From sunshine at public.kherson.ua Thu Jun 1 09:11:24 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 01:11:30 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: Message-ID: On 6/1/06 7:47 AM, "Dave Parizek" wrote: >> Hey Dave, what does the crash log say? > > I am not sure where the crash log is. Is this something that goes to > the Console? A system.log file somewhere? Sorry, Cocoa and XCode > newbie here... Sometimes crash log is show right after crash. If no shown you can find it in Console.app => Logs => CrashLog -- 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 Jun 1 09:15:23 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 01:15:31 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: Message-ID: On 6/1/06 7:47 AM, "Dave Parizek" wrote: > I did have a quick design question though. You have a Valentina > Cursor, and you feed it to an Array controller so you can use > bindings to feed the tableview. Does that mean the data is being fed > into the cursor plus also into an array for the array controller? No, data are not copied into RAM, like this do REALbasic. > Wouldn't it be faster / less memory if it were possible to have the > cursor and the tableview connect to each other directly without the > addition of the array? I believe this exactly happens in background of binding. > Or is bindings just such a great thing the additional overhead (if such > exists, I may just not be understanding how things work) is worth it? Looks to be :) In fact idea of binding is very simple. This is a contract. Contract between our classes (V4CC in this case) and Cocoa. In this contract we promise name special functions in special way. So Cocoa can self find them and call. -- 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 Jun 1 09:17:58 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 01:18:06 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: Message-ID: On 6/1/06 8:02 AM, "Dave Parizek" wrote: Hi Dave, > Some more info on the "EXC_BAD_ACCESS" error problem: > > I re-downloaded installer for Ruslan's example, ran the installer. > Re-tried his example and it worked, no crash now. But then I retried > your NSTableView example and now it crashes with same error message. > Whereas before yours worked and his crashed. > > So this seems for sure it is just a case of some necessary files are > being put in different places? But they cannot be duplicated and put > in both places because then somehow they conflict? Tell me please, if Justin have send you also V4CC.Framework? If yes, then I think you just have one hands 2 different builds, Which are not compatible. > So I think I will just look at and try to learn from both yours and > Ruslan's source code, and start working on my own example. I do not > necessarily need to be able to run both your examples anymore. I > just need to know what is final decision about where files should be > placed so that I can make sure my example (and eventually my project) > works with that decision. > > Ruslan, did you make a final decision? Or rather a decision for now? About which decision you asks? -- 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 Jun 1 09:36:55 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 01:37:05 2006 Subject: Stan's slow problem [located?] In-Reply-To: <9692DBD4-7FDD-47A5-A9CD-6AAA0FE6A1B6@mac.com> Message-ID: On 5/31/06 10:20 PM, "Stan Busk" wrote: Hi Stan, Okay, I was able to use Apple's Performance Tools to see where is problem in your case. With that Country Rating. I see that is called Find_NotIndexed 94% of time !!!! FindValue I.e. You should see about 20 times slow down, right ? And reason is that you do something as WHERE fld = xxx Where fld field is not indexed. Hmm, not clear why 2.2 is fast for you ... -------------------- I recommend you download and install V4RB 2.4 b10 Then try to use new feature EXPLAIN. For this before do your SQL query you need turn ON The WARNING mode. To do this you need simply add one line: db.SqlExecute( "SET PROPERTY WarningMode TO TRUE" ) db.SqlExecute( your query ) Now after your query done, go to your app folder You should see Warnings_log. Open it Here you should see very detailed info about each step of query. I expect that you will see WHICH field do that not indexed search. Let me know result. -- 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 Jun 1 09:50:41 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 01:50:48 2006 Subject: [MAC] CrashReporterPrefs.app and others useful things Message-ID: Hi guys, If you work on MAC, I recommend install xcode (latest 2.3) Because with xcode you will get also folder /Developer/Applications And here a lots of useful things. For example /Developer/Applications/Utilities/CrashReporterPrefs.app /Developer/Applications/ Performance Tools here Sampler have show me where have problem Stan's query in the V4RB MACHO -- 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 dev at macfoh.com Thu Jun 1 00:04:40 2006 From: dev at macfoh.com (Shaun Wexler) Date: Thu Jun 1 02:04:50 2006 Subject: [MAC] CrashReporterPrefs.app and others useful things In-Reply-To: References: Message-ID: <748C2DC5-BB4C-4027-8F5D-57C5626222C2@macfoh.com> On May 31, 2006, at 11:50 PM, Ruslan Zasukhin wrote: > Hi guys, > > If you work on MAC, I recommend install xcode (latest 2.3) > Because with xcode you will get also folder > > /Developer/Applications > > And here a lots of useful things. > > For example > /Developer/Applications/Utilities/CrashReporterPrefs.app > > /Developer/Applications/ Performance Tools > here Sampler have show me where have problem Stan's query > in the V4RB MACHO Be sure to install the latest CHUD 4.3.3 profiling tools: -- Shaun Wexler MacFOH http://www.macfoh.com From sunshine at public.kherson.ua Thu Jun 1 10:09:31 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 02:09:37 2006 Subject: [MAC] CrashReporterPrefs.app and others useful things In-Reply-To: <748C2DC5-BB4C-4027-8F5D-57C5626222C2@macfoh.com> Message-ID: On 6/1/06 10:04 AM, "Shaun Wexler" wrote: Hi Shaun, >> If you work on MAC, I recommend install xcode (latest 2.3) >> Because with xcode you will get also folder >> >> /Developer/Applications >> >> And here a lots of useful things. >> >> For example >> /Developer/Applications/Utilities/CrashReporterPrefs.app >> >> /Developer/Applications/ Performance Tools >> here Sampler have show me where have problem Stan's query >> in the V4RB MACHO > > Be sure to install the latest CHUD 4.3.3 profiling tools: > It is not included into xcode 2.3 ? -- 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 dev at macfoh.com Thu Jun 1 00:14:12 2006 From: dev at macfoh.com (Shaun Wexler) Date: Thu Jun 1 02:14:20 2006 Subject: [MAC] CrashReporterPrefs.app and others useful things In-Reply-To: References: Message-ID: <88B16664-2FB9-4F2A-BEBD-C711D86D4503@macfoh.com> On Jun 1, 2006, at 12:09 AM, Ruslan Zasukhin wrote: > On 6/1/06 10:04 AM, "Shaun Wexler" wrote: >> >> Be sure to install the latest CHUD 4.3.3 profiling tools: >> > > It is not included into xcode 2.3 ? No, you must also install CHUD, which provides Shark, Saturn, and several CPU-level performance and debugging tools (which are mostly only useful to tweak-heads like me). Shark is the profiling tool you need to use to find hotspots and bottlenecks when writing Valentina code. Anyone with Shark questions, feel free to ask me and/or post on the mailing list for help. I love Shark and use it all the time. -- Shaun Wexler MacFOH http://www.macfoh.com "I don't even see the code. All I see is blond, brunette, red-head..." -Cypher, The Matrix From sunshine at public.kherson.ua Thu Jun 1 18:03:20 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 10:03:26 2006 Subject: Stan's slow problem [located?] In-Reply-To: <21DC35EB-668E-4223-B68E-AC5F7F4BDFC9@mac.com> Message-ID: On 6/1/06 1:32 PM, "Stan Busk" wrote: > > I have updated the DB (structure + classes) and I have reindexed. I > get a 3 seconds enhancement only. > It looks like there is something else... Okay I know where is problem. Search on Boolean fields. Transaction_Refunded, Transaction_ChargeBack. They are marked as Indexed, But they do not have real index... What I cannot understand, where are my algorithms to work on Boolean field. They have disappear ... Black magic. Okay, know I know where to search for ... -- 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 Jun 1 18:56:37 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 10:56:43 2006 Subject: Stan's slow problem [located?] In-Reply-To: Message-ID: On 6/1/06 6:03 PM, "Ruslan Zasukhin" wrote: > On 6/1/06 1:32 PM, "Stan Busk" wrote: > >> >> I have updated the DB (structure + classes) and I have reindexed. I >> get a 3 seconds enhancement only. >> It looks like there is something else... > > Okay I know where is problem. > > Search on Boolean fields. > Transaction_Refunded, Transaction_ChargeBack. > > They are marked as Indexed, > But they do not have real index... > > What I cannot understand, where are my algorithms to work on Boolean field. > They have disappear ... Black magic. > > Okay, know I know where to search for ... I have find that change that did happens in 2.3: I have found that Boolean field do have index and disable creation of index for it. Probably I have learn 1.x sources and have see that for Boolean field I did have overridden search methods. And most probably I have forget implement them for Boolean. As we see this step have drop down speed. And right - at least 32 times, may be even 40 times looks to be true... So issue not in COUNT() and not in SUM(), but in WHERE fldBool = xxx -- 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 Jun 1 19:01:02 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 11:01:08 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: Message-ID: On 6/1/06 6:56 PM, "Ruslan Zasukhin" wrote: Btw, this is how looks EXPLAIN for Stan's query FOR SQL QUERY TIME = 63 : DO SELECT : TIME = 63 : ROWS = 1 DO SIMPLE SELECT : TIME = 63 : ROWS = 1 DO Single Table Without GROUP BY : TIME = 63 : ROWS = 1 FOR Table [tblTransactions] DO resolve WHERE statement : TIME = 63 : ROWS = 0 FOR Predicate [AND] TIME = 63 : ROWS = 0 DO AND Indexed operands : TIME = 63 : ROWS = 219 FOR Field [Transaction_Date] DO FindRange [01/01/2000 00:00:00:000, 12/31/2006 00:00:00:000] : TIME = 0 : ROWS = 38727 FOR Field [Transaction_Refunded] DO FindValue(1) : TIME = 25 : ROWS = 219 FOR Field [Transaction_ChargedBack] DO FindValue(0) : TIME = 38 : ROWS = 38684 DO AND Non-Indexed operands : TIME = 0 : ROWS = 0 FOR Predicate [OR] TIME = 0 : ROWS = 0 DO OR Indexed operands : TIME = 0 : ROWS = 21 FOR Field [Transaction_Country] DO FindLike : TIME = 0 : ROWS = 21 FOR Field [Transaction_Country] DO FindLike : TIME = 0 : ROWS = 0 -- 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/20060601/fe0185af/attachment.html From dave at Parizek.com Thu Jun 1 09:38:16 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 11:38:20 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: References: Message-ID: <15441C0A-9795-4D01-8D8E-4673E6984748@Parizek.com> I do not see any info about xcode or my app in any of the logs available from Console.app. None of the logs is titled "CrashLog". There are some "CrashReporter" logs in ~/Library/Logs but these all are listed by application and neither XCode or the TableViewDemo app is listed... Do I need to enable something to get a crash log? Since the debugger is catching it when I Build and Run is it maybe not printing any info to the log? --Dave __________________________________________ Dave Parizek On May 31, 2006, at 11:11 PM, Ruslan Zasukhin wrote: > On 6/1/06 7:47 AM, "Dave Parizek" wrote: > >>> Hey Dave, what does the crash log say? >> >> I am not sure where the crash log is. Is this something that goes to >> the Console? A system.log file somewhere? Sorry, Cocoa and XCode >> newbie here... > > Sometimes crash log is show right after crash. > > If no shown you can find it in Console.app => Logs => CrashLog > > > -- > 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 dave at Parizek.com Thu Jun 1 09:39:54 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 11:39:53 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: References: Message-ID: Cursors work fundamentally different in Cocoa versus RealBasic? V4RB puts cursor into RAM but Cocoa cursor works differently? --Dave __________________________________________ Dave Parizek On May 31, 2006, at 11:15 PM, Ruslan Zasukhin wrote: > On 6/1/06 7:47 AM, "Dave Parizek" wrote: > >> I did have a quick design question though. You have a Valentina >> Cursor, and you feed it to an Array controller so you can use >> bindings to feed the tableview. Does that mean the data is being fed >> into the cursor plus also into an array for the array controller? > > No, data are not copied into RAM, like this do REALbasic. From dave at Parizek.com Thu Jun 1 09:42:17 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 11:42:18 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: References: Message-ID: <7615C917-C854-4699-BA92-440361760F7D@Parizek.com> I mean the decision of how/where Valentina frameworks should be installed for V4CC. Like Justin way with symlink, or other way. Remember Justin wrote out the pros and cons of 2 different ways? And you said maybe create a script for something (building or installing, I'm not sure). --Dave __________________________________________ Dave Parizek On May 31, 2006, at 11:17 PM, Ruslan Zasukhin wrote: > On 6/1/06 8:02 AM, "Dave Parizek" wrote: > > Hi Dave, > >> Some more info on the "EXC_BAD_ACCESS" error problem: >> >> I re-downloaded installer for Ruslan's example, ran the installer. >> Re-tried his example and it worked, no crash now. But then I retried >> your NSTableView example and now it crashes with same error message. >> Whereas before yours worked and his crashed. >> >> So this seems for sure it is just a case of some necessary files are >> being put in different places? But they cannot be duplicated and put >> in both places because then somehow they conflict? > > Tell me please, if Justin have send you also V4CC.Framework? > > If yes, then I think you just have one hands 2 different builds, > Which are not compatible. > >> So I think I will just look at and try to learn from both yours and >> Ruslan's source code, and start working on my own example. I do not >> necessarily need to be able to run both your examples anymore. I >> just need to know what is final decision about where files should be >> placed so that I can make sure my example (and eventually my project) >> works with that decision. >> >> Ruslan, did you make a final decision? Or rather a decision for now? > > About