From sunshine at public.kherson.ua Sat May 1 20:26:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 1 12:26:43 2004 Subject: VSRV, new demo license files for May uploaded Message-ID: -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sun May 2 21:03:42 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun May 2 13:03:50 2004 Subject: V4RB2, CHG, ValentinaInit() In-Reply-To: <76389EEE-6BBB-11D8-A255-003065BB0634@desuetude.com> Message-ID: On 3/1/04 10:03 PM, "Charles Yeomans" wrote: Hi Charles, So you prefer, ValentinaInit( s ) real_size = ValentinaCacheSize(); What other people think? I agree that for GUI tools it is good to have function ValentinaCacheSize() >>>> ValentinaIit() now returns the actual size of cache that Valentina have >>>> allocate. >>>> >>>> >>>> NOTE: for modern OS that have virtual file system, even if you ask 100MB >>>> for cache, OS can satisfy your request. But Valentina self can decide that >>>> you ask too big cache size and allocate smaller cache. >>>> >>>> In case of any problems, ValentinaInit() will return ZERO. >>>> >>> >>> Again, why return this value? This is old C-style design. Instead, >>> why not have a separate function ValentinaCacheSize() as Integer? >> >> 1) ValentinaInit() can fail. > > This is what exceptions are for. >> >> if return nothing, then you need ask ValentinaError >> and we need add one more error code. >> >> 2) it can allocate less than you ask. >> >> one more function... >> >> >> One return value solve 2 tasks at once. > > And this is what I'm complaining about. I think that it's simpler to > have two things (functions, properties, etc), each of which solves one > task. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sun May 2 21:10:06 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun May 2 13:10:09 2004 Subject: V4RB 2.0, debug and release versions Message-ID: Hi All, As I have told, in V4RB/V4Md 2.0 we will extract Valentina engine itself into separate DLL. His give us very small plugin file. So we can easy produce 2 versions of plugins 1) debug version, have as many as possible alerts, checks, protection, ... To help developer find bugs as fast as possible. 2) release version of plugin. Where all this code is just REMOVED. In Valentina for RB 1.x that code still was in the plugin, it was wrapped by if(). So functions of RELESE plugin will be smaller and even little faster. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Mon May 3 14:27:45 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Mon May 3 13:28:14 2004 Subject: V4RB2, CHG, ValentinaInit() In-Reply-To: References: Message-ID: <926547AA-9D2F-11D8-A3C9-000A95BE1A8E@mindstarprods.com> On May 2, 2004, at 2:03 PM, Ruslan Zasukhin wrote: > So you prefer, > > ValentinaInit( s ) I don't mind this returning an integer, but C was one of my first languages after Atari basic. Using exceptions is definitely the more modern approach though. > real_size = ValentinaCacheSize(); I like this idea. > What other people think? I agree with Charles. Embracing OOP is a good idea. We already will have to modify our code when we convert to Valentina 2. Best regards, Frank From sunshine at public.kherson.ua Mon May 3 21:31:01 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 3 13:31:13 2004 Subject: V4RB2, CHG, ValentinaInit() In-Reply-To: <926547AA-9D2F-11D8-A3C9-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/3/04 9:27 PM, "Frank Schima" wrote: >> So you prefer, >> >> ValentinaInit( s ) > > I don't mind this returning an integer, but C was one of my first > languages after Atari basic. Using exceptions is definitely the more > modern approach though. > >> real_size = ValentinaCacheSize(); > > I like this idea. Already implemented. >> What other people think? > > I agree with Charles. Embracing OOP is a good idea. We already will > have to modify our code when we convert to Valentina 2. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon May 3 21:46:31 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 3 13:46:56 2004 Subject: V4RB2, CHG, ValentinaInit() In-Reply-To: <3B144BC0-9D31-11D8-BD2F-000A959E9B12@abdatatools.com> Message-ID: On 5/3/04 9:39 PM, "Aaron Bratcher" wrote: > I tend to agree with Charles. I am going to assign a cache that I > think is appropriate, but realistically what are my chances of checking > for the real cache size? If I want it, I will ask for it. > > Looking at the subject, does this mean a beta for REALbasic users is > available or is it still in-process? In process. I think I will upload archives for V4RB may be even tomorrow. I want developers become used to new design of plugin. And must be ready first 2 simplest examples. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 4 09:19:23 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 4 01:19:45 2004 Subject: How to create primary keys, constraints in 2.x...? In-Reply-To: <4096C823.1020100@accessitsoftware.com> Message-ID: On 5/4/04 1:30 AM, "Phil" wrote: Hi Phil, > Hi, I'm getting started with Valentina, trying to convert a large-ish > database schema to be compatible with Valentina. It isn't far off... > > However something I haven't got going yet is when doing a create table, > after creating the columns what is the correct syntax for creating > primary keys and constraints? > I saw in a message from 3-4 months ago that constraints & primary keys > will be in 2.0 (seems they weren't in 1.x?). In Valentina 1.x we have no RDBMS Primary Key. Instead we have RecID + ObjectPtr functionality. This way eat less disk space, require less developer work, and works 4 times faster of RDB way. In Valentina 2.0 yes, we have add Sql92 Primary and Foreign key. Syntax as EXACTLY according to SQL92 standard. But if your task allow, then still you may prefer to use ObjectPtrs. Here you can see grammar of Valentina 2.0 http://paradigmasoft.com/VSQL_Parser.html > Also I have a ValentinaSQL.pdf file but it seems quite out-of-date... is > there any recent documentation about SQL syntax in Valentina? It is not out of date. It corresponds to Valentina 1.10 > I'm using XP SP1, Valentina Server (2.0a62, Protocol 1.1, Kernel 1.10), > Valentina C++ Client SDK 2.0a62 (Win32) Please note, Valentina Server based on 1.x kernel means yet Valentina 1.x -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 7 23:46:43 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 7 15:46:57 2004 Subject: Suggestion for new Feature In-Reply-To: <003501c43471$80236a90$6701a8c0@kent> Message-ID: On 5/7/04 11:25 PM, "Simon White" wrote: > Hi Ruslan > > I do alot of work dynamically generating web sites from VFP Tables. Often > there is a need to dynamically create drop down lists using the HTML Select > control. This requires writing code that wraps data from a table in > something like the code below. > > Function CreateDropDown > Local lcHTML > lcHTML="" > Scan > lcHTML=lcHTML+[] > EndScan > Return lcHTML > EndFunc > > My thought was that the IVCursor and IVBaseObject needed a similar method > that could quickly return the HTML code something like. > > oVCursor.GetHTML_List(inIDField,inDescField,inSelectedItemID) As String > > Where > inIDField was the field name used in for the Value > inDescField was the field name used for the visible text > inSelectedItemID was the RecID of the item to be marked as selected > > This would allows rapid creation of drop down list from data in Valentina Well, Simon. May be. For example, Valentina Application (VAPP) which was oritneted few years ago for use on WEB have had "format" fucntion to produce HTML output. On the other hand, Valentina engine itself is not oriented only on WEB. You idea can be implemented as not part of Vcursor, but simply as library of some functions. I wonder, what worry you: You think about better speed if this will be written in C++, Or you simply want avoid such code in VFP ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 01:24:05 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 10 17:51:56 2004 Subject: V4RB NEW, Vdatabase.IsOpen property Message-ID: -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 01:25:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 10 17:51:58 2004 Subject: V4RB NEW Vdatabase.ThrowOut method Message-ID: DESCRIPTION: This method allow you delete files of database from HDD. --------------- OFF Of course, this is not what you need often :-) But just for completeness of methods -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 01:27:54 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 10 17:52:05 2004 Subject: V4RB NEW: Vdatabase constructor Message-ID: Now it looks as "VDatabase( inStorageType as Integer = 0 )" inStorageType - this parameter allow you create DISK-based or RAM-based database. 0 - disk based 1 - RAM based. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 00:17:22 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 10 17:52:17 2004 Subject: Valentina 2.0 for REALbasic b1 Message-ID: Hi All, http://paradigmasoft.com/download/v2_beta/v2_beta.html Here you will find links and instructions how to download and work with Our first beta of Valentina 2.0 for REALbasic (OS X) I think tomorrow, we will upload V4MD (Win). ---------- Please do not expect too many from this first beta !!! We have made and test only first 2 examples. Now will continue create more and more examples. Before we upload an example that demonstrate some feature and say that it works on our side, please do not complain that something not works. Ok? I ask you first of all to see how Valentina 2.0 will looks on HDD. I ask you take in part in the discussion of new functions, changes and so on. I ask you give comments about our examples. We try now build system of examples that will be 100% the same to all Valentina products. And which will totally complete from point of view demonstration of features. We going to have A LOTS of small examples, and each example demonstrate SINGLE and separate feature of Valentina. ---------- This is the first and last notification to Valentina list, All next letters will go only to Valentina-beta list. So if you are not yet subscribed but you going take part in beta testing please subscribe. ----------- I am not sure that it is good to use "beta" for this versions. May be more correct to say alpha. Also we will need synchronize in future Valentina Server numeration to Valentina engine 2.0, when we will switch Vserver to a new engine. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 01:54:32 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 10 17:54:39 2004 Subject: Igor, .tmp volume bug Message-ID: Igor, As I see in V4RB Examples, after db.Close() .tmp volume still here. It must be deleted on Close(). -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Mon May 10 19:03:00 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Mon May 10 18:02:35 2004 Subject: V4RB NEW Vdatabase.ThrowOut method In-Reply-To: References: Message-ID: <2F10EF68-A2D6-11D8-9BC2-003065BB0634@desuetude.com> On May 10, 2004, at 6:25 PM, Ruslan Zasukhin wrote: > > > DESCRIPTION: > > This method allow you delete files of database from HDD. > > > --------------- > OFF > > Of course, this is not what you need often :-) > But just for completeness of methods Why not call it VDatabase.Delete? If you call it ThrowOut, then I'd expect to find the files in the trash. Charles Yeomans From sunshine at public.kherson.ua Tue May 11 02:06:27 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 10 18:06:34 2004 Subject: V4RB NEW Vdatabase.ThrowOut method In-Reply-To: <2F10EF68-A2D6-11D8-9BC2-003065BB0634@desuetude.com> Message-ID: On 5/11/04 2:03 AM, "Charles Yeomans" wrote: >> DESCRIPTION: >> >> This method allow you delete files of database from HDD. >> >> >> --------------- >> OFF >> >> Of course, this is not what you need often :-) >> But just for completeness of methods > > > Why not call it VDatabase.Delete? If you call it ThrowOut, then I'd > expect to find the files in the trash. 1) Because 'Delete' is very close to C++ operator. delete pointer Which delete only C++ instance of RAM. 2) may be then db.Erase() ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Mon May 10 19:12:45 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Mon May 10 18:12:17 2004 Subject: V4RB NEW Vdatabase.ThrowOut method In-Reply-To: References: Message-ID: <8C106553-A2D7-11D8-9BC2-003065BB0634@desuetude.com> On May 10, 2004, at 7:06 PM, Ruslan Zasukhin wrote: > On 5/11/04 2:03 AM, "Charles Yeomans" wrote: > >>> DESCRIPTION: >>> >>> This method allow you delete files of database from HDD. >>> >>> >>> --------------- >>> OFF >>> >>> Of course, this is not what you need often :-) >>> But just for completeness of methods >> >> >> Why not call it VDatabase.Delete? If you call it ThrowOut, then I'd >> expect to find the files in the trash. > > 1) Because 'Delete' is very close to C++ operator. > > delete pointer > > Which delete only C++ instance of RAM. > > > 2) may be then db.Erase() ? 3) Or maybe db.DeleteFiles()? -------------- Charles Yeomans From yeomans at desuetude.com Mon May 10 19:30:10 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Mon May 10 18:29:42 2004 Subject: V4Rb 2.0 first problem Message-ID: I've installed V4Rb 2.0 instead of working on my own behind-schedule project. The first example quits when ValentinaInit is called -- no crash log, no console output. Charles Yeomans From yeomans at desuetude.com Mon May 10 19:35:12 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Mon May 10 18:34:43 2004 Subject: ValentinaDebugLevel Message-ID: Is there some technical reason why ValentinaDebugLevel cannot be a property? The current version is awkward. Instead, I propose a global property ValentinaDebugLevel as Integer, plus constants representing the debug levels. Also, you might consider using the new syntax ModuleName.Method, as in Valentina.DebugLevel = Valentina.LogMethodCalls. Charles Yeomans From frank-list2 at mindstarprods.com Mon May 10 21:14:34 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Mon May 10 20:14:40 2004 Subject: V4Rb 2.0 first problem In-Reply-To: References: Message-ID: <9086AF8D-A2E8-11D8-8C6F-000A95BE1A8E@mindstarprods.com> On May 10, 2004, at 7:30 PM, Charles Yeomans wrote: > I've installed V4Rb 2.0 instead of working on my own behind-schedule > project. The first example quits when ValentinaInit is called -- no > crash log, no console output. I see the same problem with both examples. I'm running Mac OS X 10.3.3. RB 5.5.2fc1. I followed the directions to install all the components. Best regards, Frank From sunshine at public.kherson.ua Tue May 11 08:10:08 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 00:10:16 2004 Subject: V4RB NEW Vdatabase.ThrowOut method In-Reply-To: <8C106553-A2D7-11D8-9BC2-003065BB0634@desuetude.com> Message-ID: On 5/11/04 2:12 AM, "Charles Yeomans" wrote: >> 2) may be then db.Erase() ? > > 3) Or maybe db.DeleteFiles()? Ok, this is good name. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 02:11:50 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 00:12:21 2004 Subject: [KERNEL] NEW// .tmp volume Message-ID: FYI: * in Valentina 2.0 we will use special .tmp volume. it will contain all temporary files - tmp tables - join - locks * It is ALWAYS separate file, even if you specify db.mode = 1. * we use it because now, in case of system failure, Valentina will need simply create new EMPTY .tmp volume. I.e. We get ZERO time to cleanup tmp files from the past session. ----------- Valentina 1.x in this case have spend time to check each volume on tmp files and deletion of them. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- _______________________________________________ Vdevelopers mailing list Vdevelopers@paradigma.ukrcom.kherson.ua http://192.168.2.1/mailman/listinfo/vdevelopers From sunshine at public.kherson.ua Tue May 11 08:14:56 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 00:15:01 2004 Subject: ValentinaDebugLevel In-Reply-To: Message-ID: On 5/11/04 2:35 AM, "Charles Yeomans" wrote: > Is there some technical reason why ValentinaDebugLevel cannot be a > property? The current version is awkward. Instead, I propose a global > property ValentinaDebugLevel as Integer, plus constants representing > the debug levels. Also, you might consider using the new syntax > ModuleName.Method, as in > > Valentina.DebugLevel = Valentina.LogMethodCalls. Yes Charles, we also think on this. Yes this looks logical. Only problem we can catch: Valentina.ERR_FIELD_NOT_FOUND Little long names. Is this acceptable ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 08:15:17 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 00:17:59 2004 Subject: V4Rb 2.0 first problem In-Reply-To: <9086AF8D-A2E8-11D8-8C6F-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 4:14 AM, "Frank Schima" wrote: > On May 10, 2004, at 7:30 PM, Charles Yeomans wrote: > >> I've installed V4Rb 2.0 instead of working on my own behind-schedule >> project. The first example quits when ValentinaInit is called -- no >> crash log, no console output. > > > I see the same problem with both examples. > > I'm running Mac OS X 10.3.3. RB 5.5.2fc1. I followed the directions to > install all the components. Ok, I will check again. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 13:57:34 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 06:12:21 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <7F470D70-A333-11D8-9F3E-000393D43560@parsec.info> Message-ID: On 5/11/04 1:10 PM, "Guillermo" wrote: Hi Guilermo, > One cuestion: Will the final version of Valentina 2.0 require the > installation of several components? or will be possible to pack all in > the application executable? Valentina 2.0 will looks as V4RB plguin + Vcomponents folder. Your final application will look now as app.exe + Vcomponents folder. Valentina 2.0 is much more complex by structure, So it is many advantages to keep components separately. Valentina 2.0 have own plugin system. > One of the advantages of the current > version of V4RB is that it not requiere any installation in the final > user computer. This was true only for V4RB MAC. Right? Even V4RB WIN have install dll into users' system. V4MD developers even do not pack Xtras into projectors, Xtras always are separate DLLs in Director. The same is true for Visual BASIC and so on. Keep all inside of one file is not so great idea actually. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue May 11 07:21:01 2004 From: jda at his.com (jda) Date: Tue May 11 06:21:10 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: References: Message-ID: > > >Your final application will look now as > > app.exe + Vcomponents folder. > Can Vcomponents be in the same folder as the final application? Jon From sunshine at public.kherson.ua Tue May 11 14:24:19 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 06:32:58 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: Message-ID: On 5/11/04 2:21 PM, "jda" wrote: >> Your final application will look now as >> >> app.exe + Vcomponents folder. >> > > Can Vcomponents be in the same folder as the final application? Yes! It MUST BE there. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 14:50:56 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 07:01:03 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: <9086AF8D-A2E8-11D8-8C6F-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 4:14 AM, "Frank Schima" wrote: > On May 10, 2004, at 7:30 PM, Charles Yeomans wrote: > >> I've installed V4Rb 2.0 instead of working on my own behind-schedule >> project. The first example quits when ValentinaInit is called -- no >> crash log, no console output. > > > I see the same problem with both examples. > > I'm running Mac OS X 10.3.3. RB 5.5.2fc1. I followed the directions to > install all the components. Hi Frank, Hi Charles, I have double check all on my side and it works. I have check also in clean environment on OS X 10.3.3 and 10.2.3 All works. So let's search for mistake on your side. 1) you must use REALbasic 5.5.2b1 or better because V4RB 2.0 do use default parameters. and before 5.5.2b1 REALbasic have had bug. I see that Frank uses 5.5bfc1. This is good/ what about you Charles? 2) Just to clarify. You need copy into Vcomponents folder not the folder Vengine, but files inside of Vengine. i.e. You should copy file Kernel_Carbon_Debug.shlb into Vcomponents folder. 