc8 for Windows uploaded. Message-ID: Hi All, You can download this next big build of Vstudio. This build contains all fixes of kernel up to 2.3 b3 And the following list of own issues. * I?d like underline that now VSTudio have ability to show easy wrong methods as icon of field or table with red (!) * Also we have introduce auto-completion for SQL browser. It is stupid for now, but we improve it in next build more. ---------------------------------------------------------------------------- ---------------- - 0001371: [Schema Browser] Method with wrong value and parent table now show with special icon (igor_nikita) - 0001370: [Application Menu] Autocompletion now work in SQL Editor (igor_nikita) - 0001369: [SQL Editor] SQL Editor now use tooltips as well as in DataBrowser (igor_nikita) - 0001359: [SQL Editor] List "Recent..." contains wrong ( duplicate ) strings (igor_nikita) - 0001358: [Schema Browser] Schema browser show temporary tables (igor_nikita) - 0001356: [Property Inspector] Property Inspector remember collapsed( expanded ) state of preview for pictures in DataBrowser (igor_nikita) - 0001355: [Property Inspector] All window use one Property Inspector now (igor_nikita) - 0001354: [Property Inspector] User can change "Nullable" flag for fields with types Blob, Text, Picture (igor_nikita) - 0001353: [Schema Browser] Search now work for current column (igor_nikita) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20060310/8b812d54/attachment.html From sunshine at public.kherson.ua Fri Mar 10 16:55:19 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 10 08:55:25 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/9/06 3:10 AM, "Kem Tekinay" wrote: Kem, I have open your database in viSQL using latest V4RB 2.3 b3 Make Diagnose Database VeryHigh -- no problems. Strange. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From maxprog at mac.com Fri Mar 10 19:23:08 2006 From: maxprog at mac.com (Stan Busk) Date: Fri Mar 10 12:23:20 2006 Subject: [V4RB] I found out how to corrupt a database with RB IDE Message-ID: Hi, I have a case here that always corrupt the database and is fully reproducible. First I reindex my database with myDatabase.SQLExecute ( "REINDEX DATABASE" ), once done I switch to RB IDE and press Stop. If I try to reopen the database, several tables are corrupt. Actually I think it is not usable anymore. Trashing index doesn't help and when opening Valentina Studio you get an error and then only half tables. It is quite interesting as all the corrupt databases sent by my clients so far use to have the same symptoms. Note that switching to RB IDE and pressing the Stop button without reindexing the database first doesn't cause any problems, it is only if you reindex and then press stop. Does it help? Could it be possible to look at that? I can open a Mantis ticket and a sample project of course. Stan From ktekinay at mactechnologies.com Fri Mar 10 15:18:38 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Fri Mar 10 14:18:45 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/10/06 9:55 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > I have open your database in viSQL using latest V4RB 2.3 b3 > > Make Diagnose Database VeryHigh -- no problems. I included the report from Valentina Studio 2.2fc7 and just ran it again on that file with the same result. Is the report of a problem faulty? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sat Mar 11 09:55:18 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 01:55:25 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/10/06 10:18 PM, "Kem Tekinay" wrote: >> I have open your database in viSQL using latest V4RB 2.3 b3 >> >> Make Diagnose Database VeryHigh -- no problems. > > I included the report from Valentina Studio 2.2fc7 and just ran it again on > that file with the same result. > > Is the report of a problem faulty? At least now I do not see such problems. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Sat Mar 11 09:56:51 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 01:56:58 2006 Subject: [V4RB] I found out how to corrupt a database with RB IDE In-Reply-To: Message-ID: On 3/10/06 8:23 PM, "Stan Busk" wrote: Hi Stan, We will try reproduce. How many tables you have? What size of db files ? Do you keep all in one file or in few ? > I have a case here that always corrupt the database and is fully > reproducible. First I reindex my database with myDatabase.SQLExecute > ( "REINDEX DATABASE" ), once done I switch to RB IDE and press Stop. > If I try to reopen the database, several tables are corrupt. Actually > I think it is not usable anymore. Trashing index doesn't help and > when opening Valentina Studio you get an error and then only half > tables. It is quite interesting as all the corrupt databases sent by > my clients so far use to have the same symptoms. > > Note that switching to RB IDE and pressing the Stop button without > reindexing the database first doesn't cause any problems, it is only > if you reindex and then press stop. > > Does it help? Could it be possible to look at that? I can open a > Mantis ticket and a sample project of course. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Sat Mar 11 10:26:00 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 02:26:13 2006 Subject: [INFO] VSDK C++ WIN and Visual 2005 Message-ID: Hi Guys, I have test just only, the archive of VSDK WIN which you can find on our site is build with help of Visual 2003 But it works with Visual 2005 without problems as far as I see. I have open Examples.sln, Visual have convert projects. And I am able compile, link and run all examples. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From maxprog at mac.com Sat Mar 11 12:26:04 2006 From: maxprog at mac.com (Stan Busk) Date: Sat Mar 11 05:26:15 2006 Subject: [V4RB] I found out how to corrupt a database with RB IDE In-Reply-To: References: Message-ID: Hi, I have tried with 2 different Databases that have the EVDbMode.kDscDatBlbInd and EVDbMode.kDsc_Dat_Blb_Ind properties, same results. First has eleven tables (67 MB) and second twenty (2.7 MB). I am trying to create a sample project but it is harder than expected. Stan > On 3/10/06 8:23 PM, "Stan Busk" wrote: > > Hi Stan, > > We will try reproduce. > > How many tables you have? > What size of db files ? > > Do you keep all in one file or in few ? > >> I have a case here that always corrupt the database and is fully >> reproducible. First I reindex my database with myDatabase.SQLExecute >> ( "REINDEX DATABASE" ), once done I switch to RB IDE and press Stop. >> If I try to reopen the database, several tables are corrupt. Actually >> I think it is not usable anymore. Trashing index doesn't help and >> when opening Valentina Studio you get an error and then only half >> tables. It is quite interesting as all the corrupt databases sent by >> my clients so far use to have the same symptoms. >> >> Note that switching to RB IDE and pressing the Stop button without >> reindexing the database first doesn't cause any problems, it is only >> if you reindex and then press stop. >> >> Does it help? Could it be possible to look at that? I can open a >> Mantis ticket and a sample project of course. > > -- > Best regards, > > Ruslan Zasukhin > VP Engineering and New Technology > Paradigma Software, Inc > > Valentina - Joining Worlds of Information > http://www.paradigmasoft.com > > [I feel the need: the need for speed] > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From ktekinay at mactechnologies.com Sat Mar 11 06:57:59 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sat Mar 11 05:58:05 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/11/06 2:55 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > At least now I do not see such problems. If I use Studio to reindex that database, it goes away. But any new database I create shows the error as soon as I add one record to that table. The line that makes the difference is this: fldEnabled = new VBoolean( "Enabled", EVFlag.fIndexed ) There is no problem if I change it to: fldEnabled = new VBoolean( "Enabled", EVFlag.fNone ) Do I remember correctly that you said that a boolean field cannot be indexed anyway? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sat Mar 11 20:53:18 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 12:53:26 2006 Subject: [V4RB] I found out how to corrupt a database with RB IDE In-Reply-To: Message-ID: On 3/11/06 1:26 PM, "Stan Busk" wrote: Hi Stan, > I have tried with 2 different Databases that have the > EVDbMode.kDscDatBlbInd and EVDbMode.kDsc_Dat_Blb_Ind properties, same > results. First has eleven tables (67 MB) and second twenty (2.7 MB). > > I am trying to create a sample project but it is harder than expected. No need, we will try something that we have. So it can happens on small db also. Ok. