Re: PYTHON, ODBC

From: Pierre-Frédéric Caillaud <lists(at)boutiquenumerique(dot)com>
To: "Matthew Story" <matthewstory(at)gmail(dot)com>
Cc: mstory(at)uchicago(dot)edu, pgsql-general(at)postgresql(dot)org
Subject: Re: PYTHON, ODBC
Date: 2005-01-09 19:47:08
Message-ID: opskco8uiacq72hf@musicbox
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

> due to the complicated nature of the database, and inability of zope

Well, I've found that Zope is very good to do a few things, and very bad
at the rest.

> to do what we need, and the problems with overhead we've been
> experiencing with rails due to the size of the database. Hope this

I like psycopy because selects with a lot of rows are processed
efficiently.

Look at the pydo library in the skunkweb project (google !)

I'd advise you to start writing a base class for your database object
(class DbObject) with a set of methods for setting instance methods from
query results. SELECT * then use cursor.dictfetchall() which preserves the
fields names !
Add methods for inserting and updating.
Add a mapping of fields to type converters (which are functions) to
convert non-standard types like arrays to python lists if you need them.

Then derive your DbObject class for each table, with class variables
containing the field names and types converters, which will be used by the
base class. This way you can have a derived class with almost no code.

Enjoy !

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Matthew Story 2005-01-09 20:02:45 Re: PYTHON, ODBC
Previous Message Tom Lane 2005-01-09 18:43:28 Re: [GENERAL] PostgreSQL 8.0.0 Release Candidate 4