3) when I have come to clean OS X 10.2.3 then in the folder "/Library" there was no folder "CFMSupport". I have made it manually. --------------------- Jochen, do you have realbasic 5.5.2fc1? Can you test on your side also. http://paradigmasoft.com/download/v2_beta/v2_beta.html -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Tue May 11 08:18:05 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Tue May 11 07:18:10 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: References: Message-ID: <416E0576-A345-11D8-986B-000A95BE1A8E@mindstarprods.com> Hi Ruslan, On May 11, 2004, at 7:50 AM, Ruslan Zasukhin wrote: > On 5/11/04 4:14 AM, "Frank Schima" > wrote: > >> On May 10, 2004, at 7:30 PM, Charles Yeomans wrote: >> >>> I've installed V4Rb 2.0 instead of working on my own behind-schedule >>> project. The first example quits when ValentinaInit is called -- no >>> crash log, no console output. >> >> >> I see the same problem with both examples. >> >> I'm running Mac OS X 10.3.3. RB 5.5.2fc1. I followed the directions to >> install all the components. > > Hi Frank, > Hi Charles, > > I have double check all on my side and it works. > I have check also in clean environment on OS X 10.3.3 and 10.2.3 > All works. > > So let's search for mistake on your side. > > 1) you must use REALbasic 5.5.2b1 or better > because V4RB 2.0 do use default parameters. > and before 5.5.2b1 REALbasic have had bug. > > I see that Frank uses 5.5bfc1. This is good/ > what about you Charles? > To add to this, V4RB 2.0 was the only file in my RB plug-in folder. > > 2) Just to clarify. You need copy into Vcomponents folder > not the folder Vengine, but files inside of Vengine. > > i.e. You should copy file Kernel_Carbon_Debug.shlb > into Vcomponents folder. I double checked and I had done it this way. > > 3) when I have come to clean OS X 10.2.3 then in the folder > "/Library" there was no folder "CFMSupport". > I have made it manually. The CFMSupport folder already exists in /Library and has other files inside it. And to clarify, I am inside the global /Library folder, not the local user ~/Library folder. Can someone else test this please? Ruslan, are you sure you posted the right files? I can test on many more Macs when I get back into the office on Thursday. Best regards, Frank From jda at his.com Tue May 11 08:27:02 2004 From: jda at his.com (jda) Date: Tue May 11 07:27:11 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: References: Message-ID: > > >So let's search for mistake on your side. > >1) you must use REALbasic 5.5.2b1 or better > because V4RB 2.0 do use default parameters. > and before 5.5.2b1 REALbasic have had bug. > > I see that Frank uses 5.5bfc1. This is good/ > what about you Charles? > > >2) Just to clarify. You need copy into Vcomponents folder > not the folder Vengine, but files inside of Vengine. > > i.e. You should copy file Kernel_Carbon_Debug.shlb > into Vcomponents folder. > > >3) when I have come to clean OS X 10.2.3 then in the folder > "/Library" there was no folder "CFMSupport". > I have made it manually. > Ruslan, I've tried this on 2 different Macs, each time it crashes in Init. I've put the VComponenets folder in the global Library (i.e. Mac HD:Libary:CFMSupport:) (CFMSupport was already there). Mac OS X 10.3.3 RB 5.5.2fc1 Jon From sunshine at public.kherson.ua Tue May 11 15:32:16 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 07:45:36 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: <416E0576-A345-11D8-986B-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 3:18 PM, "Frank Schima" wrote: >> 3) when I have come to clean OS X 10.2.3 then in the folder >> "/Library" there was no folder "CFMSupport". >> I have made it manually. > > The CFMSupport folder already exists in /Library and has other files > inside it. And to clarify, I am inside the global /Library folder, not > the local user ~/Library folder. > > Can someone else test this please? > > Ruslan, are you sure you posted the right files? Yes, because I have use THE SAME archives which are on FTP When I have go to clean OS X 10.2.3 Strange. > I can test on many more Macs when I get back into the office on > Thursday. Also strange that you do not get any crash log. Hmm. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 15:34:22 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 07:49:17 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: <416E0576-A345-11D8-986B-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 3:18 PM, "Frank Schima" wrote: > The CFMSupport folder already exists in /Library and has other files > inside it. And to clarify, I am inside the global /Library folder, not > the local user ~/Library folder. > > Can someone else test this please? > > Ruslan, are you sure you posted the right files? > > I can test on many more Macs when I get back into the office on > Thursday. Hi Gyus, So it crashes when you do RUN for project, yes? Can you check if code completion show Valentina plugin methods? If yes, then REALbasic see plugin. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From j.peters at valentina-db.de Tue May 11 14:55:46 2004 From: j.peters at valentina-db.de (Jochen Peters) Date: Tue May 11 07:56:02 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: References: <9086AF8D-A2E8-11D8-8C6F-000A95BE1A8E@mindstarprods.com> Message-ID: <44046.62.154.199.179.1084280146.squirrel@webmail3.server-einstellung.de> Hi Ruslan, > > --------------------- > Jochen, do you have realbasic 5.5.2fc1? > Can you test on your side also. > > http://paradigmasoft.com/download/v2_beta/v2_beta.html > Ok - i will test here also and let you know! -- Best regards, Jochen Peters PIIT GmbH ------------------------ http://www.valentina-db.de From jda at his.com Tue May 11 08:56:03 2004 From: jda at his.com (jda) Date: Tue May 11 07:56:10 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: References: Message-ID: > > >So it crashes when you do RUN for project, yes? Yes. > > >Can you check if code completion show Valentina plugin methods? >If yes, then REALbasic see plugin. > There is no crash dialog and no log. But a text file (empty) is created, named VSQL_TreeParser.txt Jon From sunshine at public.kherson.ua Tue May 11 16:01:48 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 08:06:26 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: Message-ID: On 5/11/04 3:56 PM, "jda" wrote: >> So it crashes when you do RUN for project, yes? > > Yes. > >> >> >> Can you check if code completion show Valentina plugin methods? >> If yes, then REALbasic see plugin. >> > > There is no crash dialog and no log. But a text file (empty) is created, named > > VSQL_TreeParser.txt Aha! So plugin was found, application was compiled application was started Vengine DLL was found Vengine have produce text file And crash. Interesting. Okay, it is great to know that problem not in DLLs. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Tue May 11 09:34:14 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Tue May 11 08:34:21 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: References: Message-ID: On May 11, 2004, at 9:01 AM, Ruslan Zasukhin wrote: > On 5/11/04 3:56 PM, "jda" wrote: > >>> So it crashes when you do RUN for project, yes? >> >> Yes. Yes. Same for built applications too. >>> >>> >>> Can you check if code completion show Valentina plugin methods? >>> If yes, then REALbasic see plugin. >>> >> >> There is no crash dialog and no log. But a text file (empty) is >> created, named >> >> VSQL_TreeParser.txt I forgot to mention that I see that too. > Aha! > > So plugin was found, > application was compiled > application was started > Vengine DLL was found > Vengine have produce text file > > And crash. > > Interesting. > > > Okay, it is great to know that problem not in DLLs. From sunshine at public.kherson.ua Tue May 11 16:38:00 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 08:46:04 2004 Subject: V4Rb 2.0 first problem. It works here In-Reply-To: Message-ID: On 5/11/04 4:34 PM, "Frank Schima" wrote: > I forgot to mention that I see that too. > >> Aha! >> >> So plugin was found, >> application was compiled >> application was started >> Vengine DLL was found >> Vengine have produce text file >> >> And crash. >> >> Interesting. >> >> >> Okay, it is great to know that problem not in DLLs. I think I know where is problem! Inside of my code I have specify path to MY HDD. So it not works of course on your HDDs. Although not clear why it still work here from other HDD. Okay, I will check this issue now. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 17:25:41 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 09:26:35 2004 Subject: ICU / icudt26b.dat file on OS X Message-ID: Hi All, Wow, I have found this big 8 MB file in the OS X ! It is located in the /usr/share/icu Please check that you also have it on your computers. Interesting, it looks that OS X 10.2 already have it also. This is interesting, because then we have chance to NOT distribute this 8MB file with Vcomponents folder for OS X. On the other hand may be better have it. -- then we get Vcomponents folder THE SAME for all platforms and products. BTW, that file can be made smaller. Not 8 MB but just 3.5 MB. May be even less. TIP: I have try kill that file. NEVER DO THIS! :-) OS X have become crazy. Nothing works. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue May 11 10:48:23 2004 From: jda at his.com (jda) Date: Tue May 11 09:48:30 2004 Subject: ICU / icudt26b.dat file on OS X In-Reply-To: References: Message-ID: >Hi All, > >Wow, I have found this big 8 MB file in the OS X ! >It is located in the > > /usr/share/icu > >Please check that you also have it on your computers. >Interesting, it looks that OS X 10.2 already have it also. > >This is interesting, because then we have chance to NOT distribute this 8MB >file with Vcomponents folder for OS X. > >On the other hand may be better have it. >-- then we get Vcomponents folder THE SAME for all platforms and products. > >BTW, that file can be made smaller. >Not 8 MB but just 3.5 MB. May be even less. > Can you put the pathname in non-Unix terms please? I can only find an empty folder called ICU in the Application Support folder. As to your question, I would very much like to *avoid* distributing the unicode library for OS X if possible! So I would certainly vote for allowing us to use the OS X version (we have to build different distribution packages for each platform anyway, so why not let us tailor them to best take advantage of the libraries already installed by the OS? Jon From yeomans at desuetude.com Tue May 11 11:29:58 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 10:29:36 2004 Subject: ValentinaDebugLevel In-Reply-To: References: Message-ID: <100A17BB-A360-11D8-A4C6-003065BB0634@desuetude.com> On May 11, 2004, at 1:14 AM, Ruslan Zasukhin wrote: > On 5/11/04 2:35 AM, "Charles Yeomans" wrote: > >> Is there some technical reason why ValentinaDebugLevel cannot be a >> property? The current version is awkward. Instead, I propose a >> global >> property ValentinaDebugLevel as Integer, plus constants representing >> the debug levels. Also, you might consider using the new syntax >> ModuleName.Method, as in >> >> Valentina.DebugLevel = Valentina.LogMethodCalls. > > Yes Charles, we also think on this. > > Yes this looks logical. > > Only problem we can catch: > Valentina.ERR_FIELD_NOT_FOUND > > Little long names. > > Is this acceptable ? It is to me, if only you can omit those ugly underscores; Valentina.ErrFieldNotFound is so much more pleasant to the eye. Charles Yeomans From yeomans at desuetude.com Tue May 11 11:32:36 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 10:32:09 2004 Subject: ICU / icudt26b.dat file on OS X In-Reply-To: References: Message-ID: <6E1D1644-A360-11D8-A4C6-003065BB0634@desuetude.com> On May 11, 2004, at 10:25 AM, Ruslan Zasukhin wrote: > Hi All, > > Wow, I have found this big 8 MB file in the OS X ! > It is located in the > > /usr/share/icu > > Please check that you also have it on your computers. > Interesting, it looks that OS X 10.2 already have it also. Yep; it's on my machine (10.3.2). Perhaps your code can search that path for icudt26b.dat as well; then we could trash the file in the Valentine distribution. -------------- Charles Yeomans From sunshine at public.kherson.ua Tue May 11 20:00:21 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 12:02:22 2004 Subject: ICU / icudt26b.dat file on OS X In-Reply-To: Message-ID: On 5/11/04 5:48 PM, "jda" wrote: >> Hi All, >> >> Wow, I have found this big 8 MB file in the OS X ! >> It is located in the >> >> /usr/share/icu >> >> Please check that you also have it on your computers. >> Interesting, it looks that OS X 10.2 already have it also. >> >> This is interesting, because then we have chance to NOT distribute this 8MB >> file with Vcomponents folder for OS X. >> >> On the other hand may be better have it. >> -- then we get Vcomponents folder THE SAME for all platforms and products. >> >> BTW, that file can be made smaller. >> Not 8 MB but just 3.5 MB. May be even less. >> > > Can you put the pathname in non-Unix terms please? I can only find an > empty folder called ICU in the Application Support folder. John, I think you should in search add "visible all" Path must be system volume:usr:share:icu > As to your question, I would very much like to *avoid* distributing > the unicode library for OS X if possible! > So I would certainly vote for allowing us to use the OS X version (we have to > build different distribution packages for each platform anyway, so why not let > us tailor them to best take advantage of the libraries already installed by > the OS? Agree, but you sure you will make different archives? V4RB is CARBON, i.e. ready for OS 9 also. So you will make 2 Carbon archives? For OS X and OS 9? I think we will not have Classic target for 2.0. Because nothing modern not works in Classic: threads, icu, ... -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 20:27:09 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 12:27:24 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <92F7E1A4-A35C-11D8-9ECB-0003935B6750@comcast.net> Message-ID: On 5/11/04 6:05 PM, "Brendan Murphy" wrote: ATTENTION: Please go to Valentina-beta list for 2.0 discussion http://lists.macserve.net/mailman/listinfo/valentina-beta ------------------------ >> Valentina 2.0 is much more complex by structure, >> So it is many advantages to keep components separately. >> Valentina 2.0 have own plugin system. > > Please explain in more detail, what are these advantages? I want > to use 2.0 as I use 1.10 today as a totally embedded system within > my application. Having a separate folder next to my application is > just another support hassle I would rather avoid. I believe that we get problems similar to DLL hell, Only if installation of Vcomponents folder is made into some __central__ place on computer. But IF Valentina developer will put Vcomponents folder into app folder, Then no problems. On the same computer can live SEVERAL Valentina-made applications and they will not conflict in any case, even if they will use different versions of Valentina. IF you SELF will not distribute separate .dlls from Vcomponents folder to your users then also no problems. If you decide to provide your users with some updates then this is up to you. You can make installer to do that. >>> One of the advantages of the current >>> version of V4RB is that it not requiere any installation in the final >>> user computer. >> >> This was true only for V4RB MAC. Right? >> Even V4RB WIN have install dll into users' system. > > Under Win32 it is done automatically for us by the RS runtime > library! When it is done, it removes them. As I have see .DLL stay in the Windows folder. > Without further explanation of the benefits for you, my first > inclination is to say thumbs down on a separate components folder. > Adding complexity for my end users is not a feature in my eyes. Where you see complexity? Explain me please. And I can say we have no other way. Because Valentina 2.0 is SEVERAL times more complex projects of Valentina 1.x. Valentina 2.0 depend on several BIG third party libraries. The main problem that NOT all libraries allow link them, they are build only in the form of DLLs. For example IBM ICU library or ImageMagic. Even if assume we can link all in one, then we will get V4RB plugin for single platform about 12-15MB of size. As you know REALbasic require that all 3 or even 4 plugins for each platform was inside of single plugin file. Image the size of this V4RB plugin! Close to 50MB. Yes this will affect only you Brendan, but not your users. Although your users also will get always YOUR APP with size your size + 15MB of Valentina size. Although Valentina 2.0 itself is only 3.5MB. ----------- We believe that it is much better to have component based system. At last of end look around. Where you have see complex systems ALL IN ONE? Why all big systems use CORBA, COM, .NET ? Just because when system grow to some level it is not possible and not effective keep all in one. Component system in ideal must allow YOU to skip some parts of system which you do not use. For example, if you not use Pictures, then simply remove from YOUR Vcomponents folder the image library. If you want VERY VERY small and simple database without SQL then remove HUGE SQL parser and executor overhead, remove ANTLR.dll and make Valentina engine 2 times smaller for your users. Can you do such tricks if all in one ? ------------ And we believe (hope) that if Valentina will have own plugin system, Then third party talented C++ developers will be able develop own versions of fields, functions, indexes, importers, compressors ... How many people dream about DBF import/export. But we want not able yet implement it. No time. With plugin system this can do somebody else. We believe that OPEN SYSTEM grow faster of closed private one. And so on and so on. ------------- I was sure that everybody on this list have told: we want better SQL parser we want better SQL commands more features -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Tue May 11 13:53:49 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Tue May 11 12:53:56 2004 Subject: ICU / icudt26b.dat file on OS X In-Reply-To: References: Message-ID: <283B7D3A-A374-11D8-986B-000A95BE1A8E@mindstarprods.com> Hi Ruslan, On May 11, 2004, at 10:25 AM, Ruslan Zasukhin wrote: > Hi All, > > Wow, I have found this big 8 MB file in the OS X ! > It is located in the > > /usr/share/icu > > Please check that you also have it on your computers. > Interesting, it looks that OS X 10.2 already have it also. Yes, it's on my Macs. Running 10.3.3 (also on 10.3.3 Server). I can ask around about what OS versions it is supposed to appear if you want? --Frank From frank-list2 at mindstarprods.com Tue May 11 14:09:17 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Tue May 11 13:09:22 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: References: Message-ID: <518A4F24-A376-11D8-986B-000A95BE1A8E@mindstarprods.com> Hi Ruslan, On May 11, 2004, at 1:27 PM, Ruslan Zasukhin wrote: > On 5/11/04 6:05 PM, "Brendan Murphy" wrote: > > > ATTENTION: > > Please go to Valentina-beta list for 2.0 discussion > > http://lists.macserve.net/mailman/listinfo/valentina-beta > > > ------------------------ >>> Valentina 2.0 is much more complex by structure, >>> So it is many advantages to keep components separately. >>> Valentina 2.0 have own plugin system. >> >> Please explain in more detail, what are these advantages? I want >> to use 2.0 as I use 1.10 today as a totally embedded system within >> my application. Having a separate folder next to my application is >> just another support hassle I would rather avoid. > > I believe that we get problems similar to DLL hell, > Only if installation of Vcomponents folder is made into some > __central__ place on computer. > > But IF Valentina developer will put Vcomponents folder into app folder, > Then no problems. On the same computer can live SEVERAL Valentina-made > applications and they will not conflict in any case, even if they will > use > different versions of Valentina. The problem is that now we cannot make a Valentina app that is a standalone file. Now we have to create a folder for our Valentina app and place this VComponents folder inside it. Now users will wonder what that is. I think we need to be able to optionally hide this folder from the user in some way. Maybe allow it to be in the "Application Support" folder (either from /Library or ~/Library) for the application? Also, I would rather be able to just add the VComponents inside my Realbasic app so it is not visible at all! Is this possible, at least as an option? As you know, a Mac OS X application is simply a folder. I know it will make the RB (or whatever) app much bigger, but I'm not too worried about that sometimes. So can we somehow have these 4 options (1. Your original method - in /Library/CFMSupport/), 2. Your proposal - Inside the same folder as the application, 3. Inside the "Application Support" folder, 4. Inside the app itself)? > IF you SELF will not distribute separate .dlls from Vcomponents > folder to > your users then also no problems. > If you decide to provide your users with some updates then this is up > to > you. You can make installer to do that. I agree with the philosophy, although it is generally not "Mac-like". I think with the alternatives I propose it might make everyone happy? >> Without further explanation of the benefits for you, my first >> inclination is to say thumbs down on a separate components folder. >> Adding complexity for my end users is not a feature in my eyes. > > Where you see complexity? > Explain me please. Most users expect a single clickable app. But honestly, many important Mac applications are complex like this: Bias Peak, Adobe Photoshop, Illustrator, InDesign, ok... every Adobe app :^). > And I can say we have no other way. Because Valentina 2.0 is SEVERAL > times > more complex projects of Valentina 1.x. Valentina 2.0 depend on > several BIG > third party libraries. > > The main problem that NOT all libraries allow link them, they are > build only > in the form of DLLs. For example IBM ICU library or ImageMagic. > > Even if assume we can link all in one, then we will get V4RB plugin for > single platform about 12-15MB of size. As you know REALbasic require > that > all 3 or even 4 plugins for each platform was inside of single plugin > file. > Image the size of this V4RB plugin! Close to 50MB. Yes this will > affect only > you Brendan, but not your users. > > Although your users also will get always YOUR APP with size > your size + 15MB of Valentina size. > > Although Valentina 2.0 itself is only 3.5MB. > > ----------- > We believe that it is much better to have component based system. > At last of end look around. Where you have see complex systems ALL IN > ONE? > Why all big systems use CORBA, COM, .NET ? > Just because when system grow to some level it is not possible and not > effective keep all in one. > > Component system in ideal must allow YOU to skip some parts of system > which > you do not use. For example, if you not use Pictures, then simply > remove > from YOUR Vcomponents folder the image library. If you want VERY VERY > small > and simple database without SQL then remove HUGE SQL parser and > executor > overhead, remove ANTLR.dll and make Valentina engine 2 times smaller > for > your users. > > Can you do such tricks if all in one ? > > > ------------ > And we believe (hope) that if Valentina will have own plugin system, > Then third party talented C++ developers will be able develop own > versions > of fields, functions, indexes, importers, compressors ... > > How many people dream about DBF import/export. > But we want not able yet implement it. No time. > With plugin system this can do somebody else. > > We believe that OPEN SYSTEM grow faster of closed private one. > And so on and so on. > > > ------------- > I was sure that everybody on this list have told: > we want better SQL parser > we want better SQL commands > more features From sunshine at public.kherson.ua Tue May 11 21:15:03 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 13:15:11 2004 Subject: ICU / icudt26b.dat file on OS X In-Reply-To: <283B7D3A-A374-11D8-986B-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 8:53 PM, "Frank Schima" wrote: > Hi Ruslan, > > > On May 11, 2004, at 10:25 AM, Ruslan Zasukhin wrote: > >> Hi All, >> >> Wow, I have found this big 8 MB file in the OS X ! >> It is located in the >> >> /usr/share/icu >> >> Please check that you also have it on your computers. >> Interesting, it looks that OS X 10.2 already have it also. > > Yes, it's on my Macs. Running 10.3.3 (also on 10.3.3 Server). I can ask > around about what OS versions it is supposed to appear if you want? Yes Frank, On IBM list I have read that Apple have start use it in 10.3 But I see it on my 10.2 So that guys can mistake. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue May 11 14:22:35 2004 From: jda at his.com (jda) Date: Tue May 11 13:22:46 2004 Subject: ICU / icudt26b.dat file on OS X In-Reply-To: References: Message-ID: > > > >> Can you put the pathname in non-Unix terms please? I can only find an >> empty folder called ICU in the Application Support folder. > >John, I think you should in search add "visible all" > >Path must be > > system volume:usr:share:icu Ah, thanks. Yes, it's there (OS X 10.3.3) > >> As to your question, I would very much like to *avoid* distributing >> the unicode library for OS X if possible! > >> So I would certainly vote for allowing us to use the OS X version >>(we have to >> build different distribution packages for each platform anyway, so >>why not let >> us tailor them to best take advantage of the libraries already installed by >> the OS? > >Agree, but you sure you will make different archives? > >V4RB is CARBON, i.e. ready for OS 9 also. Not me -- I'm abandoning Classic in my next major upgrade. There are just too many cool things you can do easily in OS X that are a pain in Classic. FWIW, when I first started selling my Carbon app in 2001 the downloads were split evenly (1:1) between OS X and Classic. Now they are at least 10:1 in favor of OS X. So my users are also telling me what to do... > >So you will make 2 Carbon archives? For OS X and OS 9? >I think we will not have Classic target for 2.0. >Because nothing modern not works in Classic: threads, icu, ... > Agree. Jon From sunshine at public.kherson.ua Tue May 11 21:23:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 13:23:54 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <518A4F24-A376-11D8-986B-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 9:09 PM, "Frank Schima" wrote: >> I believe that we get problems similar to DLL hell, >> Only if installation of Vcomponents folder is made into some >> __central__ place on computer. >> >> But IF Valentina developer will put Vcomponents folder into app folder, >> Then no problems. On the same computer can live SEVERAL Valentina-made >> applications and they will not conflict in any case, even if they will >> use >> different versions of Valentina. > > The problem is that now we cannot make a Valentina app that is a > standalone file. Now we have to create a folder for our Valentina app > and place this VComponents folder inside it. Now users will wonder what > that is. > > I think we need to be able to optionally hide this folder from the user > in some way. Maybe allow it to be in the "Application Support" folder > (either from /Library or ~/Library) for the application? You talk about central place. > Also, I would rather be able to just add the VComponents inside my > Realbasic app so it is not visible at all! Is this possible, at least > as an option? As you know, a Mac OS X application is simply a folder. I > know it will make the RB (or whatever) app much bigger, but I'm not too > worried about that sometimes. > > So can we somehow have these 4 options (1. Your original method - in > /Library/CFMSupport/), 2. Your proposal - Inside the same folder as the > application, 3. Inside the "Application Support" folder, 4. Inside the > app itself)? NO PROBLEMS, Frank. Yes we can make this easy. For this we will have 1) auto-search of Vcomponents folder in expected places 2) Valentina developer will get function to say DIRECTLY where it is located. btw, this is nice to do in Valentina module Valentina.VComponentsPath = ... Valentina.icu8mbfilepath = ... Valentina.Init( cache, mac serial, win, ... ) --------- Another note. I have say that central place is bad idea. But not always. If developer develop some solution in RB and Driector Then using central place for Vcomponents folder he will reduce total size of solution. The right way to do this: - create folder "MyCompany support" - keep here Vcomponents folder So your Vcomponents folder will not conflict to others. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 21:26:14 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 13:26:22 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <518A4F24-A376-11D8-986B-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/11/04 9:09 PM, "Frank Schima" wrote: > Also, I would rather be able to just add the VComponents inside my > Realbasic app so it is not visible at all! Is this possible, at least > as an option? As you know, a Mac OS X application is simply a folder. I > know it will make the RB (or whatever) app much bigger, but I'm not too > worried about that sometimes. ......... >> Where you see complexity? >> Explain me please. > > Most users expect a single clickable app. But honestly, many important > Mac applications are complex like this: Bias Peak, Adobe Photoshop, > Illustrator, InDesign, ok... every Adobe app :^). Btw, Adobe also use ICU and ACE frameworks. Guys, look on Valentina Studio for OS X. It Is made as package. Inside exists many folders. So do not worry. We will have nice MAC - way. And exactly as Apple propose -- using app packages. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From Claudius at sailer-online.de Tue May 11 20:28:37 2004 From: Claudius at sailer-online.de (Claudius Sailer) Date: Tue May 11 13:28:42 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: References: Message-ID: <04CDE382-A379-11D8-90C2-00039365848C@sailer-online.de> Am 11. Mai 2004 um 19:27 Uhr schrieb Ruslan Zasukhin: > Component system in ideal must allow YOU to skip some parts of system > which > you do not use. For example, if you not use Pictures, then simply > remove > from YOUR Vcomponents folder the image library. If you want VERY VERY > small > and simple database without SQL then remove HUGE SQL parser and > executor > overhead, remove ANTLR.dll and make Valentina engine 2 times smaller > for > your users. > > Can you do such tricks if all in one ? When I can generate an app and also hold the VComponents in the Application Folder, than I can say to the users this is necessary that this is all together, but when they have to install something in /Library/CFMSupport this will be a little bit a problem, because it brings much more possible problems to me to support users by handling their problems. But to shorten this discussion, the best solution would be a description what possibilities would work. For me as an Single-User-App Programmer it is necessary that as less as possible files the user need to handle und install. At the moment he has only to unpack a file and move it to the place he wants. Klick on the Icon and it works. Moves the folder to an other place and it works. Easy handling. I believe!! bye Claudius From jda at his.com Tue May 11 15:09:40 2004 From: jda at his.com (jda) Date: Tue May 11 14:09:52 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <13F6576F-A37C-11D8-9ECB-0003935B6750@comcast.net> References: <20040511172724.E1DBF120C71@edison.macserve.net> <13F6576F-A37C-11D8-9ECB-0003935B6750@comcast.net> Message-ID: I'm redirecting this to the betas list, where it belongs. >Ruslan, you are making the some the same mistakes here that RS has >made in the past. They choose an implementation without regard to >how it impacts the customers since it serves their perspective. > >WE NEED this component architecture to be an option! Having files >spread out all over the place is an invitation to a support >nightmare. End user's screw with their systems is a plain fact of >life. The more split up an application is, the more possible >points of failure you have. If a customer sees the component >folder one day and says, "What's this?" and decides to do some >creative deleting or updating, suddenly my application doesn't >work and I get a call. This cost me money!!! > >If you don't do something to solve this problem, I will be a very >pissed off customer and I am sure there are a few others out there >who will feel the same way. I can't express what a negative >feeling in me this component architecture creates since you are >making unilateral design decision for my product. Bad idea. Very >bad idea! Ruslan has already said you'll have an option, so relax. And frankly, I don't think it's a bad idea at all. Speaking from a Mac-centric point of view at least. Lots of apps (including mine) have shared libraries and resources. They reside in the same folder as the app. I have *never* had anyone complain about this, and *never* had a tech support problem with it in three years (and thousands of users). The only issue I see is one of scattered files across the system if one removes the app. But that's not a problem if they are all in the same folder. The sky is not falling. Jon From sunshine at public.kherson.ua Tue May 11 22:29:35 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 14:29:42 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: Message-ID: On 5/11/04 10:16 PM, "Stan Busk" wrote: Stan, Brandan, I wonder if you have read suggestion of Frank (which work in Apple as I know) and my answer: that you will be able provide to your users SINGLE FILE of application as Package. This is a MODERN WAY of Apple and MAC today. What problems guys ??? Yes, this will require A LITTLE more work from YOU as developer. > I second this. All my Valentina-based apps are currently compact and > self contained. With v2 the only external libraries will be > Valentina's! Why? Why can't I continue creating compact products? Why > don't RB starts doing the same with plugins... this would be a > nightmare.... BIG MISTAKE RUSLAN! VERY BIG! That is Mac not Windows. > For external junk lets do Window... > > ~/Stan > > >> Ruslan, you are making the some the same mistakes here that RS has >> made in the past. They choose an implementation without regard to >> how it impacts the customers since it serves their perspective. >> >> WE NEED this component architecture to be an option! Having files >> spread out all over the place is an invitation to a support >> nightmare. End user's screw with their systems is a plain fact of >> life. The more split up an application is, the more possible >> points of failure you have. If a customer sees the component >> folder one day and says, "What's this?" and decides to do some >> creative deleting or updating, suddenly my application doesn't >> work and I get a call. This cost me money!!! >> >> If you don't do something to solve this problem, I will be a very >> pissed off customer and I am sure there are a few others out there >> who will feel the same way. I can't express what a negative >> feeling in me this component architecture creates since you are >> making unilateral design decision for my product. Bad idea. Very >> bad idea! >> >> I offer to you an idea for you to implement your component idea >> and satisfy those who need an all in one plugin. Create a helper >> application that will take the contents of the components folder >> etc. and package them into one plugin. So when I download the V4RB >> SDK, I will customize my components folder and then run the helper >> application which will produce a V4RB plugin that I will use in >> RB. This way I control what my end user sees and you get to have >> your component architecture. Problem solved. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 22:38:03 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 14:38:11 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: Message-ID: On 5/11/04 10:09 PM, "jda" wrote: > I'm redirecting this to the betas list, where it belongs. I once again have sent this to Valentina list Because it seems Brendan still not subscribed to beta list. Brendan? >> Ruslan, you are making the some the same mistakes here that RS has >> made in the past. They choose an implementation without regard to >> how it impacts the customers since it serves their perspective. >> >> WE NEED this component architecture to be an option! Having files >> spread out all over the place is an invitation to a support >> nightmare. End user's screw with their systems is a plain fact of >> life. The more split up an application is, the more possible >> points of failure you have. If a customer sees the component >> folder one day and says, "What's this?" and decides to do some >> creative deleting or updating, suddenly my application doesn't >> work and I get a call. This cost me money!!! >> >> If you don't do something to solve this problem, I will be a very >> pissed off customer and I am sure there are a few others out there >> who will feel the same way. I can't express what a negative >> feeling in me this component architecture creates since you are >> making unilateral design decision for my product. Bad idea. Very >> bad idea! > > Ruslan has already said you'll have an option, so relax. And frankly, > I don't think it's a bad idea at all. Speaking from a Mac-centric > point of view at least. Lots of apps (including mine) have shared > libraries and resources. They reside in the same folder as the app. I > have *never* had anyone complain about this, and *never* had a tech > support problem with it in three years (and thousands of users). The > only issue I see is one of scattered files across the system if one > removes the app. But that's not a problem if they are all in the same > folder. The sky is not falling. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 23:02:54 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 15:03:03 2004 Subject: RB, how to define system volume? Message-ID: Anybody know if we can in REALbasic get the name of system OS X volume? I.e. I ask about function getSystemVolume() That returns "MacOS X" name? Or folder item for it. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Tue May 11 22:51:41 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 15:17:39 2004 Subject: [V4RB] Valentina 2.0 installation // RESUME. In-Reply-To: <04CDE382-A379-11D8-90C2-00039365848C@sailer-online.de> Message-ID: On 5/11/04 9:28 PM, "Claudius Sailer" wrote: > When I can generate an app and also hold the VComponents in the > Application Folder, than I can say to the users this is necessary that > this is all together, but when they have to install something in > /Library/CFMSupport this will be a little bit a problem, because it > brings much more possible problems to me to support users by handling > their problems. > > But to shorten this discussion, the best solution would be a > description what possibilities would work. For me as an Single-User-App > Programmer it is necessary that as less as possible files the user need > to handle und install. > At the moment he has only to unpack a file and move it to the place he > wants. Klick on the Icon and it works. Moves the folder to an other > place and it works. Easy handling. I believe!! Okay guys: RESUME: We will have at least 3 options A) install Vcomponents folder into app folder. but note this is old OS 9 way. Apple for OS X have invent new way - Packages. So B) PACKAGES. If somebody do now know what is this then please do next: find app Valentina Studio, or iTunes or iMovie or REALbasic 5.5 itself click by mouse on app to open context menu choose here item "Show package context" you will see a miracle :-) file of application will become a folder. go inside of that folder and you will find here how modern Apple Application must looks. Btw, if talk about Packages. It is task of REALbasic to build applications as packages. If REALbasic 5.5 cannot do that then we all must press them to implement this feature. CodeWarrior and xCode do this. C) Vcomponents folder is in some central places. Some developers/app/solutions may find this better. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Tue May 11 16:30:54 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 15:30:26 2004 Subject: RB, how to define system volume? In-Reply-To: References: Message-ID: <19FF5574-A38A-11D8-AB99-003065BB0634@desuetude.com> On May 11, 2004, at 4:02 PM, Ruslan Zasukhin wrote: > > Anybody know if we can in REALbasic get the name of system OS X volume? > > I.e. I ask about function > > getSystemVolume() > > That returns "MacOS X" name? Or folder item for it. > Sure; I have code to do this at . -------------- Charles Yeomans From maxprog at mac.com Tue May 11 23:08:36 2004 From: maxprog at mac.com (Stan Busk) Date: Tue May 11 16:08:47 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: References: Message-ID: <5E63CE70-A38F-11D8-A30F-000A95C377AA@mac.com> Hi, I think this is too complex and time consuming as you need to create the packages for each distribution... I develop in 7 languages and 3 platforms, imagine what I need to do. This is getting each day worst and worst. Not to say new people coming to Valentina for the first time, I think most of them will stick to RB built-in database capabilities as I should have done myself *or* you should ask RB to include the possibility to create packages at compile time. ~/stan > On 5/11/04 10:16 PM, "Stan Busk" wrote: > > Stan, Brandan, > > I wonder if you have read suggestion of Frank (which work in Apple as I > know) and my answer: that you will be able provide to your users > > SINGLE FILE of application as Package. > > This is a MODERN WAY of Apple and MAC today. > > What problems guys ??? > > Yes, this will require A LITTLE more work from YOU as developer. > > >> I second this. All my Valentina-based apps are currently compact and >> self contained. With v2 the only external libraries will be >> Valentina's! Why? Why can't I continue creating compact products? Why >> don't RB starts doing the same with plugins... this would be a >> nightmare.... BIG MISTAKE RUSLAN! VERY BIG! That is Mac not Windows. >> For external junk lets do Window... >> >> ~/Stan >> >> >>> Ruslan, you are making the some the same mistakes here that RS has >>> made in the past. They choose an implementation without regard to >>> how it impacts the customers since it serves their perspective. >>> >>> WE NEED this component architecture to be an option! Having files >>> spread out all over the place is an invitation to a support >>> nightmare. End user's screw with their systems is a plain fact of >>> life. The more split up an application is, the more possible >>> points of failure you have. If a customer sees the component >>> folder one day and says, "What's this?" and decides to do some >>> creative deleting or updating, suddenly my application doesn't >>> work and I get a call. This cost me money!!! >>> >>> If you don't do something to solve this problem, I will be a very >>> pissed off customer and I am sure there are a few others out there >>> who will feel the same way. I can't express what a negative >>> feeling in me this component architecture creates since you are >>> making unilateral design decision for my product. Bad idea. Very >>> bad idea! >>> >>> I offer to you an idea for you to implement your component idea >>> and satisfy those who need an all in one plugin. Create a helper >>> application that will take the contents of the components folder >>> etc. and package them into one plugin. So when I download the V4RB >>> SDK, I will customize my components folder and then run the helper >>> application which will produce a V4RB plugin that I will use in >>> RB. This way I control what my end user sees and you get to have >>> your component architecture. Problem solved. > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > > From sunshine at public.kherson.ua Wed May 12 00:21:08 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 16:21:17 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <5E63CE70-A38F-11D8-A30F-000A95C377AA@mac.com> Message-ID: On 5/12/04 12:08 AM, "Stan Busk" wrote: > Hi, > > I think this is too complex and time consuming as you need to create > the packages for each distribution... I develop in 7 languages and 3 > platforms, imagine what I need to do. This is getting each day worst > and worst. Not to say new people coming to Valentina for the first > time, I think most of them will stick to RB built-in database > capabilities as I should have done myself *or* you should ask RB to > include the possibility to create packages at compile time. I think we all should talk with them on this. In any case this is important feature for RB. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From maxprog at mac.com Tue May 11 23:34:45 2004 From: maxprog at mac.com (Stan Busk) Date: Tue May 11 16:34:55 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: References: Message-ID: <056444B4-A393-11D8-A30F-000A95C377AA@mac.com> Well, you can always compile to Mach-O format with RB as it generates the package automatically but you lose Mac OS 9 compatibility so you need to create separate applications for each system. I just wonder if the Mach-O format is currently bug free. It would be a solution. ~/Stan > On 5/12/04 12:08 AM, "Stan Busk" wrote: > >> Hi, >> >> I think this is too complex and time consuming as you need to create >> the packages for each distribution... I develop in 7 languages and 3 >> platforms, imagine what I need to do. This is getting each day worst >> and worst. Not to say new people coming to Valentina for the first >> time, I think most of them will stick to RB built-in database >> capabilities as I should have done myself *or* you should ask RB to >> include the possibility to create packages at compile time. > > I think we all should talk with them on this. > In any case this is important feature for RB. > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > > From yeomans at desuetude.com Tue May 11 17:40:42 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 16:40:14 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <056444B4-A393-11D8-A30F-000A95C377AA@mac.com> References: <056444B4-A393-11D8-A30F-000A95C377AA@mac.com> Message-ID: I've been using MachO compiles without any problems. And you can use the bundle created by a MachO build as the basis for a Carbon PEF app; simply remove the MachO app and Rb framework library, and insert the Carbon PEF app. Charles Yeomans On May 11, 2004, at 5:34 PM, Stan Busk wrote: > > Well, you can always compile to Mach-O format with RB as it generates > the package automatically but you lose Mac OS 9 compatibility so you > need to create separate applications for each system. I just wonder if > the Mach-O format is currently bug free. It would be a solution. > > ~/Stan > >> On 5/12/04 12:08 AM, "Stan Busk" wrote: >> >>> Hi, >>> >>> I think this is too complex and time consuming as you need to create >>> the packages for each distribution... I develop in 7 languages and 3 >>> platforms, imagine what I need to do. This is getting each day worst >>> and worst. Not to say new people coming to Valentina for the first >>> time, I think most of them will stick to RB built-in database >>> capabilities as I should have done myself *or* you should ask RB to >>> include the possibility to create packages at compile time. >> From yeomans at desuetude.com Tue May 11 17:43:11 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 16:42:40 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: References: Message-ID: <3329A015-A394-11D8-AB99-003065BB0634@desuetude.com> On May 11, 2004, at 5:21 PM, Ruslan Zasukhin wrote: > On 5/12/04 12:08 AM, "Stan Busk" wrote: > >> Hi, >> >> I think this is too complex and time consuming as you need to create >> the packages for each distribution... I develop in 7 languages and 3 >> platforms, imagine what I need to do. This is getting each day worst >> and worst. Not to say new people coming to Valentina for the first >> time, I think most of them will stick to RB built-in database >> capabilities as I should have done myself *or* you should ask RB to >> include the possibility to create packages at compile time. > > I think we all should talk with them on this. > In any case this is important feature for RB This would be a good time to do so, since RS is rewriting the IDE from the ground up for Rb 6; the previews they showed at RW were very interesting. Charles Yeomans From sunshine at public.kherson.ua Wed May 12 00:45:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 16:45:56 2004 Subject: [V4RB] Valentina 2.0 installation In-Reply-To: <056444B4-A393-11D8-A30F-000A95C377AA@mac.com> Message-ID: On 5/12/04 12:34 AM, "Stan Busk" wrote: Actually I suggest next point. Right now our primary task is - develop and polish algorithms, formats, examples, tests, docs. Packaging is THE LAST issue in this game. I hope you all know, that IF somebody of you will offer some great way, and that is technically possible then we will implement that. So do not worry. We do not forget that you are developers and we make TOOLS for developers. We always keep this in mind. > Well, you can always compile to Mach-O format with RB as it generates > the package automatically but you lose Mac OS 9 compatibility so you > need to create separate applications for each system. I just wonder if > the Mach-O format is currently bug free. It would be a solution. > > ~/Stan > >> On 5/12/04 12:08 AM, "Stan Busk" wrote: >> >>> Hi, >>> >>> I think this is too complex and time consuming as you need to create >>> the packages for each distribution... I develop in 7 languages and 3 >>> platforms, imagine what I need to do. This is getting each day worst >>> and worst. Not to say new people coming to Valentina for the first >>> time, I think most of them will stick to RB built-in database >>> capabilities as I should have done myself *or* you should ask RB to >>> include the possibility to create packages at compile time. >> >> I think we all should talk with them on this. >> In any case this is important feature for RB. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed May 12 00:53:09 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 16:53:17 2004 Subject: FW: realbasic-nug Digest, Vol 2, Issue 71 In-Reply-To: <20040511214210.CB5C9149441@lists.realsoftware.com> Message-ID: Claudius already have got answer from REAL on this question. ------ Forwarded Message From: realbasic-nug-request@lists.realsoftware.com Reply-To: realbasic-nug@lists.realsoftware.com Date: Tue, 11 May 2004 16:42:10 -0500 (CDT) To: realbasic-nug@lists.realsoftware.com Subject: realbasic-nug Digest, Vol 2, Issue 71 Message: 9 Subject: Re: Making Packages From: "Joseph J. Strout" Date: Tue, 11 May 2004 15:19:22 -0500 At 10:00 PM +0200 5/11/04, Claudius Sailer wrote: >I want to make packages out of my application. What do I have to do, >that my Application looks like a package, like other MacOS X >Applications. Is there any description online? Any link? Or is this >impossible with RB-Applications? If you make a Mach-O app, then it's a package already. If you make a PEF app, then you have to do it yourself. One easy way to do that is to first make a Mach-O app, then reach into it and replace the executable with your PEF version. Another way is to use Thomas Reed's AppBundler (that's from memory, hopefully I got it right!). Finally, of course, you could just do it manually based on Apple's docs -- it's just a little folder hierarchy with a certain structure. Best, - Joe ------ End of Forwarded Message From sunshine at public.kherson.ua Wed May 12 01:47:58 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 11 17:48:05 2004 Subject: V4RB, get_CacheSize Message-ID: If I will use syntax as z = Valentina.get_CacheSize And say Valentina.put_CacheSize( z ) Anybody see problem wish names? I mean get_ put_ In Valentina module we cannot have properties, So I want to use such names. This will show that virtually this is property "CacheSize" We use such names in our C++ code. This is similar to COM names. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue May 11 18:52:43 2004 From: jda at his.com (jda) Date: Tue May 11 17:52:50 2004 Subject: V4RB, get_CacheSize In-Reply-To: References: Message-ID: >If I will use syntax as > > > z = Valentina.get_CacheSize > >And say > > Valentina.put_CacheSize( z ) > > >Anybody see problem wish names? > >I mean get_ put_ > > >In Valentina module we cannot have properties, >So I want to use such names. >This will show that virtually this is property "CacheSize" > > > >We use such names in our C++ code. >This is similar to COM names. > > > How about "set" instead of "put"? And as Charles said, why the ugly underscores? Why not GetCacheSize and SetCacheSize? Jon From frank-list2 at mindstarprods.com Tue May 11 19:03:10 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Tue May 11 18:03:19 2004 Subject: V4RB, get_CacheSize In-Reply-To: References: Message-ID: <5F4844D2-A39F-11D8-92CD-000A95BE1A8E@mindstarprods.com> On May 11, 2004, at 6:52 PM, jda wrote: > > And as Charles said, why the ugly underscores? Why not GetCacheSize > and SetCacheSize? Agreed. -Frank From yeomans at desuetude.com Tue May 11 19:59:16 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 18:58:48 2004 Subject: V4RB, get_CacheSize In-Reply-To: References: Message-ID: <35E7BF90-A3A7-11D8-AE46-003065BB0634@desuetude.com> On May 11, 2004, at 6:47 PM, Ruslan Zasukhin wrote: > > If I will use syntax as > > > z = Valentina.get_CacheSize > > And say > > Valentina.put_CacheSize( z ) > > > Anybody see problem wish names? > > I mean get_ put_ > > > In Valentina module we cannot have properties, > So I want to use such names. > This will show that virtually this is property "CacheSize" > Really? Is it true that one cannot define a module property in a plugin? Charles Yeomans From yeomans at desuetude.com Tue May 11 20:46:02 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 11 19:45:35 2004 Subject: V4RB, get_CacheSize In-Reply-To: References: Message-ID: On May 11, 2004, at 6:47 PM, Ruslan Zasukhin wrote: > > If I will use syntax as > > > z = Valentina.get_CacheSize > > And say > > Valentina.put_CacheSize( z ) > > > Anybody see problem wish names? > > I mean get_ put_ Probably GetCacheSize and SetCacheSize would sound better to Rb users. Can the cache size be changed while Valentina is active? > > > In Valentina module we cannot have properties, > So I want to use such names. > This will show that virtually this is property "CacheSize" > > > > We use such names in our C++ code. > This is similar to COM names. > I thought you were discouraging people from confusing Paradigma with Microsoft. Charles Yeomans From sunshine at public.kherson.ua Wed May 12 08:59:40 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 12 00:59:47 2004 Subject: V4RB, get_CacheSize In-Reply-To: Message-ID: On 5/12/04 3:46 AM, "Charles Yeomans" wrote: >> If I will use syntax as >> >> >> z = Valentina.get_CacheSize >> >> And say >> >> Valentina.put_CacheSize( z ) >> >> >> Anybody see problem wish names? >> >> I mean get_ put_ > > Probably GetCacheSize and SetCacheSize would sound better to Rb users. Okay. And when REAL implement properties we will be able change this. > Can the cache size be changed while Valentina is active? No, that was example. We can have DebugLevel as property. >> We use such names in our C++ code. >> This is similar to COM names. >> > > I thought you were discouraging people from confusing Paradigma with > Microsoft. Okay okay, get_ set_ -> Microsoft Get Put -> CORBA :-) I do not remember exactly where how. BTW, in our C++ code we use // get_Name() put_Name() // methods NextRecord() C++ do not have properties. So we imitate them in this way. And get_ put_ very clean show us where is property. Mathods go from capital letter. Although for RB language case - insensitive. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu May 13 13:00:45 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 13 05:01:16 2004 Subject: V4MD WIN 2.0 b1. Wanted? Message-ID: Hi Guys, We are ready to upload V4MD WIN 2.0 first beta. I have see a lots of activity of RB developers on Valentina beta list, So I wonder if somebody from V4MD developers is going take part in beta testing? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu May 13 13:13:49 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 13 05:13:57 2004 Subject: [ANN] V4RB 2.0b2 Message-ID: Hi All, You need download Vengine archive and V4RB. If you have already Vcomponents then no need download it. ----------------- 1) now plugin must work, it self must locate Vcomponents folder on your computer. 2) I have add "Valentina" module. So all examples updated to syntax Valentina.Init Valentina.shutDown .. Mars have show how to simulate properties, I will do this in the next build. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu May 13 13:24:05 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 13 05:24:19 2004 Subject: V4MD WIN 2.0 b1. Wanted? In-Reply-To: <001b01c438d3$8d550230$1d0a11ac@wgregoire> Message-ID: On 5/13/04 1:17 PM, "Gregoire William" wrote: >> For now we get from Director ANSI string in default system encoding >> (may be we will need add option to see encoding >> different from system) >> And we convert strings to UTF16 before send to Valentina. > > and what about VMD->Director ? We convert back to ANSI system. Even this feature, must significantly improve work with such languages as Cyrillic and others non-Western. Also we must be able store 1-byte languages into UTF8 on disk to use less space. > with this way, it's impossible to show Japanese on a French Windows for > example :( But what about Japan on Japan OS? >> Director developers, please share info about Director and >> Unicode to us. >> What news? > > DMX2004 still lacks UNICODE support I know this. >> What promises from Macromedia? > > all and nothing, as always...they spend all their times on buggy-Flash bad >> What workarounds? > > use Flash INSIDE Director with XML and Arial Unicode installed > but it's good to SHOW text, not to INPUT them > (it's impossible to dev a Translation Engine for exemple) > >> Any other Xtra can use at least UTF8? >> Advices? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From pschwarz at lexialearning.com Thu May 13 06:33:57 2004 From: pschwarz at lexialearning.com (Paul Schwarz) Date: Thu May 13 05:34:10 2004 Subject: V4MD WIN 2.0 b1. Wanted? In-Reply-To: <20040513100121.1E779122D9A@edison.macserve.net> Message-ID: <001d01c438d5$cc7b03e0$0300a8c0@testLab8> Hi Ruslan, We will! We will! :> We're very excited to get started. Thanks Ruslan, -Paul > Date: Thu, 13 May 2004 13:00:45 +0300 > From: Ruslan Zasukhin > Subject: V4MD WIN 2.0 b1. Wanted? > To: , > > Message-ID: > Content-Type: text/plain; charset="US-ASCII" > > Hi Guys, > > We are ready to upload V4MD WIN 2.0 first beta. > > I have see a lots of activity of RB developers on Valentina > beta list, So I wonder if somebody from V4MD developers is > going take part in beta testing? > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] From rbarber at yhb.att.ne.jp Thu May 13 20:27:57 2004 From: rbarber at yhb.att.ne.jp (ron barber) Date: Thu May 13 06:27:05 2004 Subject: V4MD WIN 2.0 b1. Wanted? In-Reply-To: References: Message-ID: <9553E7F8-A4D0-11D8-A152-000A95DAEEF0@yhb.att.ne.jp> Hi Ruslan sometime, please be sure to provide a v2.0 for Revolution. Thanks Ron On May 13, 2004, at 7:00 PM, Ruslan Zasukhin wrote: > Hi Guys, > > We are ready to upload V4MD WIN 2.0 first beta. > > I have see a lots of activity of RB developers on Valentina beta list, > So I wonder if somebody from V4MD developers is going take part in beta > testing? > > > -- > Best regards, > Ruslan Zasukhin [ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > From jda at his.com Thu May 13 07:51:34 2004 From: jda at his.com (jda) Date: Thu May 13 06:51:43 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: > >1) now plugin must work, it self must locate Vcomponents folder on your >computer. > Sorry, Ruslan. It crashes in the same place as before: Valentina.Init( 8 * 1024 * 1024 ) No log, but the empty file VSQL_TreeParser.txt is created. Jon From frank-list2 at mindstarprods.com Thu May 13 09:00:07 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Thu May 13 08:00:37 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: <75B5EBE9-A4DD-11D8-92CD-000A95BE1A8E@mindstarprods.com> On May 13, 2004, at 7:51 AM, jda wrote: >> >> 1) now plugin must work, it self must locate Vcomponents folder on >> your >> computer. >> > > Sorry, Ruslan. It crashes in the same place as before: > > Valentina.Init( 8 * 1024 * 1024 ) > > No log, but the empty file VSQL_TreeParser.txt is created. I see the same thing :^( Plus, the Example "Database_CreateCloseOpen" does not even compile until you change "ValentinaInit" to "Valentina.Init". -Frank From sunshine at public.kherson.ua Thu May 13 17:52:40 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 13 09:52:49 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: <75B5EBE9-A4DD-11D8-92CD-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/13/04 4:00 PM, "Frank Schima" wrote: > > On May 13, 2004, at 7:51 AM, jda wrote: > >>> >>> 1) now plugin must work, it self must locate Vcomponents folder on >>> your >>> computer. >>> >> >> Sorry, Ruslan. It crashes in the same place as before: >> >> Valentina.Init( 8 * 1024 * 1024 ) >> >> No log, but the empty file VSQL_TreeParser.txt is created. > > I see the same thing :^( > > Plus, the Example "Database_CreateCloseOpen" does not even compile > until you change "ValentinaInit" to "Valentina.Init". Hmm. Very bad. Then I need add file log to debug this problem. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Thu May 13 17:53:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 13 09:53:42 2004 Subject: V4MD WIN 2.0 b1. Wanted? In-Reply-To: <9553E7F8-A4D0-11D8-A152-000A95DAEEF0@yhb.att.ne.jp> Message-ID: On 5/13/04 2:27 PM, "ron barber" wrote: > Hi Ruslan > > sometime, please be sure to provide a v2.0 for Revolution. Of course, Ron. But V4RB and V4MD high priority. We will prove functionality there at first. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 14 02:21:01 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 13 18:21:07 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/13/04 2:51 PM, "jda" wrote: >> >> 1) now plugin must work, it self must locate Vcomponents folder on your >> computer. >> > > Sorry, Ruslan. It crashes in the same place as before: > > Valentina.Init( 8 * 1024 * 1024 ) > > No log, but the empty file VSQL_TreeParser.txt is created. Please download 2 archives again. Run example 1 Now you should see V4RB_Log.txt file after push of Init button. Check what it have. Must be something as MacOS X:Library:CFMSupport:Vcomponents: ^^^^^^ must be name of your system volume. Let me know -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Thu May 13 19:34:01 2004 From: jda at his.com (jda) Date: Thu May 13 18:34:11 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: > > >Please download 2 archives again. > >Run example 1 > >Now you should see V4RB_Log.txt file after push of Init button. > >Check what it have. >Must be something as > > MacOS X:Library:CFMSupport:Vcomponents: > ^^^^^^ > must be name of your system volume. > >Let me know I see the log file -- it is empty. Jon From frank-list2 at mindstarprods.com Thu May 13 22:14:47 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Thu May 13 21:15:03 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: <7913A8CE-A54C-11D8-B2D3-000A95BE1A8E@mindstarprods.com> On May 13, 2004, at 7:34 PM, jda wrote: >> >> >> Please download 2 archives again. >> >> Run example 1 >> >> Now you should see V4RB_Log.txt file after push of Init button. >> >> Check what it have. >> Must be something as >> >> MacOS X:Library:CFMSupport:Vcomponents: >> ^^^^^^ >> must be name of your system volume. >> >> Let me know > > I see the log file -- it is empty. Same result for me. -Frank From sunshine at public.kherson.ua Fri May 14 10:29:15 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 14 02:29:25 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/14/04 2:34 AM, "jda" wrote: >> Please download 2 archives again. >> >> Run example 1 >> >> Now you should see V4RB_Log.txt file after push of Init button. >> >> Check what it have. >> Must be something as >> >> MacOS X:Library:CFMSupport:Vcomponents: >> ^^^^^^ >> must be name of your system volume. >> >> Let me know > > I see the log file -- it is empty. Forget to say: you need press button shutdown also and quite app Then only check log file -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 14 13:04:07 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 14 05:04:32 2004 Subject: V4RB Changes in debug level and logs Message-ID: Now V4RB produce output NOT into console or Dcon, But simply into V4RB_Log file. It looks as following. As you see we will see now PARAM valus for each function And what it returns. --------------------------------------------- Valentina_SetDebugLevel... PARAM: inLevel = 2 return. Valentina_GetDebugLevel... return gVRB_DebugMode = 2 Valentina_Init... PARAM: inCacheSize = 8388608 PARAM: inMacSN->CString() = PARAM: inWinSN->CString() = VAR: outPath = G4_MacOS X:Library:CFMSupport:VComponents: return. Valentina_GetCacheSize... return. Valentina_ShutDown... return. Valentina_GetCacheSize... return. Valentina_ShutDown... --------------------------------------------- -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From frank-list2 at mindstarprods.com