summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* GtkSnapshot: Always use int for the translationwip/alexl/snapshot-int-translateAlexander Larsson2017-01-119-31/+40
| | | | | | | | | | | | We already take ints when setting the translation, so it can't currently take any other values. Additionally, I was seeing large costs in int -> double -> int for the rects in gtk_snapshot_clips_rect(), as all callers really are ints (widget allocations) and the clip region is int-based. This change completely cleared a 2% rectangle_init_from_graphene from the profile and is likely to have nice performance effects elsewhere too.
* widget: Avoid typechecks when accessing ->clip and ->allocationAlexander Larsson2017-01-111-9/+9
|
* GtkCSSImage: Avoid some type checksAlexander Larsson2017-01-111-6/+0
| | | | | | The width/height/aspect getters are called a lot, and almost all callers already verify it from _gtk_css_image_get_concrete_size (), so just skip these checks.
* Avoid some more type checks for internal callsAlexander Larsson2017-01-1112-41/+40
|
* gtkcssgadget: Use private non-checking versions of gtk_widget callsAlexander Larsson2017-01-111-2/+3
| | | | This avoids a lot of checking overhead.
* GtkSnapshot: Reuse snapshot state objectsAlexander Larsson2017-01-112-11/+30
| | | | | Rather than allocate new ones all the time we reuse the previous ones. We just clear them and save them in the parent for later reuse.
* GtkSnapshot: Move collect data to a union in the stateAlexander Larsson2017-01-112-174/+188
| | | | | | This means we allocate the collect data with the state, avoiding an extra allocation. Also, a union means every state object is the same size and we could reuse the state objects.
* Snapshot: Only record names if inspector is recordingAlexander Larsson2017-01-1114-38/+79
| | | | | Otherwise we do a lot of allocations and vprintf calls which are not used.
* gsk_rounded_rect_init_copy: Don't normalizeAlexander Larsson2017-01-111-4/+2
| | | | | | | | This was showing up quite high on the profiles, and there is no real reason for copy to normalize, as the source is a GskRoundedRect which should be normalized already unless you did something very strange (and then you should have normalized manually).
* Skip dynamic type check in css value gettersAlexander Larsson2017-01-113-6/+4
| | | | | | This gets called a lot during snapshotting, and this change alone brings down snapshot time by almost 10% in a syntethic snapshot test.
* Some updates to the migration guideMatthias Clasen2017-01-101-2/+49
|
* Document -gtk-icon-filterMatthias Clasen2017-01-101-2/+10
| | | | Not much detail here yet, but the syntax is documented.
* cssimage: Fix cross-fadeBenjamin Otte2017-01-111-2/+2
| | | | | | | Simgle image cross-fade opacity was computed the wrong way, which caused weird fade-in/out animations, for example in flat buttons. I messed this up when porting cross-fades to snapshot().
* Remove an unnecessary checkMatthias Clasen2017-01-101-7/+0
| | | | | | Since the demise of theme engines, we can no longer hit the case of id >= GTK_CSS_PROPERTY_N_PROPERTIES. So don't check for this in a very frequently called function.
* Use an internal parameter check here as wellMatthias Clasen2017-01-101-1/+2
| | | | | We should not slow down the core parts of the css machinery with type checks.
* Use gtk_internal_return_val_if_fail hereMatthias Clasen2017-01-101-3/+3
| | | | | It seems odd to have checks in just a few functions, so switch everything over to use the internal versions.
* Use _gtk_widget_get_window moreMatthias Clasen2017-01-103-5/+6
| | | | This avoids type checks in places where we know it is safe.
* Refactor some css transition code slightlyMatthias Clasen2017-01-101-22/+15
| | | | | Reshuffling things a bit to avoid a bunch of NULL and type checks.
* Use the .symbolic.png assets in AdwaitaMatthias Clasen2017-01-101-8/+4
| | | | | | | Using an image() fallback from svg to png doesn't make too much sense, since the svg is always used (unless librsvg is not present), while the png icon is faster and cheaper to load and thus preferable.
* gdk/wayland: Handle non-existant gsettings keysRui Matos2017-01-101-36/+41
| | | | | | | Since we're a library, crashing on gsettings keys, whose presence is out of our control, isn't appropriate. https://bugzilla.gnome.org/show_bug.cgi?id=775846
* gdk/wayland: Add support for the gtk-enable-primary-paste gsettingRui Matos2017-01-101-0/+1
| | | | | | | The gsetting was recently added so that we can have this configurable on the wayland backend too. https://bugzilla.gnome.org/show_bug.cgi?id=775846
* mir: fix compile-time warningsWilliam Hua2017-01-093-0/+16
|
* revealer: Fix a typo in a function docDaniel Boles2017-01-091-1/+1
| | | | | | Also, "ie" wasn't very clear, but fixing that to "i.e." would cause truncation of the summary when processed by bindings using doxygen. So, I replaced it with "in other words", which is no _less_ clear, at least.
