summaryrefslogtreecommitdiffstats
path: root/gi
Commit message (Collapse)AuthorAgeFilesLines
* Manually type-check the instance variable during method invocations on Python 3HEADpy3kDavid Malcolm2010-04-201-0/+59
|
* Restrict GI_TYPE_TAG_GHASH on Python 3 to dicts, not mappings, due to ↵David Malcolm2010-04-201-0/+11
| | | | | | | http://bugs.python.org/issue5945 This fixes a failure in test_ghashtable_int_none_in (test_gi.TestGHashTable) where it would raise "AttributeError: keys" rather than a TypeError when invoked with a string
* Fix test cases by decoding to correctly sized typeJohn (J5) Palmieri2010-04-161-7/+15
| | | | | | | | * Make work on both Python 3.0 and 2.x by using the Long object instead of Int. Since we are decoding to a C type and this code is internal we can assume UnsignedLongLong python type for unsigned values to avoid issues where we were mistakenly decoding large values into a long.
* make sure we correctly chose between string and unicodeJohn (J5) Palmieri2010-04-151-2/+14
|
* use AsLong instead of AS_LONGJohn (J5) Palmieri2010-04-151-2/+2
|
* Initial attempts at porting to Python 3.1David Malcolm2010-04-1510-95/+217
| | | | Doesn't yet work
* override that wasn't checked in - fixes some test casesJohn (J5) Palmieri2010-04-141-0/+69
|
* [Makefile.am] Clean up CFLAGS handling, don't override all: targetColin Walters2010-04-141-6/+9
| | | | | | | | | | First, we should move the CFLAGS into AM_CFLAGS, otherwise the per-target CFLAGS forces Automake to prefix object files, which is unnecessary since we only have one target. More importantly, avoid overriding the all: target here; that's owned by Automake. Use all-local instead to append things to the end of the normal build.
* Use GIMarshallingTests (old TestGI) in gobject-introspectionTomeu Vizoso2010-04-131-69/+0
|
* Always create the .so linkTomeu Vizoso2010-04-091-1/+1
|
* Add modelines and copyright information to overrides modulesSimon van der Linden2010-02-033-0/+62
|
* Fix and complete overrides testsSimon van der Linden2010-02-031-0/+49
| | | | | | Those tests were missing in the last commit https://bugzilla.gnome.org/show_bug.cgi?id=602830
* Remove support for pointers to basic types as input-only argument and return ↵Simon van der Linden2010-01-223-548/+47
| | | | | | | | | value There is no reason for an API to use such things, and g_function_info_invoke broke such features. https://bugzilla.gnome.org/show_bug.cgi?id=607759
* Restore the overrides supportSimon van der Linden2010-01-225-64/+83
| | | | | | | | | | | | | Add a ModuleProxy in front of the DynamicModule when an overrides module is present. There is no need for an overrides module to be a class; it can just be a module. Add an override decorator to override the wrapper of a registered type. Adapt Gdk and Gtk accordingly. Add tests. https://bugzilla.gnome.org/show_bug.cgi?id=602830
* Initialize struct fields to 0 when allocatingSimon van der Linden2010-01-081-1/+1
|
* Don't set a default constructor for structures.Simon van der Linden2010-01-081-28/+7
| | | | | | | | | | Update tests accordingly. The reason for this change is that setting __new__ in the metaclass doesn't let one overrides it afterwards, in a subclass (in my experience, at least, even though it seems weird). https://bugzilla.gnome.org/show_bug.cgi?id=603536
* Suppress compilation warningsSimon van der Linden2010-01-053-4/+3
|
* Add Gtk.keysyms to overridesTomeu Vizoso2009-12-223-0/+1502
|
* The array field 'length' starts to count from the C arg list, so need to ↵Tomeu Vizoso2009-12-223-4/+24
| | | | | | decrement when it's a method https://bugzilla.gnome.org/show_bug.cgi?id=602640
* Depend on GLib 2.20 rather than 2.22Anderson Lizardo2009-11-301-0/+19
| | | | | | Backport g_array_get_element_size. https://bugzilla.gnome.org/show_bug.cgi?id=603411
* Use the limit constants from glib and interpret G_MAXUINT32 as ↵Tomeu Vizoso2009-11-301-7/+7
| | | | | | PyLong_FromLongLong https://bugzilla.gnome.org/show_bug.cgi?id=602384
* Remove global checks for pointers and move them in type cases that need themSimon van der Linden2009-11-302-253/+596
|
* Pythonify. Avoid ; and () around if statementsJohan Dahlin2009-11-301-4/+4
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=602830
* Remove trailing whitespaceJohan Dahlin2009-11-301-2/+2
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=602830
* Set a default constructor for boxed structs that don't have oneTomeu Vizoso2009-11-3010-21/+249
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=602735
* Revert "Use the limit constants from glib and interpret G_MAXUINT32 as ↵Tomeu Vizoso2009-11-301-7/+7
| | | | | | PyLong_FromLongLong" This reverts commit 05a2ed55f3e5d2620de8b3b6b0d99e928ef3b041.
* Use the limit constants from glib and interpret G_MAXUINT32 as ↵Tomeu Vizoso2009-11-281-7/+7
| | | | | | PyLong_FromLongLong https://bugzilla.gnome.org/show_bug.cgi?id=602384
* Fix members initialization in metaclassesSimon van der Linden2009-11-271-2/+3
| | | | | | | | | | | | In metaclasses, the test for the name of the class was wrong, since it prevented one to create a subclass with the same name (especially annoying for overrides). Now, if a GType is available from the info, the fact that it doesn't have any wrapper yet means that the metaclass is creating the base class, which will be registerd just after its creation. This is true for objects, and for structures registered as boxed or pointer too. This patch includes a test for basic subclassing in Python. It notably tests that methods don't get overridden by the metaclass.
* Structs in arrays are not marshalled correctlyTomeu Vizoso2009-11-271-5/+24
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=602709
* Use the right variable when looking up in sys.modulesSimon van der Linden2009-11-251-1/+1
|
* Accept 0 as a valid value for flag and enum argumentsTomeu Vizoso2009-11-241-1/+18
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=602638
* Create overridden modules in two passesJohan Dahlin2009-11-231-13/+21
| | | | | | | | This patch splits overridden module creation into two passes. The first pass creates the auto-generated module normally before the overridden module is attempted to be imported. The second pass imports the overridden module and replaces the auto-generated module with the overridden. This is necessary for the overridden modules to be able to access the auto-generated ones.
* Add support for Any argumentsTomeu Vizoso2009-11-221-8/+12
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=601253
* Register interfacesTomeu Vizoso2009-11-222-1/+37
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=601181
* Treat GI_INFO_TYPE_INTERFACE same as GI_INFO_TYPE_OBJECTTomeu Vizoso2009-11-201-0/+3
|
* Initial importSimon van der Linden2009-11-0825-0/+5785
generated by cgit (git 2.34.1) at 2024-05-06 03:40:15 +0000