Thread: Future versions of psqlODBC
Hi All, I recently polled the list for feedback of the 07.03.0260 experimental release of psqlODBC vs. 08.01.0200. Based on the few responses on list, and conversations I've had with other developers off-list, it seems that the REL-07_03_ENHANCED code branch should be used as the basis for future versions of psqlODBC. I therefore propose the following: 1) The current CVS code is branched into a REL-08_01_PATCHES branch to be used for bug fixes. 2) The REL-07_03_ENHANCED branch is merged into CVS tip, whilst maintaing the current installer and documentation etc. 3) The driver is rebranded to PostgreSQL+/psqlodbcplus.so/psqlodbcplus.lib to differentiate it from all previous versions. 4) The version number is bumped to 08.02.xxxx in preparation for a future formal release. Thoughts/comments/objections? Regards, Dave.
Dave Page wrote: [...] it seems that the REL-07_03_ENHANCED code branch should be used as the basis for future versions of psqlODBC. Hi Dave, as allways when there are ways to chose from, there should be some guide nudging the ignorant user (=me) in the "right" direction. Where is 7.3.x actually better than 8.1.2 ? Who would be encouraged to switch and the 07-branch anywhere near to be considered stable for productive use? Is it faster transfering data with office apps like Access? An informed pointing finger was nice. ;)
> -----Original Message----- > From: Andreas [mailto:maps.on@gmx.net] > Sent: 20 March 2006 18:23 > To: Dave Page > Cc: pgsql-odbc@postgresql.org > Subject: Re: [ODBC] Future versions of psqlODBC > > Dave Page wrote: > > [...] it seems that > the REL-07_03_ENHANCED code branch should be used as the basis for > future versions of psqlODBC. > > > > Hi Dave, > > as allways when there are ways to chose from, there should be > some guide > nudging the ignorant user (=me) in the "right" direction. > > Where is 7.3.x actually better than 8.1.2 ? It includes features such as Updateable cursors and savepoint support. > Who would be encouraged to switch and the 07-branch anywhere > near to be > considered stable for productive use? Dunno yet - it hasn't had particularly widespread testing. > Is it faster transfering data with office apps like Access? Also unknown at this point. > An informed pointing finger was nice. ;) Well when the time comes we will point users in the right direction, but at the moment the choice is from a developers perspective, which driver provides the best basis for future versions? It appears that the enhanced 07.03 branch is most peoples choice, so we will look at basing future versions of psqlODBC on that. At the moment though 08.01.0200 is the officially supported version. Regards, Dave
> I therefore propose the following: > > 1) The current CVS code is branched into a REL-08_01_PATCHES branch to > be used for bug fixes. Sure. > 2) The REL-07_03_ENHANCED branch is merged into CVS tip, whilst > maintaing the current installer and documentation etc. Sure. > 3) The driver is rebranded to > PostgreSQL+/psqlodbcplus.so/psqlodbcplus.lib to differentiate it from > all previous versions. I disagree. If I remember it right there was psqlodbcplus already. Why confuse users more? Why we need the differentiation? The driver is based on 7.3 version ... > 4) The version number is bumped to 08.02.xxxx in preparation for a > future formal release. I agree. Regards, Luf
> -----Original Message----- > From: pgsql-odbc-owner@postgresql.org > [mailto:pgsql-odbc-owner@postgresql.org] On Behalf Of Ludek Finstrle > Sent: 22 March 2006 11:58 > To: Dave Page > Cc: pgsql-odbc@postgresql.org; Ludek Finstrle; > anoopk@pervasive-postgres.com; Hiroshi Inoue; Hiroshi Saito > Subject: Re: [ODBC] Future versions of psqlODBC > > > I therefore propose the following: > > > > 1) The current CVS code is branched into a > REL-08_01_PATCHES branch to > > be used for bug fixes. > > Sure. > > > 2) The REL-07_03_ENHANCED branch is merged into CVS tip, whilst > > maintaing the current installer and documentation etc. > > Sure. > > > 3) The driver is rebranded to > > PostgreSQL+/psqlodbcplus.so/psqlodbcplus.lib to > differentiate it from > > all previous versions. > > I disagree. If I remember it right there was psqlodbcplus already. > Why confuse users more? Why we need the differentiation? The driver > is based on 7.3 version ... Well, either way it will be different from 08.01 because there is only one driver in the 7.3E branch, so there is no ANSI/Unicode distinction. As for the name, yes, there was a psqlodbcplus, but I thought they never actually released anything. On double checking that's not the case so that isn't such a good name as you say. We could just return to plain old psqlodbc/PostgreSQL ? Regards, Dave.