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Sat Mar 11 20:55:16 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 12:55:21 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/11/06 1:57 PM, "Kem Tekinay" wrote: Hi Kem, > On 3/11/06 2:55 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > >> At least now I do not see such problems. > > If I use Studio to reindex that database, it goes away. But any new database > I create shows the error as soon as I add one record to that table. > > The line that makes the difference is this: > > fldEnabled = new VBoolean( "Enabled", EVFlag.fIndexed ) > > There is no problem if I change it to: > > fldEnabled = new VBoolean( "Enabled", EVFlag.fNone ) > > Do I remember correctly that you said that a boolean field cannot be indexed > anyway? Yes Boolean field do not have index never. Because of its nature. Hmm, so If you specify flag fIndexed problem comes... Please send me again this db with one record which diagnose complains. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Sat Mar 11 21:36:02 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 13:36:16 2006 Subject: [VSDK 2] Valentina 2 C++ SDK In-Reply-To: <7F5BBC3B7975A94F9844614AB95C3755065C3F@srvex01.se-dd.local> Message-ID: On 3/11/06 3:10 PM, "Decoster Jean Louis" wrote: > Hi Ruslan, > > Thank your very much for your fast action. It is very good and now I can work > with Valentina 2 to solve my problems. > Thanks again for your support. > Best Regards Jean-Louis Hi Jean, Good. I will be very happy to hear your comments on design and API of Valentina C++ SDK 2.x In 2.x C++ engine we have made revolutionary changes as you can see. Although public API very similar to what you did have in 1.x My main concern now about C++ developers is: VSDK 2.x do not give you CLASS WAY. So If you are used to this style of coding, I'd like to discuss what we can do here. I see one way easy enough to get this style back into game using simple wrapper class of smart-ptr kind which use trick of operator -> Only protential problem for this implementation that we will use -> for non-pointer objects class tnlPerson { VString mName; } .... mName->SetString() // C++ developer expect normally here // mName.SetString() -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From Claudius at sailer-online.de Sat Mar 11 20:46:17 2006 From: Claudius at sailer-online.de (Claudius Sailer) Date: Sat Mar 11 13:46:20 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: References: Message-ID: Hi Ruslan, Am 11.03.2006 um 19:55 schrieb Ruslan Zasukhin: > On 3/11/06 1:57 PM, "Kem Tekinay" > wrote: > > Hi Kem, > >> On 3/11/06 2:55 AM, Ruslan Zasukhin at sunshine@public.kherson.ua >> wrote: >> >>> At least now I do not see such problems. >> >> If I use Studio to reindex that database, it goes away. But any >> new database >> I create shows the error as soon as I add one record to that table. >> >> The line that makes the difference is this: >> >> fldEnabled = new VBoolean( "Enabled", EVFlag.fIndexed ) >> >> There is no problem if I change it to: >> >> fldEnabled = new VBoolean( "Enabled", EVFlag.fNone ) >> >> Do I remember correctly that you said that a boolean field cannot >> be indexed >> anyway? > > Yes Boolean field do not have index never. Because of its nature. > > Hmm, so If you specify flag fIndexed problem comes... > Please send me again this db with one record which diagnose complains. I have Index on Boolean too. How can I fix this with Class-way? DBBuchungen name = "DBBuchungen" BDatum = new VDate("BDatum", kV_Indexed) Herk_Konto = new VObjectPtr ("Herk_Konto",inDataBase.TKonten,kV_Restrict+kV_Indexed) Ziel_Konto = new VObjectPtr ("Ziel_Konto",inDataBase.TKonten,kV_Restrict+kV_Indexed) Kategorie_ID = new VobjectPtr("Kategorie_ID", inDataBase.TKategorien,kV_Restrict+kV_Indexed) Betrag = new VLong("Betrag") Kommentar = new VVarChar("Kommentar",504,kV_Nullable) Blocked = new VBoolean("Blocked", kv_Indexed) Deleted=new VBoolean("Deleted", kv_Indexed) Can I make changes in Class and everything is fine or do I have to make ALTER DATABASE? thanks for help Claudius -- iMac CD 2GHz / MacOS X 10.4de / RB 2006r1/ Valentina 2.1 Homepage http://www.ClaSai.de iChat ryhoruk RealBasic ListBoxes: [ I feel the need...the need for speed!!! ] From Claudius at sailer-online.de Sat Mar 11 20:49:05 2006 From: Claudius at sailer-online.de (Claudius Sailer) Date: Sat Mar 11 13:49:08 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: References: Message-ID: <9EA12406-BA3C-4352-9408-D30F1DF97AAE@sailer-online.de> Am 11.03.2006 um 20:46 schrieb Claudius Sailer: > Hi Ruslan, > > Am 11.03.2006 um 19:55 schrieb Ruslan Zasukhin: > >> On 3/11/06 1:57 PM, "Kem Tekinay" >> wrote: >> >> Hi Kem, >> >>> On 3/11/06 2:55 AM, Ruslan Zasukhin at sunshine@public.kherson.ua >>> wrote: >>> >>>> At least now I do not see such problems. >>> >>> If I use Studio to reindex that database, it goes away. But any >>> new database >>> I create shows the error as soon as I add one record to that table. >>> >>> The line that makes the difference is this: >>> >>> fldEnabled = new VBoolean( "Enabled", EVFlag.fIndexed ) >>> >>> There is no problem if I change it to: >>> >>> fldEnabled = new VBoolean( "Enabled", EVFlag.fNone ) >>> >>> Do I remember correctly that you said that a boolean field cannot >>> be indexed >>> anyway? >> >> Yes Boolean field do not have index never. Because of its nature. >> >> Hmm, so If you specify flag fIndexed problem comes... >> Please send me again this db with one record which diagnose >> complains. > > I have Index on Boolean too. > How can I fix this with Class-way? > > DBBuchungen > > name = "DBBuchungen" > Blocked = new VBoolean("Blocked", kv_Indexed) Blocked=new [VBoolean(]"Blocked", EVFlag.fIndexed) Sorry rhis was old coding. This one is new one. I never had problem with this coding... bye Claudius -- iMac CD 2GHz / MacOS X 10.4de / RB 2006r1/ Valentina 2.1 Homepage http://www.ClaSai.de iChat ryhoruk RealBasic ListBoxes: [ I feel the need...the need for speed!!! ] From sunshine at public.kherson.ua Sat Mar 11 22:54:58 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 11 14:55:08 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: <9EA12406-BA3C-4352-9408-D30F1DF97AAE@sailer-online.de> Message-ID: On 3/11/06 9:49 PM, "Claudius Sailer" wrote: >> I have Index on Boolean too. >> How can I fix this with Class-way? Why and what you want fix it ? Just use it as is. Again, EACH and ANY Boolean field is shown in Vstudio for example, as ALWAYS INDEXED field, but index file physically not exists. Such magic :-) So it is not important if you set or not set this flag for boolean fields. >> DBBuchungen >> >> name = "DBBuchungen" >> Blocked = new VBoolean("Blocked", kv_Indexed) > > Blocked=new [VBoolean(]"Blocked", EVFlag.fIndexed) > > Sorry rhis was old coding. This one is new one. > I never had problem with this coding... -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sat Mar 11 16:11:35 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sat Mar 11 15:11:43 2006 Subject: [ANN] V4RB 2.3 b3 MAC In-Reply-To: Message-ID: On 3/11/06 1:55 PM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > Hmm, so If you specify flag fIndexed problem comes... > Please send me again this db with one record which diagnose complains. I'll send you the same one I sent before as that shows the problem and has a report. I'll use DropLoad again since direct e-mail had a problem. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ktekinay at mactechnologies.com Sun Mar 12 09:01:59 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 08:02:04 2006 Subject: Corrupted database Message-ID: My database is corrupted. Somewhere between adding record 80,000 and record 97,000, the data got damaged. There were no errors or unexpected quits or any other problems while adding the data. The only change I made to my code was to call gDB.Flush after every 100 K of data added rather than after every record, and I did that because it was getting very slow to add new records. Valentina Studio is of no help in recovering this database as it would rather quit than clone the data, and gives me an internal error if I try to compact it. Frankly, the data itself is unimportant as I can easily reimport it (although it takes hours to do so), but I cannot use a database that can become corrupted without warning. I have no idea how to begin troubleshooting this, and my inclination is to give up on Valentina, something I really don't want to do. Help me, please. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ktekinay at mactechnologies.com Sun Mar 12 09:38:46 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 08:38:52 2006 Subject: Clone fails Message-ID: When trying to clone my corrupted DB manually in RB, I got the following VException: Invalid parameter value was specified in function "Binary_Link::put_OnDelete". Exception Error Number: 627987 Before posting this to Mantis, I wanted to make sure my cloning code was right: dim originalDB as new VDatabase dim originalFile as FolderItem dim targetFolder as FolderItem originalFile = GetOpenFolderItem( "" ) if originalFile = nil then return originalDB.Open( originalFile ) if originalDB.ErrNumber <> 0 then MsgBox( "Couldn't open DB." ) return end if targetFolder = originalFile.Parent.Child( originalFile.Name + " clone" ) if not targetFolder.Exists then targetFolder.CreateAsFolder originalDB.Clone( targetFolder.Child( originalFile.Name ), true ) if originalDB.ErrNumber <> 0 then MsgBox( "Couldn't clone DB. Error " + str( originalDB.ErrNumber ) _ + chr( 13 ) + chr( 13 ) + originalDB.ErrString ) return end if originalDB.Close __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ivan_smahin at valentina-db.com Sun Mar 12 16:48:52 2006 From: ivan_smahin at valentina-db.com (Ivan Smahin) Date: Sun Mar 12 08:48:54 2006 Subject: Clone fails In-Reply-To: References: Message-ID: <1411611843.20060312164852@valentina-db.com> Hello Kem, Sunday, March 12, 2006, 4:38:46 PM, you wrote: > When trying to clone my corrupted DB manually in RB, I got the following > VException: > Invalid parameter value was specified in function > "Binary_Link::put_OnDelete". Obviously you are trying to Clone some database which has "cascade" onDelete behavior on M::M binaryLink. It is not allowed for now. Is this database generated with some early beta? -- Best regards, Ivan mailto:ivan_smahin@valentina-db.com From ktekinay at mactechnologies.com Sun Mar 12 10:59:08 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 09:59:13 2006 Subject: Clone fails In-Reply-To: <1411611843.20060312164852@valentina-db.com> Message-ID: On 3/12/06 9:48 AM, Ivan Smahin at ivan_smahin@valentina-db.com wrote: > Obviously you are trying to Clone some database which has > "cascade" onDelete behavior on M::M binaryLink. If that's true, it wasn't done intentionally, nor does V Studio show it. Here are all my links: linkAccountEmails = new VBinaryLink( "linkAccountEmails", tblAccount, _ tblEmail, EVLinkType.kOne, EVLinkType.kMany, EVOnDelete.kCascade ) linkEmailAttachments = new VBinaryLink( "linkEmailAttachments", _ tblEmail, tblAttachment, EVLinkType.kMany, EVLinkType.kMany, _ EVOnDelete.kDefault ) linkEmailHeaders = new VBinaryLink( "linkEmailHeader", tblEmail, _ tblHeader, EVLinkType.kOne, EVLinkType.kMany, EVOnDelete.kCascade ) linkEmailMessage = new VBinaryLink( "linkEmailMessage", tblEmail, _ tblMessageText, EVLinkType.kOne, EVLinkType.kMany, _ EVOnDelete.kCascade ) linkToAddresses = new VBinaryLink( "linkToAddresses", tblEmail, _ tblEmailAddress, EVLinkType.kMany, EVLinkType.kMany, _ EVOnDelete.kDefault ) linkCCAddresses = new VBinaryLink( "linkCCAddresses", tblEmail, _ tblEmailAddress, EVLinkType.kMany, EVLinkType.kMany, _ EVOnDelete.kDefault ) linkBCCAddresses = new VBinaryLink( "linkBCCAddresses", tblEmail, _ tblEmailAddress, EVLinkType.kMany, EVLinkType.kMany, _ EVOnDelete.kDefault ) > Is this database generated with some early beta? No, it was generated on March 5, probably with 2.3b2. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ivan_smahin at valentina-db.com Sun Mar 12 18:18:07 2006 From: ivan_smahin at valentina-db.com (Ivan Smahin) Date: Sun Mar 12 10:18:09 2006 Subject: Clone fails In-Reply-To: References: <1411611843.20060312164852@valentina-db.com> Message-ID: <1457636752.20060312181807@valentina-db.com> Hello Kem, Sunday, March 12, 2006, 5:59:08 PM, you wrote: > On 3/12/06 9:48 AM, Ivan Smahin at ivan_smahin@valentina-db.com wrote: Could you send me this database for testing? -- Best regards, Ivan mailto:ivan_smahin@valentina-db.com From akritselis at earthlink.net Sun Mar 12 11:51:08 2006 From: akritselis at earthlink.net (Anna Kritselis) Date: Sun Mar 12 10:51:14 2006 Subject: Clone fails Message-ID: <7071055.1142182269207.JavaMail.root@elwamui-francias.atl.sa.earthlink.net> -----Original Message----- Hi, I am also having database corruption problems, for what seems like no reason. Last week, three out of three users had databases that got corrupt - I use RB2006r1, v4rb 2.2 on Win Xp with Monkeybread plugins. I have zipped the databases, but they are each a little under 2 MB - where do you want me to send them? Anna From: Ivan Smahin Sent: Mar 12, 2006 11:18 AM To: Valentina Beta Subject: Re[2]: Clone fails Hello Kem, Sunday, March 12, 2006, 5:59:08 PM, you wrote: > On 3/12/06 9:48 AM, Ivan Smahin at ivan_smahin@valentina-db.com wrote: Could you send me this database for testing? -- Best regards, Ivan mailto:ivan_smahin@valentina-db.com _______________________________________________ Valentina-beta mailing list Valentina-beta@lists.macserve.net http://lists.macserve.net/mailman/listinfo/valentina-beta From ktekinay at mactechnologies.com Sun Mar 12 11:53:24 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 10:53:28 2006 Subject: Clone fails In-Reply-To: <1457636752.20060312181807@valentina-db.com> Message-ID: On 3/12/06 11:18 AM, Ivan Smahin at ivan_smahin@valentina-db.com wrote: > Could you send me this database for testing? Do you mean the one that's corrupted, or an uncorrupted new copy? I ask because the one that is corrupted is almost 6 GB compressed. If you'd like me to upload it somewhere, I can. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ivan_smahin at valentina-db.com Sun Mar 12 19:05:03 2006 From: ivan_smahin at valentina-db.com (Ivan Smahin) Date: Sun Mar 12 11:05:06 2006 Subject: Clone fails In-Reply-To: References: <1457636752.20060312181807@valentina-db.com> Message-ID: <435104765.20060312190503@valentina-db.com> Hello Kem, Sunday, March 12, 2006, 6:53:24 PM, you wrote: > On 3/12/06 11:18 AM, Ivan Smahin at ivan_smahin@valentina-db.com wrote: >> Could you send me this database for testing? > Do you mean the one that's corrupted, or an uncorrupted new copy? I'm loosing the thread... So - you have some database (let it be database A) which is fine as you think. BTW - did you try to diagnose this database? Now you want to clone database A for getting database B and get some exception during Clone. Did you finally get database B? What about diagnose? Also Clone produce some error log - is any message there? > I ask because the one that is corrupted is almost 6 GB compressed. If you'd > like me to upload it somewhere, I can. No, it is too huge for download. -- Best regards, Ivan mailto:ivan_smahin@valentina-db.com From ktekinay at mactechnologies.com Sun Mar 12 12:46:29 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 11:46:33 2006 Subject: Clone fails In-Reply-To: <435104765.20060312190503@valentina-db.com> Message-ID: On 3/12/06 12:05 PM, Ivan Smahin at ivan_smahin@valentina-db.com wrote: > I'm loosing the thread... > So - you have some database (let it be database A) which is fine as > you think. BTW - did you try to diagnose this database? No, database A was fine yesterday with about 80,000 records. Last night I added another 16,000 records, then checked it in Valentina Studio using DIAGNOSE. I found it is now severely corrupted even though there were no errors while adding the new records. I then tried to clone the corrupted database in an effort to recover the data and got the errors I mentioned. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ktekinay at mactechnologies.com Sun Mar 12 18:20:23 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 17:20:29 2006 Subject: Corruption - more info Message-ID: I can now get my database to become corrupted and I can reproduce it. The secret is the cache. If I reduce the cache to 1 MB, the database will become corrupted after just 66 records added. If I leave it at 100 MB (my default), it will get corrupted sometime after adding record 1900 and before 2000. Either situation is reproducible. I discovered this by reducing the cache, removing Flush from my code, and putting in DIAGNOSE after every record added. Now the question is, is this enough information to get this fixed or do I have to provide the code? Also, should I file this in Mantis? __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From macsforever2000 at goodeast.com Sun Mar 12 18:42:01 2006 From: macsforever2000 at goodeast.com (Frank Schima) Date: Sun Mar 12 19:42:17 2006 Subject: Corruption - more info In-Reply-To: References: Message-ID: Hi Kem, On Mar 12, 2006, at 4:20 PM, Kem Tekinay wrote: > I can now get my database to become corrupted and I can reproduce > it. The > secret is the cache. If I reduce the cache to 1 MB, the database > will become > corrupted after just 66 records added. If I leave it at 100 MB (my > default), > it will get corrupted sometime after adding record 1900 and before > 2000. > Either situation is reproducible. > > I discovered this by reducing the cache, removing Flush from my > code, and > putting in DIAGNOSE after every record added. > > Now the question is, is this enough information to get this fixed > or do I > have to provide the code? Also, should I file this in Mantis? It is best to provide the minimum amount of code required to reproduce the problem and file on Mantis. In my experience, it pretty much guarantees a quick fix. Best regards, Frank Schima From ktekinay at mactechnologies.com Sun Mar 12 20:50:33 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 12 19:50:40 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/12/06 8:42 PM, Frank Schima at macsforever2000@goodeast.com wrote: > It is best to provide the minimum amount of code required to > reproduce the problem and file on Mantis. In my experience, it pretty > much guarantees a quick fix. Of course, but that will take time, and I have already put a great deal of it into this project. I'd rather avoid more if I can. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Mon Mar 13 13:31:38 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 05:31:46 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 1:20 AM, "Kem Tekinay" wrote: Hi Kem, But question is: 1) ho many tables/fields/indexes do you have ? if you have a lots - then just set cache to 10MB > I can now get my database to become corrupted and I can reproduce it. The > secret is the cache. If I reduce the cache to 1 MB, the database will become > corrupted after just 66 records added. If I leave it at 100 MB (my default), > it will get corrupted sometime after adding record 1900 and before 2000. > Either situation is reproducible. 