Thread: ODBC changes

ODBC changes

From
Hiroshi Inoue
Date:
Hi all,

I've committed the following changes.
Now psqlodbc driver is 7.01.0007.

1) Most driver options are now DSN options.
2) Add a new DSN option *Disallow Premature*.
   (I hope this would fix bugs reported by
   Keith Millard and Michael Rudolph).
3) Improve FE/BE conversation a little(will
   fix a bug reported by Kristis Markis).
4) Handle esacaped date/time parameters.
5) Improve declare/fetch mode a little.
    (select .. for update/select .. into ..)
6) Fix SQLForeignKeys() in multibyte mode.
7) Improve procedure calls.
8) Fix '\\' handling for bytea type.
9) Improve the memory usage in tuple allocation
   a little.
10) Fix a few bugs in parse_statement().

[Internal change]
1) Removed all internal ODBC API function calls.
   (ODBC 3.0 a little for test).
2) Renamed psqlodbc.def to psqlodbc_win32.def so as to
   avoid a conflict with cygwin port.

regards,
Hiroshi Inoue

Re: ODBC changes

From
Dave Page
Date:

> -----Original Message-----
> From: Hiroshi Inoue [mailto:Inoue@tpf.co.jp]
> Sent: 11 September 2001 04:52
> To: pgsql-odbc@postgresql.org
> Subject: [ODBC] ODBC changes
>
>
> Hi all,
>
> I've committed the following changes.
> Now psqlodbc driver is 7.01.0007.
>
> 1) Most driver options are now DSN options.
> 2) Add a new DSN option *Disallow Premature*.
>    (I hope this would fix bugs reported by
>    Keith Millard and Michael Rudolph).
> 3) Improve FE/BE conversation a little(will
>    fix a bug reported by Kristis Markis).
> 4) Handle esacaped date/time parameters.
> 5) Improve declare/fetch mode a little.
>     (select .. for update/select .. into ..)
> 6) Fix SQLForeignKeys() in multibyte mode.
> 7) Improve procedure calls.
> 8) Fix '\\' handling for bytea type.
> 9) Improve the memory usage in tuple allocation
>    a little.
> 10) Fix a few bugs in parse_statement().
>
> [Internal change]
> 1) Removed all internal ODBC API function calls.
>    (ODBC 3.0 a little for test).
> 2) Renamed psqlodbc.def to psqlodbc_win32.def so as to
>    avoid a conflict with cygwin port.

I've built and packaged this new version now and have included a Merge
Module for those that need it. I'm having a little trouble getting it onto
the ftp server at present, but once it's available for download I'll let the
list know.

Regards, Dave.

Re: ODBC changes

From
Dave Page
Date:

> -----Original Message-----
> From: Hiroshi Inoue [mailto:Inoue@tpf.co.jp]
> Sent: 11 September 2001 04:52
> To: pgsql-odbc@postgresql.org
> Subject: [ODBC] ODBC changes
>
>
> Hi all,
>
> I've committed the following changes.
> Now psqlodbc driver is 7.01.0007.
>
> 1) Most driver options are now DSN options.
> 2) Add a new DSN option *Disallow Premature*.
>    (I hope this would fix bugs reported by
>    Keith Millard and Michael Rudolph).
> 3) Improve FE/BE conversation a little(will
>    fix a bug reported by Kristis Markis).
> 4) Handle esacaped date/time parameters.
> 5) Improve declare/fetch mode a little.
>     (select .. for update/select .. into ..)
> 6) Fix SQLForeignKeys() in multibyte mode.
> 7) Improve procedure calls.
> 8) Fix '\\' handling for bytea type.
> 9) Improve the memory usage in tuple allocation
>    a little.
> 10) Fix a few bugs in parse_statement().
>
> [Internal change]
> 1) Removed all internal ODBC API function calls.
>    (ODBC 3.0 a little for test).
> 2) Renamed psqlodbc.def to psqlodbc_win32.def so as to
>    avoid a conflict with cygwin port.