From: "Ludek Finstrle" > > 4) The version number is bumped to 08.02.xxxx in preparation for a > > future formal release. > > I agree. Yea, I want your achievement to be poured there. You are admired by many users. Furthermore, i wishes so again. Regards, Hiroshi Saito
> > > 3) The driver is rebranded to > > > PostgreSQL+/psqlodbcplus.so/psqlodbcplus.lib to > > > differentiate it from > > > all previous versions. > > > > I disagree. If I remember it right there was psqlodbcplus already. > > Why confuse users more? Why we need the differentiation? The driver > > is based on 7.3 version ... > > Well, either way it will be different from 08.01 because there is only > one driver in the 7.3E branch, so there is no ANSI/Unicode distinction. I don't know that we want release only one driver. It's ok to have only one driver? You wrote that there were many problems pointed by users when psqlodbc switch to Unicode driver. > As for the name, ... > > We could just return to plain old psqlodbc/PostgreSQL ? It sounds good for me. Regards, Luf
> -----Original Message----- > From: pgsql-odbc-owner@postgresql.org > [mailto:pgsql-odbc-owner@postgresql.org] On Behalf Of Ludek Finstrle > Sent: 22 March 2006 14:36 > To: Dave Page > Cc: Ludek Finstrle; pgsql-odbc@postgresql.org; > anoopk@pervasive-postgres.com; Hiroshi Inoue; Hiroshi Saito > Subject: Re: [ODBC] Future versions of psqlODBC > > I don't know that we want release only one driver. It's ok to > have only > one driver? > You wrote that there were many problems pointed by users when psqlodbc > switch to Unicode driver. Yes, but they didn't have Hiroshi around to fix things for them then (he wrote the Unicode code), and none of the rest of us fully understood the issues. I have asked Hiroshi about the need for two drivers, and he told me there should be no problem with the 7.3E branch with just the one driver. Regards Dave.
"Dave Page" <dpage@vale-housing.co.uk> writes:
> We could just return to plain old psqlodbc/PostgreSQL ?
Please. This is going to be confusing enough anyway.
Do you expect to resurrect the separate ANSI/Unicode driver versions
someday? If so, please consider making the build create both driver
names right away, even if there's not any distinction at the moment.
Otherwise you'll be creating useless thrash in the packaging and
people's odbcinst.ini files, as they have to change the driver name
only to have to change it again later.
regards, tom lane
> -----Original Message----- > From: Tom Lane [mailto:tgl@sss.pgh.pa.us] > Sent: 22 March 2006 14:48 > To: Dave Page > Cc: Ludek Finstrle; pgsql-odbc@postgresql.org > Subject: Re: [ODBC] Future versions of psqlODBC > > "Dave Page" <dpage@vale-housing.co.uk> writes: > > We could just return to plain old psqlodbc/PostgreSQL ? > > Please. This is going to be confusing enough anyway. > > Do you expect to resurrect the separate ANSI/Unicode driver versions > someday? If so, please consider making the build create both driver > names right away, even if there's not any distinction at the moment. > Otherwise you'll be creating useless thrash in the packaging and > people's odbcinst.ini files, as they have to change the driver name > only to have to change it again later. Hiroshi tells me we won't need two drivers and that the 07 branch should not suffer the problems we had with 08.00 that caused us to move back to two. My only concern is the *nix build - other than OSX Panther which doesn't have a Unicode DM, does anyone know if having a Unicode only driver is going to cause any problems on any particular platforms? Regards, Dave.
Dave Page wrote: > > > > >>-----Original Message----- >>From: Tom Lane [mailto:tgl@sss.pgh.pa.us] >>Sent: 22 March 2006 14:48 >>To: Dave Page >>Cc: Ludek Finstrle; pgsql-odbc@postgresql.org >>Subject: Re: [ODBC] Future versions of psqlODBC >> >>"Dave Page" <dpage@vale-housing.co.uk> writes: >> >> >>>We could just return to plain old psqlodbc/PostgreSQL ? >>> >>> >>Please. This is going to be confusing enough anyway. >> >>Do you expect to resurrect the separate ANSI/Unicode driver versions >>someday? If so, please consider making the build create both driver >>names right away, even if there's not any distinction at the moment. >>Otherwise you'll be creating useless thrash in the packaging and >>people's odbcinst.ini files, as they have to change the driver name >>only to have to change it again later. >> >> > >Hiroshi tells me we won't need two drivers and that the 07 branch should >not suffer the problems we had with 08.00 that caused us to move back to >two. > AFAIR I didn't tell you such a thing. Though I've heard no problem from Japanese( e.g. about the use with MS Access) so far , I don't know if it solves the problems e.g. about BDE or under other languages' environment. regards, Hiroshi Inoue
> -----Original Message----- > From: Hiroshi Inoue [mailto:inoue@tpf.co.jp] > Sent: 22 March 2006 17:18 > To: Dave Page > Cc: Tom Lane; Ludek Finstrle; pgsql-odbc@postgresql.org > Subject: Re: [ODBC] Future versions of psqlODBC > > > AFAIR I didn't tell you such a thing. Though I've heard no > problem from > Japanese( e.g. about the > use with MS Access) so far , I don't know if it solves the problems > e.g. about BDE or under > other languages' environment. Oh, I apologise - I was misremembering what you said: > What do you mean by *non-latin1* ? As for the combination of > Japanase and > MS Access, it seems to work well. Can anyone who was suffering the latin character problems with the old Unicode driver try the 07.03 demo version please? Regards, Dave.
