2017-03-02  Rico Tzschichholz  <[email protected]>

       Release 0.34.5

2017-03-02  Niko Tyni  <[email protected]>

       gtkmodule: Preserve order of required-gtype-classes for reproducibility
       The elements of current_required_app_classes end up in the generated
       code in g_type_ensure() statements. The order of these can vary
       between builds when using a HashSet, breaking reproducibilty.

       See https://wiki.debian.org/ReproducibleBuilds

       https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=802528

2017-03-02  Rico Tzschichholz  <[email protected]>

       glib-2.0: Fix g_bookmark_file_load_from_*/to_file bindings
       BookmarkFile.load_from_file, load_from_data_dirs and to_file may
       also throw FileError in addition to BookmarkFileError.

2017-03-01  Carlos Garnacho  <[email protected]>

       tests: Add test for dbus fd passing failures
       The server will exhaust all fds before the fd list in the dbus
       request is opened. We do expect it to fail in the client.

       https://bugzilla.gnome.org/show_bug.cgi?id=778540

       dbus: Move send_message() call before _error label
       Commit 09e6818d01 changed dbus server paths so requests fell back
       to freeing allocated memory on errors. However certain kinds of
       dbus replies issue a final g_dbus_connection_send_message() there.

       In case of errors, _reply_message will be NULL, and the error paths
       would have already issued g_dbus_method_invocation_return_gerror(),
       so the g_dbus_connection_send_message() call is both unnecessary and
       warns on the NULL argument. So  move the DBus reply before the _error
       label.

       https://bugzilla.gnome.org/show_bug.cgi?id=778540

       gdbus: Handle errors when extracting fds from dbus messages
       On the right situations (eg. fd exhaustion), both
       g_dbus_message_get_unix_fd_list() and g_unix_fd_list_get() should be
       considered failable here. Add proper error handling to avoid triggering
       glib warnings.

       https://bugzilla.gnome.org/show_bug.cgi?id=778540

       gdbus: Ensure extracted data from the GVariant is freed on error
       This will make extracted variables properly freed when extracting those
       fails at some point.

       https://bugzilla.gnome.org/show_bug.cgi?id=778540

       gdbus: Initialize intermediate variables
       Those may be left uninitialized after error handling paths are introduced,
       so invalid data might be attempted to be freed later on.

       https://bugzilla.gnome.org/show_bug.cgi?id=778540

2017-03-01  Michael James Gratton  <[email protected]>

       codegen: Don't return void for non-nullable simple-type structs
       https://bugzilla.gnome.org/show_bug.cgi?id=778224

2017-03-01  Rico Tzschichholz  <[email protected]>

       codegen: Default to void-return for vfunc pointer cast of *_async methods
       This make it match vala's generated code of async-module.

       To properly account for non-void *_async methods we require some additional
       metadata since only the return-type of *_finish method is known.

       codegen: Fix vfunc pointer cast for async method implementation/override

       codegen: Cast destroy_data for g_cclosure_new to GClosureNotify

       codegen: Cast *_destroy_func arg of generic-types calls to GDestroyNotify

       gobject-2.0: Add some missing "array_length_type" annotations

       codegen: Cast vfunc pointer assigns of abstract overrides
       https://bugzilla.gnome.org/show_bug.cgi?id=777958

2017-03-01  Nikolay Orlyuk  <[email protected]>

       codegen: Fix instance parameter for property setter in SimpleType structs
       Add test-case and fix code generator for call to the setter to pass
       instance as a value rather than as a pointer to value.

       Based on patches by Nathan Summers.

       https://bugzilla.gnome.org/show_bug.cgi?id=657346

2017-03-01  Rico Tzschichholz  <[email protected]>

       codegen: Fix memory leak when using object initializer for properties
       Assigning values to properties this way leads to a ref/copy of the source
       and therefore requires a unref/destroy afterwards.

       https://bugzilla.gnome.org/show_bug.cgi?id=766739

2017-03-01  Richard Wiedenhöft  <[email protected]>

       Fix broken destruction of null-terminated arrays in async data structs
       https://bugzilla.gnome.org/show_bug.cgi?id=762819

2017-03-01  Rico Tzschichholz  <[email protected]>

       Fix build with -Werror=switch

       codegen: Reference struct fields with their actual cname
       https://bugzilla.gnome.org/show_bug.cgi?id=775761

2017-02-04  Rico Tzschichholz  <[email protected]>

       gdk-2.0: Include gdk/gdkkeysyms.h for Gdk.Key
       Required since gdk.h doesn't include gdk/gdkkeysyms.h as gdk-3.0 does.

       https://bugzilla.gnome.org/show_bug.cgi?id=778173

       gdk-3.0: Update to 3.22.7

       vapi: "ref" functions must return an unowned value
       Otherwise vala would automatically add a corresponding "unref" call.

2016-12-29  Rico Tzschichholz  <[email protected]>

       gtk+-3.0: Update to 3.22.5+a1aee460

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.