Hiroshi,

I've been playing with this new version a little and am seeing some
differences with the error messages raised. It's difficult to tell exactly
what's going on because my own code is somewhat 'in limbo' at the moment,
but with 07.01.0006, if I try to connect to a server that's not running I
get:

2001-09-12 08:51:14 - Error in pgAdmin II:frmConnect.cmdConnect_Click:
-2147467259 - Could not connect to the server; Could not connect to remote
socket.

But with 07.01.0007 I get:

2001-09-12 08:54:46 - Error in pgAdmin II:frmConnect.cmdConnect_Click:
-2147467259 - Unspecified error

The errors reported by pgAdmin II are as returned by ADO from the ODBC
driver. Any ideas?

Regards, Dave.

Re: ODBC changes

From
Hiroshi Inoue
Date:
Dave Page wrote:
>
> Hiroshi,
>
> I've been playing with this new version a little and am seeing some
> differences with the error messages raised. It's difficult to tell exactly
> what's going on because my own code is somewhat 'in limbo' at the moment,
> but with 07.01.0006, if I try to connect to a server that's not running I
> get:
>
> 2001-09-12 08:51:14 - Error in pgAdmin II:frmConnect.cmdConnect_Click:
> -2147467259 - Could not connect to the server; Could not connect to remote
> socket.
>
> But with 07.01.0007 I get:
>
> 2001-09-12 08:54:46 - Error in pgAdmin II:frmConnect.cmdConnect_Click:
> -2147467259 - Unspecified error
>
> The errors reported by pgAdmin II are as returned by ADO from the ODBC
> driver. Any ideas?

Hmm it seems to work if I remove ODBC3.0 functions from
psqlodbc_win32.def.

regards,
Hiroshi Inoue

Re: ODBC changes

From
Dave Page
Date:

> -----Original Message-----
> From: Hiroshi Inoue [mailto:Inoue@tpf.co.jp]
> Sent: 12 September 2001 11:31
> To: Dave Page
> Cc: pgsql-odbc@postgresql.org
> Subject: Re: [ODBC] ODBC changes
>
>
> Dave Page wrote:
> >
> > Hiroshi,
> >
> > I've been playing with this new version a little and am seeing some
> > differences with the error messages raised. It's difficult to tell
> > exactly what's going on because my own code is somewhat 'in
> limbo' at
> > the moment, but with 07.01.0006, if I try to connect to a server
> > that's not running I
> > get:
> >
> > 2001-09-12 08:51:14 - Error in pgAdmin
> II:frmConnect.cmdConnect_Click:
> > -2147467259 - Could not connect to the server; Could not connect to
> > remote socket.
> >
> > But with 07.01.0007 I get:
> >
> > 2001-09-12 08:54:46 - Error in pgAdmin
> II:frmConnect.cmdConnect_Click:
> > -2147467259 - Unspecified error
> >
> > The errors reported by pgAdmin II are as returned by ADO
> from the ODBC
> > driver. Any ideas?
>
> Hmm it seems to work if I remove ODBC3.0 functions from
> psqlodbc_win32.def.

Ahh, could it be that ADO is using SQLGetDiagRec instead of SQLError because
it thinks it's ODBC3.0? If so, that would tie in with some of the other
oddities I've seen with my code (which is calling SQLGetDiagRec directly).

Regards, Dave.