which decision you asks? > > -- > 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 Thu Jun 1 20:45:41 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 12:45:49 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: Message-ID: On 6/1/06 7:39 PM, "Dave Parizek" wrote: > Cursors work fundamentally different in Cocoa versus RealBasic? V4RB > puts cursor into RAM but Cocoa cursor works differently? Not cursor, but listbox of REALbasic. > On May 31, 2006, at 11:15 PM, Ruslan Zasukhin wrote: > >> On 6/1/06 7:47 AM, "Dave Parizek" wrote: >> >>> I did have a quick design question though. You have a Valentina >>> Cursor, and you feed it to an Array controller so you can use >>> bindings to feed the tableview. Does that mean the data is being fed >>> into the cursor plus also into an array for the array controller? >> >> No, data are not copied into RAM, like this do REALbasic. -- 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 Jun 1 20:47:16 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 12:47:22 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: <7615C917-C854-4699-BA92-440361760F7D@Parizek.com> Message-ID: On 6/1/06 7:42 PM, "Dave Parizek" wrote: > I mean the decision of how/where Valentina frameworks should be > installed for V4CC. Like Justin way with symlink, or other way. I believe that I have forward to beta list 2 weeks ago Note from Jochen, where he say that xcode 2.3 fix bug When we was need add sym links into SDK 10.4 > Remember Justin wrote out the pros and cons of 2 different ways? And > you said maybe create a script for something (building or installing, > I'm not sure). This not polished yet and for now this is not important. -- 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 Jun 1 20:48:29 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 12:48:58 2006 Subject: V4CC - Cocoa examples problem In-Reply-To: <15441C0A-9795-4D01-8D8E-4673E6984748@Parizek.com> Message-ID: On 6/1/06 7:38 PM, "Dave Parizek" wrote: > I do not see any info about xcode or my app in any of the logs > available from Console.app. None of the logs is titled "CrashLog". > There are some "CrashReporter" logs in ~/Library/Logs but these all > are listed by application and neither XCode or the TableViewDemo app > is listed... > > Do I need to enable something to get a crash log? Since the debugger > is catching it when I Build and Run is it maybe not printing any info > to the log? Well, very rare can be crash when crash log is not produced. * open console * click Logs * open ~/Library/Logs * here CrashReporter -- 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 Jun 1 20:52:32 2006 From: maxprog at mac.com (Stan Busk) Date: Thu Jun 1 13:52:44 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: References: Message-ID: <2D3AFC12-D6F3-4B97-93BE-DD4D51E70DAE@mac.com> Hi, With my other problematic app I get huge timings in 'Predicate' line (compared to previous app): FOR SQL QUERY TIME = 2 : DO SELECT : TIME = 2 : ROWS = 1 DO SIMPLE SELECT : TIME = 2 : ROWS = 1 DO Single Table Without GROUP BY : TIME = 2 : ROWS = 1 FOR Table [tblArticle] DO resolve WHERE statement : TIME = 2 : ROWS = 0 FOR Predicate [AND] TIME = 4290251425 : TIME = 4290251425 : FOR Field [Article_Status] DO FindLike : TIME = 2 : ROWS = 0 and so on... Always very big amounts for TIME. Does it correspond to ticks? Stan > On 6/1/06 6:56 PM, "Ruslan Zasukhin" > wrote: > > Btw, this is how looks EXPLAIN for Stan's query > > > FOR SQL QUERY TIME = 63 : > DO SELECT : TIME = 63 : ROWS = 1 > DO SIMPLE SELECT : TIME = 63 : ROWS = 1 > DO Single Table Without GROUP BY : TIME = 63 : ROWS = 1 > FOR Table [tblTransactions] DO resolve WHERE > statement : TIME = 63 : ROWS = 0 > FOR Predicate [AND] TIME = 63 : ROWS = 0 > DO AND Indexed operands : TIME = 63 : ROWS > = 219 > FOR Field [Transaction_Date] DO > FindRange [01/01/2000 00:00:00:000, 12/31/2006 00:00:00:000] : TIME > = 0 : ROWS = 38727 > FOR Field [Transaction_Refunded] DO > FindValue(1) : TIME = 25 : ROWS = 219 > FOR Field [Transaction_ChargedBack] DO > FindValue(0) : TIME = 38 : ROWS = 38684 > DO AND Non-Indexed operands : TIME = 0 : > ROWS = 0 > FOR Predicate [OR] TIME = 0 : ROWS = 0 > DO OR Indexed operands : TIME = > 0 : ROWS = 21 > FOR Field > [Transaction_Country] DO FindLike : TIME = 0 : ROWS = 21 > FOR Field > [Transaction_Country] DO FindLike : TIME = 0 : ROWS = 0 > > > -- > 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 Thu Jun 1 21:54:59 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 13:55:05 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: <2D3AFC12-D6F3-4B97-93BE-DD4D51E70DAE@mac.com> Message-ID: On 6/1/06 9:52 PM, "Stan Busk" wrote: > Hi, > > With my other problematic app I get huge timings in 'Predicate' line > (compared to previous app): > > FOR SQL QUERY TIME = 2 : > DO SELECT : TIME = 2 : ROWS = 1 > DO SIMPLE SELECT : TIME = 2 : ROWS = 1 > DO Single Table Without GROUP BY : TIME = 2 : ROWS = 1 > FOR Table [tblArticle] DO resolve WHERE statement : TIME = 2 : > ROWS = 0 > FOR Predicate [AND] TIME = 4290251425 : > TIME = 4290251425 : > FOR Field [Article_Status] DO FindLike : TIME = 2 : ROWS = 0 > > and so on... > > Always very big amounts for TIME. Does it correspond to ticks? I think this is a garbage value. Probably somewhere bug 0 - somevalue = negative = big integer -- 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 Jun 1 21:04:59 2006 From: maxprog at mac.com (Stan Busk) Date: Thu Jun 1 14:05:14 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: References: Message-ID: <906CB3F0-4568-4F02-8D56-9F7480183403@mac.com> I get some like: FOR SQL QUERY TIME = 69 : DO SELECT : TIME = 69 : ROWS = 1 DO SIMPLE SELECT : TIME = 69 : ROWS = 1 DO Single Table Without GROUP BY : TIME = 69 : ROWS = 1 FOR Table [tblArticle] DO resolve WHERE statement : TIME = 69 : ROWS = 3765 FOR Field [Article_Newsgroups] DO FindLike : TIME = 69 : ROWS = 3765 and most as: FOR SQL QUERY TIME = 12 : DO SELECT : TIME = 12 : ROWS = 1 DO SIMPLE SELECT : TIME = 12 : ROWS = 1 DO Single Table Without GROUP BY : TIME = 12 : ROWS = 1 FOR Table [tblArticle] DO resolve WHERE statement : TIME = 12 : ROWS = 3254 FOR Predicate [AND] TIME = 12 : ROWS = 3254 DO AND Indexed operands : TIME = 0 : ROWS = 29234 FOR Field [Article_Status] DO FindLike : TIME = 0 : ROWS = 29234 DO AND Non-Indexed operands : TIME = 12 : ROWS = 3254 FOR Field [Article_Newsgroups] DO FindLike : TIME = 12 : ROWS = 3254 Meanwhile I am looking for that Non-Indexed operand. Are they normal timings? No boolean involved here however queries are extremely slow. Stan > On 6/1/06 9:52 PM, "Stan Busk" wrote: > >> Hi, >> >> With my other problematic app I get huge timings in 'Predicate' line >> (compared to previous app): >> >> FOR SQL QUERY TIME = 2 : >> DO SELECT : TIME = 2 : ROWS = 1 >> DO SIMPLE SELECT : TIME = 2 : ROWS = 1 >> DO Single Table Without GROUP BY : TIME = 2 : ROWS = 1 >> FOR Table [tblArticle] DO resolve WHERE statement : TIME = 2 : >> ROWS = 0 >> FOR Predicate [AND] TIME = 4290251425 : >> TIME = 4290251425 : >> FOR Field [Article_Status] DO FindLike : TIME = 2 : ROWS = 0 >> >> and so on... >> >> Always very big amounts for TIME. Does it correspond to ticks? > > I think this is a garbage value. Probably somewhere bug > > 0 - somevalue = negative = big integer > > > > -- > 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 Thu Jun 1 22:23:49 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 14:23:53 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: <906CB3F0-4568-4F02-8D56-9F7480183403@mac.com> Message-ID: On 6/1/06 10:04 PM, "Stan Busk" wrote: > I get some like: > > FOR SQL QUERY TIME = 69 : > DO SELECT : TIME = 69 : ROWS = 1 > DO SIMPLE SELECT : TIME = 69 : ROWS = 1 > DO Single Table Without GROUP BY : TIME = 69 : ROWS = 1 > FOR Table [tblArticle] DO resolve WHERE statement : TIME = 69 : > ROWS = 3765 > FOR Field [Article_Newsgroups] DO FindLike : TIME = 69 : ROWS = > 3765 I think it will be good idea show type of 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 Thu Jun 1 22:24:38 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 14:24:46 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: <906CB3F0-4568-4F02-8D56-9F7480183403@mac.com> Message-ID: On 6/1/06 10:04 PM, "Stan Busk" wrote: > Meanwhile I am looking for that Non-Indexed operand. > > Are they normal timings? No boolean involved here however queries are > extremely slow. Stan, lets step by step. I will fix booleans (lets hope tomorrow), then I will check next your build, okay ? -- 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 Jun 1 21:31:08 2006 From: maxprog at mac.com (Stan Busk) Date: Thu Jun 1 14:31:24 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: References: Message-ID: No problem, I was just pinpointing the queries. Right now I think I am ready to send stuff on your request. > On 6/1/06 10:04 PM, "Stan Busk" wrote: > >> Meanwhile I am looking for that Non-Indexed operand. >> >> Are they normal timings? No boolean involved here however queries are >> extremely slow. > > Stan, lets step by step. > > I will fix booleans (lets hope tomorrow), > then I will check next your build, okay ? > > -- > 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 fitzbew at nc.rr.com Thu Jun 1 15:54:30 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Thu Jun 1 14:54:37 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: Message-ID: On 6/1/06 3:31 PM, "Stan Busk" wrote: > > No problem, I was just pinpointing the queries. Right now I think I > am ready to send stuff on your request. > >> On 6/1/06 10:04 PM, "Stan Busk" wrote: >> >>> Meanwhile I am looking for that Non-Indexed operand. >>> >>> Are they normal timings? No boolean involved here however queries are >>> extremely slow. >> >> Stan, lets step by step. >> >> I will fix booleans (lets hope tomorrow), >> then I will check next your build, okay ? >> Not trying to "hijack" this effort --- I'm really trying to be helpful --- but my mantis issue: 0001575: Particular Query Performance Slow With Trivial Join is probably the result of some problem similar to Stan's. Although I had thought perhaps the parser was the issue. -- Russ Tyndall Wake Forest, NC From sunshine at public.kherson.ua Thu Jun 1 23:01:41 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 15:01:46 2006 Subject: Stan's slow problem EXPLAIN In-Reply-To: Message-ID: On 6/1/06 10:54 PM, "Russ Tyndall" wrote: >>> I will fix booleans (lets hope tomorrow), >>> then I will check next your build, okay ? >>> > > Not trying to "hijack" this effort --- I'm really trying to be helpful --- > but my mantis issue: > > 0001575: Particular Query Performance Slow With Trivial Join > > is probably the result of some problem similar to Stan's. Although I had > thought perhaps the parser was the issue. Yes Russ, I have think about this also. I think you will need test new build and see if it help. Then inform me. -- 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 Jun 1 23:48:36 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 15:48:43 2006 Subject: Stan's slow problem -- NEW BUILD of 2.4 b10 In-Reply-To: Message-ID: On 6/1/06 10:31 PM, "Stan Busk" wrote: Hi Stan, please download new build of V4RB MAC Still marked as 2.4 b10 Let me know results about these: > Run the app, select the 'Reports' tab, select 'Country Ranking' > report and press the button on the right. > You can also select the 'Chart' tab. Switch between 'Expenses' and > 'Sales'. It was instantaneous before. -- 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 Jun 1 23:10:16 2006 From: maxprog at mac.com (Stan Busk) Date: Thu Jun 1 16:10:52 2006 Subject: Stan's slow problem -- NEW BUILD of 2.4 b10 In-Reply-To: References: Message-ID: <91B63086-2F47-41FE-B3F4-CC1939FD2AF0@mac.com> Back to normal speed!... wow... 4s rather than 40s for the report and chart are again instantaneous! Really cool! I have checked with the other app and no joy though, seems it has another different problem. Stan > On 6/1/06 10:31 PM, "Stan Busk" wrote: > > Hi Stan, > > please download new build of V4RB MAC > Still marked as 2.4 b10 > > Let me know results about these: > >> Run the app, select the 'Reports' tab, select 'Country Ranking' >> report and press the button on the right. > >> You can also select the 'Chart' tab. Switch between 'Expenses' and >> 'Sales'. It was instantaneous before. > > > -- > 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 Jun 2 00:15:27 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 1 16:15:34 2006 Subject: Stan's slow problem -- NEW BUILD of 2.4 b10 In-Reply-To: <91B63086-2F47-41FE-B3F4-CC1939FD2AF0@mac.com> Message-ID: On 6/2/06 12:10 AM, "Stan Busk" wrote: > Back to normal speed!... wow... 4s rather than 40s for the report and > chart are again instantaneous! > Really cool! Good. Although I think you can rewrite your report to use BitSets instead of SQL And you can be much faster of 4 second, it seems . > I have checked with the other app and no joy though, seems it has > another different problem. Okay, prepare for me new build please. If you can guess self, add Valentina.Profile calls -- 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 dave at Parizek.com Thu Jun 1 18:44:25 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 20:44:31 2006 Subject: V4RB converting to 2 from 1.x, Intel, V4CC In-Reply-To: References: Message-ID: <3F31CE6A-7D1D-46D6-8379-D59C6501DF87@Parizek.com> Is there any problem with having V4RB Macho running on the same machine as V4CC? --Dave From dave at Parizek.com Thu Jun 1 18:45:48 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 20:45:47 2006 Subject: searchable archive of list In-Reply-To: References: Message-ID: <3960171F-F9C3-49B0-AEA9-145BDD8D0247@Parizek.com> The link for the mailing list archive takes you to pages with no search facility. You can download a huge file and then search it, but that seems rather inefficient. Is there still a web searchable copy of the list archives somewhere? --Dave From dave at Parizek.com Thu Jun 1 18:51:22 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 20:51:22 2006 Subject: problem converting from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: <1B93FC2C-000D-4D09-B42D-95355C336683@Parizek.com> I am converting a 1.x app in V4RB to 2.3. I get all the required changes made, it compiles, but crashes on startup. I get this error: Runtime Error 4: Failed Assertion Location: plugin.cpp:7417 Failure Condition: pluginEntryTable.GetEntry (entrypointName, out) Failure Message: can't find plugin method Valentina.DebugLevel(Assigns inLevel as integer) I apologize I have not searched the list archives yet, wondering what happened to the web searchable archives I used to turn to... Is this a common