Re: for psqlodbc developers: [ psqlodbc-Bugs-1000507 ]

Agree with Dave and Luf's instinct whole heartedly.
The driver should support things in the SQL and ODBC standards, particularly DML (INSERT,UPDATE, DELETE,
SELECT).

Hopefully the application developer builds a Data Access Layer to handle the variations of PostgreSQL, MS
SQL Sever, Oracle, MySQL, etc.

Dave Page wrote:
> This would require fully parsing statements and would lead to an endless
> stream of similar requests for rewriting other quirky syntaxes IMHO. This is
> not something we can support.
>
> Regards, Dave
>
>
> On 27/12/05 9:11 am, "Ludek Finstrle" <luf@pzkagis.cz> wrote:
>
>
>>I ask other developers: what's your opinion?
>>
>>I know the man may need it but I don't think we want it. We
>>have to support Oracle, Informix, MySQL, ... special commands than.
>>
>>I hope we don't need parse statements before execute them
>>in new development version so it could be slowdown in the future.
>>
>>Regards,
>>
>>Luf
>>
>>
>>>Summary: MSSQL syntax
>>>
>>>Initial Comment:
>>>MSSQL have a command that Postgresql don't support:
>>>
>>>@@spid
>>>
>>>This command returns MSSQL connection pid.
>>>
>>>I've a application that uses this query. I can't change code, but
>>>i need that psqlodbc ignore this command.
>>>
>>>----------------------------------------------------------------------
>>>
>>>
>>>>Comment By: Ludek Finstrle (luf)
>>>
>>>Date: 2005-12-27 10:01
>>>
>>>Message:
>>>I don't know if we want this behaviour. When you move to other DB
>>>do you think they will add support for this?
>>>
>>>I need to talk about it with other developers. I'm againist this
>>>kind of support.
>>>
>>>@@spid is whole command or is it located inside statements?
>>>
>>>If you're a little bit familiar with C you can hack it into psqlodbc
>>>code yourself (I think in PGAPI_Prepare and PGAPI_ExecDirect).
>>>
>>>----------------------------------------------------------------------
>>>
>>>Comment By: Nobody (None)
>>>Date: 2005-12-27 02:25
>>>
>>>Message:
>>>can psqlodbc to transform all "@@xyz" syntax in:
>>>
>>>xyz()
>>>
>>>?????
>>
>>---------------------------(end of broadcast)---------------------------
>>TIP 6: explain analyze is your friend
>
>
> Regards, Dave
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 3: Have you checked our extensive FAQ?
>
>                http://www.postgresql.org/docs/faq