100MB cache this is too many I think. 10-20MB will work good for 99% of cases. > I discovered this by reducing the cache, removing Flush from my code, and > putting in DIAGNOSE after every record added. > > Now the question is, is this enough information to get this fixed or do I > have to provide the code? Also, should I file this in Mantis? No this is not enough of course. Because as you know hundreds developers use it without such problem. I think secret is somehow in your tables and fields. So we need to get at least your compiled application, And steps to add that 66 record or 1900 records. I prefer test on 3MB cache and amount of record enough to get problem. Send all this to us please -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Mon Mar 13 13:33:51 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 05:33:57 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 1:20 AM, "Kem Tekinay" wrote: > I can now get my database to become corrupted and I can reproduce it. The > secret is the cache. If I reduce the cache to 1 MB, the database will become > corrupted after just 66 records added. If I leave it at 100 MB (my default), > it will get corrupted sometime after adding record 1900 and before 2000. > Either situation is reproducible. > > I discovered this by reducing the cache, removing Flush from my code, and > putting in DIAGNOSE after every record added. And what is corrupted btw? What diagnose says ? Again index of boolean field ? -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Mon Mar 13 13:52:06 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 05:52:11 2006 Subject: Clone fails In-Reply-To: Message-ID: On 3/12/06 7:46 PM, "Kem Tekinay" wrote: >> I'm loosing the thread... >> So - you have some database (let it be database A) which is fine as >> you think. BTW - did you try to diagnose this database? > > No, database A was fine yesterday with about 80,000 records. Last night I > added another 16,000 records, then checked it in Valentina Studio using > DIAGNOSE. I found it is now severely corrupted even though there were no > errors while adding the new records. Kem, Send to us at least diagnose file. I wonder, 80K records and so huge database? What you keep there? Bit text? Pictures? Contact us off-list. We will give you our ICQ So we will communicate faster. ========================= We have found problem with Boolean fields. In fact in 2.0 they have work as BYTE field physically: - 1 byte per record - index file So I re-write now Boolean field to work correctly as it should: - 1 bit per record - no index file at all. ========================= Ken, About Clone of your db. Please make for us your db with only 2-4 records And try clone it. Works or now. I assume it will not, because of some Link problem as you say. Then send us this SMALL db. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Mon Mar 13 14:24:23 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 06:24:32 2006 Subject: ktekinay@mactechnologies.com has sent you a file. In-Reply-To: <20060311215250.7D475FA418A@s3.dropload.com> Message-ID: On 3/11/06 11:52 PM, "dropload" wrote: > > A file named "Afterthought DB.zip" has been placed in a temporary location for > you by Kem Tekinay (ktekinay@mactechnologies.com). If you trust this person > and wish to accept this file, click or paste this link into your browser: > > http://www.dropload.com/redeem.php?t=1aec8d8b9ca4297b36d6b2a6938da444 > > Kem has attached a note: > This is exactly the same database I sent earlier. Kem, I take this database, 4 files, 1 MB total, Run viSQL with 2.3 b3 Open database Do "diagnose database veryhigh" Diagnose show All right! ( Database: ) -------------------------------------------- So I do not understand how you get diagnose file with info about problems?? I do this on MAC, do you also run MAC? What version of V4RB you use ? ------------------------- Also, when you mention about corruption for that huge db in many GBs , You also see this from similar diagnose ? I tend to think that there is no any corruption on your side. On some reason you get wrong diagnose. And I still think that you have discover some problem with Clone + BinaryLinks. So please send us small your db to reproduce this. Again, please contact us via ICQ ASAP. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Mon Mar 13 08:40:45 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Mon Mar 13 07:40:53 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 6:31 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > 1) ho many tables/fields/indexes do you have ? > if you have a lots - then just set cache to 10MB It's the same structure that I sent to you. 6 tables, 7 BinaryLinks. > 100MB cache this is too many I think. > 10-20MB will work good for 99% of cases. That's fine, but if I make the cache larger, it just takes longer to corrupt. > I think secret is somehow in your tables and fields. > > So we need to get at least your compiled application, I'd be happy to send you the whole app along with a username and password to access the account that holds the records. Or do you need me to trim it down? Also, my app uses Einhugur's UtilLib. Is that a problem? > I prefer test on 3MB cache and amount of record enough to get problem. 3 MB was the same as 1 MB. The database corrupted after 66 records: > ==========> Field: AddressTo , type VarChar > ==========> IndexFile > PAGE (0, 0) Inventory pages OK. > Index Inventory pages OK. > > PAGE (3, 2) OK. > > All right! ( IndexFile ) > > .....> VarCharFile > (!) FirstRecID is not correct Must be 66 but it is 33 > (!) RefFile is not correct. Must be rebuilt. > (!) Total number of records on VarChar pages does not match table's record > count. > > Problems found! ( VarCharFile ) > > > Problems found! ( Field: AddressTo , type VarChar ) __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Mon Mar 13 15:47:37 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 07:47:41 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 3:40 PM, "Kem Tekinay" wrote: >> 1) ho many tables/fields/indexes do you have ? >> if you have a lots - then just set cache to 10MB > > It's the same structure that I sent to you. 6 tables, 7 BinaryLinks. We have reproduce on your db problem with Clone + binary link. Ivan now try to fix it. Now we need fix this mystery with Diagnose + index report on boolean field. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Mon Mar 13 08:48:42 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Mon Mar 13 07:48:48 2006 Subject: Clone fails In-Reply-To: Message-ID: On 3/13/06 6:52 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > I wonder, 80K records and so huge database? > What you keep there? Bit text? Pictures? It is a database for storing e-mail, so VTEXT for the message body and source, BLOB for attachments. > Contact us off-list. We will give you our ICQ > So we will communicate faster. OK. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Mon Mar 13 15:49:36 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 07:49:41 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 3:40 PM, "Kem Tekinay" wrote: >> I prefer test on 3MB cache and amount of record enough to get problem. > > 3 MB was the same as 1 MB. The database corrupted after 66 records: Then THIS sounds like reasonable to give us something to reproduce !!! PLEASE collect all we need to reproduce this. You work on MacOS? With RB? Then make for us COMPILE app which create new fresh db then import that 66 records and Bum. Okay ? >> ==========> Field: AddressTo , type VarChar >> ==========> IndexFile >> PAGE (0, 0) Inventory pages OK. >> Index Inventory pages OK. >> >> PAGE (3, 2) OK. >> >> All right! ( IndexFile ) >> >> .....> VarCharFile >> (!) FirstRecID is not correct Must be 66 but it is 33 >> (!) RefFile is not correct. Must be rebuilt. >> (!) Total number of records on VarChar pages does not match table's record >> count. >> >> Problems found! ( VarCharFile ) >> >> >> Problems found! ( Field: AddressTo , type VarChar ) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Mon Mar 13 15:50:41 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 07:50:45 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 3:40 PM, "Kem Tekinay" wrote: >> I prefer test on 3MB cache and amount of record enough to get problem. > > 3 MB was the same as 1 MB. The database corrupted after 66 records: > >> ==========> Field: AddressTo , type VarChar >> ==========> IndexFile >> PAGE (0, 0) Inventory pages OK. >> Index Inventory pages OK. >> One more: may be you use VarChar field with length > 1022 ??????? Is this true? If yes then it explains. I will recommend replace that field(s) on TEXT fields instead. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Mon Mar 13 08:53:44 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Mon Mar 13 07:53:50 2006 Subject: Clone fails In-Reply-To: Message-ID: On 3/13/06 6:52 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > About Clone of your db. > > Please make for us your db with only 2-4 records > And try clone it. Works or now. > > I assume it will not, because of some Link problem as you say. > Then send us this SMALL db. No, it did not work. I am sending it to you offlist. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From ktekinay at mactechnologies.com Mon Mar 13 08:59:40 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Mon Mar 13 07:59:48 2006 Subject: Corruption - more info In-Reply-To: Message-ID: On 3/13/06 8:50 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > may be you use VarChar field with length > 1022 ??????? No, the largest I have is 1022. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Tue Mar 14 02:41:26 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 13 18:41:33 2006 Subject: [ATTN] We are going limit VarChar to 4Kb of data ! Message-ID: Hi All, I have spent few hours today in code of VarChar field... My Resume is: it was evil idea try to make VarChar that support data > 4Kb. * I am going tomorrow re-write VarChar to simplify code and make it faster. After this we will have * VarChar[2044] - maximal length for UTF16 * VarChar[4088] - maximal length for UTF8 or other single byte This remake should not affect you if you not use VarChars > 4K. And as far as I know there is no such developers. If you will ask me, if this limit Valentina, answer is - not so much. IF you look on MS SQL, they have limit on Table Record size = 8 Kb. We give you limit on single VarChar = 4Kb. But you can have many VarChar as you want. And you can use TEXT field for long strings. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From justin at crunch-recording.com Tue Mar 14 07:19:56 2006 From: justin at crunch-recording.com (Justin Drury) Date: Tue Mar 14 06:20:02 2006 Subject: Does flush really flush? In-Reply-To: <20060313113151.F23264031C5@edison.macserve.net> References: <20060313113151.F23264031C5@edison.macserve.net> Message-ID: <84BF5CA6-6CF6-4403-8BFB-EE3B3FB5B526@crunch-recording.com> Here's my situation, I'm seeing a few corrupted databases. Its pretty much happening whenever new records are added. After adding records I call the database method flush, if the user force quits or my app crashes, then the database get's corrupted, its hard to tell its corrupted apart from the fact that that certain sql queries (like a distinct on a field) don't behave properly. However if the app has closed and the user again quits or force quits in the future the database is fine... I'm storing the database in 4 files deleting the index file has no effect. Previously I tried using vstudio to recover the database(it crashed), (dumps would crash[both sql and xml]my only solution was to duplicate the database layout to another database and walk the records copying from one record to another. Ruslan:I have this database still if you are interested... So my question is... does database flush really clean up everything, I'm adding records API way, is there anything else I can do(apart from closing and reopening?) tnx justin From sunshine at public.kherson.ua Tue Mar 14 15:22:30 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 14 07:22:38 2006 Subject: Does flush really flush? In-Reply-To: <84BF5CA6-6CF6-4403-8BFB-EE3B3FB5B526@crunch-recording.com> Message-ID: On 3/14/06 2:19 PM, "Justin Drury" wrote: Hi Justin, > Here's my situation, I'm seeing a few corrupted databases. Its > pretty much happening whenever new records are added. After adding > records I call the database method flush, if the user force quits or > my app crashes, then the database get's corrupted, its hard to tell > its corrupted apart from the fact that that certain sql queries (like > a distinct on a field) don't behave properly. However if the app has > closed and the user again quits or force quits in the future the > database is fine... ----- > I'm storing the database in 4 files deleting the index file has no > effect. Previously I tried using vstudio to recover the database(it > crashed), (dumps would crash[both sql and xml]my only solution was to > duplicate the database layout to another database and walk the > records copying from one record to another. Ruslan:I have this > database still if you are interested... > So my question is... does database flush really clean up everything, > I'm adding records API way, is there anything else I can do(apart > from closing and reopening?) Yes db.Flush() must flush everything If you have way to reproduce write, db.flush, force quite => corruption Then please give us such way. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Wed Mar 15 00:24:21 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 14 16:24:37 2006 Subject: [ANN] Vstudio 2.3 fc9 for Windows uploaded. Message-ID: Hi All, In this build you can find: * significant improvements of SQL Editor: - Auto-completion when you type SQL query. Quite good as I see. Works even for 1.x databases. - Ctrl+Space show auto-completion if it is not shown. - you can now open several TABs with results. - Ctrl + E do EXECUTE of query - Ctrl + Shift + E EXECUTE query and put result in new TAB - in result tab you can see its SQL query, select and copy that query. - ToolTips now work in SqlEditor as well as in Data Browser - now works usual Edit commands in SQL Editors as Ctrl + C, + V + X +A * In Schema browser - Ctrl + W close current database - Improved work with encrypted databases. * v1 plugin updated to work with fc9 -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Wed Mar 15 02:24:04 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 14 18:24:10 2006 Subject: [ANN] VCOM 2.3 b4 uploaded. Message-ID: Hi All, Hi Jeffrey, You can download 2.3 b4 build. * Ivan have fix that problem with SUM(f) / 100 * this build is first after major change in VarChar code. our tests are DONE, so I hope all should be all right here. Let me know -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Thu Mar 16 23:13:39 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 16 15:13:46 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 Message-ID: Hi All, You can download 2.3 b5 build for many products. (if somebody need other products let me know) ---------------------------------------------------------------------- * This build contains big remake of VarChar field - now we limit it to VarChar[1022] for UTF16 in 2.4 we will move this up to VarChar[2044] in general, limit is 4Kb per VarChar. - VarChar now is much faster. I have not test it separately. but in V4RB/Examples/V1/SingleTable project present 9 numeric fields and one VarChar. 100K adds of records was 3.5 sec now 3.2 sec. ---------------- 10% speedup. but VarChar itself has bigger growing up. - fixed bug in VarChar which could come if you add a lots of records in small cache. ---------------------------------------------------------------------- * V4RB and VNET products now contains quite complete example TestProject Its target -- to test code of plugin itself. for kernel we have other C++ tests. I think you can find useful learn this project, and help us periodically test it and improve endlessly :-) Right now we develop such project for V4MD, then will be made for VCOM and V4REV. ---------------------------------------------------------------------- * We have add new feature to kernel: you can now turn ON/OFF warning.log read our Wiki for new SQL commands for this. there is not many yet info here, but we will constantly improve it. for example, we going in nearest time add here explanation of query execution with time of EACH operation. We believe that such feature will help to you and to us easy discover bottlenecks of slow queries, and also see what in query on in db structure should be optimized to get faster results. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From jda at his.com Thu Mar 16 16:27:29 2006 From: jda at his.com (jda) Date: Thu Mar 16 15:27:35 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 In-Reply-To: References: Message-ID: >* This build contains big remake of VarChar field > > - now we limit it to VarChar[1022] for UTF16 > in 2.4 we will move this up to VarChar[2044] > Hi Ruslan, Just to be clear, I already have several VarChar fields of [2000]. I won't be able to use any more beta2.3 versions, right? I'll have to stick with beta 4 unto 2.4 betas are ready. Jon From sunshine at public.kherson.ua Thu Mar 16 23:55:12 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 16 15:55:18 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 In-Reply-To: Message-ID: On 3/16/06 11:27 PM, "jda" wrote: >> * This build contains big remake of VarChar field >> >> - now we limit it to VarChar[1022] for UTF16 >> in 2.4 we will move this up to VarChar[2044] >> > > Hi Ruslan, > > Just to be clear, I already have several VarChar fields of [2000]. I > won't be able to use any more beta2.3 versions, right? I'll have to > stick with beta 4 unto 2.4 betas are ready. Hi Jon, Well, Then I think we need enable VarChar[2044] right now ... Actually not big problem I think. Only algorithms should be changed... ANYBODY ELSE use VarChar bigger of 2044 ? If yes, then you should ship new version of your app using 2.2, and convert that fields into VarChar[2044] or into TEXT. only after this you should use v2.3+ of Valentina -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From maxprog at mac.com Thu Mar 16 23:02:33 2006 From: maxprog at mac.com (Stan Busk) Date: Thu Mar 16 16:02:44 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 In-Reply-To: References: Message-ID: <49F01878-E3B5-4B89-8471-3652D5D7A4C9@mac.com> It seems old index is not compatible with new one. I mean I need to to rebuild indexes in order to be able to get data. Is this expected effect? > Hi All, > > You can download 2.3 b5 build for many products. > (if somebody need other products let me know) > > ---------------------------------------------------------------------- > * This build contains big remake of VarChar field > > - now we limit it to VarChar[1022] for UTF16 > in 2.4 we will move this up to VarChar[2044] > > in general, limit is 4Kb per VarChar. > > - VarChar now is much faster. I have not test it separately. > but in V4RB/Examples/V1/SingleTable project present > 9 numeric fields and one VarChar. 