thing someone can tell me what I have setup wrong and save me a lot of time hunting thru the list archives? --Dave __________________________________________ Dave Parizek On Jun 1, 2006, at 10:48 AM, Ruslan Zasukhin wrote: > On 6/1/06 7:38 PM, "Dave Parizek" wrote: > >> I do not see any info about xcode or my app in any of the logs >> available from Console.app. None of the logs is titled "CrashLog". >> There are some "CrashReporter" logs in ~/Library/Logs but these all >> are listed by application and neither XCode or the TableViewDemo app >> is listed... >> >> Do I need to enable something to get a crash log? Since the debugger >> is catching it when I Build and Run is it maybe not printing any info >> to the log? > > Well, very rare can be crash when crash log is not produced. > > * open console > > * click Logs > > * open ~/Library/Logs > > * here CrashReporter > > > -- > 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 fitzbew at nc.rr.com Thu Jun 1 22:40:48 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Thu Jun 1 21:40:58 2006 Subject: problem converting from V4RB 1.x to 2.3 In-Reply-To: <1B93FC2C-000D-4D09-B42D-95355C336683@Parizek.com> Message-ID: On 6/1/06 9:51 PM, "Dave Parizek" wrote: > I am converting a 1.x app in V4RB to 2.3. I get all the required > changes made, it compiles, but crashes on startup. I get this error: > > Runtime Error 4: Failed Assertion > > Location: plugin.cpp:7417 > Failure Condition: pluginEntryTable.GetEntry > (entrypointName, out) > Failure Message: can't find plugin method > Valentina.DebugLevel(Assigns inLevel as integer) > > I apologize I have not searched the list archives yet, wondering what > happened to the web searchable archives I used to turn to... > You mean "compiles" as in compiles within the debugger? Or as a standalone app? I got similar messages when I (incorrectly) tried to debug in Mach-O. Currently, V4RB can only be debugged in the RB IDE using the V4RB carbon plugin. You can't debug using the V4RB_Macho.rbx plugin. If you are talking about a compiled app, then I assume you are copying in the Mach-o components into the package? If it is some other problem, fear not, dawn will soon be breaking over the Black Sea and Ruslan will soon be up and eager to help us with our Valentina questions. (Hey, it's 10:50pm here, it's acceptable to wax poetic.) -- Russ Tyndall Wake Forest, NC From fitzbew at nc.rr.com Thu Jun 1 22:42:14 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Thu Jun 1 21:42:19 2006 Subject: searchable archive of list In-Reply-To: <3960171F-F9C3-49B0-AEA9-145BDD8D0247@Parizek.com> Message-ID: On 6/1/06 9:45 PM, "Dave Parizek" wrote: > The link for the mailing list archive takes you to pages with no > search facility. You can download a huge file and then search it, > but that seems rather inefficient. Is there still a web searchable > copy of the list archives somewhere? This link works for me: Russ From dave at Parizek.com Thu Jun 1 19:44:21 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 1 21:44:22 2006 Subject: problem converting from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: I meant compiles as within the debugger. You answered my question - no Mach-O for debugging. Thanks!! I've been sitting here going thru the archives and my head was starting to hurt, and you saved me! --dave __________________________________________ Dave Parizek On Jun 1, 2006, at 7:40 PM, Russ Tyndall wrote: > On 6/1/06 9:51 PM, "Dave Parizek" wrote: > >> I am converting a 1.x app in V4RB to 2.3. I get all the required >> changes made, it compiles, but crashes on startup. I get this error: >> >> Runtime Error 4: Failed Assertion >> >> Location: plugin.cpp:7417 >> Failure Condition: pluginEntryTable.GetEntry >> (entrypointName, out) >> Failure Message: can't find plugin method >> Valentina.DebugLevel(Assigns inLevel as integer) >> >> I apologize I have not searched the list archives yet, wondering what >> happened to the web searchable archives I used to turn to... >> > > You mean "compiles" as in compiles within the debugger? Or as a > standalone > app? > > I got similar messages when I (incorrectly) tried to debug in Mach-O. > Currently, V4RB can only be debugged in the RB IDE using the V4RB > carbon > plugin. You can't debug using the V4RB_Macho.rbx plugin. > > If you are talking about a compiled app, then I assume you are > copying in > the Mach-o components into the package? > > If it is some other problem, fear not, dawn will soon be breaking > over the > Black Sea and Ruslan will soon be up and eager to help us with our > Valentina > questions. > > (Hey, it's 10:50pm here, it's acceptable to wax poetic.) > > -- > Russ Tyndall > Wake Forest, NC > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From maxprog at mac.com Fri Jun 2 13:37:36 2006 From: maxprog at mac.com (Stan Busk) Date: Fri Jun 2 06:37:54 2006 Subject: Stan's slow problem -- NEW BUILD of 2.4 b10 In-Reply-To: References: Message-ID: <86D8065C-3F4D-41A0-AA37-023EABBD8746@mac.com> Hi, I have been looking at that but most queries start looking for record in a date range. myField.FindRange( True, "03/01/2006", "03/31/2006", True ) is that valid? or should it be like in the EXPLAIN log (and Kem Tekinay Apr 21, 2005 messages): myField.FindRange( True, "03/01/2006 00:00:00:000", "03/31/2006 00:00:00:000", True ) Is that format for VDateTime and VDate or VDateTime only? Also is there a way to SUM a field with that or should I go thru all VSet records? And for Count? I don't see in the doc. a VSet.Count but I guess it exists right? > Although I think you can rewrite your report to use BitSets instead > of SQL > And you can be much faster of 4 second, it seems . Stan From sunshine at public.kherson.ua Fri Jun 2 15:04:43 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 2 07:04:47 2006 Subject: Russ, your slow query Message-ID: Hi Russ, The first thing I have discover - you have join on state fields select * from locations, statetable where locations.state = statetable.state But in table locations it is not indexed. This cause non-indexed searches. This is not main show stopper. I see another non-indexed scan of table, which take 70% of time. I will check it now and let you 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 Fri Jun 2 16:06:32 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 2 08:06:34 2006 Subject: Stan's slow problem -- NEW BUILD of 2.4 b10 In-Reply-To: <86D8065C-3F4D-41A0-AA37-023EABBD8746@mac.com> Message-ID: On 6/2/06 2:37 PM, "Stan Busk" wrote: Hi Stan, > I have been looking at that but most queries start looking for record > in a date range. My main concern was that all your N queries do the same search on that boolean fields. Point is that if you use Sets, you can do that searches only once before loop and keep that sets to be used in loops many times. > myField.FindRange( True, "03/01/2006", "03/31/2006", True ) > > is that valid? I think yes > or should it be like in the EXPLAIN log (and Kem Tekinay Apr 21, 2005 > messages): > > myField.FindRange( True, "03/01/2006 00:00:00:000", "03/31/2006 > 00:00:00:000", True ) This is DateTime I think > Is that format for VDateTime and VDate or VDateTime only? > Also is there a way to SUM a field with that or should I go thru all > VSet records? > And for Count? I don't see in the doc. a VSet.Count but I guess it > exists right? 1) do you ask about SELECT COUNT(f) or about Vset.Count ? Vset.Count as Integer - exists such property 2) I think you ask about SQL aggregate functions SUM(), COUNT(), AVG(), yes? I am sure we have in C++ kernel methods for this .. but they are not simple to open for SQL. We need to think how to optimize your loop. Please show me very detailed what you do in that loop. -- 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 fitzbew at nc.rr.com Fri Jun 2 09:20:27 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Fri Jun 2 08:20:32 2006 Subject: Russ, your slow query In-Reply-To: Message-ID: On 6/2/06 8:04 AM, "Ruslan Zasukhin" wrote: > Hi Russ, > > The first thing I have discover - you have join on state fields > > > select * > from locations, statetable > where locations.state = statetable.state > > > But in table locations it is not indexed. > This cause non-indexed searches. > Somehow, I made a mistake when I sent you the test db; the "real" db has both of those fields indexed. But you have perhaps taught me something else that I need to know: Do both fields in a join need to be indexed? What if one of the tables in the join has only 2-3 records? In other words, is indexing worthwhile in tables with only a few records? > This is not main show stopper. > No, although I somehow made a blunder with the test db I sent you, the "real" dbs have both of those fields indexed and the problem still occurs (the query with "and state = 'NC' " takes too long to run). > I see another non-indexed scan of table, which take 70% of time. > I will check it now and let you know > Great. Identical query works blazing fast in Valentina 1.x version of project, so I presume some tiny glitch causing a "stall" or such. Just in case it is helpful to you: Sometimes (not in all cases), I have regained normal Valentina speed by changing such statements "if state = 'NC' " to syntax like this: "if state in('NC')". Russ From dave at Parizek.com Fri Jun 2 07:23:25 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 09:23:33 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: <1A286AA6-ACFB-4A7D-85AB-52E28B52A19F@Parizek.com> I converted an RB project (2006r2) from 1.x to 2.3. After compiling in the debugger I get this error as it starts to run (in the debugger): Application Launch Failure The Application "(null)" could not be launched because of a shared library error: "
<>" I removed all Valentina (V4CC and V4RB) items from my machine and reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get that error. I checked the mailing list archives under "Application Launch Failure" and "shared library error" but did not see anything relevant. Suggestions? --Dave From fitzbew at nc.rr.com Fri Jun 2 10:30:25 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Fri Jun 2 09:30:31 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <1A286AA6-ACFB-4A7D-85AB-52E28B52A19F@Parizek.com> Message-ID: On 6/2/06 10:23 AM, "Dave Parizek" wrote: > I converted an RB project (2006r2) from 1.x to 2.3. > > After compiling in the debugger I get this error as it starts to run > (in the debugger): > > Application Launch Failure > The Application "(null)" could not be launched because of a shared > library error: "
<>" > > I removed all Valentina (V4CC and V4RB) items from my machine and > reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get > that error. > > I checked the mailing list archives under "Application Launch > Failure" and "shared library error" but did not see anything relevant. > > Suggestions? > I'm sure you've ruled this out, but just to be safe: Mach-o is not still checked in the RB build settings is it? Carbon/Classic only is checked? Russ From dave at Parizek.com Fri Jun 2 08:05:55 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 10:06:01 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: <3A0F07F2-4D54-40D8-90CF-47F64401EBCD@Parizek.com> Yes, under build settings I have checked the Mac OSX/Classic option. I also tried classic only option, same result. and always the osx only option unchecked. --dave __________________________________________ Dave Parizek On Jun 2, 2006, at 7:30 AM, Russ Tyndall wrote: > On 6/2/06 10:23 AM, "Dave Parizek" wrote: > >> I converted an RB project (2006r2) from 1.x to 2.3. >> >> After compiling in the debugger I get this error as it starts to run >> (in the debugger): >> >> Application Launch Failure >> The Application "(null)" could not be launched because of a shared >> library error: "
<>" >> >> I removed all Valentina (V4CC and V4RB) items from my machine and >> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >> that error. >> >> I checked the mailing list archives under "Application Launch >> Failure" and "shared library error" but did not see anything >> relevant. >> >> Suggestions? >> > > I'm sure you've ruled this out, but just to be safe: Mach-o is not > still > checked in the RB build settings is it? Carbon/Classic only is > checked? > > Russ > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From fred.stephenson at communication-unltd.com Fri Jun 2 17:23:57 2006 From: fred.stephenson at communication-unltd.com (Fred) Date: Fri Jun 2 10:25:07 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <1A286AA6-ACFB-4A7D-85AB-52E28B52A19F@Parizek.com> References: <1A286AA6-ACFB-4A7D-85AB-52E28B52A19F@Parizek.com> Message-ID: <4480580D.8060000@communication-unltd.com> Are you using any declares ?? Fred Dave Parizek wrote: > I converted an RB project (2006r2) from 1.x to 2.3. > > After compiling in the debugger I get this error as it starts to run > (in the debugger): > > Application Launch Failure > The Application "(null)" could not be launched because of a shared > library error: "
<>" > > I removed all Valentina (V4CC and V4RB) items from my machine and > reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get that > error. > > I checked the mailing list archives under "Application Launch Failure" > and "shared library error" but did not see anything relevant. > > Suggestions? > > --Dave > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > From dave at Parizek.com Fri Jun 2 08:29:11 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 10:29:16 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <4480580D.8060000@communication-unltd.com> References: <1A286AA6-ACFB-4A7D-85AB-52E28B52A19F@Parizek.com> <4480580D.8060000@communication-unltd.com> Message-ID: <65388A04-4AA8-47C6-941C-1D1E6CABD1B9@Parizek.com> I am using HierPop contextual menu extension which does some declares. I am using Einhugur's DataGrid and TreeView plugins, so they might do some declares? --Dave __________________________________________ Dave Parizek On Jun 2, 2006, at 8:23 AM, Fred wrote: > Are you using any declares ?? > > Fred > > Dave Parizek wrote: >> I converted an RB project (2006r2) from 1.x to 2.3. >> >> After compiling in the debugger I get this error as it starts to >> run (in the debugger): >> >> Application Launch Failure >> The Application "(null)" could not be launched because of a shared >> library error: "