* mir: properly handle empty clipboardWilliam Hua2017-01-091-3/+9
| | | | https://bugzilla.gnome.org/show_bug.cgi?id=775732
* Visual Studio builds: Move project files to win32/Chun-wei Fan2017-01-0961-644/+636
| | | | | It was suggested that the project files to be moved to win32/, so that we can have one less layer of directories we need to go down into to reach the project files.
* build/Makefile.msvcproj: Improve documentationChun-wei Fan2017-01-091-2/+7
| | | | | | Tell people about what happens when generating projects when Visual Studio 2013 or later is required, and mention that the .headers are only needed when headers need to be copied.
* cssgadget: Compute clip correctlyBenjamin Otte2017-01-091-15/+4
| | | | | We were computing it relative to the gadget allocation, but it should be relative to the widget allocation.
* widget: Document child/sibling accessorsTimm Bäder2017-01-081-0/+32
|
* switch: Use a widget as sliderTimm Bäder2017-01-081-33/+10
|
* widget: Add construct-only css-name propertyTimm Bäder2017-01-081-1/+24
| | | | | | | So we can set the css name of a widget to something that's not related to the class name. If the css-name property is set to NULL, we will still fall back to the one set using gtk_widget_class_set_css_name which is alwasys non-NULL since GtkWidget itself sets it to "widget".
* actionbar: Fix class and instance struct parent memberTimm Bäder2017-01-081-3/+3
|
* actionbar: Add revealed propertyTimm Bäder2017-01-084-3/+110
| | | | | So we can show and hide it with a transition as well as bind another property to it.
* testsuite/cssprovider: Use actual existing style propertyTimm Bäder2017-01-081-4/+4
| | | | | All the style properties in GtkTreeView are gone, so use one from GtkScrollbar for now.
* builderparser: Shuffle if-statements around once moreTimm Bäder2017-01-081-14/+14
| | | | The previous reordering broke the builderparser test case.
* testsuite/builder: Use g_assert_no_error to check GErrorTimm Bäder2017-01-081-2/+1
| | | | So we get a proper error message.
* printunixdialog: Remove leftover draw handlerTimm Bäder2017-01-081-1/+0
|
* actionbar: Add revealer as internal childTimm Bäder2017-01-083-61/+18
|
* actionbar: Remove show() and hide() implementationsTimm Bäder2017-01-081-30/+0
| | | | | They only show/hide the widget in a delayed fashion which doesn't work, just like it doesn't work with infobars and popovers.
* Update Brazilian Portuguese translationRafael Fontenelle2017-01-081-3955/+3534
|
* Update Brazilian Portuguese translationRafael Fontenelle2017-01-081-1057/+953
|
* gdk: Remove testing functionsBenjamin Otte2017-01-0824-762/+0
| | | | They were unused and unimplemented.
* tests: Remove gtk_widget_send_key()Benjamin Otte2017-01-083-65/+0
| | | | It's unused in GTK.
* tests: Remove widget find functionsBenjamin Otte2017-01-083-194/+0
| | | | | | | They are all unused by GTK. And other people can write their own find functions if they need any in their tests.
* Make GTK_DEBUG=interactive work betterMatthias Clasen2017-01-071-1/+12
| | | | | | | | We currently have various ways to initialize GTK+, and not all of them were supporting this way of bringing up the inspector. Fix this. https://bugzilla.gnome.org/show_bug.cgi?id=776807
* build: Fix vulkan detectionPavel Grunt2017-01-071-1/+1
| | | | | | Add missing 'test' https://bugzilla.gnome.org/show_bug.cgi?id=776736
* docs: Remove a line that's not true anymoreBenjamin Otte2017-01-081-2/+1
|
* viewport: Remove API to query GdkWindowsBenjamin Otte2017-01-084-48/+2
|
* treeview: Remove gtk_tree_view_get_bin_window()Benjamin Otte2017-01-087-51/+28
| | | | We don't want to expose GdkWindows in the public API.
* gdk: Remove unused debug categoryBenjamin Otte2017-01-082-7/+5
|
* gtkfishbowl: Remove gtk_container_snapshot_child callTimm Bäder2017-01-071-3/+3
|
View Lorry Controller Status