Re: ODBC changes

From
"Hiroshi Inoue"
Date:
> -----Original Message-----
> From: Dave Page
> > -----Original Message-----
> > From: Hiroshi Inoue [mailto:Inoue@tpf.co.jp]
> > Sent: 12 September 2001 11:31
> > To: Dave Page
> > Cc: pgsql-odbc@postgresql.org
> > Subject: Re: [ODBC] ODBC changes
> >
> >
> > Dave Page wrote:
> > >
> > > Hiroshi,
> > >
> > > I've been playing with this new version a little and am seeing some
> > > differences with the error messages raised. It's difficult to tell
> > > exactly what's going on because my own code is somewhat 'in
> > limbo' at
> > > the moment, but with 07.01.0006, if I try to connect to a server
> > > that's not running I
> > > get:
> > >
> > > 2001-09-12 08:51:14 - Error in pgAdmin
> > II:frmConnect.cmdConnect_Click:
> > > -2147467259 - Could not connect to the server; Could not connect to
> > > remote socket.
> > >
> > > But with 07.01.0007 I get:
> > >
> > > 2001-09-12 08:54:46 - Error in pgAdmin
> > II:frmConnect.cmdConnect_Click:
> > > -2147467259 - Unspecified error
> > >
> > > The errors reported by pgAdmin II are as returned by ADO
> > from the ODBC
> > > driver. Any ideas?
> >
> > Hmm it seems to work if I remove ODBC3.0 functions from
> > psqlodbc_win32.def.
>
> Ahh, could it be that ADO is using SQLGetDiagRec instead of
> SQLError because
> it thinks it's ODBC3.0? If so, that would tie in with some of the other
> oddities I've seen with my code (which is calling SQLGetDiagRec directly).

OK I would change the driver to not export ODBC3.0 functions
together with a bug fix.
Anyway ODBC3.0 stuff is only for test.

regards,
Hiroshi Inoue


Re: ODBC changes

From
Hiroshi Inoue
Date:
Hiroshi Inoue wrote:
>
> > -----Original Message-----
> > From: Dave Page

[snip]

> > >
> > > Hmm it seems to work if I remove ODBC3.0 functions from
> > > psqlodbc_win32.def.
> >
> > Ahh, could it be that ADO is using SQLGetDiagRec instead of
> > SQLError because
> > it thinks it's ODBC3.0? If so, that would tie in with some of the other
> > oddities I've seen with my code (which is calling SQLGetDiagRec directly).
>
> OK I would change the driver to not export ODBC3.0 functions
> together with a bug fix.

I've just committed the change to cvs.

regards,
Hiroshi Inoue

Re: ODBC changes

From
Dave Page
Date:

> -----Original Message-----
> From: Hiroshi Inoue [mailto:Inoue@tpf.co.jp]
> Sent: 13 September 2001 01:48
> To: Dave Page; pgsql-odbc@postgresql.org
> Subject: Re: [ODBC] ODBC changes
>
>
> Hiroshi Inoue wrote:
> >
> > > -----Original Message-----
> > > From: Dave Page
>
> [snip]
>
> > > >
> > > > Hmm it seems to work if I remove ODBC3.0 functions from
> > > > psqlodbc_win32.def.
> > >
> > > Ahh, could it be that ADO is using SQLGetDiagRec instead
> of SQLError
> > > because it thinks it's ODBC3.0? If so, that would tie in
> with some
> > > of the other oddities I've seen with my code (which is calling
> > > SQLGetDiagRec directly).
> >
> > OK I would change the driver to not export ODBC3.0
> functions together
> > with a bug fix.
>
> I've just committed the change to cvs.
>
> regards,
> Hiroshi Inoue

Just tried that - it compiles and works fine here. I'll redo the 07.01.0007
distribution with this version, unfortunately I still can't log into the ftp
server to upload it though - does anyone know if Marc's on holiday
(vacation)?

Regards, Dave.

Re: ODBC changes

From
Dave Page
Date:

> -----Original Message-----
> From: Hiroshi Inoue [mailto:Inoue@tpf.co.jp]
> Sent: 13 September 2001 01:48
> To: Dave Page; pgsql-odbc@postgresql.org
> Subject: Re: [ODBC] ODBC changes
>
>
> Hiroshi Inoue wrote:
> >
> > > -----Original Message-----
> > > From: Dave Page
>
> [snip]
>
> > > >
> > > > Hmm it seems to work if I remove ODBC3.0 functions from
> > > > psqlodbc_win32.def.
> > >
> > > Ahh, could it be that ADO is using SQLGetDiagRec instead
> of SQLError
> > > because it thinks it's ODBC3.0? If so, that would tie in
> with some
> > > of the other oddities I've seen with my code (which is calling
> > > SQLGetDiagRec directly).
> >
> > OK I would change the driver to not export ODBC3.0
> functions together
> > with a bug fix.
>
> I've just committed the change to cvs.

OK, thanks. I'll give it a go.