100K adds of records > > was 3.5 sec > now 3.2 sec. > ---------------- > 10% speedup. > > but VarChar itself has bigger growing up. > > - fixed bug in VarChar which could come if you add a lots of > records > in small cache. > > > ---------------------------------------------------------------------- > * V4RB and VNET products now contains quite complete > example TestProject > > Its target -- to test code of plugin itself. > for kernel we have other C++ tests. > > I think you can find useful learn this project, and help us > periodically test it and improve endlessly :-) > > Right now we develop such project for V4MD, then will be made > for VCOM and V4REV. > > ---------------------------------------------------------------------- > * We have add new feature to kernel: > you can now turn ON/OFF warning.log > read our Wiki for new SQL commands for this. > > there is not many yet info here, but we will constantly improve > it. > for example, we going in nearest time add here explanation of > query execution with time of EACH operation. We believe that such > feature will help to you and to us easy discover bottlenecks > of slow queries, and also see what in query on in db structure > should be optimized to get faster results. > > -- > Best regards, > > Ruslan Zasukhin > VP Engineering and New Technology > Paradigma Software, Inc > > Valentina - Joining Worlds of Information > http://www.paradigmasoft.com > > [I feel the need: the need for speed] > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From sunshine at public.kherson.ua Fri Mar 17 00:09:48 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 16 16:09:54 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 In-Reply-To: <49F01878-E3B5-4B89-8471-3652D5D7A4C9@mac.com> Message-ID: On 3/17/06 12:02 AM, "Stan Busk" wrote: Hi Stan, > It seems old index is not compatible with new one. I mean I need to > to rebuild indexes in order to be able to get data. Is this expected > effect? I can think only about boolean field. Do you have them? As I have told I have disable creation of index for bool fields. What happens before reindex? Wrong searches? Or some error codes? Else? -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From maxprog at mac.com Thu Mar 16 23:12:56 2006 From: maxprog at mac.com (Stan Busk) Date: Thu Mar 16 16:13:08 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 In-Reply-To: References: Message-ID: <1DA13329-BB81-4229-AD51-722955C3771F@mac.com> Only wrong searches, no errors. Yes, I used indexed booleans everywhere. Estic fotut. Once reindexed all works fine again. > On 3/17/06 12:02 AM, "Stan Busk" wrote: > > Hi Stan, > >> It seems old index is not compatible with new one. I mean I need to >> to rebuild indexes in order to be able to get data. Is this expected >> effect? > > I can think only about boolean field. Do you have them? > As I have told I have disable creation of index for bool fields. > > What happens before reindex? Wrong searches? Or some error codes? > Else? > > > -- > Best regards, > > Ruslan Zasukhin > VP Engineering and New Technology > Paradigma Software, Inc > > Valentina - Joining Worlds of Information > http://www.paradigmasoft.com > > [I feel the need: the need for speed] > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From sunshine at public.kherson.ua Fri Mar 17 00:31:58 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 16 16:32:03 2006 Subject: [ANN] V4RB, V4MD mac+win and V4REV, Vservers win 2.3 b5 In-Reply-To: <1DA13329-BB81-4229-AD51-722955C3771F@mac.com> Message-ID: On 3/17/06 12:12 AM, "Stan Busk" wrote: > Only wrong searches, no errors. Yes, I used indexed booleans > everywhere. Estic fotut. > Once reindexed all works fine again. Ok. Actually in near future I will need yet one more step with boolean fields. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Fri Mar 17 17:38:22 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 17 09:38:28 2006 Subject: [ANN] V4MD and V4REV Reference.pdf updated Message-ID: Hi All, Corrected few missing things, Thanks to Jo?l Guillod, found many enough corrections in V4REV docs -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Fri Mar 17 20:58:53 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 17 12:59:00 2006 Subject: [ANN] Vstudio win 2.3 fc10 uploaded Message-ID: Hi All, You can download 2.3 fc10 build. * Attention! now 1.x plugin is right here in folder "plugins_disabled" If you need it, copy it into folder "plugins" * Download archive now not 10.3MB but 6.6 MB thanks to Innosetup installer. * fixed few GUI glitches in fc10 -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Sat Mar 18 10:43:12 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Mar 18 02:43:29 2006 Subject: REDIRECT to valentina-db.com // [ANN] Vstudio win 2.3 fc10 In-Reply-To: <7.0.1.0.2.20060318141531.0395ca40@paradise.net.nz> Message-ID: On 3/18/06 3:17 AM, "Sean Wilson" wrote: > >>> Your site seems to be down ATM >> >> Paradigmasoft is ok > > Not for me it isn't! Really. Hmm, we have some black line with sites.. >> Valentina-db yes glitches > > Seems fine but is showing 2.2fc6 for download I have set temporary redirect from paradigmasoft.com to Valentina-db.com and update all archives on Valentina-db.com. It is good that we have physical mirror of site now :-) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sun Mar 19 01:31:40 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 19 00:31:47 2006 Subject: Good news and bad news Message-ID: The good news is that 2.3b5 seems to have solved one corruption problem. The bad news is that there is something else. When my database got to about 98,000 records, I quit my program and ran DIAGNOSE from V-Studio, and it failed. The relevant part of the report is below. I tried to clone the database and V-Studio quit without warning. Reindexing the entire database didn't help either. Any idea where this corruption would be coming from? *************************************************************************** Database: *************************************************************************** .....> DBStorage: .....> VolumeOnFile: /Users/ktekinay/Documents/Afterthought DB/Afterthought.dat All right! ( VolumeOnFile: /Users/ktekinay/Documents/Afterthought DB/Afterthought.dat ) .....> VolumeOnFile: /Users/ktekinay/Documents/Afterthought DB/Afterthought.blb File {12885360649}: the tree corrupted: balance violated. File {17180327946}: the tree corrupted: balance violated. Problems found! ( VolumeOnFile: /Users/ktekinay/Documents/Afterthought DB/Afterthought.blb ) .....> VolumeOnFile: /Users/ktekinay/Documents/Afterthought DB/Afterthought.ind All right! ( VolumeOnFile: /Users/ktekinay/Documents/Afterthought DB/Afterthought.ind ) .....> VolumeOnFile: /private/var/tmp/folders.501/TemporaryItems/91036962 All right! ( VolumeOnFile: /private/var/tmp/folders.501/TemporaryItems/91036962 ) Problems found! ( DBStorage: ) __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From sunshine at public.kherson.ua Sun Mar 19 11:23:28 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Mar 19 03:23:35 2006 Subject: Good news and bad news In-Reply-To: Message-ID: On 3/19/06 8:31 AM, "Kem Tekinay" wrote: Hi Kem, > The good news is that 2.3b5 seems to have solved one corruption problem. The > bad news is that there is something else. When my database got to about > 98,000 records, I quit my program and ran DIAGNOSE from V-Studio, and it > failed. The relevant part of the report is below. You have quite it in normal way? Or by force quite? > I tried to clone the database and V-Studio quit without warning. Reindexing > the entire database didn't help either. > > Any idea where this corruption would be coming from? I can think about internal file tree (ala-FAT) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From ktekinay at mactechnologies.com Sun Mar 19 09:46:29 2006 From: ktekinay at mactechnologies.com (Kem Tekinay) Date: Sun Mar 19 08:46:33 2006 Subject: Good news and bad news In-Reply-To: Message-ID: On 3/19/06 4:23 AM, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: >> The good news is that 2.3b5 seems to have solved one corruption problem. The >> bad news is that there is something else. When my database got to about >> 98,000 records, I quit my program and ran DIAGNOSE from V-Studio, and it >> failed. The relevant part of the report is below. > > You have quite it in normal way? Or by force quite? Everything was normal. I chose to