<>" >> >> I removed all Valentina (V4CC and V4RB) items from my machine and >> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >> that error. >> >> I checked the mailing list archives under "Application Launch >> Failure" and "shared library error" but did not see anything >> relevant. >> >> Suggestions? >> >> --Dave >> >> _______________________________________________ >> Valentina-beta mailing list >> Valentina-beta@lists.macserve.net >> http://lists.macserve.net/mailman/listinfo/valentina-beta >> > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From sunshine at public.kherson.ua Fri Jun 2 19:16:32 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 2 11:16:32 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <65388A04-4AA8-47C6-941C-1D1E6CABD1B9@Parizek.com> Message-ID: On 6/2/06 6:29 PM, "Dave Parizek" wrote: >>> I checked the mailing list archives under "Application Launch >>> Failure" and "shared library error" but did not see anything >>> relevant. >>> >>> Suggestions? This happens only in your project ? V4RB Examples do work? Or If to make new empty project and do call only Valentina Init Compiles? -- 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 fitzbew at nc.rr.com Fri Jun 2 12:17:41 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Fri Jun 2 11:17:45 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <3A0F07F2-4D54-40D8-90CF-47F64401EBCD@Parizek.com> Message-ID: On 6/2/06 11:05 AM, "Dave Parizek" wrote: > Yes, under build settings I have checked the Mac OSX/Classic option. > I also tried classic only option, same result. and always the osx > only option unchecked. > > --dave > __________________________________________ > Dave Parizek > > > On Jun 2, 2006, at 7:30 AM, Russ Tyndall wrote: > >> On 6/2/06 10:23 AM, "Dave Parizek" wrote: >> >>> I converted an RB project (2006r2) from 1.x to 2.3. >>> >>> After compiling in the debugger I get this error as it starts to run >>> (in the debugger): >>> >>> Application Launch Failure >>> The Application "(null)" could not be launched because of a shared >>> library error: "
<>" >>> >>> I removed all Valentina (V4CC and V4RB) items from my machine and >>> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >>> that error. >>> >>> I checked the mailing list archives under "Application Launch >>> Failure" and "shared library error" but did not see anything >>> relevant. >>> >>> Suggestions? >>> >> >> I'm sure you've ruled this out, but just to be safe: Mach-o is not >> still >> checked in the RB build settings is it? Carbon/Classic only is >> checked? >> I downloaded the current V4RB b10 and installed it, and everything seems to be working fine on my machine (X.4.6, RB2006r2, PPC DP G4). These are my plugins: *CoreClasses (Einhugur) #TypeLib (Einhuger) CSV Plugin (RS) E-Cryptit Engine (Einhugur) Micono RbJPEGLib PictureEffects (Einhugur) SQLitePluginPro (You-know-who) StyleGrid (Einhugur) TimeControl (Einhugur) WindowSplitter (Einhugur) You may wish to consider a corruption issue, possibly even with your RB installation. (Your old Debug compiles are getting cleared ok out of the project directory? Has something been locked? Have you restarted the whole machine?) Can you step into the App.Open event? (Sorry for the reckless brainstorming.) Will Ruslan's sample projects compile and launch? Your CFMSupport folder is in the "right" place "/Library/" and you only have one of them? Is it possible your Macho vComponents folder and your carbon folder got mixed up? Russ From dave at Parizek.com Fri Jun 2 10:16:13 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 12:16:20 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: <325860D7-9DAA-41BB-965B-F8292DF08A42@Parizek.com> OK, Ruslan's example projects worked (although I can't find "Picture1" anywhere in the example folder, and at least a couple of the examples want it). So based on your question about declares I removed everything associated with HierPop contextual menu classes and now it works. HierPop worked well with V4RB 1.x but I guess not 2.x. Should I add that to the Wiki? Where is the best place to put info on potential conflicts? --Dave __________________________________________ Dave Parizek On Jun 2, 2006, at 9:17 AM, Russ Tyndall wrote: > On 6/2/06 11:05 AM, "Dave Parizek" wrote: > >> Yes, under build settings I have checked the Mac OSX/Classic option. >> I also tried classic only option, same result. and always the osx >> only option unchecked. >> >> --dave >> __________________________________________ >> Dave Parizek >> >> >> On Jun 2, 2006, at 7:30 AM, Russ Tyndall wrote: >> >>> On 6/2/06 10:23 AM, "Dave Parizek" wrote: >>> >>>> I converted an RB project (2006r2) from 1.x to 2.3. >>>> >>>> After compiling in the debugger I get this error as it starts to >>>> run >>>> (in the debugger): >>>> >>>> Application Launch Failure >>>> The Application "(null)" could not be launched because of a shared >>>> library error: "
<>" >>>> >>>> I removed all Valentina (V4CC and V4RB) items from my machine and >>>> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >>>> that error. >>>> >>>> I checked the mailing list archives under "Application Launch >>>> Failure" and "shared library error" but did not see anything >>>> relevant. >>>> >>>> Suggestions? >>>> >>> >>> I'm sure you've ruled this out, but just to be safe: Mach-o is not >>> still >>> checked in the RB build settings is it? Carbon/Classic only is >>> checked? >>> > > I downloaded the current V4RB b10 and installed it, and everything > seems to > be working fine on my machine (X.4.6, RB2006r2, PPC DP G4). > > These are my plugins: > *CoreClasses (Einhugur) > #TypeLib (Einhuger) > CSV Plugin (RS) > E-Cryptit Engine (Einhugur) > Micono RbJPEGLib > PictureEffects (Einhugur) > SQLitePluginPro (You-know-who) > StyleGrid (Einhugur) > TimeControl (Einhugur) > WindowSplitter (Einhugur) > > You may wish to consider a corruption issue, possibly even with > your RB > installation. (Your old Debug compiles are getting cleared ok out > of the > project directory? Has something been locked? Have you restarted > the whole > machine?) > > Can you step into the App.Open event? (Sorry for the reckless > brainstorming.) > > Will Ruslan's sample projects compile and launch? > > Your CFMSupport folder is in the "right" place "/Library/" and you > only have > one of them? Is it possible your Macho vComponents folder and your > carbon > folder got mixed up? > > Russ > > > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From sunshine at public.kherson.ua Fri Jun 2 20:44:26 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 2 12:44:35 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <325860D7-9DAA-41BB-965B-F8292DF08A42@Parizek.com> Message-ID: On 6/2/06 8:16 PM, "Dave Parizek" wrote: > OK, Ruslan's example projects worked (although I can't find > "Picture1" anywhere in the example folder, and at least a couple of > the examples want it). > > So based on your question about declares I removed everything > associated with HierPop contextual menu classes and now it works. > HierPop worked well with V4RB 1.x but I guess not 2.x. Should I add > that to the Wiki? Where is the best place to put info on potential > conflicts? Well, we have PUBLIC area in WIKI where everybody from Valentina developers can put any info other people can find useful. ------ Deal is that I do not see how they can conflict. Valentina 2 name all its classes from V all global constants also kV_ enums EV Where they can conflict...may be only if other plugin also use .e.g VTable class -- 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 dave at Parizek.com Fri Jun 2 10:50:17 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 12:50:24 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <325860D7-9DAA-41BB-965B-F8292DF08A42@Parizek.com> References: <325860D7-9DAA-41BB-965B-F8292DF08A42@Parizek.com> Message-ID: <3237CBC2-C081-415D-A931-32FEC15F7835@Parizek.com> Oh, and I apologize to the list for not trying the examples first... next time I will think of that... haven't been programming in awhile... Thanks for your help Russ! --Dave __________________________________________ Dave Parizek On Jun 2, 2006, at 10:16 AM, Dave Parizek wrote: > OK, Ruslan's example projects worked (although I can't find > "Picture1" anywhere in the example folder, and at least a couple of > the examples want it). > > So based on your question about declares I removed everything > associated with HierPop contextual menu classes and now it works. > HierPop worked well with V4RB 1.x but I guess not 2.x. Should I > add that to the Wiki? Where is the best place to put info on > potential conflicts? > > --Dave > __________________________________________ > Dave Parizek > > > On Jun 2, 2006, at 9:17 AM, Russ Tyndall wrote: > >> On 6/2/06 11:05 AM, "Dave Parizek" wrote: >> >>> Yes, under build settings I have checked the Mac OSX/Classic option. >>> I also tried classic only option, same result. and always the osx >>> only option unchecked. >>> >>> --dave >>> __________________________________________ >>> Dave Parizek >>> >>> >>> On Jun 2, 2006, at 7:30 AM, Russ Tyndall wrote: >>> >>>> On 6/2/06 10:23 AM, "Dave Parizek" wrote: >>>> >>>>> I converted an RB project (2006r2) from 1.x to 2.3. >>>>> >>>>> After compiling in the debugger I get this error as it starts >>>>> to run >>>>> (in the debugger): >>>>> >>>>> Application Launch Failure >>>>> The Application "(null)" could not be launched because of a shared >>>>> library error: "
<>" >>>>> >>>>> I removed all Valentina (V4CC and V4RB) items from my machine and >>>>> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >>>>> that error. >>>>> >>>>> I checked the mailing list archives under "Application Launch >>>>> Failure" and "shared library error" but did not see anything >>>>> relevant. >>>>> >>>>> Suggestions? >>>>> >>>> >>>> I'm sure you've ruled this out, but just to be safe: Mach-o is not >>>> still >>>> checked in the RB build settings is it? Carbon/Classic only is >>>> checked? >>>> >> >> I downloaded the current V4RB b10 and installed it, and everything >> seems to >> be working fine on my machine (X.4.6, RB2006r2, PPC DP G4). >> >> These are my plugins: >> *CoreClasses (Einhugur) >> #TypeLib (Einhuger) >> CSV Plugin (RS) >> E-Cryptit Engine (Einhugur) >> Micono RbJPEGLib >> PictureEffects (Einhugur) >> SQLitePluginPro (You-know-who) >> StyleGrid (Einhugur) >> TimeControl (Einhugur) >> WindowSplitter (Einhugur) >> >> You may wish to consider a corruption issue, possibly even with >> your RB >> installation. (Your old Debug compiles are getting cleared ok out >> of the >> project directory? Has something been locked? Have you restarted >> the whole >> machine?) >> >> Can you step into the App.Open event? (Sorry for the reckless >> brainstorming.) >> >> Will Ruslan's sample projects compile and launch? >> >> Your CFMSupport folder is in the "right" place "/Library/" and you >> only have >> one of them? Is it possible your Macho vComponents folder and >> your carbon >> folder got mixed up? >> >> Russ >> >> >> >> >> _______________________________________________ >> Valentina-beta mailing list >> Valentina-beta@lists.macserve.net >> http://lists.macserve.net/mailman/listinfo/valentina-beta > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From fitzbew at nc.rr.com Fri Jun 2 14:02:55 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Fri Jun 2 13:03:03 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <325860D7-9DAA-41BB-965B-F8292DF08A42@Parizek.com> Message-ID: On 6/2/06 1:16 PM, "Dave Parizek" wrote: > OK, Ruslan's example projects worked (although I can't find > "Picture1" anywhere in the example folder, and at least a couple of > the examples want it). > > So based on your question about declares I removed everything > associated with HierPop contextual menu classes and now it works. > HierPop worked well with V4RB 1.x but I guess not 2.x. Curious about this, I added a couple V4RB calls to the example HierPop project from wireframesoftware and had no problem compiling or running the app. Hierpop (hopefully we are talking about the same thing) is not a plugin, just a class you drag into your project. I'm stumped. You are on a macintel machine? My instinct is your problem has nothing to do with V4RB, but I'm sure I have already bored you enough with my input. Russ From fred.stephenson at communication-unltd.com Fri Jun 2 21:30:58 2006 From: fred.stephenson at communication-unltd.com (Fred) Date: Fri Jun 2 14:32:09 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <65388A04-4AA8-47C6-941C-1D1E6CABD1B9@Parizek.com> References: <1A286AA6-ACFB-4A7D-85AB-52E28B52A19F@Parizek.com> <4480580D.8060000@communication-unltd.com> <65388A04-4AA8-47C6-941C-1D1E6CABD1B9@Parizek.com> Message-ID: <448091F2.5070400@communication-unltd.com> Dave Parizek wrote: > I am using HierPop contextual menu extension which does some declares. > > I am using Einhugur's DataGrid and TreeView plugins, so they might do > some declares? > > --Dave I ran into the same compile message today when I was testing some declares with system events without using V4RB all the best Fred From dave at Parizek.com Fri Jun 2 13:36:42 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 15:36:48 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: I have 2 versions I am working with. One without HierPop runs perfect now. HierPop one does not. And that is the only difference between them. I also have a version with HierPop and V4RB 1.x that works perfect. Maybe I need a newer version of HierPop... Yes, i am on a macintel macbook pro. I may not worry about it. If Ruslan says HierPop should not be causing the problem -- then I would like to understand it better -- but for now I was planning on stopping using HierPop anyway. --Dave __________________________________________ Dave Parizek On Jun 2, 2006, at 11:02 AM, Russ Tyndall wrote: > On 6/2/06 1:16 PM, "Dave Parizek" wrote: > >> OK, Ruslan's example projects worked (although I can't find >> "Picture1" anywhere in the example folder, and at least a couple of >> the examples want it). >> >> So based on your question about declares I removed everything >> associated with HierPop contextual menu classes and now it works. >> HierPop worked well with V4RB 1.x but I guess not 2.x. > > Curious about this, I added a couple V4RB calls to the example HierPop > project from wireframesoftware and had no problem compiling or > running the > app. > > Hierpop (hopefully we are talking about the same thing) is not a > plugin, > just a class you drag into your project. I'm stumped. You are on a > macintel machine? > > My instinct is your problem has nothing to do with V4RB, but I'm > sure I have > already bored you enough with my input. > > Russ > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From dave at Parizek.com Fri Jun 2 13:42:26 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 15:42:31 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: <22899E1D-8C8D-4F36-A6D2-D86B7239A5B4@Parizek.com> Yes, it