Matthias Clasen [Wed, 13 Mar 2013 10:01:57 +0000 (06:01 -0400)]
Add a missing runtime check
The cursor theme and size settting code was ifdefed to only
be compiled with the X11 backend, but it didn't check for
running under X at runtime. Fix that.
http://bugzilla.gnome.org/695495
Chun-wei Fan [Thu, 21 Feb 2013 07:54:18 +0000 (15:54 +0800)]
gdk/gdkwindow.c, gtk/gtkwidget.c: Include fallback-c89.c
Both of them started to make use of round(), a C99 function. So, include
fallback-c89.c to provide a fallback implementation for round() for
compilers that don't have round()
filechooserbutton: Do not propagate state from the dialog unless it is active
Change of plans to match the tests from the previous commit.
The state of the underlying dialog is never reflected by GtkFileChooserButton's API,
as the dialog is a transient thing. The file chooser button only updates its state from the dialog,
and reflects the dialog's state, when the dialog has been confirmed and dismissed by the user.
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
filechooserbutton: For the open-dialog-cancel-* tests, add a real nomenclature and test all the combinations
We used to have numeric names, which are a pain to maintain when new tests are added.
Now we have a real nomenclature (see the comment at the beginning of the open-dialog-cancel-* tests),
which lets us see easily if we have tested all the combinations.
Also, added all the combinations that were missing and removed redundant tests.
Not all the tests pass currently.
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
filechooserbutton: Update the button's state only on programmatic changes to the filechooser's state
The idea is that the button will only update its state of the selection and current folder
when changes to those are done either by the calling program (with the filechooser's API)
or when the user actually confirms and dismisses the underlying GtkFileChooserDialog.
If the user makes changes to the dialog but has not dismissed it yet, those changes
will not be reflected in the button (as one would expect).
This commit also makes sure the current-folder-changed and selection-changed signals
are emitted at the right times.
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
filechooserbutton: Emit 'selection-changed' when changing the selection programmatically
We only emitted that signal when the selection changed through the underlying GtkFileChooserDialog.
To do this when the dialog is not active and the selection is changed by the calling program
(instead of by the user), we need to wait until the GtkFileChooserButton's UI has been updated
via an async callback from GIO. So, we keep track of whether an entry point into the
button's API caused a programmatic change in the selection.
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
Benjamin Otte [Tue, 12 Mar 2013 11:44:03 +0000 (12:44 +0100)]
stylecontext: If we force invalidate, invalidate all properties
We only FORCE_INVALIDATE when something weird changes that the CSS
machinery can't detect. But now that our style_updated functions skip
recomputations when some properties don't change we want to make sure
these recomputations are still run. So we just claim all properties
changed.
combobox: popdown the combobox when on 'grab-broken-event'
And also explicitly remove pointer/keyboard grabs from the display.
Whenever the grab is reported lost, we should popdown the combobox, so that the
GDK_WINDOW_TEMP window is hidden and removed from the toplevel, as done with
the menu for example.
Leaving the GDK_WINDOW_TEMP window open when re-activating the application
triggers several issues in the win32 backend, due to restacking windows of the
non-toplevel group into the toplevel group:
filechooserbutton: HACK: Set a reasonable size for the dialog in the tests' second iteration
Something is causing the GtkFileChooserDialog to be resized really small on the second time it is run
during each test for GtkFileChooserButton. So as a temporary hack we set it to 500x500 pixels on
the second run, so the size allocation code doesn't bomb on us.
Cosimo Cecchi [Fri, 8 Mar 2013 22:56:08 +0000 (17:56 -0500)]
treeview: fix a critical warning
gtk_tree_view_column_unset_tree_view() resets column->priv->tree_view to
NULL.
