ODBC Driver Date/Text Bug

Hi,

Is there any indication as to when a new release of the Windows ODBC
driver will be available that doesn't suffer from the date/time fields
being mangled to text? The latest version I could find is 3.51.11, and
that still doesn't work. Is there a pre-release binary snapshot in which
this has been fixed that I could get? Sadly I con't have a C compiler
for Windows, which makes it more difficult. :-(

Many thanks.

Gordan

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org

Re: ODBC Driver Date/Text Bug

Gordan Bobic wrote:

> Hi,
>
> Is there any indication as to when a new release of the Windows ODBC
> driver will be available that doesn't suffer from the date/time fields
> being mangled to text? The latest version I could find is 3.51.11, and
> that still doesn't work. Is there a pre-release binary snapshot in
> which this has been fixed that I could get? Sadly I con't have a C
> compiler for Windows, which makes it more difficult. :-(
>
3.51.06 used to work well for me.
Otherwise look at bug #8363 on the MySQL bug system for a pre-release
that seems to fix this issue.
Unfortunately I'm starting to get the impression that MyODBC is a bit of
an afterthought. I suppose most people who use MySQL are using PHP or Perl.

--
Daniel Kasak
IT Developer
NUS Consulting Group
Level 5, 77 Pacific Highway
North Sydney, NSW, Australia 2060
T: (+61) 2 9922-7676 / F: (+61) 2 9922 7989
email: dkasak@nusconsulting.com.au
website: http://www.nusconsulting.com.au

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org

Re: ODBC Driver Date/Text Bug

Daniel Kasak wrote:

>>Is there any indication as to when a new release of the Windows ODBC
>>driver will be available that doesn't suffer from the date/time fields
>>being mangled to text? The latest version I could find is 3.51.11, and
>>that still doesn't work. Is there a pre-release binary snapshot in
>>which this has been fixed that I could get? Sadly I con't have a C
>>compiler for Windows, which makes it more difficult. :-(
>>
>
> 3.51.06 used to work well for me.

Indeed, and I think that 3.51.09 also works. Sadly, the support for the
new authentication method used by 5.x is only implemented in the ODBC
drivers v3.51.10 and later, and both 3.51.10 and 3.51.11 have broken
date fields.

> Otherwise look at bug #8363 on the MySQL bug system for a pre-release
> that seems to fix this issue.

Sadly, the 3.51.11-2(-win) driver for windows doesn't seem to be
available. Only the Linux x86 RPM version of 11-2 is available for
download. Any chance that somebody could email me a copy or a download
link to it? I am also lacking a Windows C compiler, so can't rebuild the
driver myself. :-(

That is all assuming that this would fix the problem on Win2K.

> Unfortunately I'm starting to get the impression that MyODBC is a bit of
> an afterthought. I suppose most people who use MySQL are using PHP or Perl.

That's probably a bit harsh. Personally, I'm more surprised that a bug
this serious has managed to get through two releases. It badly breaks
applications that try to do date/time operations on fields.

Gordan

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org

Re: ODBC Driver Date/Text Bug

Hi,
you can download MyODBC 3.51.11-2 from this link

http://www.peterharvey.org/Downloads/MySQL/MyODBC/

Regards,
Christian

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org

Re: ODBC Driver Date/Text Bug

Christian Surya wrote:
> Hi,
> you can download MyODBC 3.51.11-2 from this link
>
> http://www.peterharvey.org/Downloads/MySQL/MyODBC/

Many thanks. :-)

Gordan

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org

Re: ODBC Driver Date/Text Bug

Gordan Bobic wrote:

> Indeed, and I think that 3.51.09 also works. Sadly, the support for
> the new authentication method used by 5.x is only implemented in the
> ODBC drivers v3.51.10 and later, and both 3.51.10 and 3.51.11 have
> broken date fields.
>
No kidding.

>> Otherwise look at bug #8363 on the MySQL bug system for a pre-release
>> that seems to fix this issue.
>
>
> Sadly, the 3.51.11-2(-win) driver for windows doesn't seem to be
> available. Only the Linux x86 RPM version of 11-2 is available for
> download. Any chance that somebody could email me a copy or a download
> link to it? I am also lacking a Windows C compiler, so can't rebuild
> the driver myself. :-(
>
> That is all assuming that this would fix the problem on Win2K.
>
In the bug I referred to there is a link to the pre-release binaries.
OK fine then. Here's a direct link:
http://www.peterharvey.org/Downloads/MySQL/MyODBC/

>> Unfortunately I'm starting to get the impression that MyODBC is a bit of
>> an afterthought. I suppose most people who use MySQL are using PHP or
>> Perl.
>
>
> That's probably a bit harsh.