is just a class. But it does do declares. I can send you a copy of my project if you want to play with it. I am sort of embarrassed though to have someone else see my code - I am not the most skilled programmer by an incredibly long shot... But now that I have the project working without HierPop it is not critical to me to fix the matter - more of just a learning thing now. --Dave __________________________________________ Dave Parizek On Jun 2, 2006, at 11:02 AM, Russ Tyndall wrote: > > Curious about this, I added a couple V4RB calls to the example HierPop > project from wireframesoftware and had no problem compiling or > running the > app. > > Hierpop (hopefully we are talking about the same thing) is not a > plugin, > just a class you drag into your project. I'm stumped. You are on a > macintel machine? > > My instinct is your problem has nothing to do with V4RB, but I'm > sure I have > already bored you enough with my input. > > Russ > > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From steve at steve-albin.com Fri Jun 2 16:49:26 2006 From: steve at steve-albin.com (Steve Albin) Date: Fri Jun 2 15:49:30 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <3A0F07F2-4D54-40D8-90CF-47F64401EBCD@Parizek.com> Message-ID: FWIW, I get this error when I have the Mac OSX/Classic option or Classic only option checked. The only way I can make a debug build is have ONLY the OSX only option checked. I don't plan on supporting Classic, so I never worried about this before, but maybe I'm headed for trouble. I'll admit I don't have a firm grasp of the difference between Mach-o, carbon and all that jazz. But for now, everything works for me..... I have no trouble building for Windows either. Using V4RB 2.4 b7 and RB 2006r2. Steve -- Steve Albin - Montclair, NJ http://www.jazzdiscography.com/ http://www.steve-albin.com/ On Friday, June 2, 2006 dave@Parizek.com (Dave Parizek) wrote: >Yes, under build settings I have checked the Mac OSX/Classic option. >I also tried classic only option, same result. and always the osx >only option unchecked. > >--dave >__________________________________________ >Dave Parizek > > >On Jun 2, 2006, at 7:30 AM, Russ Tyndall wrote: > >> On 6/2/06 10:23 AM, "Dave Parizek" wrote: >> >>> I converted an RB project (2006r2) from 1.x to 2.3. >>> >>> After compiling in the debugger I get this error as it starts to run >>> (in the debugger): >>> >>> Application Launch Failure >>> The Application "(null)" could not be launched because of a shared >>> library error: "
<>" >>> >>> I removed all Valentina (V4CC and V4RB) items from my machine and >>> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >>> that error. >>> >>> I checked the mailing list archives under "Application Launch >>> Failure" and "shared library error" but did not see anything >>> relevant. >>> >>> Suggestions? >>> >> >> I'm sure you've ruled this out, but just to be safe: Mach-o is not >> still >> checked in the RB build settings is it? Carbon/Classic only is >> checked? >> >> Russ >> >> _______________________________________________ >> Valentina-beta mailing list >> Valentina-beta@lists.macserve.net >> http://lists.macserve.net/mailman/listinfo/valentina-beta > >_______________________________________________ >Valentina-beta mailing list >Valentina-beta@lists.macserve.net >http://lists.macserve.net/mailman/listinfo/valentina-beta From dave at Parizek.com Fri Jun 2 14:00:38 2006 From: dave at Parizek.com (Dave Parizek) Date: Fri Jun 2 16:00:45 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: References: Message-ID: <23ADDFE4-6797-4F9C-9A3F-A67AFB65B96C@Parizek.com> That fixed it. Using V4RB 2.3 and the Carbon (NOT Mach-O) V4RB plugin: The project without HierPop works regardless of build settings. The project with HierPop works ONLY if set to Mac OS X only in the build settings. THANKS! --Dave __________________________________________ Dave Parizek On Jun 2, 2006, at 1:49 PM, Steve Albin wrote: > FWIW, I get this error when I have the Mac OSX/Classic option or > Classic only option checked. The only way I can make a debug build > is have ONLY the OSX only option checked. I don't plan on > supporting Classic, so I never worried about this before, but maybe > I'm headed for trouble. I'll admit I don't have a firm grasp of > the difference between Mach-o, carbon and all that jazz. But for > now, everything works for me..... I have no trouble building for > Windows either. Using V4RB 2.4 b7 and RB 2006r2. > > Steve > > -- > Steve Albin - Montclair, NJ > http://www.jazzdiscography.com/ > http://www.steve-albin.com/ > > > On Friday, June 2, 2006 dave@Parizek.com (Dave Parizek) wrote: > >> Yes, under build settings I have checked the Mac OSX/Classic option. >> I also tried classic only option, same result. and always the osx >> only option unchecked. >> >> --dave >> __________________________________________ >> Dave Parizek >> >> >> On Jun 2, 2006, at 7:30 AM, Russ Tyndall wrote: >> >>> On 6/2/06 10:23 AM, "Dave Parizek" wrote: >>> >>>> I converted an RB project (2006r2) from 1.x to 2.3. >>>> >>>> After compiling in the debugger I get this error as it starts to >>>> run >>>> (in the debugger): >>>> >>>> Application Launch Failure >>>> The Application "(null)" could not be launched because of a shared >>>> library error: "
<>" >>>> >>>> I removed all Valentina (V4CC and V4RB) items from my machine and >>>> reinstalled V4RB Carbon (no Mach-O)(latest Beta), but I still get >>>> that error. >>>> >>>> I checked the mailing list archives under "Application Launch >>>> Failure" and "shared library error" but did not see anything >>>> relevant. >>>> >>>> Suggestions? >>>> >>> >>> I'm sure you've ruled this out, but just to be safe: Mach-o is not >>> still >>> checked in the RB build settings is it? Carbon/Classic only is >>> checked? >>> >>> Russ >>> >>> _______________________________________________ >>> Valentina-beta mailing list >>> Valentina-beta@lists.macserve.net >>> http://lists.macserve.net/mailman/listinfo/valentina-beta >> >> _______________________________________________ >> Valentina-beta mailing list >> Valentina-beta@lists.macserve.net >> http://lists.macserve.net/mailman/listinfo/valentina-beta > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta From fitzbew at nc.rr.com Fri Jun 2 17:09:19 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Fri Jun 2 16:09:25 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <22899E1D-8C8D-4F36-A6D2-D86B7239A5B4@Parizek.com> Message-ID: On 6/2/06 4:42 PM, "Dave Parizek" wrote: > Yes, it is just a class. But it does do declares. I can send you a > copy of my project if you want to play with it. I am sort of > embarrassed though to have someone else see my code - I am not the > most skilled programmer by an incredibly long shot... > LOL! I thought I was the only one that ever felt that way! Seriously, all I can do is see if I can duplicate the problem on my PPC machine. I'd be glad to try, if you want to email it [directly] to me. > But now that I have the project working without HierPop it is not > critical to me to fix the matter - more of just a learning thing now. Understood. -- Russ Tyndall Wake Forest, NC From sunshine at public.kherson.ua Sat Jun 3 08:30:12 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat Jun 3 00:30:20 2006 Subject: error switching from V4RB 1.x to 2.3 In-Reply-To: <23ADDFE4-6797-4F9C-9A3F-A67AFB65B96C@Parizek.com> Message-ID: On 6/3/06 12:00 AM, "Dave Parizek" wrote: > That fixed it. > > Using V4RB 2.3 and the Carbon (NOT Mach-O) V4RB plugin: > > The project without HierPop works regardless of build settings. > > The project with HierPop works ONLY if set to Mac OS X only in the > build settings. Dave, May be you need contact HierPop developer with ask find what do conflict ? -- 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 Jun 4 09:56:09 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Jun 4 01:56:18 2006 Subject: Russ, your slow query In-Reply-To: Message-ID: On 6/2/06 4:20 PM, "Russ Tyndall" wrote: Hi Russ, I have locate source of problem. We have no yet optimization for case when joins present WHERE have at least 2 conditions from one table and at least one condition from other join table In this case Valentina go todo SCAN instead of indexed searches. I will add this optimization. May be today. -- 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 fitzbew at nc.rr.com Sun Jun 4 04:03:11 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Sun Jun 4 03:03:19 2006 Subject: Russ, your slow query In-Reply-To: Message-ID: On 6/4/06 2:56 AM, "Ruslan Zasukhin" wrote: > On 6/2/06 4:20 PM, "Russ Tyndall" wrote: > > Hi Russ, > > I have locate source of problem. > > We have no yet optimization for case when > joins present > WHERE have at least 2 conditions from one table > and at least one condition from other join table > > In this case Valentina go todo SCAN instead of indexed searches. > > I will add this optimization. May be today. > Great news! I look forward to trying it out. -- Russ Tyndall Wake Forest, NC From jda at his.com Sun Jun 4 08:54:25 2006 From: jda at his.com (jda) Date: Sun Jun 4 07:54:49 2006 Subject: Mantis In-Reply-To: References: Message-ID: Hi Ruslan, For some reason Mantis no longer alerts me when an issue I reported has changed. I have found a few cases where notes were added asking questions, and I wasn't notified. In case this is happening for you, too, please note that I added information at your request to bug http://valentina-db.com/bt/view.php?id=1620 Thanks, Jon From sunshine at public.kherson.ua Sun Jun 4 15:58:12 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Jun 4 07:58:17 2006 Subject: Mantis In-Reply-To: Message-ID: On 6/4/06 3:54 PM, "jda" wrote: Hi Jon, > For some reason Mantis no longer alerts me when an issue I reported > has changed. I have found a few cases where notes were added asking > questions, and I wasn't notified. > > In case this is happening for you, too, please note that I added > information at your request to bug > > http://valentina-db.com/bt/view.php?id=1620 I see your note in my list of letters from 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 snw at paradise.net.nz Mon Jun 5 09:06:31 2006 From: snw at paradise.net.nz (Sean Wilson) Date: Sun Jun 4 16:06:37 2006 Subject: Mantis In-Reply-To: References: Message-ID: <7.0.1.0.2.20060605090514.03cc1ea8@paradise.net.nz> >For some reason Mantis no longer alerts me when an issue I reported >has changed. I have found a few cases where notes were added asking >questions, and I wasn't notified. I have asked/expected it to mail me a couple of times in the last few days too and heard nothing. Perhaps there's a disconnect between Mantis and its email interface? Cheers, -Sean. From sunshine at public.kherson.ua Mon Jun 5 09:03:40 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Jun 5 01:03:51 2006 Subject: Mantis In-Reply-To: <7.0.1.0.2.20060605090514.03cc1ea8@paradise.net.nz> Message-ID: On 6/5/06 12:06 AM, "Sean Wilson" wrote: >> For some reason Mantis no longer alerts me when an issue I reported >> has changed. I have found a few cases where notes were added asking >> questions, and I wasn't notified. > > I have asked/expected it to mail me a couple of times in the last few > days too and heard nothing. Perhaps there's a disconnect between > Mantis and its email interface? For me it works, and I know for other people also May be its letters are filtered as spam on your side? -- 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 ernestogiannotta at tiscalinet.it Mon Jun 5 18:29:14 2006 From: ernestogiannotta at tiscalinet.it (erne) Date: Mon Jun 5 11:29:20 2006 Subject: Unicode Path Bite Message-ID: Hello Ruslan, I've just released my first app with Vale 2 Mach-O and already got the first report of the app not starting when path has unicode characters Please, I know to be bothering but this one is really urgent Cool Runnings, Erne. -- | e r | Ernesto Giannotta | n e | Musical Box - a media store From sunshine at public.kherson.ua Mon Jun 5 19:48:05 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Jun 5 11:48:11 2006 Subject: Unicode Path Bite --> Valentina Light - all in one. In-Reply-To: Message-ID: On 6/5/06 7:29 PM, "erne" wrote: > Hello Ruslan, > > I've just released my first app with Vale 2 Mach-O > > and already got the first report of the app not starting when path has > unicode characters > > Please, I know to be bothering > but this one is really urgent Problem comes from ICU. Apple says that it should work, but I do not see this so far. --------------------------- To inform all you guys. Shaun Wexler - cool MAC OS X / C++ / Cocoa / and so on developer Have join us in development of Valentina engine. * Shaun already have got cool success in building of Valentina engine as SINGLE DLL and/or LIB. * no more ICU inside of Valentina. Instead Shaun have found way to use OS X ICU. So now we know that it is really possible to make "Valentina All In One". ---------- I very hope that THIS month we will be able introduce new product line "Valentina Light" Only I have no idea so far about V4MD and V4RB. Deal is that we have got this success using xcode and GCC. V4MD and V4RB still are on CodeWarrior. * Director is in deep stagnation. So far there is no any info, if and when Adobe will produce new version of Director. * When REALbasic will become universal, we switch to xcode also. I.e., so far only C++/Cococa/Revolution developers will get benefit. MAY BE I will find way to do trick with CW also. I see one way. MAY BE it will take 2-3 days only... -- 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 ernestogiannotta at tiscalinet.it Mon Jun 5 19:25:20 2006 From: ernestogiannotta at tiscalinet.it (erne) Date: Mon Jun 5 12:25:28 2006 Subject: Unicode Path Bite --> Valentina Light - all in one. In-Reply-To: Message-ID: Thank you Ruslan, sincerely, it's a real pleasure to deal with such a responsive and gentle person. Yes, it would be real cool to get rid of the ICU heavy load especially for me selling just small utilities it's difficult to justufy 12mb downloads And yes, you got my vote (FWIW :-) for putting this issue on high priority on 5-06-2006 18:48, Ruslan Zasukhin at sunshine@public.kherson.ua wrote: > On 6/5/06 7:29 PM, "erne" wrote: > >> Hello Ruslan, >> >> I've just released my first app with Vale 2 Mach-O >> >> and already got the first report of the app not starting when path has >> unicode characters >> >> Please, I know to be bothering >> but this one is really urgent > > Problem comes from ICU. > Apple says that it should work, but I do not see this so far. > > > --------------------------- > To inform all you guys. > > Shaun Wexler - cool