stop the program at an arbitrary point and quit it using File -> Quit. There was NO indication of trouble at all until I ran DIAGNOSE. __________________________________________________________________________ Kem Tekinay (212) 201-1465 MacTechnologies Consulting Fax (914) 242-7294 545 Eighth Avenue, Suite 401 Pager (917) 491-5546 New York, New York 10018 http://www.mactechnologies.com To join the MacTechnologies Consulting mailing list, send an e-mail to: mactechnologies_consulting-subscribe@yahoogroups.com From jda at his.com Tue Mar 21 11:21:45 2006 From: jda at his.com (jda) Date: Tue Mar 21 10:24:19 2006 Subject: Diagnose problem In-Reply-To: References: Message-ID: Hi Ruslan, I have a database that seems fine, and when I do a diagnose (via viSQL) the report comes back OK. But if I do this: result = myDatabase.Diagnose I get an "Internal error" exception thrown (the same diagnose worked many times before with this database, this is new as of today). May I send you the db and my updated project? You can perform the diagnose in memory with my project. Jon Sonny Software From sunshine at public.kherson.ua Tue Mar 21 18:41:41 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 21 10:41:51 2006 Subject: Diagnose problem In-Reply-To: Message-ID: On 3/21/06 6:21 PM, "jda" wrote: Hi Jochen, > I have a database that seems fine, and when I do a diagnose (via > viSQL) the report comes back OK. > > But if I do this: > > result = myDatabase.Diagnose > > I get an "Internal error" exception thrown (the same diagnose worked > many times before with this database, this is new as of today). > May I send you the db and my updated project? You can perform the > diagnose in memory with my project. May be you use different levels of diagnose? -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From jda at his.com Tue Mar 21 11:45:48 2006 From: jda at his.com (jda) Date: Tue Mar 21 10:56:12 2006 Subject: Diagnose problem In-Reply-To: References: Message-ID: >On 3/21/06 6:21 PM, "jda" wrote: > >Hi Jochen, > >> I have a database that seems fine, and when I do a diagnose (via >> viSQL) the report comes back OK. >> >> But if I do this: >> >> result = myDatabase.Diagnose >> >> I get an "Internal error" exception thrown (the same diagnose worked >> many times before with this database, this is new as of today). > >> May I send you the db and my updated project? You can perform the >> diagnose in memory with my project. > >May be you use different levels of diagnose? > No, they are the same: SQL: Diagnose database in memory: result = myDatabase.diagnose And Ruslan, see the error, it is a VException! It should return true or false, not generate an exception. Jon From sunshine at public.kherson.ua Tue Mar 21 20:44:19 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 21 12:44:26 2006 Subject: AW: Follow Up with Jean-Louis In-Reply-To: <7F5BBC3B7975A94F9844614AB95C3755065C4C@srvex01.se-dd.local> Message-ID: On 3/21/06 7:40 PM, "Decoster Jean Louis" wrote: Hi Jean, > Hi Russlan, > > Thank you for last mail. I wanted some days to try to work with VSDK_2; but > till now I could not make many things. > When you write VSDK 2.x do not give you CLASS WAY. It's for me not very easy. > I have been writing applications using Valentina, Corba and C++ and Java > applications. Using VSDK_1 and factory design pattern, I can declare classes > which extend VDKBaseObject and Corba like: > vdb_xyztable::vdb_xyztable():VDK_BaseObject("XyzTable"), table(), > xyzCorba() { > // Create Fields > xyzName = new VDK_String ( "Xyz_Name", 32 ); > Value = new VDK_ULong ( "Value" ); > } > You wrote >> So If you are used to this style of coding, I'd like to discuss what we >> can do here. >> >> I see one way easy enough to get this style back into game using simple >> wrapper class of smart-ptr kind which use trick of operator ->. >> Only potential problem for this implementation that we will use -> >> for non-pointer objects >> class tnlPerson >> { >> VString mName; >> } >> .... >> mName->SetString() // C++ developer expect normally here >> // mName.SetString() > Is it possible for you to make something to provide with VSDK_2 the same > possibitlities given by VSDK_1? > I am not sure that I can get them at this moment. > Perhaps it's also possible for you to sent a little more completed > examples using load, save, add, delete records from tables using wrapper > classes with VSDK_2. IT will be not exactly as in v1, but very similar. For other v2 products as V4RB, VNET we still support Class way. Okay, we will try make such example in nearest 1-2 days. > Please, can you add my name to your MemberList.? > Thanks a lot for your next answer. Memberlist? Which one? On Valentina list you can subscribe from our SUPPORT page on site -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Tue Mar 21 21:50:19 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 21 13:50:28 2006 Subject: Diagnose problem In-Reply-To: Message-ID: On 3/21/06 6:45 PM, "jda" wrote: >>> result = myDatabase.Diagnose >>> >>> I get an "Internal error" exception thrown (the same diagnose worked >>> many times before with this database, this is new as of today). >> >>> May I send you the db and my updated project? You can perform the >>> diagnose in memory with my project. >> >> May be you use different levels of diagnose? >> > > No, they are the same: > > SQL: Diagnose database > > in memory: result = myDatabase.diagnose I think enough only db > And Ruslan, see the error, it is a VException! It should return true > or false, not generate an exception. Agree with this, please add into Mantis -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From jda at his.com Tue Mar 21 14:55:37 2006 From: jda at his.com (jda) Date: Tue Mar 21 14:05:21 2006 Subject: Diagnose problem In-Reply-To: References: Message-ID: > > >Agree with this, please add into Mantis > Already done, with db uploaded (#1457). Jon From sunshine at public.kherson.ua Wed Mar 22 18:48:09 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Mar 22 10:48:21 2006 Subject: [ANN] Vstudio 2.3 fc11 + VCOM 2.3 b7 uploaded. Message-ID: Hi All, * 2.3 b7 contains fix of nasty bug 1456 * Vstudio fc11 include 2.3b7 engine and have few cosmetic fixes. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Thu Mar 23 10:55:38 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 23 02:55:59 2006 Subject: [ANN] Valentina Studio for MAC 2.3fc11 (Universal format) Message-ID: Hi MAC developers, You can download 2.3 fc11 build of Valentina Studio for MAC! Few weeks MAC build was on hold comparing to Windows builds. At last of end it is here! You can see in the Manti change log all features and fixed provided. http://www.valentina-db.com/bt/changelog_page.php VStudio - 2.3 fc10 ================== - 0001453: [Schema Browser] Added shortcuts Ctrl-K, Ctrl-M and Ctrl-B for SchemaBrowser (igor_nikita) - 0001445: [Schema Browser] Crash when closing Studio and SchemaBrowser is open (igor_nikita) - 0001280: [Application Menu] Newly created 1.x databases are not shown in Schema Browser (igor_nikita) - 0001448: [Property Inspector] It is impossible to change length of VarChar field in PropertyInspector (igor_nikita) - 0001446: [Schema Browser] Buttons in SchemaBrowser looks terrible (igor_nikita) - 0001441: [Table Browser] DataBrowser shows wrong columns for V1 databases (igor_nikita) - 0000850: [ODBC Import] Trying to Import an ODBC data source on Mac OS X in Valentina Studio (jochen) - 0001191: [SQL Editor] Boolean fields do not correctly show state in SQL Editor Result (jochen) VStudio - 2.3 fc9 ================= - 0001386: [Application Menu] Plug-in vs_plugin_kernel_1.dll does not load (ruslan) - 0001426: [SQL Editor] Improved display of query text in result tab (igor_nikita) - 0001427: [SQL Editor] Shortcuts Ctrl-C, Ctrl-V, Ctrl-X, Ctrl-A, Ctrl-Space work now (igor_nikita) - 0001428: [SQL Editor] Some small Improvements of SQL Editor (igor_nikita) - 0001383: [SQL Editor] Query result panel's query's frame should be scrollable. (igor_nikita) VStudio - 2.2 fc8 ================= - 0001385: [SQL Editor] Improved work with query results (igor_nikita) - 0001384: [Schema Browser] Wrong icons for children of encrypted object (igor_nikita) - 0001359: [SQL Editor] List "Recent..." contains wrong ( duplicate ) strings (igor_nikita) - 0001358: [Schema Browser] Schema browser show temporary tables (igor_nikita) - 0001356: [Property Inspector] Property Inspector remember collapsed( expanded ) state of preview for pictures in DataBrowser (igor_nikita) - 0001355: [Property Inspector] All window use one Property Inspector now (igor_nikita) - 0001354: [Property Inspector] User can change "Nullable" flag for fields with types Blob, Text, Picture (igor_nikita) - 0001353: [Schema Browser] Search now work for current column (igor_nikita) VStudio - 2.2 fc7 ================= - 0001371: [Schema Browser] Method with wrong value and parent table now show with special icon (igor_nikita) - 0001370: [Application Menu] Auto completion now work in SQL Editor (igor_nikita) - 0001369: [SQL Editor] SQL Editor now use tool tips as well as in DataBrowser (igor_nikita) -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20060323/a8295f2f/attachment.html From sunshine at public.kherson.ua Thu Mar 23 10:57:05 