On Wed, 2011-08-31 at 13:50 -0700, John Fabiani wrote:
> signer, or AppWizard yet.
>
> I'm wondering if it might be a better idea to create unique odbc interfaces
> for each of the databases we are going to support using pyodbc. IOW's I'd
> like to see:
> dbMsSqlODBC.py
> dbDbfODBC.py
> dbOracleODBC.py
> etc...
>
> Rather than attempt to create a universal dbODBC.py where I attempt to deal
> with each database unique requirements, i.e. data types, getting the last
> insert PK, limits, etc.. This would be a match to what we are currently doing
> for each of the supported databases (postgres, mssql, mysql) currently.
Hmm, I was going to say that each of those databases has it own native
API - but that's not quite right because we are using pythons DBAPI.
I think we want to keep as much common as possible - there are close
to standard ways for doing most of that. So I think we should write an
Odbc layer which uses the standard ways where possible and then override
it to fix it for specific cases when it doesn't work.
In cases where it is known to work we can always document that as well
with say:
dbOracleODBC = dbOdbcCommon
in dbOdbc.py or somewhere as likely to be imported.
--
Roger Gammans <[email protected]>
--- StripMime Report -- processed MIME parts ---
multipart/signed
text/plain (text body -- kept)
application/pgp-signature
---
_______________________________________________
Post Messages to: [email protected]
Subscription Maintenance: http://leafe.com/mailman/listinfo/dabo-dev
Searchable Archives: http://leafe.com/archives/search/dabo-dev
This message: http://leafe.com/archives/byMID/1314825007.7512.6.camel@localhost