I don't think my comment was harsh. In fact I think it was right on the
ball. And now I will expand on it. 4.1.x has been 'stable' for how long
now? If MyODBC were *not* an afterthought, surely there'd be a stable
driver that could talk to the stable server? It's not like the MyODBC
developer(s) didn't know about the authentication changes, and it was
somehow 'sprung' on them with the onset of a stable 4.1 release. 4.1 has
been out for *ages*.

> Personally, I'm more surprised that a bug this serious has managed to
> get through two releases. It badly breaks applications that try to do
> date/time operations on fields.

That's right. A stable release is a stable release. People shouldn't
have to cherry-pick between them. Ever used MyODBC-3.51.07? Or 3.51.08?
Or 3.51.10? Or 3.51.11? Then you know what I'm talking about. I have no
problem with development talking place that sometimes breaks things ...
but this is what the *unstable* branch is for. However there are *no*
unstable releases ... apart from the stable releases of course. As for
the phantom unstable releases - 3.53 has been 'coming soon' for how
long? 2 years?

Instead of swapping installation splash screens, icons and configuration
interfaces in and out and frigging around with new wizbang MSI
installers, I would like to see some good old-fashioned features of
traditional open source releases ... like a changelog for instance. Or
new features ( new authentication system in a stable release ). Or
increased stability.

--
Daniel Kasak
IT Developer
NUS Consulting Group
Level 5, 77 Pacific Highway
North Sydney, NSW, Australia 2060
T: (+61) 2 9922-7676 / F: (+61) 2 9922 7989
email: dkasak@nusconsulting.com.au
website: http://www.nusconsulting.com.au

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org

Re: ODBC Driver Date/Text Bug

On Fri, Apr 01, 2005 at 09:08:06AM +1000, Daniel Kasak wrote:

> Gordan Bobic wrote:
>
> > That's probably a bit harsh.
>
> I don't think my comment was harsh. In fact I think it was right on the
> ball. And now I will expand on it. 4.1.x has been 'stable' for how long
> now? If MyODBC were *not* an afterthought, surely there'd be a stable
> driver that could talk to the stable server? It's not like the MyODBC
> developer(s) didn't know about the authentication changes, and it was
> somehow 'sprung' on them with the onset of a stable 4.1 release. 4.1 has
> been out for *ages*.
>
> > Personally, I'm more surprised that a bug this serious has managed to
> > get through two releases. It badly breaks applications that try to do
> > date/time operations on fields.
>
> That's right. A stable release is a stable release. People shouldn't
> have to cherry-pick between them. Ever used MyODBC-3.51.07? Or 3.51.08?
> Or 3.51.10? Or 3.51.11? Then you know what I'm talking about. I have no
> problem with development talking place that sometimes breaks things ...
> but this is what the *unstable* branch is for. However there are *no*
> unstable releases ... apart from the stable releases of course. As for
> the phantom unstable releases - 3.53 has been 'coming soon' for how
> long? 2 years?
>
> Instead of swapping installation splash screens, icons and configuration
> interfaces in and out and frigging around with new wizbang MSI
> installers, I would like to see some good old-fashioned features of
> traditional open source releases ... like a changelog for instance. Or
> new features ( new authentication system in a stable release ). Or
> increased stability.

Time to chime in!

I have used the odbc and jdbc connectors for years, and there has been a
huge difference in the experience. As an annointed devil's advocate:

o The relative difficulties in developing each type of connector should
not propagate to me. Trying to configure myodbc for maximum performance
should not be an arcane art.

o Access, Oracle, MSSQL, they seem to be a lot better at performing well
with crap database/application design. An end user is only going to
understand that the application ran a lot better when it was connected
to SQL Server. "That's a bad way to use a database" is a short sighted
answer. We are all painfully aware that your average developer sucks.
It's true, there's a lot of us out there. Every joker that connects a
button in VB to a database query thinks it's time to put software
engineer on their resume.

Ok, must jet. Someone mention the mysql "clustering" solution if you
want to get me foaming at the mouth.

p.s. No, I don't pay for support.
p.p.s. Yes, if I could afford Oracle, I would switch.


--
Chris Horn, CIO
Metrix Matrix Inc.
785 Elmgrove Road, Building 1, Rochester, NY 14624
chris.horn@metrixmatrix.com
888.865.0065 x200

--
MySQL ODBC Mailing List
For list archives: http://lists.mysql.com/myodbc
To unsubscribe: http://lists.mysql.com/myodbc?unsub=gcdmo-myodbc@m.gmane.org