Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 16:47:34 +0200] rev 7630
closes #1784285: [rql2sql] crash when using a relation for inner and outer join
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 16:22:08 +0200] rev 7629
[range facet] closes #1806937: simply retrieve min/max values when computing vocabulary/possible_values, this is enough
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 13:53:00 +0200] rev 7628
[facet] fix facet values update, broken by 7626:56300bec75f8
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:34:15 +0200] rev 7627
[facet slider] quote attribute value, else may break with earlier jquery
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:34:05 +0200] rev 7626
[facets] use unique dom id
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:33:55 +0200] rev 7625
[facet] with rql path facet rel may be None, handle it
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:33:39 +0200] rev 7624
[facet] closes #1806932: test and fix facet bug w/ having query (need rql update)
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:10:50 +0200] rev 7623
[ajax js loading] when everything is already loaded, don't do anything
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:10:37 +0200] rev 7622
backport stable
Sylvain Thénault <sylvain.thenault@logilab.fr> [Thu, 07 Jul 2011 11:08:58 +0200] rev 7621
[IE, plot] ship with jqplot'excanvas version (include http://code.google.com/p/explorercanvas/issues/detail?id=82 for instance)
Sylvain Thénault <sylvain.thenault@logilab.fr> [Tue, 05 Jul 2011 14:49:17 +0200] rev 7620
[etwist, modconcat] closes #1806935: directly write data, don't put the whole concatenated thing in memory
Sylvain Thénault <sylvain.thenault@logilab.fr> [Tue, 05 Jul 2011 12:09:53 +0200] rev 7619
[etwist] clear ui cache at server startup time. Closes #1806933
this avoid pb when cubicweb/cubes provide new version of some js/css and also accumulating junk in the cache directories
Sylvain Thénault <sylvain.thenault@logilab.fr> [Tue, 05 Jul 2011 10:54:52 +0200] rev 7618
[facet] closes #1806931: new facet type, based on arbitrary rql path
Sylvain Thénault <sylvain.thenault@logilab.fr> [Tue, 05 Jul 2011 10:52:34 +0200] rev 7617
[selectors, test] remove duplicated test and add some for relation_possible selector w/ rqlst
Sylvain Thénault <sylvain.thenault@logilab.fr> [Tue, 05 Jul 2011 10:48:40 +0200] rev 7616
[repository] catch Exception instead of everything
Sylvain Thénault <sylvain.thenault@logilab.fr> [Mon, 04 Jul 2011 16:46:29 +0200] rev 7615
[facet] fix argument inversion in date range facet
Aurelien Campeas <aurelien.campeas@logilab.fr> [Mon, 04 Jul 2011 12:37:44 +0200] rev 7614
[facetbox] put the facet box in a box
Nicolas Chauvat <nicolas.chauvat@logilab.fr> [Fri, 01 Jul 2011 20:14:20 +0200] rev 7613
[doc] fix broken links
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 18:44:03 +0200] rev 7612
backport stable
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 18:43:16 +0200] rev 7611
[test] fix test, code from 3.13 gone in 3.12 accidentaly
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 18:42:36 +0200] rev 7610
[selectors, facet] make is_instance and other EClassSelector subclass gather entity classes from 'select' and 'filtered_variable' context information. Closes #1796732
This allow for instance to use the filter form on rqlst where the filtered variable is not selected
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 16:20:57 +0200] rev 7609
[facets] DontUpdateConstraint mecanism introduced a few commit ago isn't needed: we already have the 'needs_update' property controlling this
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 16:20:45 +0200] rev 7608
[facet] more api cleanup: support_and should be a property, so it's easier to deactivate.
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 15:26:45 +0200] rev 7607
[range facet] refactor to ease overriding, possible_values raise DontUpdateFacet (not handled by js actually)
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 15:26:44 +0200] rev 7606
[facets] allow to specify on which rqlst's variable facets should be applied. Closes #1795915
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 15:26:43 +0200] rev 7605
[facets] refactor / cleanup facet api: more consistent variable naming and easier to reuse function. Closes #1796804
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 15:26:33 +0200] rev 7604
backport stable
Florent Cayré <florent.cayre@gmail.com> [Fri, 01 Jul 2011 15:26:12 +0200] rev 7603
[schema / rql] fix mainvars of RRQLExpression with parens
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 15:23:25 +0200] rev 7602
[facets js] better cubicweb:facetargs handling and avoid initializing a same form twice (that may occurs)
Sylvain Thénault <sylvain.thenault@logilab.fr> [Fri, 01 Jul 2011 15:22:43 +0200] rev 7601
[facets] should remove groupby from base rqlst. Closes #1796196
as we search for values for a variable, there seems to be no case
where we would be interested in groups, though there are clearly
some were leaving groups cause errors.