MAC OS X / C++ / Cocoa / and so on developer > Have join us in development of Valentina engine. > > * Shaun already have got cool success in building of Valentina engine as > SINGLE DLL and/or LIB. > > * no more ICU inside of Valentina. Instead Shaun have found way > to use OS X ICU. > > So now we know that it is really possible to make "Valentina All In One". > > ---------- > I very hope that THIS month we will be able introduce new product line > "Valentina Light" > > Only I have no idea so far about V4MD and V4RB. Deal is that we have got > this success using xcode and GCC. > > V4MD and V4RB still are on CodeWarrior. > > * Director is in deep stagnation. So far there is no any info, if and when > Adobe will produce new version of Director. > > * When REALbasic will become universal, we switch to xcode also. > > I.e., so far only C++/Cococa/Revolution developers will get benefit. > > MAY BE I will find way to do trick with CW also. I see one way. > MAY BE it will take 2-3 days only... > Cool Runnings, Erne. -- / |\ |\ | / ? |/ | \ | ? \ |\ | \| \ From sunshine at public.kherson.ua Wed Jun 7 10:36:23 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Jun 7 02:36:30 2006 Subject: [REFAQ C++] Renamed VSQL::Database => VSQL::SqlDatabase Message-ID: The same for interface VSQL::I_Database => VSQL::I_SqlDatabase -- 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 Jun 7 10:39:22 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Jun 7 02:39:30 2006 Subject: [REFAC C++] New files Message-ID: publ/Interfaces/FBL_I_Unknown.cpp publ/Interfaces/FBL_Public_IIDs.h prot/Interfaces/FBL_Protected_IIDs.h priv//Interfaces/FBL_Private_IIDs.h -- 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 Jun 7 10:40:29 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Jun 7 02:40:36 2006 Subject: [REFAC C++] Renamed class/file FBL_AspectHost => AspectHost_imp Message-ID: -- 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 Jun 7 10:46:28 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Jun 7 02:46:36 2006 Subject: [NEW C++] I_Unknown::QueryInterfaces() + macros Message-ID: Now we have full-featured QueryInterface() method, like in COM. QueryInterface() this is analog of dynamic_cast<>() Example: I_FldObjectPtr pObjectPtr = ... // now you want work with ObjectPtr as with Link. // old way - and it still works: I_Link_Ptr pLink = fbl_dynamic_cast( pObjectPtr ); // new way I_Link_Ptr pLink = (I_Link*) pObjectPtr->QueryInterface( IID_I_Link ); // or with help of macro: I_Link_Ptr pLink = QUERY_INTERFACE( pObjectPtr, I_Link ); // or even more short macro: I_Link_Ptr pLink = QI( pObjectPtr, I_Link ); -- 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 Jun 7 12:35:20 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Jun 7 04:35:25 2006 Subject: [REFAC C++] FBL_I_Aspects.h file splited on 2 files I_Aspect.h and I_AspectHost Message-ID: FBL_I_Aspects.h file splited on 2 files I_Aspect.h and I_AspectHost -- 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 Wed Jun 7 11:43:14 2006 From: jda at his.com (jda) Date: Wed Jun 7 10:43:20 2006 Subject: Error codes In-Reply-To: References: Message-ID: Hi Ruslan, A user is reporting an error # 301 (v1.x engine). I can't find error codes anywere in the Valentina docs (I swear they used to be there, the kernal docs I think). Anyway, can you tell me what this means? Thanks, Jon From sunshine at public.kherson.ua Wed Jun 7 19:35:50 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed Jun 7 11:35:56 2006 Subject: Error codes In-Reply-To: Message-ID: On 6/7/06 6:43 PM, "jda" wrote: Hi Jon, > Hi Ruslan, > > A user is reporting an error # 301 (v1.x engine). I can't find error > codes anywere in the Valentina docs (I swear they used to be there, > the kernal docs I think). Anyway, can you tell me what this means? kFBL_DataFileError = 301, This is corruption. -- 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 Jun 8 11:52:50 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 8 03:52:57 2006 Subject: Attention To All C++ developers. In-Reply-To: Message-ID: On 6/6/06 6:14 PM, "Matthew Jew" wrote: Hi Mathew, Hi All C++ Developers. I'd like to invite all C++ developers that use Valentina SDK to join IRC chat of our team. You can use this chat to discuss any C++ related issues. If you want, please contact me off list, I will send you info how to connect to our IRC channel. -- Mathew, I am starting check your project. -- 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 Jun 8 21:35:37 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 8 13:35:44 2006 Subject: vClient Secure Connections In-Reply-To: <000001c68b1a$53d9a1a0$9601a8c0@Brandon> Message-ID: On 6/8/06 7:41 PM, "Brandon" wrote: Hi Brandon, > Ruslan, > > We are building a client server application with V4RB. It must use a > secure connection. > > Valentina does not use a secure connection (ssl or something else), > right? Right. > Will this be added in the future? I think not a problem. Please add feature request into our Mantis. I think we should be able implement this in July - Aug. Also we consider integration with Rendezvous. Anybody think this is good idea? > Thank You, > > Brandon Warlick -- 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 yeomans at desuetude.com Thu Jun 8 16:47:51 2006 From: yeomans at desuetude.com (Charles Yeomans) Date: Thu Jun 8 15:47:57 2006 Subject: vClient Secure Connections In-Reply-To: References: Message-ID: <3C34B6C2-C299-4C9B-92E0-3D79BE06D701@desuetude.com> On Jun 8, 2006, at 2:35 PM, Ruslan Zasukhin wrote: > On 6/8/06 7:41 PM, "Brandon" wrote: > > Hi Brandon, > >> Ruslan, >> >> We are building a client server application with V4RB. It must use a >> secure connection. >> >> Valentina does not use a secure connection (ssl or something else), >> right? > > Right. > >> Will this be added in the future? > > I think not a problem. > Please add feature request into our Mantis. > > I think we should be able implement this in July - Aug. > > Also we consider integration with Rendezvous. > Anybody think this is good idea? This would be great for me. Charles Yeomans From dave at Parizek.com Thu Jun 8 14:17:32 2006 From: dave at Parizek.com (Dave Parizek) Date: Thu Jun 8 16:17:36 2006 Subject: vClient Secure Connections In-Reply-To: <3C34B6C2-C299-4C9B-92E0-3D79BE06D701@desuetude.com> References: <3C34B6C2-C299-4C9B-92E0-3D79BE06D701@desuetude.com> Message-ID: <264D9B6E-47B4-45D3-96B7-41A20F26CEA9@Parizek.com> Rendezvous - me too Yes. --dave __________________________________________ Dave Parizek >> >> Also we consider integration with Rendezvous. >> Anybody think this is good idea? > > This would be great for me. > > Charles Yeomans > _______________________________________________ > From dev at macfoh.com Thu Jun 8 14:31:30 2006 From: dev at macfoh.com (Shaun Wexler) Date: Thu Jun 8 16:31:35 2006 Subject: vClient Secure Connections In-Reply-To: <3C34B6C2-C299-4C9B-92E0-3D79BE06D701@desuetude.com> References: <3C34B6C2-C299-4C9B-92E0-3D79BE06D701@desuetude.com> Message-ID: On Jun 8, 2006, at 1:47 PM, Charles Yeomans wrote: > On Jun 8, 2006, at 2:35 PM, Ruslan Zasukhin wrote: > >> On 6/8/06 7:41 PM, "Brandon" wrote: >> >>> We are building a client server application with V4RB. It must use a >>> secure connection. >>> >>> Valentina does not use a secure connection (ssl or something else), >>> right? >> >> Right. >> >>> Will this be added in the future? >> >> I think not a problem. >> Please add feature request into our Mantis. >> >> I think we should be able implement this in July - Aug. >> >> Also we consider integration with Rendezvous. >> Anybody think this is good idea? > > This would be great for me. I already wrote classes to manage client/server with support for Bonjour (Rendezvous), and they use CFSocket and CFStream, which both provide secure communications via SSL. I will work with Ruslan to make them compatible... -- Shaun Wexler MacFOH http://www.macfoh.com From sunshine at public.kherson.ua Fri Jun 9 00:44:24 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 9 02:11:02 2006 Subject: vClient Secure Connections In-Reply-To: Message-ID: On 6/9/06 12:31 AM, "Shaun Wexler" wrote: >> This would be great for me. > > I already wrote classes to manage client/server with support for > Bonjour (Rendezvous), and they use CFSocket and CFStream, which both > provide secure communications via SSL. I will work with Ruslan to > make them compatible... Great, Shaun! Please always keep in mind, we need solution on C++ level that work on all platforms, not only OS X. -- 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 Jun 11 16:51:39 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Jun 11 08:51:45 2006 Subject: ObjectiveC++: expected unqualified-id before '@' token In-Reply-To: <4CE9C69D-43EC-43DA-8DBF-58A747D67E59@econophone.ch> Message-ID: On 6/11/06 4:16 PM, "Philip M?tteli" wrote: Hi Philip, > I also saw, that there's a VSDK in framework style distributed with > V4CC. But am I right and this framework doesn't contain client > functions, e. g. I_Server? It contains, as any and all our ADKs do > Will there be or is there already a complete VSDK in framework style > somewhere? C++ VSDK as framework ? Well, I do not see any serious advantage of VSDK.Framework Vs VComponents folder + Headers. Btw, Shaun have got working ala-Valentina Light. So now we can make single dll all-in-one. And Shaun in his project have made VX.Framework. I have not think yet deeply, but may be VSDK Lite will be in the form of framework. I think next week we will start play with projects of ADK Lite. -- 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 Jun 12 19:18:23 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Jun 12 11:18:37 2006 Subject: ObjectiveC++: expected unqualified-id before '@' token In-Reply-To: <5239ACD5-6B15-4A4A-9F87-E82AA3960FD4@econophone.ch> Message-ID: On 6/12/06 2:38 PM, "Philip M?tteli" wrote: Hi Philip, > I have prepared a reduced test project: > . > I have the VSDK installed on the default location ? my home > directory. You should change the build variable HEADER_SEARCH_PATHS > to what you have on your system. > It's already interesting, that these dozens of warnings disappeared. > That's probably just a compiler flag. I didn't interrogate that more, > because the party-pooper itself persisted: > > /tmp/Test/PMVDBMS.h:13: error: expected unqualified-id before '@' token Okay, I have found reason. I have comment code as #import @interface PMVServerProcess : PMPDBMS { // VSQL::I_Server *server; } And it still produce error. If comment VSDK2.h then it compiles. Bingo. Problem in word interface. Ww in our FBL interfaces have made #define interface struct This make crazy the OBJC compiller To fix simply do #import #undef interface -- 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 bwillius at gmx.de Tue Jun 13 15:41:19 2006 From: bwillius at gmx.de (Beatrix Willius) Date: Tue Jun 13 08:41:26 2006 Subject: Unicode Path Bite --> Valentina Light - all in one. Message-ID: <20060613134119.304170@gmx.net> Hello Ruslan, you can count me as being very interested in this. As a user of V4RB I would like to see this feature sooner rather than later. I have to use a 56 k modem, which makes each upload of my application a pain. Ruslan wrote: >Only I have no idea so far about V4MD and V4RB. Deal is that we have got >this success using xcode and GCC. > >V4MD and V4RB still are on CodeWarrior. > >* Director is in deep stagnation. So far there is no any info, if and when >Adobe will produce new version of Director. > >* When REALbasic will become universal, we switch to xcode also. > >I.e., so far only C++/Cococa/Revolution developers will get benefit. > >MAY BE I will find way to do trick with CW also. I see one way. >MAY BE it will take 2-3 days only... Regards Trixi Willius -- Echte DSL-Flatrate dauerhaft f?r 0,- Euro*! "Feel free" mit GMX DSL! http://www.gmx.net/de/go/dsl From sunshine at public.kherson.ua Tue Jun 13 18:29:07 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 13 10:29:12 2006 Subject: Unicode Path Bite --> Valentina Light - all in one. In-Reply-To: <20060613134119.304170@gmx.net> Message-ID: On 6/13/06 4:41 PM, "Beatrix Willius" wrote: Hi Beatrix, > you can count me as being very interested in this. As a user of V4RB I would > like to see this feature sooner rather than later. I have to use a 56 k modem, > which makes each upload of my application a pain. Yes this is very interesting, because allow for MAC OS X still use ICU in full format. On Windows "Valentina Lite" - should work without ICU and therefore loose some functionality. -- 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 Jun 13 20:47:10 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 13 12:47:16 2006 Subject: [ANN] V4RB mac 2.4b12 uploaded Message-ID: Hi All http://paradigmasoft.com/en/products/download/beta **** ATTENTION !!! ***** 1) This beta include new Vclient which is not compatible to current Valentina Server. New Valentina Server expect in nearest days so, if you work with VServer DO NOT use this build !! 