Sylvain Thénault <sylvain.thenault@logilab.fr> [Mon, 21 Jun 2010 08:53:13 +0200] rev 5803
[schema migration] don't try to drop column if backend doesn't support column alteration (eg sqlite)
Sylvain Thénault <sylvain.thenault@logilab.fr> [Mon, 21 Jun 2010 08:52:13 +0200] rev 5802
[session] when rollback_on_failure is false, we should not try to reconnect if the sql raised an error
Alexandre Fayolle <alexandre.fayolle@logilab.fr> [Sun, 20 Jun 2010 15:09:42 +0000] rev 5801
Use SQL standard ALTER TABLE command (with ADD instead of ADD COLUMN)
Alexandre Fayolle <alexandre.fayolle@logilab.fr> [Sun, 20 Jun 2010 15:08:23 +0000] rev 5800
reduced logging verbosity in ldapuser source
Alexandre Fayolle <alexandre.fayolle@logilab.fr> [Sat, 19 Jun 2010 16:08:54 +0200] rev 5799
use sessionid in log message
Alexandre Fayolle <alexandre.fayolle@logilab.fr> [Sat, 19 Jun 2010 16:08:08 +0200] rev 5798
constraints can contain unicode values
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 18 Jun 2010 18:31:22 +0200] rev 5797
[entity] fix .complete so we don't crash on missing mandatory relation. Also simplify code. Closes #1058267
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 18 Jun 2010 18:49:17 +0200] rev 5796
closes #1059729: cw-ctl upgrade : i18n upgrade not performed if DB does not need upgrade
Aurelien Campeas <aurelien.campeas@logilab.fr> [Fri, 18 Jun 2010 18:35:09 +0200] rev 5795
[devtools/qunit] prefix jquery_dir with 'file://'
(as this is used from within a web browser)
Pierre-Yves David <pierre-yves.david@logilab.fr> [Fri, 18 Jun 2010 18:16:36 +0200] rev 5794
[devtools/qunit] Locate jquery through cubicwew config to handle installed setup
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 18 Jun 2010 18:14:55 +0200] rev 5793
[rql2sql] use new dbhelper.sql_concat_string function to deal with sql string concatenation. This together with lgdb 1.0.4 fixes #1055133: SQL Server support : string concatenation issue
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 18 Jun 2010 15:41:39 +0200] rev 5792
[session] update session's timestamp in session.execute, so long running transactions are not erroneously closed by the repository