Fri May 14 06:48:12 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Fri May 14 05:48:37 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: <3232A7F0-A594-11D8-B2D3-000A95BE1A8E@mindstarprods.com> On May 14, 2004, at 3:29 AM, Ruslan Zasukhin wrote: > Forget to say: > > you need press button shutdown also > and quite app > > Then only check log file The apps crashes after hitting the Init button, so we cannot press the Shutdown button. -Frank From jda at his.com Fri May 14 07:37:07 2004 From: jda at his.com (jda) Date: Fri May 14 06:37:14 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: > > > > > I see the log file -- it is empty. > >Forget to say: > > you need press button shutdown also > and quite app > >Then only check log file > Um, it crashes immediately on Init -- no chance to press shutdown! Either it crashes before you've written anything to the file, or you didn't flush the file to disk before the crash. Jon From lists at praxisworks.com Fri May 14 07:51:18 2004 From: lists at praxisworks.com (Erik Mueller-Harder) Date: Fri May 14 06:51:28 2004 Subject: V4RB Changes in debug level and logs In-Reply-To: Message-ID: Hi, Ruslan! Ruslan Zasukhin wrote: > Now V4RB produce output NOT into console or Dcon, > But simply into V4RB_Log file. As others have noted, the "(1) Init_ShutDown" example crashes on the "Valentina.Init( 8 * 1024 * 1024 )" step, without writing (or at least not flushing) to the log file. Is it relevant that, as Jon reported before, the name of the log file that's created is not "V4RB_Log," but rather "VSQL_TreeParser.txt"? I'm running OS X 10.3.3, Rb 5.5.2fc2, and V4Rb 2.0b2. Erik From jda at his.com Fri May 14 08:14:35 2004 From: jda at his.com (jda) Date: Fri May 14 07:14:46 2004 Subject: V4RB Changes in debug level and logs In-Reply-To: References: Message-ID: >Hi, Ruslan! > >Ruslan Zasukhin wrote: > >> Now V4RB produce output NOT into console or Dcon, >> But simply into V4RB_Log file. > >As others have noted, the "(1) Init_ShutDown" example crashes on the >"Valentina.Init( 8 * 1024 * 1024 )" step, without writing (or at >least not flushing) to the log file. > >Is it relevant that, as Jon reported before, the name of the log >file that's created is not "V4RB_Log," but rather >"VSQL_TreeParser.txt"? > >I'm running OS X 10.3.3, Rb 5.5.2fc2, and V4Rb 2.0b2. > >Erik Erik, The file V4RB_Log is now created by b2 -- it wasn't by b1. The VSQL_TreeParser.txt is created in b2 as well. Jon From lists at praxisworks.com Fri May 14 08:26:46 2004 From: lists at praxisworks.com (Erik Mueller-Harder) Date: Fri May 14 07:26:54 2004 Subject: V4RB Changes in debug level and logs In-Reply-To: Message-ID: jda wrote: > The file V4RB_Log is now created by b2 -- it wasn't by b1. The > VSQL_TreeParser.txt is created in b2 as well. Thanks, Jon -- Actually, b2 is not giving me a V4RB_Log file at all (either empty or populated); though it does give me an empty VSQL_TreeParser.txt file. I never downloaded b1, so I'm not sure what results I would have gotten with it. Erik From sunshine at public.kherson.ua Fri May 14 16:01:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 14 08:04:16 2004 Subject: V4RB Changes in debug level and logs In-Reply-To: Message-ID: On 5/14/04 3:14 PM, "jda" wrote: >> Hi, Ruslan! >> >> Ruslan Zasukhin wrote: >> >>> Now V4RB produce output NOT into console or Dcon, >>> But simply into V4RB_Log file. >> >> As others have noted, the "(1) Init_ShutDown" example crashes on the >> "Valentina.Init( 8 * 1024 * 1024 )" step, without writing (or at >> least not flushing) to the log file. >> >> Is it relevant that, as Jon reported before, the name of the log >> file that's created is not "V4RB_Log," but rather >> "VSQL_TreeParser.txt"? >> >> I'm running OS X 10.3.3, Rb 5.5.2fc2, and V4Rb 2.0b2. >> >> Erik > > Erik, > > The file V4RB_Log is now created by b2 -- it wasn't by b1. The > VSQL_TreeParser.txt is created in b2 as well. Look, V4RB_Log -- this is new log for V4RB. I will upload right now new build that must produce log with new info as I have show. Other log files -- are created by our internal test system, they will be disabled later. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 14 16:04:08 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 14 08:04:57 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/14/04 2:37 PM, "jda" wrote: >> >>> >>> I see the log file -- it is empty. >> >> Forget to say: >> >> you need press button shutdown also >> and quite app >> >> Then only check log file >> > > Um, it crashes immediately on Init -- no chance to press shutdown! I see. I will make new build. It must flush log file one each WRITE() So we must see at least something. > Either it crashes before you've written anything to the file, or you > didn't flush the file to disk before the crash. Yes, I did not flush btw, I am going ad such parameter to Valentina module, because while there is no crashes, better work without flushes but in case we have crash, we need set flushing to see LATEST write. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Sat May 15 07:47:20 2004 From: jda at his.com (jda) Date: Sat May 15 06:47:27 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: >I see. > >I will make new build. >It must flush log file one each WRITE() >So we must see at least something. > Ruslan, I don't see a new build on your site. Am I missing something, or are you still working on it? Jon From sunshine at public.kherson.ua Sat May 15 15:10:04 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 15 07:10:15 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/15/04 2:47 PM, "jda" wrote: >> I see. >> >> I will make new build. >> It must flush log file one each WRITE() >> So we must see at least something. >> > > Ruslan, I don't see a new build on your site. Am I missing something, > or are you still working on it? I am working. One bug yet I want to fix and upload. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat May 15 17:10:57 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 15 09:11:06 2004 Subject: v4rb new build Message-ID: Hi All, Please download 2 archives Vengine V4RB 1) I have fix technical problem with C++ exceptions and DLLs. 2) DebugLevel, CacheSize now looks EXACTLY as properties 3) added db.Get_Name() into second example. 4) now works Valentina.Version() Exmaple1 show this. 5) you should see something now in V4RB_Log, let me know what exactly 6) attached is XML file that you should drop into Vcomponents/resources/English ------- You can open this file in MS IE to see it. We use it as: - documentation of errors - generator of our C++ sources with this error codes - and we load on start of V4RB strings of error description from it. Idea is that everybody can translate this XML to say German, And Valentina engine will return errors on German then. I.e. Very easy way for localization. And we should add yet some function to specify which language to use from resources. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/octet-stream Size: 5102 bytes Desc: not available Url : http://lists.macserve.net/pipermail/valentina-beta/attachments/20040515/807eda95/attachment-0001.obj -------------- next part -------------- Scanned by evaluation version of Dr.Web antivirus Daemon http://drweb.ru/unix/ From jda at his.com Sat May 15 10:26:37 2004 From: jda at his.com (jda) Date: Sat May 15 09:26:45 2004 Subject: v4rb new build In-Reply-To: References: Message-ID: >Hi All, > >Please download 2 archives > Vengine > V4RB > OK, it still crashes on Init, but now I can report some log info: Valentina_GetVersion... return res.get() = 2.0b1 Valentina_SetDebugLevel... PARAM: inLevel = 2 return Valentina_GetDebugLevel... return gVRB_DebugMode = 2 Valentina_Init... PARAM: inCacheSize = 8388608 PARAM: inMacSN->CString() = PARAM: inWinSN->CString() = VAR: outPath = Sonny's HD:Library:CFMSupport:VComponents: FYI, the outPath path is correct. FBL_Asserts.txt is empty. VSQL_TreeParser.txt is empty. HTH! Jon From lists at praxisworks.com Sat May 15 16:25:06 2004 From: lists at praxisworks.com (Erik Mueller-Harder) Date: Sat May 15 15:25:16 2004 Subject: v4rb new build In-Reply-To: Message-ID: My results are somewhat different, FYI. Perhaps the error number will help? Valentina_GetVersion... return res.get() = 2.0b1 Valentina_SetDebugLevel... PARAM: inLevel = 2 return Valentina_GetDebugLevel... return gVRB_DebugMode = 2 Valentina_Init... PARAM: inCacheSize = 8388608 PARAM: inMacSN->CString() = PARAM: inWinSN->CString() = error 365826. return Erik jda wrote: > >Hi All, > > > >Please download 2 archives > > Vengine > > V4RB > > > > OK, it still crashes on Init, but now I can report some log info: > > Valentina_GetVersion... > return res.get() = 2.0b1 > > Valentina_SetDebugLevel... > PARAM: inLevel = 2 > return > > Valentina_GetDebugLevel... > return gVRB_DebugMode = 2 > > Valentina_Init... > PARAM: inCacheSize = 8388608 > PARAM: inMacSN->CString() = > PARAM: inWinSN->CString() = > VAR: outPath = Sonny's HD:Library:CFMSupport:VComponents: > > FYI, the outPath path is correct. > > > FBL_Asserts.txt is empty. > VSQL_TreeParser.txt is empty. > > HTH! From sunshine at public.kherson.ua Sun May 16 00:29:04 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 15 16:29:12 2004 Subject: v4rb new build In-Reply-To: Message-ID: On 5/15/04 11:25 PM, "Erik Mueller-Harder" wrote: Hi Erik, > My results are somewhat different, FYI. Perhaps the error number will help? > > Valentina_GetVersion... > return res.get() = 2.0b1 > > Valentina_SetDebugLevel... > PARAM: inLevel = 2 > return > > Valentina_GetDebugLevel... > return gVRB_DebugMode = 2 > > Valentina_Init... > PARAM: inCacheSize = 8388608 > PARAM: inMacSN->CString() = > PARAM: inWinSN->CString() = > error 365826. > > return Interesting, this means that folder with ICU data file was not found. Not clear why VAR was not logged in your case. In Next build I will add more debug logs, And I will add ability specify location of ICU dat file. Something very odd happens here. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sun May 16 01:59:52 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 15 18:00:01 2004 Subject: V4RB, Next build for test Message-ID: Hi All, Please download 2 archives. 1) now you MUST drop xml file into vcomponents/resources/english 2) now V4RB_Log produce strings in "" 3) now FBL_Tests.txt file will have some records. let me know what records it have. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Sat May 15 19:07:27 2004 From: jda at his.com (jda) Date: Sat May 15 18:07:36 2004 Subject: V4RB, Next build for test In-Reply-To: References: Message-ID: >Hi All, > >Please download 2 archives. > >1) now you MUST drop xml file into > vcomponents/resources/english > >2) now V4RB_Log produce strings in "" > >3) now FBL_Tests.txt file will have some records. > let me know what records it have. > > Hey, example 1 seems to work! FBL_Texts.txt: DONE: FBL_ValentinaInit.cpp:init_ICU:95 U_SUCCESS(status) DONE: FBL_ValentinaInit.cpp:ValentinaInit:179 true DONE: FBL_ValentinaInit.cpp:ValentinaInit:182 true DONE: FBL_ValentinaInit.cpp:ValentinaInit:186 true DONE: FBL_ValentinaInit.cpp:ValentinaInit:189 true V4RB_Log: Valentina_GetVersion... return res.get() = "2.0b1" Valentina_SetDebugLevel... PARAM: inLevel = 2 return Valentina_GetDebugLevel... return gVRB_DebugMode = 2 Valentina_Init... PARAM: inCacheSize = 8388608 PARAM: inMacSN->CString() = "" PARAM: inWinSN->CString() = "" VAR: outPath = "Sonny's HD:Library:CFMSupport:VComponents:" return Valentina_GetCacheSize... return AllocatedSize = 8392704 I'll play with the other examples shortly. Jon From jda at his.com Sat May 15 19:10:57 2004 From: jda at his.com (jda) Date: Sat May 15 18:11:01 2004 Subject: V4RB, Next build for test In-Reply-To: References: Message-ID: Example 2 seems to work, except after .throwaway I get a NilObjectError in the ShowStatistic method: if( mDatabase.IsOpen ) then -> NilObjectError stDbName.Text = mDatabase.Name stIsOpen.Text = "db is opened" ... You have to check for nil here, of course. The good news is that the database files were deleted. Jon From frank-list2 at mindstarprods.com Sat May 15 20:08:06 2004 From: frank-list2 at mindstarprods.com (Frank Schima) Date: Sat May 15 19:08:20 2004 Subject: V4RB, Next build for test In-Reply-To: References: Message-ID: <1B1A1698-A6CD-11D8-9A1D-000A95BE1A8E@mindstarprods.com> Hi Ruslan, > Please download 2 archives. > > 1) now you MUST drop xml file into > vcomponents/resources/english > > 2) now V4RB_Log produce strings in "" > > 3) now FBL_Tests.txt file will have some records. > let me know what records it have. It works for me now! Here is the FBL_Tests.txt file for all 3 examples: DONE: FBL_ValentinaInit.cpp:init_ICU:95 U_SUCCESS(status) DONE: FBL_ValentinaInit.cpp:ValentinaInit:179 true DONE: FBL_ValentinaInit.cpp:ValentinaInit:182 true DONE: FBL_ValentinaInit.cpp:ValentinaInit:186 true DONE: FBL_ValentinaInit.cpp:ValentinaInit:189 true BTW, I saw the same NilObjectError Exception that Jon saw in the 2nd example. Best regards, Frank From sunshine at public.kherson.ua Sun May 16 10:10:38 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun May 16 02:10:47 2004 Subject: V4RB, Next build for test In-Reply-To: <1B1A1698-A6CD-11D8-9A1D-000A95BE1A8E@mindstarprods.com> Message-ID: On 5/16/04 3:08 AM, "Frank Schima" wrote: > Hi Ruslan, > > >> Please download 2 archives. >> >> 1) now you MUST drop xml file into >> vcomponents/resources/english >> >> 2) now V4RB_Log produce strings in "" >> >> 3) now FBL_Tests.txt file will have some records. >> let me know what records it have. > > It works for me now! Here is the FBL_Tests.txt file for all 3 examples: > > DONE: FBL_ValentinaInit.cpp:init_ICU:95 U_SUCCESS(status) > DONE: FBL_ValentinaInit.cpp:ValentinaInit:179 true > DONE: FBL_ValentinaInit.cpp:ValentinaInit:182 true > DONE: FBL_ValentinaInit.cpp:ValentinaInit:186 true > DONE: FBL_ValentinaInit.cpp:ValentinaInit:189 true Great! Although I wonder, may be in prev time XML file was not found. Okay, I am glad it have start to work! > BTW, I saw the same NilObjectError Exception that Jon saw in the 2nd > example. Yes I also have see, just did not fix before upload. Okay, now I can start move forward. On the other hand thanks to this problem we have improve log file system for Valentina engine and V4RB. Now we can do a lots of debug output as from engine, so from V4RB. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From vidal_olivier at yahoo.fr Mon May 17 10:10:33 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Mon May 17 03:06:04 2004 Subject: V4RB Valentina 1 -> valentina 2 ? Message-ID: Hello, I am novice in data bases. I begin just to study Valentina 1 (V4RB 1.10 realbasic 5.5) I began a project with Valentina 1. According to you, which is the best solution: - I makes my project at once with Valentina 2 to have no problems of compatibility later, even if there is still bugs in beta Valentina 2 ? - I waits that Valentina 2 is more stable and ended. But in that case it will be necessary that I change my code to adapt Valentina 2 to my project ? - Valentina 2 is he an evolution of Valentina 1 or it is a completely different product (for a usage or a different purpose)? - What are the differences among Valentina 1 and 2? Thank you very much Olivier From sunshine at public.kherson.ua Mon May 17 11:11:55 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 03:20:34 2004 Subject: V4RB Valentina 1 -> valentina 2 ? In-Reply-To: Message-ID: On 5/17/04 11:10 AM, "olivier" wrote: Hi Oliver, > I am novice in data bases. > I begin just to study Valentina 1 (V4RB 1.10 realbasic 5.5) > I began a project with Valentina 1. > > According to you, which is the best solution: > > - I makes my project at once with Valentina 2 to have no problems of > compatibility later, even if there is still bugs in beta Valentina 2 ? > - I waits that Valentina 2 is more stable and ended. But in that case > it will be necessary that I change my code to adapt Valentina 2 to my > project ? > > > - Valentina 2 is he an evolution of Valentina 1 or it is a completely > different product (for a usage or a different purpose)? > > - What are the differences among Valentina 1 and 2? It is revolutionary evolution :-) So I think you can start with Valentina 1.x. Old code must work in most cases. Although V4RB 2.0 must introduce exceptions, This feature changes style of coding. In Valentina engine 2.0 main features: - new SQL parser and engine so you will have more SQL commands - unicode support ... In V4RB main changes -- exceptions. auto-conversion of strings. Keep eye on this list. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From vidal_olivier at yahoo.fr Mon May 17 11:15:28 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Mon May 17 04:10:58 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: References: Message-ID: > Although V4RB 2.0 must introduce exceptions, It is true that a exceptions handling would be good ! In realbasic, one can manage the exceptions and avoid the crash but in realbasic / Valentina, if there is an exception in Valentina, it is the crash. Not good. Generally, by beginning to study sql, I realized that the commands of security were missing in Valentina: commit/rollback and journaling. That seems to be a problem. My customers often have electric breakdowns. The journaling-commit-rollback group seems to be very effective. How do you make without these functions ? Only backups ? These commands are really effective ? > - unicode support - auto-conversion of strings. great ! > Although V4RB 2.0 must introduce exceptions, > This feature changes style of coding. Now I know that a exceptions handling is included in Valentina 2, I am very tried to begin again my project with Valentina 2! The security is important in my project. Please , think you that it is necessary all the same that I wait that Valentina 2 is more stable? According to you, how long it will set? Thank you olivier Ps: also ' HAVING ' seems also important for the requests. It is not the problem that there is this command in Valentina? Le 17 mai 04, ? 10:11, Ruslan Zasukhin a ?crit : > It is revolutionary evolution :-) > > So I think you can start with Valentina 1.x. > Old code must work in most cases. > > Although V4RB 2.0 must introduce exceptions, > This feature changes style of coding. > > In Valentina engine 2.0 main features: > > - new SQL parser and engine > so you will have more SQL commands > - unicode support > ... > > > In V4RB main changes -- exceptions. > auto-conversion of strings. > > Keep eye on this list. From sunshine at public.kherson.ua Mon May 17 11:58:46 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 04:15:48 2004 Subject: V4RB. Db.TablesCount Message-ID: Renamed to TableCount -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon May 17 12:16:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 04:33:06 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: Message-ID: On 5/17/04 12:15 PM, "olivier" wrote: >> Although V4RB 2.0 must introduce exceptions, > > It is true that a exceptions handling would be good ! > In realbasic, one can manage the exceptions and avoid the crash but in > realbasic / Valentina, if there is an exception in Valentina, it is the > crash. Not good. > > Generally, by beginning to study sql, I realized that the commands of > security were missing in Valentina: commit/rollback and journaling. > > That seems to be a problem. For now hundreds developers live with this and no big problems. MySQL have live without transactions 20 years. > My customers often have electric breakdowns. The > journaling-commit-rollback group seems to be very effective. How do you > make without these functions ? Only backups ? You should do flush after each modification. After flush() any crash is not scary. You can consider flush as commit() -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon May 17 12:17:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 04:43:48 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: Message-ID: On 5/17/04 12:15 PM, "olivier" wrote: >> Although V4RB 2.0 must introduce exceptions, >> This feature changes style of coding. > > Now I know that a exceptions handling is included in Valentina 2, I am > very tried to begin again my project with Valentina 2! > > The security is important in my project. > Please , think you that it is necessary all the same that I wait that > Valentina 2 is more stable? > According to you, how long it will set? Few weeks yet. I still think you can start become used to Valentina now with 1.x > Thank you > > olivier > > Ps: also ' HAVING ' seems also important for the requests. It is not > the problem that there is this command in Valentina? 2.0 will have it -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From vidal_olivier at yahoo.fr Mon May 17 11:51:30 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Mon May 17 04:47:11 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: References: Message-ID: Le 17 mai 04, ? 11:17, Ruslan Zasukhin a ?crit : > > I still think you can start become used to Valentina now with 1.x > ok, thank you Ruslan ! From vidal_olivier at yahoo.fr Mon May 17 12:03:59 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Mon May 17 04:59:26 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: References: Message-ID: <84170FC9-A7E9-11D8-AFCE-00039310B7DA@yahoo.fr> Le 17 mai 04, ? 