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 23 02:57:14 2006 Subject: [ANN] Vserver and V4REV 2.3b7 for MAC uploaded. In-Reply-To: Message-ID: Hi All, You can download 2.3b7 builds of V4REV and Valentina Server For Mac OS -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Thu Mar 23 11:10:33 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Mar 23 03:10:52 2006 Subject: RSS sources for Valentina Database. Message-ID: Hi Guys, On our home page you can see now News (RSS) Wiki (RSS) So you can subscribe to this channels to watch for any news from us. -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Fri Mar 24 09:51:24 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 24 01:51:33 2006 Subject: [VSDK 2] Class way implemented. In-Reply-To: <7F5BBC3B7975A94F9844614AB95C3755065C4C@srvex01.se-dd.local> Message-ID: On 3/21/06 7:40 PM, "Decoster Jean Louis" wrote: Hi Jean, I have got it working in the first example. Looks very good I should say. I will send you example + few files in near hours. > Hi Ruslan, > > First thanks for your mail. I don't know why but the last days I couln't send > any mail to you. > -------------------- > > Hi Russlan, > > Thank you for last mail. I wanted some days to try to work with VSDK_2; but > till now I could not make many things. > When you write VSDK 2.x do not give you CLASS WAY. It's for me not very easy. > I have been writing applications using Valentina, Corba and C++ and Java > applications. Using VSDK_1 and factory design pattern, I can declare classes > which extend VDKBaseObject and Corba like: > vdb_xyztable::vdb_xyztable():VDK_BaseObject("XyzTable"), table(), > xyzCorba() { > // Create Fields > xyzName = new VDK_String ( "Xyz_Name", 32 ); > Value = new VDK_ULong ( "Value" ); > } > You wrote > So If you are used to this style of coding, I'd like to discuss what > we can do here. > I see one way easy enough to get this style back into game using > simple wrapper class of smart-ptr kind which use trick of operator ->. > Only protential problem for this implementation that we will use -> for > non-pointer objects > class tnlPerson > { > VString mName; > } > .... > mName->SetString() // C++ developer expect normally here > // mName.SetString() > Is it possible for you to make something to provide with VSDK_2 the same > possibitlities given by VSDK_1? I am not sure that I can get them at this > moment. Perhaps it's also possible for you to sent a little more completed > examples using load, save, add, delete records from tables using wrapper > classes with VSDK_2. Please, can you add my name to your MemberList. ? Thanks > a lot for your next answer. > Regards, jean-Louis -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From bkeeney at everestkc.net Fri Mar 24 16:27:14 2006 From: bkeeney at everestkc.net (Bob Keeney) Date: Fri Mar 24 16:27:35 2006 Subject: [V4RB] 2.3B5 Difference Between Mac and Windows? In-Reply-To: <20060323091047.97483408291@edison.macserve.net> References: <20060323091047.97483408291@edison.macserve.net> Message-ID: I have a functioning Mac version of an application that uses this code on a regular basis and has no issues. Same application compiled for Windows using the same database (created on the Mac and used in Windows) oSet is always nil when running this code. I've verified via VStudio that the data exists. I'm using the class way. What, if any, differences are there between the Mac and Windows versions that would cause this? Thanks, Bob Keeney ================================================================== Function AccountExists(sAcct as string, byRef TheRecordID as integer) As Boolean dim oSet as vSet dim iter as VSetIterator dim recID as integer sAcct = left(sAcct, 7) oSet = me.acct.FindValue(sAcct) if oSet = nil then return false end if iter = oSet.MakeNewIterator if oSet.IsEmpty then return false RecID = iter.FirstItem() if RecID = 0 then return false end if TheRecordID = recid if me.RecordExists(recid) then return true else TheRecordID = 0 return false end End Function From sunshine at public.kherson.ua Sat Mar 25 00:47:17 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Mar 24 16:47:22 2006 Subject: [V4RB] 2.3B5 Difference Between Mac and Windows? In-Reply-To: Message-ID: On 3/25/06 12:27 AM, "Bob Keeney" wrote: Hi Bob, > I have a functioning Mac version of an application that uses this > code on a regular basis and has no issues. Same application compiled > for Windows using the same database (created on the Mac and used in > Windows) oSet is always nil when running this code. I've verified > via VStudio that the data exists. I'm using the class way. > > What, if any, differences are there between the Mac and Windows > versions that would cause this? Of course behavior should be the same. Try REINDEX please. If db is small ? I think it is enough to have this db and do this FindValue() On it, so test project with 5 lines of code should reproduce. Right ? -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From sunshine at public.kherson.ua Mon Mar 27 16:16:32 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 27 07:16:43 2006 Subject: [IMPROVED] in 2.3 b8 will be speed improve 10-30% Message-ID: Hi All, Improved speed of Table.AddRecord() on cache level. Test with one field in table was 111 sec now 77 sec Test with 13 fields in tables was 3.49 sec now 3.17 -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From jda at his.com Mon Mar 27 08:21:31 2006 From: jda at his.com (jda) Date: Mon Mar 27 07:21:42 2006 Subject: [IMPROVED] in 2.3 b8 will be speed improve 10-30% In-Reply-To: References: Message-ID: >Hi All, > >Improved speed of Table.AddRecord() on cache level. > >Test with one field in table > was 111 sec now 77 sec > > >Test with 13 fields in tables > was 3.49 sec now 3.17 > Hi Ruslan, Can we use VarChar of 2000 with this beta now? Thanks, Jon From sunshine at public.kherson.ua Mon Mar 27 18:04:25 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Mar 27 09:04:36 2006 Subject: [IMPROVED] in 2.3 b8 will be speed improve 10-30% In-Reply-To: Message-ID: On 3/27/06 3:21 PM, "jda" wrote: >> Hi All, >> >> Improved speed of Table.AddRecord() on cache level. >> >> Test with one field in table >> was 111 sec now 77 sec >> >> >> Test with 13 fields in tables >> was 3.49 sec now 3.17 >> > > Hi Ruslan, > > Can we use VarChar of 2000 with this beta now? Yes VarChar[2044] -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed] From bkeeney at everestkc.net Tue Mar 28 17:04:05 2006 From: bkeeney at everestkc.net (Bob Keeney) Date: Tue Mar 28 17:04:11 2006 Subject: V4RB 2.3b5 Too Big Maximal Length Error In-Reply-To: <20060323091047.97483408291@edison.macserve.net> References: <20060323091047.97483408291@edison.macserve.net> Message-ID: I have a pre-2.2 database file that has a String field length set at 1024. Using b5 I get the following error: You have specify too big maximal length for VarChar field "sNotes". No biggie, right? Well, my database is encrypted, both data and structure and I can't even get to a point where I can alter the table. I have this code: dim tempDB as new VDataBase try tempDB.Open f catch tempDB.UseEncryptionKey kStructureEncryption, EVDataKind.kStructureAndRecords tempDB.Open f <--Throws error above. end ...check schema version here The first open is for the occasional non-encrypted database that we have (very old versions). But I get the above error message and subsequent hard crash of the application even before I can modify the table to fix the problem. Is there better way of doing this? This is very big problem with production databases out in the field so any assistance is appreciated. BTW, VStudio lets you make this field 1024 with no errors. Thanks, Bob K From sunshine at public.kherson.ua Wed Mar 29 02:18:37 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Mar 28 17:18:43 2006 Subject: V4RB 2.3b5 Too Big Maximal Length Error In-Reply-To: Message-ID: On 3/29/06 2:04 AM, "Bob Keeney" wrote: > I have a pre-2.2 database file that has a String field length set at > 1024. Using b5 I get the following error: > > You have specify too big maximal length for VarChar field "sNotes". > > No biggie, right? Well, my database is encrypted, both data and > structure and I can't even get to a point where I can alter the > table. I have this code: > > dim tempDB as new VDataBase > > try > tempDB.Open f > catch > tempDB.UseEncryptionKey kStructureEncryption, > EVDataKind.kStructureAndRecords > tempDB.Open f <--Throws error above. > end > > ...check schema version here > > The first open is for the occasional non-encrypted database that we > have (very old versions). But I get the above error message and > subsequent hard crash of the application even before I can modify the > table to fix the problem. > > Is there better way of doing this? This is very big problem with > production databases out in the field so any assistance is appreciated. > > BTW, VStudio lets you make this field 1024 with no errors. > > Thanks, > Bob K b5 set limit to 1022 Next beta will increase size to VarChar[2044] IF you have longer VarChar or String please redo on TEXT that field -- Best regards, Ruslan Zasukhin VP Engineering and New Technology Paradigma Software, Inc Valentina - Joining Worlds of Information http://www.paradigmasoft.com [I feel the need: the need for speed]