2) This build change format of database to 2.4 version. converted are Boolean fields to use really one bit on disk. There was mistake, so in 2.0 - 2.3 boolean field have use 1 byte --------------------- [!] - Added optimization of queries with Joins. Russ, please check your db. Should be fine 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 joakim at astrocalc.com Tue Jun 13 22:26:16 2006 From: joakim at astrocalc.com (Joakim Schramm) Date: Tue Jun 13 15:26:17 2006 Subject: Unicode Path Bite --> Valentina Light - all in one. In-Reply-To: Message-ID: <20060613201924.6BAF881A8A3@merc.astrocalc.net> Ruslan, What about COM? I wouldn't mind to get rid of some of that 8 Mb's as well :-) Regards, Joakim > -----Original Message----- > From: valentina-beta-bounces@lists.macserve.net > [mailto:valentina-beta-bounces@lists.macserve.net] On Behalf > Of Ruslan Zasukhin > Sent: 13 June 2006 17:29 > To: valentina-beta@lists.macserve.net > Subject: Re: Unicode Path Bite --> Valentina Light - all in one. > > On 6/13/06 4:41 PM, "Beatrix Willius" wrote: > > Hi Beatrix, > > > you can count me as being very interested in this. As a > user of V4RB I > > would like to see this feature sooner rather than later. I > have to use > > a 56 k modem, which makes each upload of my application a pain. > > Yes this is very interesting, because allow for MAC OS X > still use ICU in full format. > > On Windows "Valentina Lite" - should work without ICU and > therefore loose some functionality. > > -- > 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 Tue Jun 13 23:30:28 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 13 15:30:33 2006 Subject: Unicode Path Bite --> Valentina Light - all in one. In-Reply-To: <20060613201924.6BAF881A8A3@merc.astrocalc.net> Message-ID: On 6/13/06 11:26 PM, "Joakim Schramm" wrote: > Ruslan, > > What about COM? I wouldn't mind to get rid of some of that 8 Mb's as well > :-) If we get success then all ADKs will get Lite Edition -- 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 Wed Jun 14 18:34:39 2006 From: jda at his.com (jda) Date: Wed Jun 14 17:34:52 2006 Subject: [ANN] V4RB mac 2.4b12 uploaded In-Reply-To: References: Message-ID: >Hi All > > http://paradigmasoft.com/en/products/download/beta > Hi Ruslan, I just installed b12, and I'm getting all sorts of "record locked" errors when trying to read databases. Even database *created* with b12. You have my project -- please run it with b12. Jon From macsforever2000 at goodeast.com Wed Jun 14 18:02:09 2006 From: macsforever2000 at goodeast.com (Frank Schima) Date: Wed Jun 14 19:02:24 2006 Subject: [ANN] V4RB mac 2.4b12 uploaded In-Reply-To: References: Message-ID: On Jun 14, 2006, at 4:34 PM, jda wrote: >> Hi All >> >> http://paradigmasoft.com/en/products/download/beta > > I just installed b12, and I'm getting all sorts of "record locked" > errors when trying to read databases. Even database *created* with > b12. You have my project -- please run it with b12. Same problems for me, but I am importing data. This all worked fine in b10. Best regards, Frank Schima From sunshine at public.kherson.ua Thu Jun 15 12:37:28 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 15 04:37:34 2006 Subject: [ANN] V4RB mac 2.4b12 uploaded In-Reply-To: Message-ID: On 6/15/06 3:02 AM, "Frank Schima" wrote: >>> http://paradigmasoft.com/en/products/download/beta >> >> I just installed b12, and I'm getting all sorts of "record locked" >> errors when trying to read databases. Even database *created* with >> b12. You have my project -- please run it with b12. > > Same problems for me, but I am importing data. This all worked fine > in b10. I see we have catch leaked objects. Probably problem comes from this...debugging. BTW, Frank, I have found reason of your problem Boolean + Vstudio mac Reason is that in 2.x Boolean did work as byte :( In 2.4 b12 we fix this. In Vstudio MAC -- GCC consider boolean as 4 bytes instead of 1 as CW do. This was reason of wrong 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 sunshine at public.kherson.ua Fri Jun 16 00:05:05 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 15 16:05:17 2006 Subject: [ANN] V4RB mac+win 2.4 b13 uploaded Message-ID: Hi All, * Fixed leaks introduced in b13, I think this should fix problem with locks. * fixed bug in UPDATE WHERE .. for 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 jda at his.com Thu Jun 15 18:24:18 2006 From: jda at his.com (jda) Date: Thu Jun 15 17:24:32 2006 Subject: [ANN] V4RB mac+win 2.4 b13 uploaded In-Reply-To: References: Message-ID: >Hi All, > >* Fixed leaks introduced in b13, > I think this should fix problem with locks. > Hi Ruslan, The record locked bug seems to be fixed. But I can't open db's made with b12 with an app compiled with b11. Is this expected behavior? Jon From sunshine at public.kherson.ua Fri Jun 16 08:57:32 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 16 00:57:45 2006 Subject: [ANN] V4RB mac+win 2.4 b13 uploaded In-Reply-To: Message-ID: On 6/16/06 1:24 AM, "jda" wrote: >> Hi All, >> >> * Fixed leaks introduced in b13, >> I think this should fix problem with locks. >> > > Hi Ruslan, > > The record locked bug seems to be fixed. Good. > But I can't open db's made with b12 with an app compiled with b11. Is > this expected behavior? Yes, read below point 2) **** ATTENTION !!! ***** 1) This beta include new Vclient which is not compatible to current Valentina Server. New Valentina Server expect in nearest days so, if you work with VServer DO NOT use this build !! 2) This build change format of database to 2.4 version. converted are Boolean fields to use really one bit on disk. There was mistake, so in 2.0 - 2.3 boolean field have use 1 byte --------------------- [!] - Added optimization of queries with Joins. Russ, please check your db. Should be fine 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 jda at his.com Fri Jun 16 06:58:50 2006 From: jda at his.com (jda) Date: Fri Jun 16 05:59:00 2006 Subject: [ANN] V4RB mac+win 2.4 b13 uploaded In-Reply-To: References: Message-ID: > > But I can't open db's made with b12 with an app compiled with b11. Is >> this expected behavior? > >Yes, read below point 2) > > >2) This build change format of database to 2.4 version. > converted are Boolean fields to use really one bit on disk. > There was mistake, so in 2.0 - 2.3 boolean field have use 1 byte > Hi Ruslan, Loss of backward compatibility is a big price for me to pay for something of no benefit to my app. Are you planning on introducing the new indexing scheme for text and varchar in 2.4? I ask, because if not I'll wait until you do (or add something else I could really use) to upgrade. Thanks, Jon From sunshine at public.kherson.ua Fri Jun 16 14:01:34 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 16 06:01:39 2006 Subject: [ANN] V4RB mac+win 2.4 b13 uploaded In-Reply-To: Message-ID: On 6/16/06 1:58 PM, "jda" wrote: > Loss of backward compatibility is a big price for me to pay for > something of no benefit to my app. Are you planning on introducing > the new indexing scheme for text and varchar in 2.4? I ask, because > if not I'll wait until you do (or add something else I could really > use) to upgrade. Not in 2.4 ... No time and issue is big. -- 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 Fri Jun 16 07:08:50 2006 From: jda at his.com (jda) Date: Fri Jun 16 06:09:16 2006 Subject: Problem with RB 2006r3 beta 3 In-Reply-To: References: Message-ID: Hi Ruslan, Would you please get the latest RB beta...when I try to compile anything with it (and V4RB is in the plugin folder) I get this error message: An error has occurred while compiling this project Message: End quote missing. File: V4RB Component: Valentina Item: ImportTextAnalyse Jon From sunshine at public.kherson.ua Fri Jun 16 14:26:55 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 16 06:26:58 2006 Subject: Problem with RB 2006r3 beta 3 In-Reply-To: Message-ID: > Would you please get the latest RB beta...when I try to compile > anything with it (and V4RB is in the plugin folder) I get this error > message: > > An error has occurred while compiling this project > > Message: End quote missing. > > File: V4RB > > Component: Valentina > > Item: ImportTextAnalyse Hmm. ImportTextAnalyse( File as FolderItem, FieldDelimeter as String = \",\", <<<<<<<<<<< LineDelimeter as String = \"\", inEncoding as String = \"UTF-16\", inHasColumnHeader as Boolean = false) as Integer May be they do not like this come inside of quotes.. But it have work so long ... It is not clear: if this is a new bug in RB 2006 b3 - and REAL should fix it? or this is fix of OLD bug of RB - and we need fix our code ? -- 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 Fri Jun 23 07:56:57 2006 From: jda at his.com (jda) Date: Fri Jun 23 06:57:11 2006 Subject: MACH-O Message-ID: Hi Ruslan, I'm starting to move my app to MACH-O. I'm using the latest 2.4 beta, and there is no MACH-O build. Is it safe to mix the 2.4 beta Carbon VComponents with the older 2.3 MACH-O plug-in and components? Or do we need a MACH-O build of 2.4? Jon From sunshine at public.kherson.ua Fri Jun 23 15:37:09 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri Jun 23 07:37:15 2006 Subject: MACH-O In-Reply-To: Message-ID: On 6/23/06 2:56 PM, "jda" wrote: Hi Jon, > I'm starting to move my app to MACH-O. I'm using the latest 2.4 beta, > and there is no MACH-O build. Is it safe to mix the 2.4 beta Carbon > VComponents with the older 2.3 MACH-O plug-in and components? Or do > we need a MACH-O build of 2.4? Do not mix If I not mistake: you can use V4RB todo MACHO debug build of your app. Actually this is the only way to debug macho RB app. You need V4RB Macho only to make release of your app. -- 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 Fri Jun 23 08:55:00 2006 From: jda at his.com (jda) Date: Fri Jun 23 07:55:32 2006 Subject: MACH-O In-Reply-To: References: Message-ID: >On 6/23/06 2:56 PM, "jda" wrote: > >Hi Jon, > >> I'm starting to move my app to MACH-O. I'm using the latest 2.4 beta, >> and there is no MACH-O build. Is it safe to mix the 2.4 beta Carbon >> VComponents with the older 2.3 MACH-O plug-in and components? Or do >> we need a MACH-O build of 2.4? > >Do not mix > >If I not mistake: > >you can use V4RB todo MACHO debug build of your app. >Actually this is the only way to debug macho RB app. > >You need V4RB Macho only to make release of your app. > Hm, when I try to run my project in the IDE (as MACH-O) the debug build crashes immediately on launch. No crash log. I only have Valentina as a plug-in and my own shared Carbon library. Any clue what the problem is? Jon From jda at his.com Fri Jun 23 09:35:09 2006 From: jda at his.com (jda) Date: Fri Jun 23 08:35:25 2006 Subject: MACH-O In-Reply-To: References: Message-ID: >Do not mix > >If I not mistake: > >you can use V4RB todo MACHO debug build of your app. >Actually this is the only way to debug macho RB app. > Hi Ruslan, Still no luck, no matter what combination of plugins I use. Is anyone on this list using MACH-O (Mac OS X only) in their RB projects? If so, can you tell me what your setup is (debug builds for now is fine)? Thanks. Jon From fitzbew at nc.rr.com Fri Jun 23 10:03:17 2006 From: fitzbew at nc.rr.com (Russ Tyndall) Date: Fri Jun 23 09:03:24 2006 Subject: MACH-O In-Reply-To: Message-ID: On 6/23/06 9:35 AM, "jda" wrote: > Is anyone on this list using MACH-O (Mac OS X only) in their RB > projects? I have used V4RB 2.3x in Mach-O for desktop dbs successfully (no problems). The only problems I have had are using Mach-O with Vserver, BUT those problems are unverified by Paradigma and they may very well be specific to me. > If so, can you tell me what your setup is (debug builds for > now is fine)? You cannot debug in Mach-O; you must build the Mach-O app in RB and then manually add the mach-o format components to the Package. Eventually, we will be able to debug in Mach-o when RB itself is Mach-o, so I have been told. You can only debug using the carbon-format V4RB plugin. So, for now, you must debug using Carbon plugin and PEF setting in RB. When you are ready to roll out Mach-O, build your app and then add Mach-O vcomponents to built package. You can only test built mach-o v4RB app, not debug compile. -- Russ Tyndall Wake Forest, NC From jda at his.com Fri Jun 23 10:12:18 2006 From: jda at his.com (jda) Date: Fri Jun 23 09:12:32 2006 Subject: MACH-O In-Reply-To: References: Message-ID: >You cannot debug in Mach-O; you must build the Mach-O app in RB and then >manually add the mach-o format components to the Package. Eventually, we >will be able to debug in Mach-o when RB itself is Mach-o, so I have been >told. You can only debug using the carbon-format V4RB plugin. > >So, for now, you must debug using Carbon plugin and PEF setting in RB. When >you are ready to roll out Mach-O, build your app and then add Mach-O >vcomponents to built package. You can only test built mach-o v4RB app, not >debug compile. > Thanks for the info, Russ. I now think that the crash I'm getting is due to my shared library being PEF. The RB archives have a few comments on this, and it looks like you can't call a PEF library from a MACH-O build (even a debug build). It looks like I'll have to find a way to rebuild the library as MACH-O before I can revisit this. Jon From jda at his.com Sun Jun 25 09:46:32 2006 From: jda at his.com (jda) Date: Sun Jun 25 08:46:44 2006 Subject: Speeding things up In-Reply-To: References: Message-ID: Hi Ruslan, I'm trying to improve the launch time of my app, and in the course of this found one point I hope can be optimized. When I open a Valentina db I check to see the schema version so that I can update one of the tables if necessary (otherwise of course I get an exception if the class definitions don't match). I'm finding that simply opening a VDatabase to check the schemaversion takes about half a second (30 ticks) on my Intel Mac! Here's the code: vdb = new VDatabase vdb.open(f) if vdb <> nil then if vdb.SchemaVersion < 904 then //add new field to table end if vdb.close vdb = nil end if Is there a faster way to check for schema version and compatibility? Thanks, Jon From jda at his.com Sun Jun 25 09:59:35 2006 From: jda at his.com (jda) Date: Sun Jun 25 08:59:46 2006 Subject: Speeding things up In-Reply-To: References: Message-ID: >vdb = new VDatabase > vdb.open(f) > if vdb <> nil then > if vdb.SchemaVersion < 904 then > //add new field to table > end if > vdb.close > vdb = nil > end if > >Is there a faster way to check for schema version and compatibility? > To answer my own question: I found this very nice function Valentina.GetSchemaVersion(f) which is almost instantaneous. Much better! Jon From dave at Parizek.com Sun Jun 25 10:30:39 2006 From: dave at Parizek.com (Dave Parizek) Date: Sun Jun 25 12:30:42 2006 Subject: V4CC - Cocoa examples In-Reply-To: References: Message-ID: <0B397ED5-853E-4964-8123-84972B9CFE13@Parizek.com> Any idea when there might be more V4CC examples done? --Dave __________________________________________ Dave Parizek From sunshine at public.kherson.ua Sun Jun 25 20:35:36 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun Jun 25 12:35:43 2006 Subject: V4CC - Cocoa examples In-Reply-To: <0B397ED5-853E-4964-8123-84972B9CFE13@Parizek.com> Message-ID: On 6/25/06 8:30 PM, "Dave Parizek" wrote: Hi Dave, > Any idea when there might be more V4CC examples done? Sergey tomorrow Monday start FULL TIME this job, he have finish all other issues... He right now setup his mac for V4CC bulds. Sergey need made about 50 example, and I think he should be able do 5-10 example per day.. Also I want that Sergey have help you setup start of your project. --------------- And I think in nearest week Kirill will start work on V4CC_Reference.pdf -- 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 Jun 26 09:01:49 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon Jun 26 01:01:58 2006 Subject: Progress on ICU job Message-ID: Hi All, I want inform you that last 3 days I have spend on change Valentina code that use C++ classes from ICU to use C functions from ICU. This is major step to be able use ICU from OS X and to be able use it from CW projects also... I have port all such code and our tests was fine. Sunday I have re-write our String class and now I need recover our tests... -- 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 Jun 27 12:33:24 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 04:33:37 2006 Subject: Uregex.cpp C API depends on C++ ? Bad idea Message-ID: Hi All, If you remember we develop Valentina database, and on MAC OS X I was going to use icu.dylib made by Apple. I was recommended to use C only API. And this is right. So last few days we have correct all our code to switch only on C API in ICU. * But in ICU 3.0 the regex have only draft of C API. * MACOS 10.3.9 has icu.dylib which NOT include regex C methods. Argh! :( * MacOS 10.4 has icu.dylib which contains regex C methods -- good. I was in hope that I will be able add uregex.cpp into our project, and this will work...but not so easy. This file contains references on C++ API of ICU ?! ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ IMHO this is not best design. C API must depend only on C API. C API should contain main-core algorithms of regEx. C algorithms can wrap some INTERNAL for this .cpp file C++ structures/algorithms, but only internal for file. C++ of REGEX should be thing wrapper to C algorithms of regex What you think ? ---- P.S. Now I have only 2 ways: a) drop support of REGEX -- not acceptable b) drop support of OS X 10.3.9 P.S.2 Deboragh, I think it will be good idea include into Apple's ICU library as many methods/features from ICU as possible. Even if they are not used by Apple itself.. -- 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 Jun 27 13:43:39 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 05:43:44 2006 Subject: We need drop support of Panther 10.3.9 Message-ID: Hi All, 1) We have time X -- when we need drop support of 10.3 to be able do next step for xcode-made projects and use ICU from Apple included into each OS X 10.4 + How many people will be disappointed by this fact ? ----------- 2) Also I have found that V4RB and V4MD CARBON will not be able use MACHO libicucore.dylib -- 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 Jun 27 07:28:40 2006 From: jda at his.com (jda) Date: Tue Jun 27 06:28:48 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: References: Message-ID: >Hi All, > >1) We have time X -- when we need drop support of 10.3 > > to be able do next step for xcode-made projects > and use ICU from Apple included into each OS X 10.4 + > >How many people will be disappointed by this fact ? Hi Ruslan. Me. I suggest at the least there be two builds, then. One that works with 10.3.9 and earlier (my app goes back to support 10.2 now!) and requires that ICU be included and one that does not. > > >----------- >2) Also I have found that V4RB and V4MD CARBON will not be able use > MACHO libicucore.dylib > Same as above. Jon From sunshine at public.kherson.ua Tue Jun 27 14:43:17 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 06:43:23 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: Message-ID: On 6/27/06 2:28 PM, "jda" wrote: >> Hi All, >> >> 1) We have time X -- when we need drop support of 10.3 >> >> to be able do next step for xcode-made projects >> and use ICU from Apple included into each OS X 10.4 + >> >> How many people will be disappointed by this fact ? > > > Hi Ruslan. > > > Me. I suggest at the least there be two builds, then. One that works > with 10.3.9 and earlier (my app goes back to support 10.2 now!) and > requires that ICU be included and one that does not. > >> >> >> ----------- >> 2) Also I have found that V4RB and V4MD CARBON will not be able use >> MACHO libicucore.dylib >> > > Same as above. Well, V4RB CARBON users still have ICU included. so you will not be limited by 10.4 So you are safe until REAL will make RB UB, and we will must move to xcode for RB plugins. So V4RB 2.4 and V4MD will work on 10.2+ yet...... --------- Limitations comes only to Xcode - made products Valentina for Rev UB Valentina Studio UB Valentina Server UB V4RB MACHO -- if we agree to reduce its size and remove here ICU, then it also will be limited to 10.4 + -- 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 Jun 27 14:59:33 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 06:59:37 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: <3BD5A18A-2C9F-4A66-BAC9-F3D5D2E7191C@valentina-db.de> Message-ID: On 6/27/06 2:54 PM, "Jochen Peters" wrote: Hi Jochen, > reason is missing regex support in earlier libicu from Apple? > Why not include own regex? How ?! I have hope 5 days that I will be able include their own .cpp file, But they depend on C++ classes --> problems in name manglings Only solution could be if Apple in 10.2 - 10.3 Have expose something as "CFRegEx" API in CoreFoundation, But as far as I see there is nothing similar in docs. ----- I have consider usage of (man) regex... But it works on char*, and there no any info if it can accept here UTF8 unicode. May be somebody from Darwin list knows this? If it support UTF8, then I can convert UTF16 to UTF8 and be happy probably... -- 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 rjb at robelko.com Tue Jun 27 14:04:38 2006 From: rjb at robelko.com (Robert Brenstein) Date: Tue Jun 27 07:15:06 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: References: Message-ID: >--------- >Limitations comes only to Xcode - made products > > Valentina for Rev UB > Valentina Studio UB > Valentina Server UB > > >V4RB MACHO -- if we agree to reduce its size and remove here ICU, then > it also will be limited to 10.4 + > > >-- >Best regards, > >Ruslan Zasukhin >VP Engineering and New Technology >Paradigma Software, Inc I think it may be premature to drop 10.3.9 support. Despite popularity of Tiger, and the fact that many developers moved to Tiger, I believe that Panther is still the most used OSX version overall. If you want to move ahead with slimmer versions of Valentina, you should consider to fork the product lines and produce Tiger-only and Panther+ versions in parallel at least for a while. Robert From sunshine at public.kherson.ua Tue Jun 27 15:15:18 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 07:15:22 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: Message-ID: On 6/27/06 2:43 PM, "Ruslan Zasukhin" wrote: > Well, V4RB CARBON users still have ICU included. > so you will not be limited by 10.4 > So you are safe until REAL will make RB UB, > and we will must move to xcode for RB plugins. > > So V4RB 2.4 and V4MD will work on 10.2+ yet...... > > --------- > Limitations comes only to Xcode - made products > > Valentina for Rev UB > Valentina Studio UB > Valentina Server UB > > > V4RB MACHO -- if we agree to reduce its size and remove here ICU, then > it also will be limited to 10.4 + Well, correction! * we CAN provide Vstudio + Vserver to run on 10.3 because this is Pradigma products, and we can be happy with keeping own version of ICU. * original idea was remove ICU from ADK products for OS X. especially ICU become pain for UB because contains double size. * We have currently only one UB ADK -- Valentina for Revolution. so we can get for now this limitation only for V4REV 2.4. May be really if will be needed we can make separate build of V4REV that contains ICU. ---------- Well, Jochen, to be able do this, I am starting correct all our xcode projects in next way: * I duplicate "Release (Universal)" to "Release (Universal) 10.3" * in current Debug and Release configs I remove settings for 10.3 * in the new "Release (Universal) 10.3" we need correct names of dlls to be as vkernel_fat_release_gcc33.dylib * on our computers we can build all these dlls into the same /use/local/vcomponents they will not conflict because OS X know exact names of dependent dlls. * we need correct all our RUN SCRIPTS in Vstudio and Vserver projects ... What you think about this plan ? ------ Oh! Another way is: we do not touch names of current universal dlls, instead we change names of new "Release" target to be as vkernel_fat_release_gcc4.dylib then we do not need correct nothing in vstudio and vserver scripts, -- 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 Jun 27 16:01:19 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 08:01:23 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: <13342.62.154.199.179.1151412783.squirrel@webmail3.server-einstellung.de> Message-ID: On 6/27/06 3:53 PM, "Jochen Peters" wrote: >> I still thinking >> >> Idea is to have 3 configurations >> >> Debug 10.4 << here to remove gcc 3.3 settings >> Release 10.4 << here to remove gcc 3.3 settings >> Release 10.3 10.3 << this is what we have now > > I see - because all 3 are build into different build directories this > should work because vStudio scripts uses configuration name to determine > the folder. > > In any case this will add more complexity to the build process... > At least for a while since Panther support is not important any more. > Maybe these Apple Laydy has some hints about CF and regex - i have made > a quick search for unicode regex and do not find any appropiate... Well, it seems regex.h of POSIX works more or less correctly on UTF8 I was able do russian text search! -- 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 Jun 27 16:02:14 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 08:03:15 2006 Subject: We need drop support of Panther 10.3.9 In-Reply-To: <13342.62.154.199.179.1151412783.squirrel@webmail3.server-einstellung.de> Message-ID: On 6/27/06 3:53 PM, "Jochen Peters" wrote: > In any case this will add more complexity to the build process... > At least for a while since Panther support is not important any more. > Maybe these Apple Laydy has some hints about CF and regex - i have made > a quick search for unicode regex and do not find any appropiate... I think best unicode regex can do ICU only. -- 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 Jun 27 16:41:56 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 08:42:02 2006 Subject: Well, good luck! I have found regex solution! Message-ID: Hi All, Well, few tests have show that for MAC OS X, at least some time before we all agree to drop 10.3, we can use instead of ICU the man's I have made few tests on UTF8 russian text it works correctly. ------- Interesting to test yet searches with Ignore Case ... I will do this now Nope. (?i) syntax is not support -- strange So we can all all except ignore case ... -- 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 Jun 27 21:00:03 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue Jun 27 13:00:21 2006 Subject: Okay, lets use ICU and 10.3 as is some time more... Message-ID: Hi All, I have spend few days port from C++ ICU to C ICU in hope to get more simple configuration for all us. Instead small trap -- absent of regex C function in OS X 10.3 become show stopper. Attempt to workaround this trap will only add troubles. So it is reasonable to say: * we have made GOOD step to C API. this is required step in any case. * some code have become more clean * some code have become improved. * we have prepare self for next step when OS X will be ready. * we have prepare self for next step tp get Valentina-All-In-One as on MAC so on Windows. * after 2.4 release I will work mainly on text-related issues in all regards. Including try compile is STATIC library and reduce its default size Future we will consider what is better for all us: a) Valentina ADKS as now + new line Valentina Lite (all in one with some disabled features, e.g. -XML -PAINTLIB or b) Valentina All in one without limits only ICU reduced in size + new line Valentina ADK ADVANCED with VComponents as now + more. ----------- Okay, now I need finish polishing of our tests with new String class. -- 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 dev at macfoh.com Tue Jun 27 11:40:12 2006 From: dev at macfoh.com (Shaun Wexler) Date: Tue Jun 27 13:40:30 2006 Subject: Uregex.cpp C API depends on C++ ? Bad idea In-Reply-To: References: Message-ID: <17AB1B7A-D012-40ED-B6D0-05096C15DF84@macfoh.com> On Jun 27, 2006, at 2:33 AM, Ruslan Zasukhin wrote: > P.S. Now I have only 2 ways: > > a) drop support of REGEX -- not acceptable > > b) drop support of OS X 10.3.9 This should answer for b: -- Shaun Wexler MacFOH http://www.macfoh.com From ayu8 at cdc.gov Thu Jun 29 14:30:57 2006 From: ayu8 at cdc.gov (Sims, John (CDC/CCHP/NCBDDD) (CTR)) Date: Thu Jun 29 13:36:06 2006 Subject: Triggers Message-ID: Hi Ruslan, In your "Roadmap" post back in February, you mentioned that you thought you would have Triggers and views in 2.4. I was wondering if that is still the case or have Triggers been pushed back? Also, do you have any kind of write-up on how you envision triggers will be implemented in Valentina (I didn't find anything on the Wiki but I may not have looked hard enough)? Thanks, Ruslan. -John From sunshine at public.kherson.ua Thu Jun 29 22:27:21 2006 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu Jun 29 14:27:27 2006 Subject: Triggers In-Reply-To: Message-ID: On 6/29/06 9:30 PM, "Sims, John (CDC/CCHP/NCBDDD) (CTR)" wrote: Hi John, > In your "Roadmap" post back in February, you mentioned that you thought > you would have Triggers and views in 2.4. I was wondering if that is > still the case or have Triggers been pushed back? You know that our wish is always faster of our abilities :) For info: we have decide that Triggers is less important feature than views. Especially interesting looks materialized views. they really can speedup things. > Also, do you have any kind of write-up on how you envision triggers will > be implemented in Valentina (I didn't find anything on the Wiki but I > may not have looked hard enough)? For now we have put triggers on "last point in order" transactions, XMLSQL, views, ... Triggers. -- 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 ayu8 at cdc.gov Fri Jun 30 08:10:48 2006 From: ayu8 at cdc.gov (Sims, John (CDC/CCHP/NCBDDD) (CTR)) Date: Fri Jun 30 07:24:39 2006 Subject: Triggers Message-ID: Thanks for the update, Ruslan. -John > Hi John, > > > In your "Roadmap" post back in February, you mentioned that you > > thought you would have Triggers and views in 2.4. I was > wondering if > > that is still the case or have Triggers been pushed back? > > You know that our wish is always faster of our abilities :) > > For info: we have decide that Triggers is less important > feature than views. > Especially interesting looks materialized views. > they really can speedup things. > > > Also, do you have any kind of write-up on how you envision triggers > > will be implemented in Valentina (I didn't find anything on > the Wiki > > but I may not have looked hard enough)? > > For now we have put triggers on "last point in order" > > transactions, XMLSQL, views, ... Triggers. > > > -- > Best regards, > > Ruslan Zasukhin > VP Engineering and New Technology > Paradigma Software, Inc