11:16, Ruslan Zasukhin a ?crit : > > For now hundreds developers live with this and no big problems. > MySQL have live without transactions 20 years. It is true, Ruslan... But maybe that it will be replaced these next years by postgresSQL for these reasons? > You should do flush after each modification. > After flush() any crash is not scary. > > You can consider flush as commit() > It is what I made. In every modification or addition in the base, I use flush at once. But that does not resolve the problem if the computer falls at the time of the flush. The data base risks strongly to be corrupted. Without journaling, there is no means immediate and clean to get back the data base without corruption. olivier From sunshine at public.kherson.ua Mon May 17 13:11:56 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 05:22:08 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: <84170FC9-A7E9-11D8-AFCE-00039310B7DA@yahoo.fr> Message-ID: On 5/17/04 1:03 PM, "olivier" wrote: >> For now hundreds developers live with this and no big problems. >> MySQL have live without transactions 20 years. > > It is true, Ruslan... > > But maybe that it will be replaced these next years by postgresSQL for > these reasons? Never. Postgre is academic project IMHO, it do not have strong support companies, Only hobbyists. No marketing and so on. It is not true that Postgre have speed up. Today only one developer have test Postgre and say it is MUCH MUCH slower of mySQL. I have ask him inform about this RB list. And od not forget, mySQL 4.x already have transactions. >> You should do flush after each modification. >> After flush() any crash is not scary. >> >> You can consider flush as commit() > It is what I made. In every modification or addition in the base, I use > flush at once. > But that does not resolve the problem if the computer falls at the time > of the flush. Oliver, a lots of Valentina developers distribute products to thousands of clients. And corruptions are very rare. Some developers have ZERO issues with this. You can check Valentina list archive for 5 years. So as say my father: solve problems when they really come to you. but do not fill your brain by POSSIBLE problems. :-) > The data base risks strongly to be corrupted. > > Without journaling, there is no means immediate and clean to get back > the data base without corruption. In theory yes. But reality not so bad. Of course, I will NOT recommend now use Valentina for banking or ecommerce. Because this areas really require 100% of sure. On the other hand, one of our developer, which work in bank say that they use Sybase, and even such mature db sometimes cannot recover after system failure. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From vidal_olivier at yahoo.fr Mon May 17 12:45:16 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Mon May 17 05:40:24 2004 Subject: V4RB V.1 -> V.2 ? Security in valentina In-Reply-To: References: Message-ID: <48E0229A-A7EF-11D8-AFCE-00039310B7DA@yahoo.fr> Le 17 mai 04, ? 12:11, Ruslan Zasukhin a ?crit : > > Postgre is academic project IMHO, it do not have strong support > companies, > Only hobbyists. No marketing and so on. > Oh ok, I did not know. > It is not true that Postgre have speed up. > Today only one developer have test Postgre and say it is MUCH MUCH > slower of > mySQL. I have ask him inform about this RB list. > > Oliver, a lots of Valentina developers distribute products to > thousands of > clients. And corruptions are very rare. Some developers have ZERO > issues > with this. You can check Valentina list archive for 5 years. > > That reassures me ok. That reassures me. > So as say my father: > solve problems when they really come to you. > but do not fill your brain by POSSIBLE problems. > :-) it's true ! thank you Ruslan ! From lists at praxisworks.com Mon May 17 09:39:52 2004 From: lists at praxisworks.com (Erik Mueller-Harder) Date: Mon May 17 08:39:58 2004 Subject: v4rb new build In-Reply-To: Message-ID: Hi, Ruslan -- You wrote: >> Valentina_Init... >> PARAM: inCacheSize = 8388608 >> PARAM: inMacSN->CString() = >> PARAM: inWinSN->CString() = >> error 365826. >> >> return > > Interesting, this means that folder with ICU data file was not > found. Not clear why VAR was not logged in your case. > > In Next build I will add more debug logs, And I will add ability > specify location of ICU dat file. > > Something very odd happens here. I had put the VComponents folder in my local user library, rather than my global library. Your latest download gives this error in the FBL_Asserts.txt log file: FBL_ValentinaInit.cpp:ValentinaInit:162 gComponentsPath == 0L So, I moved the VComponents folder into the global library and no longer get an FBL_Asserts log file or the error #365826 on init. I know you are planning to allow us to put the VComponents folder in any of several different places -- I'm guessing that putting it in the local user library isn't allowed yet. So, now, in the Database_CreateCloseOpen example, I'm seeing the same NilObjectException as Frank and Jon. FYI, on the re-opening of the database in that example, I'm getting 21 FBL assertions. Are others seeing these, too, or do I have something set up incorrectly? FBL_DTFormat.h:SetCentury:84 inCentury2 - inCentury1 == 100 FBL_Index_imp.cpp:FindPage:433 IsGoodDataPageNumber( inContext.mRange.first ) FBL_Index_imp.cpp:FindPage:434 IsGoodDataPageNumber( inContext.mRange.second ) FBL_Index_imp.cpp:FindPage:435 inContext.mRange.first <= inContext.mRange.second FBL_Index_NotUnique.cpp:BinSearchPages:176 first == second FBL_Index_Page_Inventory.cpp:get_IsPageBig:259 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() FBL_InventoryPage.cpp:ReadItem:332 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() + 2 FBL_Index_Page_Inventory.cpp:get_IsPageBig:259 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() FBL_InventoryPage.cpp:ReadItem:332 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() + 2 FBL_Index_NotUnique_Page.h:get_LastValuePtr:392 get_ValuesCount() >= 1 FBL_Index_NotUnique_Page.h:GetValuePtr:540 inIndex < get_ValuesCount() FBL_Index_imp.cpp:FindPage:433 IsGoodDataPageNumber( inContext.mRange.first ) FBL_Index_imp.cpp:FindPage:434 IsGoodDataPageNumber( inContext.mRange.second ) FBL_Index_imp.cpp:FindPage:435 inContext.mRange.first <= inContext.mRange.second FBL_Index_NotUnique.cpp:BinSearchPages:176 first == second FBL_Index_Page_Inventory.cpp:get_IsPageBig:259 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() FBL_InventoryPage.cpp:ReadItem:332 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() + 2 FBL_Index_Page_Inventory.cpp:get_IsPageBig:259 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() FBL_InventoryPage.cpp:ReadItem:332 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() + 2 FBL_Index_Page_Inventory.cpp:get_IsPageBig:259 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() FBL_InventoryPage.cpp:ReadItem:332 1 <= inLogicalIndex && inLogicalIndex <= mPageFile->get_LogicalPageCount() + 2 -- Erik From sunshine at public.kherson.ua Mon May 17 16:43:23 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 08:50:31 2004 Subject: v4rb new build In-Reply-To: Message-ID: On 5/17/04 4:39 PM, "Erik Mueller-Harder" wrote: > download gives this error in the FBL_Asserts.txt log file: > > FBL_ValentinaInit.cpp:ValentinaInit:162 gComponentsPath == 0L > > So, I moved the VComponents folder into the global library and no longer get > an FBL_Asserts log file or the error #365826 on init. I know you are planning > to allow us to put the VComponents folder in any of several different places > -- I'm guessing that putting it in the local user library isn't allowed yet. Yes, right now I search /Library/CMFSupport > So, now, in the Database_CreateCloseOpen example, I'm seeing the same > NilObjectException as Frank and Jon. It needs in RB code simply move mDatabase = nil To last line, after ShowStatistic > FYI, on the re-opening of the database in that example, I'm getting 21 FBL > assertions. Are others seeing these, too, or do I have something set up > incorrectly? This is engine problems. All have this asserts. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Mon May 17 23:16:58 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Mon May 17 15:17:15 2004 Subject: [DONE] V4MD: log file & options In-Reply-To: <001901c43c42$1b36f0b0$3b04a8c0@giv> Message-ID: On 5/17/04 10:06 PM, "Igor Gomon" wrote: > All the methods of VDatabase & VCursor Xtra's > now contains LOG_PARAM & LOG_NAMED_PARAM > macros, which dumps input parameters to either the Message Window > or log file, or both. > Also two new functions added: > > ValentianLogToFile(int on, in flush); > ValentianLogToMsgWindow(int on); > > V4MD samples updated to call ValentianLogToMsgWindow() function. Ok, Igor. > RETURN_xxx macro still not implemened. > I think I'll rename it in V4MD to LOG_RET_VAL_xxx since it's not so > easy to implement it in such way as in V4RB. What problems? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue May 18 14:16:13 2004 From: jda at his.com (jda) Date: Tue May 18 13:16:30 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: Hi Rualan, don't be shy. We're ready for 2.0b3. :-> Jon From sunshine at public.kherson.ua Tue May 18 22:55:34 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 18 14:55:44 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/18/04 9:16 PM, "jda" wrote: > Hi Rualan, don't be shy. We're ready for 2.0b3. :-> :-) Today I have improve example 3. And have spend half of day on strange glitch in CW linking. Now I have guess why it was. I want tomorrow implement Exceptions for V4RB. --- Also we need discuss exceptions. Old code from V4RB 1.x use other style. It do command if( db.GetLastError() ) ... end if With exceptions code must change. What you will say guys? You ready redo sources for this feature? Actually we have no other way. We MUST switch to new feature of RB language. I have one hope. May be I will be able A) throw exceptions B) have flag: do not throw. developers that use EXISTED code, can set it OFF to not break code C) even if throw exc, db.LastError still will return last erro code. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Tue May 18 16:12:30 2004 From: jda at his.com (jda) Date: Tue May 18 15:12:42 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: > > >I have one hope. >May be I will be able > >A) throw exceptions > >B) have flag: do not throw. > developers that use EXISTED code, can set it OFF > to not break code > >C) even if throw exc, db.LastError still will return last erro code. > There are other folks on this list with a better OOP background than I, so I defer to them. But I like #3. There are times when a db.error is expected and you don't want an exception thrown. For example, I have a field for a unique ID. When I create a record, I assign a new unique ID. I then check the db.error property to see if this caused a "not unique" error. If so I loop through again, assigning a new unique id, until I don't get the error any more. I think this would be more cumbersome if the assignment of a non-unique ID threw an exception (and took control away from my code loop). Jon From sunshine at public.kherson.ua Tue May 18 23:28:26 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 18 15:30:21 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/18/04 11:12 PM, "jda" wrote: >> I have one hope. >> May be I will be able >> >> A) throw exceptions >> >> B) have flag: do not throw. >> developers that use EXISTED code, can set it OFF >> to not break code >> >> C) even if throw exc, db.LastError still will return last erro code. >> > > There are other folks on this list with a better OOP background than > I, so I defer to them. But I like #3. There are times when a db.error > is expected and you don't want an exception thrown. For example, I > have a field for a unique ID. When I create a record, I assign a new > unique ID. I then check the db.error property to see if this caused a > "not unique" error. If so I loop through again, assigning a new > unique id, until I don't get the error any more. I think this would > be more cumbersome if the assignment of a non-unique ID threw an > exception (and took control away from my code loop). Yes, unique value probably the ONLY case when we MAY need performance. But it is godo always use THE SAME style. It is bad idea if half of functions errors throw, But second half reurn them. Developer will get great headache with mixing of 2 styles -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Tue May 18 17:09:02 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 18 16:08:29 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: <96DAB180-A90F-11D8-91CF-003065BB0634@desuetude.com> On May 18, 2004, at 4:12 PM, jda wrote: >> >> >> I have one hope. >> May be I will be able >> >> A) throw exceptions >> >> B) have flag: do not throw. >> developers that use EXISTED code, can set it OFF >> to not break code >> >> C) even if throw exc, db.LastError still will return last erro code. >> > > There are other folks on this list with a better OOP background than > I, so I defer to them. But I like #3. There are times when a db.error > is expected and you don't want an exception thrown. For example, I > have a field for a unique ID. When I create a record, I assign a new > unique ID. I then check the db.error property to see if this caused a > "not unique" error. If so I loop through again, assigning a new unique > id, until I don't get the error any more. I think this would be more > cumbersome if the assignment of a non-unique ID threw an exception > (and took control away from my code loop). It wouldn't if you use a try block, as in While NOT IDCreated try CreateID IDCreated = true catch vExc as ValentinaException If vExc.ErrorCode <> 341 then Raise vExc //because it's an error other than uniqueness violated Else // End if end try Wend -------------- Charles Yeomans From yeomans at desuetude.com Tue May 18 17:10:51 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 18 16:10:10 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: On May 18, 2004, at 3:55 PM, Ruslan Zasukhin wrote: > On 5/18/04 9:16 PM, "jda" wrote: > >> Hi Rualan, don't be shy. We're ready for 2.0b3. :-> > > :-) > > Today I have improve example 3. > > And have spend half of day on strange glitch in CW linking. > Now I have guess why it was. > > I want tomorrow implement Exceptions for V4RB. > > --- > Also we need discuss exceptions. > > Old code from V4RB 1.x use other style. > It do > > command > if( db.GetLastError() ) > ... > end if > > > With exceptions code must change. > > What you will say guys? > You ready redo sources for this feature? > > Actually we have no other way. > We MUST switch to new feature of RB language. > > I have one hope. > May be I will be able > > A) throw exceptions > > B) have flag: do not throw. > developers that use EXISTED code, can set it OFF > to not break code > > C) even if throw exc, db.LastError still will return last erro code. Some code should still be broken. For instance, VDatabase.Create and Open should be subroutines that either succeed or return an exception. -------------- Charles Yeomans From jda at his.com Tue May 18 17:11:05 2004 From: jda at his.com (jda) Date: Tue May 18 16:11:14 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: <96DAB180-A90F-11D8-91CF-003065BB0634@desuetude.com> References: <96DAB180-A90F-11D8-91CF-003065BB0634@desuetude.com> Message-ID: >> >>There are other folks on this list with a better OOP background >>than I, so I defer to them. But I like #3. There are times when a >>db.error is expected and you don't want an exception thrown. For >>example, I have a field for a unique ID. When I create a record, I >>assign a new unique ID. I then check the db.error property to see >>if this caused a "not unique" error. If so I loop through again, >>assigning a new unique id, until I don't get the error any more. I >>think this would be more cumbersome if the assignment of a >>non-unique ID threw an exception (and took control away from my >>code loop). > >It wouldn't if you use a try block, as in > > >While NOT IDCreated > try > CreateID > IDCreated = true > catch vExc as ValentinaException > If vExc.ErrorCode <> 341 then > Raise vExc //because it's an error other than uniqueness violated > Else > // > End if > end try >Wend > Good point, Charles. I haven't used the (relatively) new try/catch feature, but it would work well here. Thanks. Jon From sunshine at public.kherson.ua Wed May 19 00:15:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 18 16:15:46 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: <96DAB180-A90F-11D8-91CF-003065BB0634@desuetude.com> Message-ID: On 5/19/04 12:09 AM, "Charles Yeomans" wrote: >> There are other folks on this list with a better OOP background than >> I, so I defer to them. But I like #3. There are times when a db.error >> is expected and you don't want an exception thrown. For example, I >> have a field for a unique ID. When I create a record, I assign a new >> unique ID. I then check the db.error property to see if this caused a >> "not unique" error. If so I loop through again, assigning a new unique >> id, until I don't get the error any more. I think this would be more >> cumbersome if the assignment of a non-unique ID threw an exception >> (and took control away from my code loop). > > It wouldn't if you use a try block, as in > > > While NOT IDCreated > try > CreateID > IDCreated = true > catch vExc as ValentinaException > If vExc.ErrorCode <> 341 then > Raise vExc //because it's an error other than uniqueness violated > Else > // > End if > end try > Wend :-) They have grab so many things from C++. I wonder why they simply did not make syntax close to c++ or Java. if() { } else { } I believe that using of {} is much better for human eye reading Then if() then else end if Because {} is __graphical__ information for eye. then, end if --> is textual information. eye must READ it, instead just SEE it. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed May 19 00:17:30 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 18 16:17:37 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: Message-ID: On 5/19/04 12:10 AM, "Charles Yeomans" wrote: >> C) even if throw exc, db.LastError still will return last erro code. > > Some code should still be broken. For instance, VDatabase.Create and > Open should be subroutines that either succeed or return an exception. If we will have way DISABLE exceptions then old code can work AS IS. No ? -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Tue May 18 17:22:02 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 18 16:21:26 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: <67539D92-A911-11D8-91CF-003065BB0634@desuetude.com> On May 18, 2004, at 5:17 PM, Ruslan Zasukhin wrote: > On 5/19/04 12:10 AM, "Charles Yeomans" wrote: > >>> C) even if throw exc, db.LastError still will return last erro code. >> >> Some code should still be broken. For instance, VDatabase.Create and >> Open should be subroutines that either succeed or return an exception. > > If we will have way DISABLE exceptions then old code can work AS IS. > No ? Are you saying that VDatabase.Create should continue to return a Boolean? -------------- Charles Yeomans From sunshine at public.kherson.ua Wed May 19 00:34:17 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Tue May 18 16:34:25 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: <67539D92-A911-11D8-91CF-003065BB0634@desuetude.com> Message-ID: On 5/19/04 12:22 AM, "Charles Yeomans" wrote: > On May 18, 2004, at 5:17 PM, Ruslan Zasukhin wrote: > >> On 5/19/04 12:10 AM, "Charles Yeomans" wrote: >> >>>> C) even if throw exc, db.LastError still will return last erro code. >>> >>> Some code should still be broken. For instance, VDatabase.Create and >>> Open should be subroutines that either succeed or return an exception. >> >> If we will have way DISABLE exceptions then old code can work AS IS. >> No ? > > Are you saying that VDatabase.Create should continue to return a > Boolean? I mean that db.LastError() should still work -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Tue May 18 18:26:08 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Tue May 18 17:25:34 2004 Subject: [ANN] V4RB 2.0b2 In-Reply-To: References: Message-ID: <5C024DD2-A91A-11D8-91CF-003065BB0634@desuetude.com> On May 18, 2004, at 5:34 PM, Ruslan Zasukhin wrote: > On 5/19/04 12:22 AM, "Charles Yeomans" wrote: > >> On May 18, 2004, at 5:17 PM, Ruslan Zasukhin wrote: >> >>> On 5/19/04 12:10 AM, "Charles Yeomans" wrote: >>> >>>>> C) even if throw exc, db.LastError still will return last erro >>>>> code. >>>> >>>> Some code should still be broken. For instance, VDatabase.Create >>>> and >>>> Open should be subroutines that either succeed or return an >>>> exception. >>> >>> If we will have way DISABLE exceptions then old code can work AS IS. >>> No ? >> >> Are you saying that VDatabase.Create should continue to return a >> Boolean? > > I mean that db.LastError() should still work I think that this makes sense. -------------- Charles Yeomans From sunshine at public.kherson.ua Wed May 19 18:06:03 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 19 10:06:14 2004 Subject: VSRV a63, and clients Message-ID: Hi Jochen, Hi All, I have upload MAC and WIN of Vserver a63 And clients C++, V4RB, V4MD, VXCMD FIXED bug with TEXT field in case language parameter is specified. Jochen, we will be many thank you if you will make build of Vstudio Today evening. Yuri need it for work. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 21 19:21:39 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 11:21:54 2004 Subject: V4RB b3 uploaded Message-ID: Hi All, You can download b3 * implemented VException class. in examples you 3 , 4 you can see its usage., and you can produce exception if click 2 times on create table. It fails create second table with the same name. * db.Open() and db.Create() now return nothing. * new example added Field_CreateDrop. it creates fields of all types. in SQL way and in API way. * New after run of example, you can look into VSQL_TreeParser.txt to see the work of SQL parser. it dumps tree that was built for a query. * You can check improved format of V4RB_Log file to see how many useful info it show now. ------- Next b4 must have new example with records add/delete/update -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Fri May 21 12:29:28 2004 From: jda at his.com (jda) Date: Fri May 21 11:29:43 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: >Hi All, > >You can download b3 > The url http://paradigmasoft.com/download/v2_beta/v2_beta.html times out. Jon From sunshine at public.kherson.ua Fri May 21 19:33:51 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 11:34:04 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/21/04 7:29 PM, "jda" wrote: >> Hi All, >> >> You can download b3 >> > > The url > > http://paradigmasoft.com/download/v2_beta/v2_beta.html > > times out. Ops. Brian, can you check please. It seems WEB site down -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Fri May 21 13:02:00 2004 From: jda at his.com (jda) Date: Fri May 21 12:02:14 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: > > >* implemented VException class. > > in examples you 3 , 4 you can see its usage., > and you can produce exception if click 2 times on > create table. It fails create second table with the same name. > >* db.Open() and db.Create() now return nothing. > >* new example added Field_CreateDrop. > it creates fields of all types. > in SQL way and in API way. > >* New after run of example, you can look into VSQL_TreeParser.txt > to see the work of SQL parser. > it dumps tree that was built for a query. > >* You can check improved format of V4RB_Log file to see how many useful info >it show now. > Everything works for me. Question: in my Valentina 1.x application, I create a database with multiple tables like this: myTable = new boTable myMisc = new boMisc Will this work in 2.0? Jon From sunshine at public.kherson.ua Fri May 21 20:24:49 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 12:25:00 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/21/04 8:02 PM, "jda" wrote: >> * implemented VException class. >> >> in examples you 3 , 4 you can see its usage., >> and you can produce exception if click 2 times on >> create table. It fails create second table with the same name. >> >> * db.Open() and db.Create() now return nothing. >> >> * new example added Field_CreateDrop. >> it creates fields of all types. >> in SQL way and in API way. >> >> * New after run of example, you can look into VSQL_TreeParser.txt >> to see the work of SQL parser. >> it dumps tree that was built for a query. >> >> * You can check improved format of V4RB_Log file to see how many useful info >> it show now. >> > > Everything works for me. Excellent. > Question: in my Valentina 1.x application, I create a database with > multiple tables like this: > > myTable = new boTable > myMisc = new boMisc > > > Will this work in 2.0? Yes of course, just this is special CLASS way, It require special magic, and I have not made it yet in 2.0. Now example show 2 ways SQL API I was going also add CLASS way for V4RB into the same examples, But I afraid they will become very complex. I will think yet. The idea is to make MANY small simple examples, Each show ONE feature. So a developer can easy pick up example and see HOW TO. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 21 20:32:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 12:32:14 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/21/04 8:02 PM, "jda" wrote: >> * implemented VException class. >> >> in examples you 3 , 4 you can see its usage., >> and you can produce exception if click 2 times on >> create table. It fails create second table with the same name. >> >> * db.Open() and db.Create() now return nothing. >> >> * new example added Field_CreateDrop. >> it creates fields of all types. >> in SQL way and in API way. >> >> * New after run of example, you can look into VSQL_TreeParser.txt >> to see the work of SQL parser. >> it dumps tree that was built for a query. >> >> * You can check improved format of V4RB_Log file to see how many useful info >> it show now. >> > > Everything works for me. Jon, You can try to play in Table_Create or Field_Create examples With names of tables/fields. Try assign non-English names may be. They will go to Valentina in UTF8 from RB. V4RB code everywhere already convert from UTF8 to UTF16 on input. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Fri May 21 14:50:07 2004 From: jda at his.com (jda) Date: Fri May 21 13:50:16 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: > >> Question: in my Valentina 1.x application, I create a database with >> multiple tables like this: >> >> myTable = new boTable >> myMisc = new boMisc >> >> >> Will this work in 2.0? > >Yes of course, just this is special CLASS way, >It require special magic, and I have not made it yet in 2.0. > >Now example show 2 ways > > SQL API > >I was going also add CLASS way for V4RB into the same examples, >But I afraid they will become very complex. >I will think yet. > >The idea is to make MANY small simple examples, >Each show ONE feature. >So a developer can easy pick up example and see HOW TO. > OK. Like most of us, I'd like to change my code as little as possible going from 1.x to 2.x (except where it adds features/performance, of course), so I'll wait a bit for the class method to be implemented. Jon From jda at his.com Fri May 21 14:57:56 2004 From: jda at his.com (jda) Date: Fri May 21 13:58:05 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: > >Jon, > >You can try to play in Table_Create or Field_Create examples >With names of tables/fields. > >Try assign non-English names may be. >They will go to Valentina in UTF8 from RB. > >V4RB code everywhere already convert from UTF8 to UTF16 on input. > Hi Ruslan, I tried -- nothing works with Unicode names. I replaced the table Person with the Hebrew letters created by typing abc (I'm sure that won't come through on email). When I tried creating a table with the SQL command (Example 4), no table was created. When I tried with the API command, a table was created, but it's name was garbage. Here's some code from V4RB_Log (the Unicode Hebrew characters show as ???): DataBase_SqlExecute... PARAM: instance = 26237492 PARAM: inQuery->CString() = "CREATE TABLE ???????????? ( firstname string(40) )" return (int) AffectedRecords = 0 DataBase_SqlExecute... PARAM: instance = 26237492 PARAM: inQuery->CString() = "DROP TABLE ????????????" return (int) AffectedRecords = 0 I can send you the whole log if you want. But you can easily do this yourself, use the Character Palette to insert Greek/Hebrew/Japanese/etc. characters. Jon From sunshine at public.kherson.ua Fri May 21 22:06:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 14:06:48 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/21/04 9:57 PM, "jda" wrote: >> Jon, >> >> You can try to play in Table_Create or Field_Create examples >> With names of tables/fields. >> >> Try assign non-English names may be. >> They will go to Valentina in UTF8 from RB. >> >> V4RB code everywhere already convert from UTF8 to UTF16 on input. >> > > Hi Ruslan, > > I tried -- nothing works with Unicode names. > > I replaced the table Person with the Hebrew > letters created by typing abc (I'm sure that > won't come through on email). > > When I tried creating a table with the SQL > command (Example 4), no table was created. When I > tried with the API command, a table was created, > but it's name was garbage. Here's some code from > V4RB_Log (the Unicode Hebrew characters show as > ???): > > DataBase_SqlExecute... > PARAM: instance = 26237492 > PARAM: > inQuery->CString() = > "CREATE TABLE ???????????? ( > firstname string(40) )" > return (int) AffectedRecords = 0 > > DataBase_SqlExecute... > PARAM: instance = 26237492 > PARAM: > inQuery->CString() = > "DROP TABLE ????????????" > return (int) AffectedRecords = 0 > > > > I can send you the whole log if you want. But you > can easily do this yourself, use the Character > Palette to insert Greek/Hebrew/Japanese/etc. > characters. Okay, most probably problem located in SQL parser of ANTLR. They still have not port it to Unicode. So we have big chances that names of tables and fields must be in English, But data inside of records must work... Some people say that ANTLR can be with some hacking improved to utf. We will investigate this later. On the other hand, you say that API also not works. Hmm. Okay we will check this. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Fri May 21 22:07:34 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 14:07:43 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/21/04 9:57 PM, "jda" wrote: > Hi Ruslan, > > I tried -- nothing works with Unicode names. > > I replaced the table Person with the Hebrew > letters created by typing abc (I'm sure that > won't come through on email). I wonder, RB show this text ? Are they able show e.g. Cyrlic text ? I have never try yet -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Fri May 21 15:11:34 2004 From: jda at his.com (jda) Date: Fri May 21 14:11:44 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: > > > >> I tried -- nothing works with Unicode names. >> >> I replaced the table Person with the Hebrew >> letters created by typing abc (I'm sure that >> won't come through on email). > >I wonder, RB show this text ? > >Are they able show e.g. Cyrlic text ? >I have never try yet > Yes -- select Russian from the International control panel and type into a RB editfield -- works. Jon From sunshine at public.kherson.ua Sat May 22 00:00:33 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 16:00:53 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/21/04 10:11 PM, "jda" wrote: >>> I replaced the table Person with the Hebrew >>> letters created by typing abc (I'm sure that >>> won't come through on email). >> >> I wonder, RB show this text ? >> >> Are they able show e.g. Cyrlic text ? >> I have never try yet >> > > Yes -- select Russian from the International control panel and type > into a RB editfield -- works. This raise question: will then all RB strings work not in UTF8 encoding, but in Cyrilic-MAC (single byte) ? What about your Herbew test? This is not questions to you. Just thinking aloud. If they are not in UTF8 then this explain. Because now V4RB simply expect UTF8 and do not check real encoding of REALstring. I will add that later. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Fri May 21 17:13:46 2004 From: jda at his.com (jda) Date: Fri May 21 16:13:56 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: > > > Yes -- select Russian from the International control panel and type >> into a RB editfield -- works. > >This raise question: > > will then all RB strings work not in UTF8 encoding, > but in Cyrilic-MAC (single byte) ? > >What about your Herbew test? > >This is not questions to you. >Just thinking aloud. > >If they are not in UTF8 then this explain. >Because now V4RB simply expect UTF8 and do not check real encoding of >REALstring. I will add that later. > The default encoding for all RB strings is now utf-8. The Cyrilic, Hebrew, Greek, etc. are all utf-8. This includes string literals, which is what you are using in the beta examples. What I meant above was NOT that Cyrilic text works in Valentina 2.0 examples -- I meant Cyrilic is entered into *RB* apps (and code editor in the IDE) as unicode correctly, which is what I thought you were asking. So, there is still a problem with 2.0. Ruslan, you should try this yourself -- it's a matter of a few seconds to add multibyte UTF-8 characters to a table or field name and run the example. Jon From sunshine at public.kherson.ua Sat May 22 00:26:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 16:26:59 2004 Subject: V4RB b3 uploaded In-Reply-To: Message-ID: On 5/22/04 12:13 AM, "jda" wrote: >>> Yes -- select Russian from the International control panel and type >>> into a RB editfield -- works. >> >> This raise question: >> >> will then all RB strings work not in UTF8 encoding, >> but in Cyrilic-MAC (single byte) ? >> >> What about your Herbew test? >> >> This is not questions to you. >> Just thinking aloud. >> >> If they are not in UTF8 then this explain. >> Because now V4RB simply expect UTF8 and do not check real encoding of >> REALstring. I will add that later. >> > > The default encoding for all RB strings is now utf-8. The Cyrilic, > Hebrew, Greek, etc. are all utf-8. This includes string literals, > which is what you are using in the beta examples. But they why in RB SDK they tell us that strings can be UTF8 ANSI it seems something else > What I meant above was NOT that Cyrilic text works in Valentina 2.0 > examples -- I meant Cyrilic is entered into *RB* apps (and code > editor in the IDE) as unicode correctly, which is what I thought you > were asking. Yes I have ask exactly this. > So, there is still a problem with 2.0. > > Ruslan, you should try this yourself -- it's a matter of a few > seconds to add multibyte UTF-8 characters to a table or field name > and run the example. Yes of course. I am busy now on other task. And I will need check what encoding really come from RB. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Fri May 21 17:33:40 2004 From: jda at his.com (jda) Date: Fri May 21 16:33:49 2004 Subject: V4RB b3 uploaded In-Reply-To: References: Message-ID: > > >But they why in RB SDK they tell us that strings can be > > UTF8 > ANSI > it seems something else > Probably because the developer can convert utf-8 to *any* encoding supported by RB if they want. But I didn't, I left it as UTF-8. Just to make sure, I re-ran the test. I added a Greek alpha to the the table name, and added the line query = ConvertEncoding(query, encodings.utf8) just to make sure that Valentina was receiving a utf-8 string. It still failed... Jon From jda at his.com Fri May 21 19:12:19 2004 From: jda at his.com (jda) Date: Fri May 21 18:12:30 2004 Subject: More Unicode incompatability In-Reply-To: References: Message-ID: Ruslan, I also re-ran the project DB_CreateCloseOpen.rb. In the project, I changed "Database1" to a name containing an accented ?, or a Greek letter. When it was saved with .create, these characters changed to question marks. You have to use FSRef to create files with Unicode names in OS X. I can send you some of my own code as an example, if that would save some time for you... Jon From sunshine at public.kherson.ua Sat May 22 02:41:03 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 21 18:41:12 2004 Subject: More Unicode incompatability In-Reply-To: Message-ID: On 5/22/04 2:12 AM, "jda" wrote: > Ruslan, > > I also re-ran the project DB_CreateCloseOpen.rb. > > In the project, I changed "Database1" to a name > containing an accented ?, or a Greek letter. When > it was saved with .create, these characters > changed to question marks. > > You have to use FSRef to create files with > Unicode names in OS X. I can send you some of my > own code as an example, if that would save some > time for you... We have found glitch in conversion. Igor work now on it. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat May 22 11:56:37 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 22 03:56:49 2004 Subject: V4RB, FIXED problem with UTF8 In-Reply-To: Message-ID: On 5/22/04 11:50 AM, "Ruslan Zasukhin" wrote: > Hi all, > > I was able create table with russian name. > And now it show its name back correctly. > > Problem was that when in plugin I create REALstring, > I must self tell it that it is UTF8. > I did expect that REALbuildString() on default make it. > I think it simply allocate RAM. > > Really looks wonderful. > > BTW, I nothing have do in RB itself. > I simply have switch to Russiam keyboard, and start type. > Editor of REALbasic show russian letters self. Ops, sorry, wrong list. Must go to Valentina-beta -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed May 26 15:19:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 07:19:59 2004 Subject: more betas? In-Reply-To: Message-ID: On 5/26/04 3:07 PM, "jda" wrote: Hi Jon, > I'm getting greedy. The supply of betas has stopped, and the last one > is still not nearly enough complete for me to compile my real app so > I can test it and start modifying my code. Is the beta period really > only going to take 1-2 months (already about 2.5 weeks into it)? I > know you're going as fast as you can... b1 was uploaded at 11 May. So we only 15 days in beta period. Yes, I have get working example with records add/delete/update first/last/prev/next Was many bugs in String, VarChar and Text fields when they work via cursor. I have spend 3 days catching them. Only one operation not works yet, and it require some thinking. Well, I can upload b4 right now. I will start package it. But it still will not give you YOUR app requirements. You use classes. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed May 26 16:59:42 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 08:59:54 2004 Subject: V4RB b4 uploaded Message-ID: http://paradigmasoft.com/download/v2_beta/v2_beta.html Hi All, You need download all 3 archives. Was added one small DLL for ICU In this beta 1) added Example "AddDeleteUpdate" it show work with Vcursor for all major field types. you can add/delete/update records, navigate them. KNOWN: deletion when are on not last error, works only once. 2) in above example if you click "select" then Delete you will see new "smart error message" when it give more useful info about error. 3) pay attention to sqlstates.xml file open it in MS Internet Explorer. you will be able collaps groups and see it. it seems we have also style file to see this XML file in the nice way. 4) Jon, now works unicode table and fields name when you use CreateTable(), ... I.e. Non SQL way. 5) Jon, you can try to modify new example to add/read unicode strings into String/Varchar and Text fields. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Wed May 26 10:20:56 2004 From: jda at his.com (jda) Date: Wed May 26 09:21:08 2004 Subject: V4RB b4 uploaded In-Reply-To: References: Message-ID: I tried to create a database with a Greek name (databaseCreateOpenClose). Two problems: res = mDataBase.Create and res = mDatabase.open -> error, this method does not return a result. If I remove the 'res =' it compiles. The database name in the Finder was garbage: ??? > > >4) Jon, now works unicode table and fields name > when you use CreateTable(), ... I.e. Non SQL way. Right. Works fine (SQL fails). > >5) Jon, you can try to modify new example to add/read unicode strings into >String/Varchar and Text fields. > Works! Jon From sunshine at public.kherson.ua Wed May 26 20:05:26 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 12:05:40 2004 Subject: V4RB b4 uploaded In-Reply-To: Message-ID: On 5/26/04 5:20 PM, "jda" wrote: > I tried to create a database with a Greek name (databaseCreateOpenClose). > > Two problems: > > res = mDataBase.Create and res = mDatabase.open -> error, this method > does not return a result. > If I remove the 'res =' it compiles. > > The database name in the Finder was garbage: ??? Db names do not support yet unicode. This is on TODO -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed May 26 20:07:02 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 12:07:12 2004 Subject: V4RB b4 uploaded In-Reply-To: Message-ID: On 5/26/04 5:20 PM, "jda" wrote: >> 4) Jon, now works unicode table and fields name >> when you use CreateTable(), ... I.e. Non SQL way. > > Right. Works fine (SQL fails). >> 5) Jon, you can try to modify new example to add/read unicode strings into >> String/Varchar and Text fields. >> > > Works! Great! :-) -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Wed May 26 20:29:30 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 12:29:42 2004 Subject: Advice please on example organization ? Message-ID: Hi All, I want get advices on example organizations for 2.0 products. The task is next: - we have in e.g. V4RB, C++, Java, 3 ways to work with Valentina engines: 1) SQL way. "CREATE TABLE", "INSERT INTO", DELETE WHERE, ... SELECT 2) API way. Many commands work with API functions. SELECT we still do as SQL, but later work with Vcursor API. 3) Classes way. It allow in nice OO way define structure of tables. in V4RB 2.0 we will introduce new Set classes and new Find() functions in Vtable class so you will be able do quite complex tasks with no SQL at all. I sure now many of you will understand why we was need that classes. Although this will work for LOCAL dbs only. At least now. - in Director and other products we have only 2 first ways. So, for V4RB from one side it is good idea do examples as Common SQL way API way Classes way. If follow this way, then current example CreateDropTable Must be splinted on 2 projects to show only SQL and only API ways. >From one side this is good, because we get simpler examples, And if developer want study SQL way, he will not see other "garbage". >From other side, if we mix all 3 ways in one example, it is more easier compare ways to see what is better for you. I tend go by the first way: splinted isolated examples. What you think? ------ Mixing ways in one project make it more complex, even with 2 ways as we have now. When I will add third way, windows will be too big. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Wed May 26 14:55:51 2004 From: jda at his.com (jda) Date: Wed May 26 13:56:07 2004 Subject: Advice please on example organization ? In-Reply-To: References: Message-ID: > > >------ >Mixing ways in one project make it more complex, even with 2 ways as we have >now. When I will add third way, windows will be too big. > Well, it's a matter of taste, but I think Valentina is confusing enough without dozens and dozens of example projects, many showing how to do the same thing but in different ways. I think it makes sense to consolidate examples by function: Example 1: 3 ways to create/destroy a table (all in one window) Example 2: etc. You can always make smaller buttons! My $0.02 Jon From sunshine at public.kherson.ua Wed May 26 22:03:15 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 14:03:27 2004 Subject: Advice please on example organization ? In-Reply-To: Message-ID: On 5/26/04 9:55 PM, "jda" wrote: >> ------ >> Mixing ways in one project make it more complex, even with 2 ways as we have >> now. When I will add third way, windows will be too big. >> > > Well, it's a matter of taste, but I think Valentina is confusing > enough without dozens and dozens of example projects, many showing > how to do the same thing but in different ways. > > I think it makes sense to consolidate examples by function: > > Example 1: 3 ways to create/destroy a table (all in one window) > > Example 2: etc. > > You can always make smaller buttons! So you prefer MIXED examples? And as you have see I try EXPLAIN each step directly on Windows. This also eat space. again: - Assume you have folders SQL_way and API_way - both folders have the same example SQL CreateDropTable() API CreateDropTable() - so you can choose which way to study. Examples are simple. not confusing. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From jda at his.com Wed May 26 15:06:11 2004 From: jda at his.com (jda) Date: Wed May 26 14:06:20 2004 Subject: Advice please on example organization ? In-Reply-To: References: Message-ID: > >So you prefer MIXED examples? > >And as you have see I try EXPLAIN each step directly on Windows. >This also eat space. > >again: > >- Assume you have folders SQL_way and API_way > >- both folders have the same example > > SQL > CreateDropTable() > > API > CreateDropTable() > > >- so you can choose which way to study. Examples are simple. > not confusing. > I don't feel strongly about this. But simpler is better, of course, and you're going to spawn a lot of folders and projects this way. What do other folks think? Jon From sunshine at public.kherson.ua Wed May 26 22:09:36 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Wed May 26 14:09:44 2004 Subject: Advice please on example organization ? In-Reply-To: Message-ID: On 5/26/04 10:06 PM, "jda" wrote: >> So you prefer MIXED examples? >> >> And as you have see I try EXPLAIN each step directly on Windows. >> This also eat space. >> >> again: >> >> - Assume you have folders SQL_way and API_way >> >> - both folders have the same example >> >> SQL >> CreateDropTable() >> >> API >> CreateDropTable() >> >> >> - so you can choose which way to study. Examples are simple. >> not confusing. >> > > I don't feel strongly about this. But simpler is better, of course, > and you're going to spawn a lot of folders and projects this way. > What do other folks think? Yes will be many folders. although we will have 3 main sublevels -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From yeomans at desuetude.com Wed May 26 16:23:21 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Wed May 26 15:23:31 2004 Subject: Advice please on example organization ? In-Reply-To: References: Message-ID: <881B2034-AF52-11D8-B127-000A957CB4CC@desuetude.com> On May 26, 2004, at 3:06 PM, jda wrote: >> >> So you prefer MIXED examples? >> >> And as you have see I try EXPLAIN each step directly on Windows. >> This also eat space. >> >> again: >> >> - Assume you have folders SQL_way and API_way >> >> - both folders have the same example >> >> SQL CreateDropTable() >> API >> CreateDropTable() >> >> >> - so you can choose which way to study. Examples are simple. >> not confusing. >> > > I don't feel strongly about this. But simpler is better, of course, > and you're going to spawn a lot of folders and projects this way. What > do other folks think? > I think that focused examples are much more useful. When I have some time, I'll be offering feedback on them. I recently got an XServe for our firm, and I am doing a lot of catching up on Unix. Charles Yeomans From vidal_olivier at yahoo.fr Thu May 27 07:40:13 2004 From: vidal_olivier at yahoo.fr (olivier) Date: Thu May 27 00:35:07 2004 Subject: Advice please on example organization ? In-Reply-To: <881B2034-AF52-11D8-B127-000A957CB4CC@desuetude.com> References: <881B2034-AF52-11D8-B127-000A957CB4CC@desuetude.com> Message-ID: <5376AD42-AFA0-11D8-BFEB-00039310B7DA@yahoo.fr> Le 26 mai 04, ? 22:23, Charles Yeomans a ?crit : > I think that focused examples are much more useful. agree. olivier From sunshine at public.kherson.ua Thu May 27 18:17:53 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Thu May 27 10:18:43 2004 Subject: V4RB b4, VField.Value as VARAINT Message-ID: Hi guys, Forget to mention. Note that example AddDeleteRecords demonstrate new feature Vfield.Value as Variant. As you remember, in 1.x (and in 2.0 of course also) Each field class have property Value of correct type Vlong.value = 5 This functions are FASTER !!! than variant usage. But variant is more simple and may be some of you will find it more comfortable for use compare: curs.ULongField("a").value = 5 To variant way curs.Field("a").value = 5 You do not worry about field type. Also Variant way make V4RB more close to RBDB in syntax. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat May 29 01:59:20 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Fri May 28 17:59:33 2004 Subject: KERNEL, NEW Message-ID: Hi guys, Just for info. The next beta of V4RB includes such new features of kenel as ------------------ Vtable class ID as integer -- each table have unique ID. TMP tables have negative ID LinkCount -- the number of links from this table. links can be -- ObjectPtr -- Foreign Key -- Binary Link Link( index ) as VLink Link( name ) as VLink -------------- Field class ID as integer ------------ Database class LinkCount -- the number of links from this table. links can be -- ObjectPtr -- Foreign Key -- Binary Link Link( index ) as VLink Link( name ) as VLink ------------- Link class. Totally new class for V4RB // Ala-properties: { (REALproc) Link_get_Name, REALnoImplementation, "Name() as String"}, { (REALproc) Link_get_IsTemporary, REALnoImplementation, "IsTemporary() as Boolean"}, { (REALproc) Link_get_BranchCount, REALnoImplementation, "BranchCount() as Integer"}, { (REALproc) Link_get_OnDelete, REALnoImplementation, "OnDelete() as Integer"}, // { (REALproc) Link_put_OnDelete, REALnoImplementation, "OnDelete( Assigns v as Integer)"}, { (REALproc) Link_get_OnUpdate, REALnoImplementation, "OnUpdate() as Integer"}, // { (REALproc) Link_put_OnUpdate, REALnoImplementation, "OnUpdate( Assigns v as Integer)"}, // Methods: { (REALproc) Link_get_IsBetween, REALnoImplementation, "IsBetween( inTableA as VTable, inTableB as VTable ) as Boolean"}, { (REALproc) Link_get_Table, REALnoImplementation, "Table( inIndex as integer ) as VTable"}, // Search Methods: { (REALproc) Link_get_LinkedCount, REALnoImplementation, "CountLinked( inRecID as Integer, inTableA as VTable, inTableB as VTable ) as Integer"}, /* { (REALproc) Link_FindLinked, REALnoImplementation, "FindLinked( inRecID as Integer, inTableA as VTable, inTableB as VTable ) as VArraySet"}, { (REALproc) Link_FindExclusivelyLinked, REALnoImplementation, "FindExclusivelyLinked( inRecID as Integer, " "inTableA as VTable, inTableB as VTable ) as VArraySet"}, { (REALproc) Link_FindAllLinked, REALnoImplementation, "FindAllLinked( inTableA as VTable, inTableB as VTable ) as VBitSet"}, */ // { (REALproc) Link_LinkRecords, REALnoImplementation, "LinkRecords( inRecIDs() as Integer )"}, { (REALproc) Link_UnlinkRecords, REALnoImplementation, "UnlinkRecords( inRecIDs() as Integer )"}, { (REALproc) Link_DeleteLinkedRecords, REALnoImplementation, "DeleteLinkedRecords( inRecID as Integer, inTableA as VTable )"}, { (REALproc) Link_DeleteAllLinkedRecords, REALnoImplementation, "DeleteAllLinkedRecords( inTableA as VTable )"}, -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.macserve.net/pipermail/valentina-beta/attachments/20040529/dabdb49e/attachment.html From yeomans at desuetude.com Fri May 28 19:15:16 2004 From: yeomans at desuetude.com (Charles Yeomans) Date: Fri May 28 18:15:29 2004 Subject: KERNEL, NEW In-Reply-To: References: Message-ID: This is great, I think --what do it mean? Charles Yeomans On May 28, 2004, at 6:59 PM, Ruslan Zasukhin wrote: > Hi guys, > > Just for info. > > The next beta of V4RB includes such new features of kenel as > > > ------------------ > Vtable class > > ????ID as integer ??-- each table have unique ID. > ????????????????????????TMP tables have negative ID > > ????LinkCount ?-- the number of links from this table. > ??????????????????links can be > ????????????????????????-- ObjectPtr > ????????????????????????-- Foreign Key > ????????????????????????-- Binary Link > > ????Link( index ) as VLink > ????Link( name ) as VLink > > > -------------- > Field class > > ????ID as integer > > > > ------------ > Database class > > ????LinkCount ?-- the number of links from this table. > ??????????????????links can be > ????????????????????????-- ObjectPtr > ????????????????????????-- Foreign Key > ????????????????????????-- Binary Link > > ????Link( index ) as VLink > ????Link( name ) as VLink > > > ------------- > Link class. Totally new class for V4RB > > > // Ala-properties: > > ????{ ???(REALproc) Link_get_Name, ???REALnoImplementation, > ????????"Name() as String"}, > > ????{ ???(REALproc) Link_get_IsTemporary, ???REALnoImplementation, > "IsTemporary() as Boolean"}, > ????{ ???(REALproc) Link_get_BranchCount, ???REALnoImplementation, > "BranchCount() as Integer"}, > > ????{ ???(REALproc) Link_get_OnDelete, ???REALnoImplementation, > "OnDelete() as Integer"}, > // ???{ ???(REALproc) Link_put_OnDelete, ???REALnoImplementation, > "OnDelete( Assigns v as Integer)"}, > > ????{ ???(REALproc) Link_get_OnUpdate, ???REALnoImplementation, > "OnUpdate() as Integer"}, > // ???{ ???(REALproc) Link_put_OnUpdate, ???REALnoImplementation, > "OnUpdate( Assigns v as Integer)"}, > > // Methods: > > ????{ ???(REALproc) Link_get_IsBetween, ???REALnoImplementation, > ????????"IsBetween( inTableA as VTable, inTableB as VTable ) as > Boolean"}, > > ????{ ???(REALproc) Link_get_Table, ???REALnoImplementation, > ????????"Table( inIndex as integer ) as VTable"}, > ???? > > // Search Methods: > > ????{ ???(REALproc) Link_get_LinkedCount, ???REALnoImplementation, > ????????"CountLinked( inRecID as Integer, inTableA as VTable, > inTableB as VTable ) as Integer"}, > ???????? > /* ???{ ???(REALproc) Link_FindLinked, ???REALnoImplementation, > ????????"FindLinked( inRecID as Integer, inTableA as VTable, inTableB > as VTable ) as VArraySet"}, > > ????{ ???(REALproc) Link_FindExclusivelyLinked, > ???REALnoImplementation, > ????????"FindExclusivelyLinked( inRecID as Integer, " > ????????????????????????????????"inTableA as VTable, inTableB as > VTable ) as VArraySet"}, > > ????{ ???(REALproc) Link_FindAllLinked, ???REALnoImplementation, > ????????"FindAllLinked( inTableA as VTable, inTableB as VTable ) as > VBitSet"}, > */ > // > > ????{ ???(REALproc) Link_LinkRecords, ???REALnoImplementation, > ????????"LinkRecords( inRecIDs() as Integer )"}, > ???????? > ????{ ???(REALproc) Link_UnlinkRecords, ???REALnoImplementation, > ????????"UnlinkRecords( inRecIDs() as Integer )"}, > ???????? > ????{ ???(REALproc) Link_DeleteLinkedRecords, ???REALnoImplementation, > ????????"DeleteLinkedRecords( inRecID as Integer, inTableA as VTable > )"}, > > ????{ ???(REALproc) Link_DeleteAllLinkedRecords, > ???REALnoImplementation, > ????????"DeleteAllLinkedRecords( inTableA as VTable )"}, > > > -- > Best regards, > Ruslan Zasukhin ?????[ I feel the need...the need for speed ] > ------------------------------------------------------------- > e-mail: ruslan@paradigmasoft.com > web: http://www.paradigmasoft.com > > To subscribe to the Valentina mail list go to: > http://lists.macserve.net/mailman/listinfo/valentina > ------------------------------------------------------------- > _______________________________________________ > Valentina-beta mailing list > Valentina-beta@lists.macserve.net > http://lists.macserve.net/mailman/listinfo/valentina-beta > -------------- Charles Yeomans From sunshine at public.kherson.ua Sat May 29 08:42:24 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 29 00:42:36 2004 Subject: KERNEL, NEW In-Reply-To: Message-ID: On 5/29/04 2:15 AM, "Charles Yeomans" wrote: > This is great, I think --what do it mean? Link -- new abstraction in Valentina data model. Link is abstraction of THE SAME level as Table. So this is 100% extension of Relational model. You can think about link as it is in ER model / Entity - Relationship. We already have had links: ObjectPtr and RDB-link, but V4RB developers did not have access to them. Also in V4RB we make in classes Set, ArraySet and BitSet. All this tools will allow use V4RB in NON-SQL way, Similar to Navigation model, but again not exactly navigational. It is Extended Navigational, I'd say. I write docs theory about all this models Navigational Relational OO OR Valentina model Valentina model combines all pluses of all models I believe. So this tools, will allow V4RB developers work in Navigational style. Using only Vdatabase, Vtable, Field, Sets, Links I.e. Not using SQL and cursor. Often this can be much faster of SQL way. The problem os this way -- it not works with client/server. At least now. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From sunshine at public.kherson.ua Sat May 29 09:45:01 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sat May 29 01:45:11 2004 Subject: KERNEL, NEW In-Reply-To: Message-ID: On 5/29/04 1:59 AM, "Ruslan Zasukhin" wrote: Another new thing: Each table now have ALWAYS 2 virtual fields RecID -- unique ID of record in table. OID -- unique ID of record in the scope of database. 8 bytes. Problem is that REALbasic do not support 8 bytes integers. OID -- this follow to ODMG standard of OO dbs. But only Valentina eat ZERO byte on disk space and in RAM for these fields!!! Just image, all other dbs, including OO DBMS, will eat in million db 8MB of disk space, and even more if present index. Having now OID, I have start develop idea of ObjectsPtr, Note, it have "Objects", not "Object" This ObjectsPtr by idea, will have size 8 byte, And it will be able point TO ANY record of ANY table in the database! This ObjectsPtr is not so simple thing. It have own traps. I have not yet complete theoretical basic for it. > Hi guys, > > Just for info. > > The next beta of V4RB includes such new features of kenel as > > > ------------------ > Vtable class > > ID as integer -- each table have unique ID. > TMP tables have negative ID > > LinkCount -- the number of links from this table. > links can be > -- ObjectPtr > -- Foreign Key > -- Binary Link > > Link( index ) as VLink > Link( name ) as VLink > > > -------------- > Field class > > ID as integer > > > > ------------ > Database class > > LinkCount -- the number of links from this table. > links can be > -- ObjectPtr > -- Foreign Key > -- Binary Link > > Link( index ) as VLink > Link( name ) as VLink > > > ------------- > Link class. Totally new class for V4RB > > > // Ala-properties: > > { (REALproc) Link_get_Name, REALnoImplementation, > "Name() as String"}, > > { (REALproc) Link_get_IsTemporary, REALnoImplementation, > "IsTemporary() as Boolean"}, > { (REALproc) Link_get_BranchCount, REALnoImplementation, > "BranchCount() as Integer"}, > > { (REALproc) Link_get_OnDelete, REALnoImplementation, "OnDelete() > as Integer"}, > // { (REALproc) Link_put_OnDelete, REALnoImplementation, "OnDelete( > Assigns v as Integer)"}, > > { (REALproc) Link_get_OnUpdate, REALnoImplementation, "OnUpdate() > as Integer"}, > // { (REALproc) Link_put_OnUpdate, REALnoImplementation, "OnUpdate( > Assigns v as Integer)"}, > > // Methods: > > { (REALproc) Link_get_IsBetween, REALnoImplementation, > "IsBetween( inTableA as VTable, inTableB as VTable ) as Boolean"}, > > { (REALproc) Link_get_Table, REALnoImplementation, > "Table( inIndex as integer ) as VTable"}, > > > // Search Methods: > > { (REALproc) Link_get_LinkedCount, REALnoImplementation, > "CountLinked( inRecID as Integer, inTableA as VTable, inTableB as > VTable ) as Integer"}, > > /* { (REALproc) Link_FindLinked, REALnoImplementation, > "FindLinked( inRecID as Integer, inTableA as VTable, inTableB as > VTable ) as VArraySet"}, > > { (REALproc) Link_FindExclusivelyLinked, REALnoImplementation, > "FindExclusivelyLinked( inRecID as Integer, " > "inTableA as VTable, inTableB as VTable ) as > VArraySet"}, > > { (REALproc) Link_FindAllLinked, REALnoImplementation, > "FindAllLinked( inTableA as VTable, inTableB as VTable ) as > VBitSet"}, > */ > // > > { (REALproc) Link_LinkRecords, REALnoImplementation, > "LinkRecords( inRecIDs() as Integer )"}, > > { (REALproc) Link_UnlinkRecords, REALnoImplementation, > "UnlinkRecords( inRecIDs() as Integer )"}, > > { (REALproc) Link_DeleteLinkedRecords, REALnoImplementation, > "DeleteLinkedRecords( inRecID as Integer, inTableA as VTable )"}, > > { (REALproc) Link_DeleteAllLinkedRecords, REALnoImplementation, > "DeleteAllLinkedRecords( inTableA as VTable )"}, > -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina ------------------------------------------------------------- From lists at praxisworks.com Sun May 30 09:08:40 2004 From: lists at praxisworks.com (Erik Mueller-Harder) Date: Sun May 30 08:08:47 2004 Subject: V4RB b4 uploaded In-Reply-To: Message-ID: jda wrote: > res = mDataBase.Create and res = mDatabase.open -> error, this > method does not return a result. > > If I remove the 'res =' it compiles. I'm getting this same result, even when using the default name of "DataBase1.vdb". V4RB 2.0b4 Rb 5.5.2 OS X 10.3.4 Erik From rjb at robelko.com Sun May 30 15:47:59 2004 From: rjb at robelko.com (Robert Brenstein) Date: Sun May 30 08:54:20 2004 Subject: Advice please on example organization ? In-Reply-To: <5376AD42-AFA0-11D8-BFEB-00039310B7DA@yahoo.fr> References: <881B2034-AF52-11D8-B127-000A957CB4CC@desuetude.com> <5376AD42-AFA0-11D8-BFEB-00039310B7DA@yahoo.fr> Message-ID: >Le 26 mai 04, ? 22:23, Charles Yeomans a ?crit : > >>I think that focused examples are much more useful. > >agree. > >olivier > In short: Me too. In long: I would keep parallel examples for each way of working with Valentina in a separate properly labelled folder. This will makes things easier for beginners and faster to find for more advanced users. And having multiple approaches in a single example can be misleading. As we have seen on the list, it is not uncommon for people to not properly distinguish between sql and direct api's and mixing them in not appropriate ways. A few simple focused examples for basic tasks is the way to go. But, it may be a good idea to include also a couple more advanced examples showing more complete (but still simple) projects to show how the basics work together. I think this split approach will also make things easier for you, Ruslan, in terms of maintaining parallel examples for different environments. Robert Brenstein From sunshine at public.kherson.ua Sun May 30 17:04:47 2004 From: sunshine at public.kherson.ua (Ruslan Zasukhin) Date: Sun May 30 09:05:09 2004 Subject: Advice please on example organization ? In-Reply-To: Message-ID: On 5/30/04 4:47 PM, "Robert Brenstein" wrote: >> >>> I think that focused examples are much more useful. >> >> agree. >> >> olivier >> > > In short: Me too. > > In long: > > I would keep parallel examples for each way of > working with Valentina in a separate properly > labelled folder. This is exactly what I am doing Robert! Common SQL_way API_way Classes_way > This will makes things easier > for beginners and faster to find for more > advanced users. And having multiple approaches in > a single example can be misleading. As we have > seen on the list, it is not uncommon for people > to not properly distinguish between sql and > direct api's and mixing them in not appropriate > ways. right > A few simple focused examples for basic tasks is > the way to go. But, it may be a good idea to > include also a couple more advanced examples > showing more complete (but still simple) projects > to show how the basics work together. Right, in ideal must be 1) Examples one for each feature, as above. so will be a lots of examples in this area. 2) few ADVANCED examples, e.g. CD finder > I think this split approach will also make things > easier for you, Ruslan, in terms of maintaining > parallel examples for different environments. Exactly! We going to grow up document of Examples list that will be neutral to IDE. Although we still will have some language-specific examples. -- Best regards, Ruslan Zasukhin [ I feel the need...the need for speed ] ------------------------------------------------------------- e-mail: ruslan@paradigmasoft.com web: http://www.paradigmasoft.com To subscribe to the Valentina mail list go to: http://lists.macserve.net/mailman/listinfo/valentina -------------------------------------------------------------