Julien Cristau <julien.cristau@logilab.fr> [Mon, 28 Apr 2014 12:54:14 +0200] rev 9707
[server/session] add missing rollback when freeing a db connection
When we let go of a cnxset and give it back to the pool, we need to make
sure it's not in the middle of a transaction.
We got bitten by a migration where the connection used by admin's
repo.connect was not rolled back before attempting the migration,
causing a deadlock because it had a lock on a table. This could
potentially also be an issue for e.g. hooks or looping tasks that didn't
explicitly rollback before calling free_cnxset or moving out of a
ensure_cnx_set block.
Julien Cristau <julien.cristau@logilab.fr> [Thu, 24 Apr 2014 17:52:20 +0200] rev 9706
[debian] add Breaks on cubicweb-bootstrap
Need to be versioned once we know which version fixes
https://www.cubicweb.org/ticket/3728939
Denis Laxalde <denis.laxalde@logilab.fr> [Tue, 08 Apr 2014 16:08:31 +0200] rev 9705
Drop duplicated content in ITreeAdapter docstring
Rémi Cardona <remi.cardona@logilab.fr> [Mon, 14 Apr 2014 13:31:19 +0200] rev 9704
an -> a (misc)
Rémi Cardona <remi.cardona@logilab.fr> [Mon, 14 Apr 2014 12:29:01 +0200] rev 9703
an utf-8 -> a utf-8
Rémi Cardona <remi.cardona@logilab.fr> [Mon, 14 Apr 2014 12:27:37 +0200] rev 9702
an URI -> a URI
Rémi Cardona <remi.cardona@logilab.fr> [Mon, 14 Apr 2014 12:07:00 +0200] rev 9701
an unicode string -> a unicode string
Rémi Cardona <remi.cardona@logilab.fr> [Mon, 14 Apr 2014 11:54:33 +0200] rev 9700
an URL -> a URL
Christophe de Vienne <christophe@unlish.com> [Tue, 01 Apr 2014 13:57:25 +0200] rev 9699
Remove a misleading comment about self.render() behavior.
The former comment suggested that View.render called with 'row' or 'col' would
set cw_row or cw_col, which is not (anymore ?) accurate.
Related to #3696871.
Julien Cristau <julien.cristau@logilab.fr> [Thu, 06 Mar 2014 18:41:23 +0100] rev 9698
[migractions] add sanity check in rename_relation_type
If the relation is still present in the fs schema, we probably shouldn't
be renaming it (because it's still in use somewhere, possibly in a
different cube). Closes #3608172
Vincent Michel <vincent.michel@logilab.fr> [Mon, 07 Apr 2014 08:46:41 +0000] rev 9697
[dataimport] Add safety belt on "gen_rtype" in MetaGenerator, closes #3712892
Some cubes may well add to the META_RTYPES set, hence
crashing the gen_<somecubemeta> call.
Vincent Michel <vincent.michel@logilab.fr> [Mon, 07 Apr 2014 08:44:56 +0000] rev 9696
[dataimport] Correctly serialize datetime objects in _create_copyfrom_buffer, closes #3712885
Rémi Cardona <remi.cardona@logilab.fr> [Thu, 03 Apr 2014 15:36:33 +0200] rev 9695
[dataimport] Prevent ucsvreader from skipping the first line when ignore_errors is True (closes #3705791)
Rémi Cardona <remi.cardona@logilab.fr> [Thu, 03 Apr 2014 15:30:37 +0200] rev 9694
[dataimport] Avoid double unicode decoding in ucsvreader (closes #3705752)
Christophe de Vienne <christophe@unlish.com> [Fri, 25 Apr 2014 12:22:07 +0200] rev 9693
[doc] Document the user_data dictionnary content
Julien Cristau <julien.cristau@logilab.fr> [Thu, 24 Apr 2014 14:44:58 +0200] rev 9692
[debian] break old cubicweb-vcsfile
We removed call_service's async argument, which it used.
Julien Cristau <julien.cristau@logilab.fr> [Wed, 23 Apr 2014 18:17:42 +0200] rev 9691
[services] catch missing group early in register_user
Better an assertion than a rql syntax error for such a programming error.
Julien Cristau <julien.cristau@logilab.fr> [Wed, 23 Apr 2014 18:16:46 +0200] rev 9690
[services] rename register_user kwargs to cwuserkwargs
They're attributes/relations for the created CWUser, this makes it clearer.
Julien Cristau <julien.cristau@logilab.fr> [Tue, 22 Apr 2014 15:18:04 +0200] rev 9689
[services] don't hardcode 'users' group for register_user
Let the caller choose which group the user should be in. Related to #3020639
Caveat: this is a possible security hole if untrusted values are allowed.
Julien Cristau <julien.cristau@logilab.fr> [Tue, 22 Apr 2014 15:03:24 +0200] rev 9688
[sobjects/test] check that register_user properly sets user name and email address
Related to #3020639
Vladimir Popescu <vladimir.popescu@logilab.fr> [Fri, 19 Jul 2013 17:38:15 +0200] rev 9687
[server] Refactor Repository.register_user into a CubicWeb service (closes #3020639)
[jcr: move commit to the caller; add item in release notes; fix NameErrors]
Julien Cristau <julien.cristau@logilab.fr> [Fri, 18 Apr 2014 14:25:36 +0200] rev 9686
[migration] set a cnxset before process_script()
Not pretty, but I don't have a better idea right now and this at least
ensures migration scripts get run in the expected environment and can
access sql stuff as before.
Julien Cristau <julien.cristau@logilab.fr> [Thu, 10 Apr 2014 11:04:13 +0200] rev 9685
[server] make internal_cnx get and keep a connection to the db
- internal_session worked that way
- it's less surprising for users if their internal connections "just
work", rather than having to deal with the cnxset when using low-level
APIs
Includes a backout of 3f62606c01a2 "[repo] Fix register_user" which is
no longer needed.
Depend on logilab-database 1.12.1 to avoid errors due to pointless check
in python's sqlite3 binding.
Julien Cristau <julien.cristau@logilab.fr> [Tue, 22 Apr 2014 14:43:11 +0200] rev 9684
[server/session] make commit/rollback obey ensure_cnx_set
Inside the ensure_cnx_set context manager, commit and rollback aren't
allowed to dispose of the cnxset.
Julien Cristau <julien.cristau@logilab.fr> [Wed, 09 Apr 2014 14:05:54 +0200] rev 9683
[repo] fix deprecation messages for stats and gc_stats
Julien Cristau <julien.cristau@logilab.fr> [Thu, 10 Apr 2014 17:31:50 +0200] rev 9682
[cwconfig] use "foo not in bar" instead of "not foo in bar"
Julien Cristau <julien.cristau@logilab.fr> [Thu, 10 Apr 2014 15:28:21 +0200] rev 9681
[debian] Add Breaks on a few cubes that need changes for 3.19
- mercurial_server is affected by composite deletion changes
- forgotpwd relied on session data to be forwarded to notification view
- registration hooks into the authentication stack
Julien Cristau <julien.cristau@logilab.fr> [Thu, 10 Apr 2014 15:03:01 +0200] rev 9680
[hooks] run cascading delete hook later
Applications may have their own before_delete_entity hooks that rely on the
composite relations still existing.