Python and ODBC?
Ryujin
ryu-jin at bigfoot.com
Mon Jan 15 11:35:49 EST 2001
More information about the Python-list mailing list
Mon Jan 15 11:35:49 EST 2001
- Previous message (by thread): Python and ODBC?
- Next message (by thread): Python and ODBC?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
In article <93upsf02pvl at news1.newsguy.com>, "Alex Martelli" <aleaxit at yahoo.com> wrote: thx for your help matte > "Ryujin" <ryu-jin at bigfoot.com> wrote in message > news:93tfo0$12p$1 at ryu.it... >> Which is the best solution to work with python and a SQL database under > Windows? > > So much depends on what _scale_ you need to work...! Multi-user, or > single? Do you need transactions, stored procedures/triggers, etc? Even > typical database size is a factor. > > For really small-scale needs, for example, Gadfly has interesting > characteristics -- you might want to look into it -- but it is not > designed to scale up to intensive multi-user accesses or really huge > databases. > > So, hard to give a 100% answer, but, if you're Windows-only, you might > want to try MSDE, the freely-redistributable version of SQL Server > (shorn of the various graphical front-ends &c); you can download and > use/redistribute it freely if you have a license for Microsoft Studio or > Office Developer. It scales well up to 5 simultaneous users, and has > all the trimmings you'd expect from a professional RDBMS, such as > transactions, stored-procedures, etc, etc. If you don't care about > redistribution, you can also download and use (but not redistribute) SQL > Server Developer Edition, which is the same thing plus the graphical > aids for DB administration/development/debugging. > > It does have an upper boundary of 2 gigabytes per database -- which was > not a problem for us, but might be in other applications. > > I have not tried out the versions of those related to SQL Server 2000, > only the ones related to SQL Server 7 (the previous release), which were > still around for download on the MS site a couple months ago > (sorry, don't have the URL and don't know if they're still there, as > I found out the same bits also were on our MSDN Universal subscription > CDROM's, so I saved downloading the big packs) -- in any case, I have no > reason to doubt that SQL Server 2000 is at least as good (but I only did > a few tests on that myself, and not on the MSDE-equivalents). > >> I mean which module do I have to install on Python 2.0? > > That's part of what I like re MSDE -- you can talk to it directly from > Python via ADO and ADOX (the client-side parts of the MSDAC, which is > also freely downloadable/redistributable), and that turns out to be VERY > fast as well as quite convenient. It may be more so in our case since > we don't _only_ use Python to talk to those DB's; ADO/ADOX can be > accessed equally well from a great variety of programming/scripting > languages. > > Really, the only serious downside I can see is the very strong Microsoft > lock-in that is implicit in all of this. Since our app was > MS-platforms-only anyway for another variety of reason, I chose MSDE > with a perfectly clear conscience. If I had to avoid platform lock-ins, > and still get transactions and stored procedures > (each of those being a real 'must' for us), then, I dunno -- maybe > Interbase, now open-source, has those, but I don't really know much > about it; maybe Postgres (again, no first-hand experience). > > >> What about under Linux? I'm trying to install mxODBC module with MySQL >> db on a Linuxppc box. > > I'll pass on this one, having no experience with MySQL (I don't even > know if it supports transactions and stored procedures!). I do know > it's very popular, so I guess one can't go TOO far wrong with it. > > > Alex > > > -- --------------------------------------- Memelli Matteo Could u imagine a world without Windoz? ryu-jin at bigfoot.com ---------------------------------------
- Previous message (by thread): Python and ODBC?
- Next message (by thread): Python and ODBC?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Python-list mailing list