As nobody said anything against forwarding my SoC reports here, I'm assuming it is OK to continue doing it :)
---------- Forwarded message ---------- From: Leo Soto M. <[EMAIL PROTECTED]> Date: Fri, Jun 6, 2008 at 11:16 PM Subject: GSoC Weekly Report (#2): Django on Jython To: JythonDevelopers <[EMAIL PROTECTED]> Hi again! So another week has passed and more progress has be done :) Basically, every problem found with the postgresql_zxjdbc driver which would also be a problem with other *_zxjdbc driver (mysql, oracle, etc) has been fixed on the proper place. On my jython.doj branch, zxJDBC automatically maps dates and times between Java and Python. And my django.doj branch worked around the problems found with respect to dates, times, booleans and even integers parameters being converted to strings before passed to the backend (JDBC didn't liked that). The missing bit is timezone support. I plan to add that really soon. Bottom line: creating an oracle and/or mysql backend for django/jython should be easiest now. I may even give them a shot next week. However, I have other goals too (see below), so this may have to wait a bit. With respect to particular issues with postgresql_zxjdbc, I wasn't able to solve the JDBC mapping for inet types. The PostgrSQL JDBC driver doesn't like String instances as values for them. As as workaround, now IPAddressFields are mapped to CHAR(15), which makes us incompatible with tables created with CPython/Django/psycopg containing inet values. As I've found some JDBC driver extensions[1] that could help here, not everything is lost yet. The goal for the next week is to pass the test suite. I advanced a bit the last couple of days, and now the failure count is around 25 (from ~35 which was our previous "record"). Now, I think that the goal is not going to be _strictly_ met, because we have problems with missing unicode data, source encoding support and problems with unicode+re which I won't be able to address right now. Other than that, by next Friday I should have fixed the following: - AttributeErrors raised from descriptors are swallowed[2] - Problems with DateQuerySet (they give datetimes with one or three hours more than expected) - Dispatcher test failures (they seem related to some GC assumptions) - SQLExceptions coming from PostgreSQL integrity errors are not mapped to DB-API standard IntegrityError And if time suffices, also: - Support for string constants bigger than 32k chars. - Make doctest a bit smarter with respect to HTML, XML and JSON output [1] http://archives.postgresql.org/pgsql-jdbc/2007-08/msg00089.php [2] http://bugs.jython.org/issue1041 -- Leo Soto M. http://blog.leosoto.com -- Leo Soto M. http://blog.leosoto.com --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django developers" group. To post to this group, send email to django-developers@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/django-developers?hl=en -~----------~----~----~----~------~----~------~--~---