The function is called when a column is removed, but later from the same
function we would call _gtk_tree_view_column_unrealize_button(), which
expects column->priv->tree_view to be != NULL, causing these critical
warnings
filechooserbutton: Update the combo box even after the dialog is cancelled
We assumed that we didn't have to update the combo box if the dialog got cancelled,
as it should simply retain its previous contents. But this assumption doesn't work
as the dialog is brought up with the 'Other...' item - we don't want the
combo box to keep showing 'Other...' if the dialog is cancelled.
The test from the previous commit now passes.
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
filechooserbutton: Don't show the (None) item when the combo box is popped up
This is surprisingly tricky, since the (None) item *has* to be a visible item while
the combo box is *not* popped up, so that it can show its contents. But the item
has to be *not* visible when the combo box is popped up.
Also, update the whole button's selection, not just the underlying dialog's, when
the combo box changes its selection - based on a patch by Paul Davis in
https://bugzilla.gnome.org/show_bug.cgi?id=691040#c20
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
filechooserbutton: Update the button/combo whenever the selection is changed programmatically
We didn't change it when the file chooser button's dialog was inactive, and so
the actual file chooser button would not visually reflect the current selection.
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
filechooserbutton: Test that the expected filename is shown
We do some gymnastics to pull the string out of the GtkButton or the GtkComboBox that is
being used in GtkFileChooserButton to show the current selection when the dialog
is inactive - namely, we look for the subwidget with the correct ATK role, and pull its
accessible name.
Currently the test fails; this is https://bugzilla.gnome.org/show_bug.cgi?id=691040#c18
Signed-off-by: Federico Mena Quintero <federico@gnome.org>
Chun-wei Fan [Wed, 6 Mar 2013 05:14:02 +0000 (13:14 +0800)]
Update Visual Studio Solution Files
Integrate the utility projects to build the introspection files into the
main solution files, so that one can build the introspection files from the
IDE. This is not built by default, so one can build the introspection
files if he/she chooses to do so.
Chun-wei Fan [Wed, 6 Mar 2013 05:12:01 +0000 (13:12 +0800)]
Visual Studio Builds: Support Building Introspection Files
Add Windows .bat and Python script to call g-ir-scanner to build
introspection files for Visual Studio builds. This will read from the
autotools files using Python REGEX functionality to determine the headers
and sources for g-ir-scanner to process, so the autotools files will not
need to be updated except to distribute the necessary files. Thils will
also enable one to build introspection files on Windows without using a
BASH-style shell such as MSYS.
Also add an utility Visual Studio project to call the Windows .bat to
build the introspection files for GTK+/GDK, for convenience.
Timothy Arceri [Sun, 18 Nov 2012 08:39:11 +0000 (19:39 +1100)]
filechooser: Show FUSE mounted locations in shortcuts
Since FUSE locations can be handled safely by applications show these mounted locations regardless of whether gtk_file_chooser_set_local_only()
is set to TRUE
Matthias Clasen [Wed, 6 Feb 2013 06:20:02 +0000 (01:20 -0500)]
Allow to populate bubbles with extra content
We do this by making the ::populate-popup signals a little more
flexible. They used to just accept a GtkMenu as argument, now
they can take a menu or a toolbar. To not break the expectations
of existing callbacks, we only emit ::populate-popup with a toolbar
if the :populate-toolbar property is TRUE.
Carlos Garnacho [Fri, 11 Jan 2013 16:22:02 +0000 (17:22 +0100)]
texthandle: Set a bigger input shape, covering the line height
Now, even if the handles being rendered are small, the handle touch
input shape will be as wide as the visible part of the rendered asset, and
high enough to cover both the handle and the height of the line where
the selection bound is.
Also, make handles have the same virtual distance to the line top/bottom
when a drag starts, so the handle doesn't jump to another line after a
too short threshold.
Carlos Garnacho [Wed, 21 Nov 2012 15:49:37 +0000 (16:49 +0100)]
Ensure the insertion handle stays on fake events (eg from IM)
Don't set handles mode to none if the event has send_event set.
For consistency with GtkEntry, also make GtkTextView keep the
handle mode on buffer changes.