Ahh, thanks for testing Miguel. Are you hanging around for a while and prepared to run any more tests if needed? Regards,Dave. -----Original Message----- From: Miguel Juan [mailto:mjuan@cibal.es] Sent: Mon 3/27/2006 9:59 AM To: Dave Page; Hiroshi Inoue; pgsql-odbc@postgresql.org Subject: Re: [ODBC] Future versions of psqlODBC Hi all, I have tested the enhanced version (7.3.2.60) with BDE, and still have the same problems with text fields that have version 7.3.2.0. Basically, all the text fields doesn't appear when you make a query. The last working version with BDE is the 7.3.0.1 Regards, Miguel Juan NOTE: I send you a direct mail becouse my messages never appear on the list. ----- Original Message ----- From: "Dave Page" <dpage@vale-housing.co.uk> To: "Hiroshi Inoue" <inoue@tpf.co.jp> Cc: "Tom Lane" <tgl@sss.pgh.pa.us>; "Ludek Finstrle" <luf@pzkagis.cz>; <pgsql-odbc@postgresql.org> Sent: Wednesday, March 22, 2006 7:30 PM Subject: Re: [ODBC] Future versions of psqlODBC > -----Original Message----- > From: Hiroshi Inoue [mailto:inoue@tpf.co.jp] > Sent: 22 March 2006 17:18 > To: Dave Page > Cc: Tom Lane; Ludek Finstrle; pgsql-odbc@postgresql.org > Subject: Re: [ODBC] Future versions of psqlODBC > > > AFAIR I didn't tell you such a thing. Though I've heard no > problem from > Japanese( e.g. about the > use with MS Access) so far , I don't know if it solves the problems > e.g. about BDE or under > other languages' environment. Oh, I apologise - I was misremembering what you said: > What do you mean by *non-latin1* ? As for the combination of > Japanase and > MS Access, it seems to work well. Can anyone who was suffering the latin character problems with the old Unicode driver try the 07.03 demo version please? Regards, Dave. ---------------------------(end of broadcast)--------------------------- TIP 5: don't forget to increase your free space map settings
Miguel Juan wrote: > Hi all, > > I have tested the enhanced version (7.3.2.60) with BDE, and still have > the same problems with text fields that have version 7.3.2.0. Basically, > all the text fields doesn't appear when you make a query. Thanks. Could you send me the Mylog output ? regards, Hiroshi Inoue
Miguel Juan wrote: > Hello Hiroshi, Hi Miguel, > here you have the MyLog output. I just opened a connection, and ha a > look for some data in a table with 2 fields (varchar, text). Thanks but it looks like you are turning on the global option only. Could you turn on the DSN option also ? regards, Hiroshi Inoue
Dave Page schrieb: > >Can anyone who was suffering the latin character problems with the old >Unicode driver try the 07.03 demo version please? > >Regards, Dave. > >---------------------------(end of broadcast)--------------------------- >TIP 5: don't forget to increase your free space map settings > > > Hi Dave, a short (late) Test with: PostgreSQL 8.0X on Debian Sarge Unicode Database Application with psqlodbc+ 7.02.02.60 on Win XP Sp2 Operation: Reading and writing of a, A, u, U, o, O-Umlaut and sz (german characters) Data type: varchar(40) Remarks: I got some odd errors when leaving my application (only once) Protocol: 7.4+ and 6.4 Result: success That never worked with any unicode version so far. The result is remarkable since I can use a unicode database now. Congratulations! I have to repeat the test with Win ME. Regards, Johann
Hi Dave, here is another one, same problem like Miguel: PostgreSQL 8.0X on Debian Sarge Unicode Database Application with psqlodbc+ 7.02.02.60 on Win XP Sp2 Operation: Reading and writing of a, A, u, U, o, O-Umlaut and sz (german characters) Data type: text Remarks: none Protocol: 7.4+ Result: failed, insert works, querying doesn't work Remark: No errors in postgres.log Regards, Johann
Miguel Juan wrote: > Hi, > >> Hi Miguel, >> >>> here you have the MyLog output. I just opened a connection, and ha a >>> look for some data in a table with 2 fields (varchar, text). >> >> >> Thanks but it looks like you are turning on the global option only. >> Could you turn on the DSN option also ? >> > > Yes, my fault. I attached the mylog file with both options checked Thanks. Judging from the log you can see 4 records for both queries. Is it right ? regards, Hiroshi Inoue