summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Bump version to 3.31.913.31.91Florian Müllner2019-02-212-1/+24
| | | | Update NEWS.
* monitor-config-manager: respect other outputs' CRTCEmilio Pozuelo Monfort2019-02-211-5/+73
| | | | | | | | | We should not only take the old CRTC for an output whenever possible, but we should also assign one that is 'free', i.e. one that another monitor (to be processed after this one) isn't using, so that that monitor can use the same CRTC. https://gitlab.gnome.org/GNOME/mutter/issues/373
* monitor-config-manager: reuse old CRTC when possibleEmilio Pozuelo Monfort2019-02-211-1/+7
| | | | | | | | | | We shouldn't change an output's CRTC if we don't have to, as that causes the output to go black. This patch depends on "monitor-unit-tests: initial crtcs in custom_lid_switch". https://gitlab.gnome.org/GNOME/mutter/issues/373
* monitor-unit-tests: initial crtcs in custom_lid_switchPekka Paalanen2019-02-211-0/+2
| | | | | | | | | | | | | | | | | This test forgot to specify the existing CRTC routings in the setup. For the first output the default 0 was ok, now it is -1 to ensure that the code will assign it correctly. For the second output the default 0 was incorrect, because possible_crtcs does not include 0. Now that CRTC is initialized to off instead, because the second output is hotplugged later and running a CRTC without an output does not make sense. This fix will keep this test passing when a future patch attempts to preserve existing CRTC routings. Assuming that any existing routing is valid, such routing will be kept. In this test case the existing routing was illegal, it should have been impossible, which then causes that future patch to fail the test by assigning the wrong CRTC. https://gitlab.gnome.org/GNOME/mutter/issues/373
* MetaPluginManager: Don't return void valueOlivier Fourdan2019-02-211-2/+2
| | | | | | | The spec for `meta_plugin_manager_confirm_display_change()` is to return `void`, no need to return a value there. https://gitlab.gnome.org/GNOME/mutter/merge_requests/456
* Update Brazilian Portuguese translationRafael Fontenelle2019-02-211-62/+60
|
* keybinding: Store flags in MetaKeyGrabAndrea Azzarone2019-02-201-1/+3
| | | | | | | | | | | The external grab handler is shared across all external bindings and external bindings have now different binding flags. For this reason, when rebuilding the binding table there could be loss of information if we assign the bindings flags of the external handler to all external bindings. Let's store the bindings flags in MetaKeyGrab too and use this when rebuilding the binding table to avoid the above issue. https://gitlab.gnome.org/GNOME/mutter/merge_requests/169
* keybindings: Add flag param to grab_acceleratorAndrea Azzarone2019-02-202-4/+7
| | | | | | | | | | Add a 'flags' parameters to meta_display_grab_accelerator. This will allow e.g. gnome-settings-daemon (through the gnome-shell's GrabAccelerator API) to create shorcuts that should descard auto-repeated key events. https://gitlab.gnome.org/GNOME/mutter/merge_requests/169 Closes: https://gitlab.gnome.org/GNOME/gnome-settings-daemon/issues/68
* Updated Spanish translationDaniel Mustieles2019-02-201-8/+3
|
* window: Do not restore shortcuts on a NULL windowOlivier Fourdan2019-02-181-0/+3
| | | | | | | | | | | The "force restore shortcuts" being triggered by a key-combo, there is no guarantee that the currently focused window is actually non-NULL in which case we would crash. Make sure there is a window currently focused before trying to restore the shortcuts on that window. Closes: https://gitlab.gnome.org/GNOME/mutter/issues/464
* Update Chinese (Taiwan) translationYi-Jyun Pan2019-02-171-45/+54
|
* Update Kazakh translationBaurzhan Muftakhidinov2019-02-171-121/+171
|
* clutter: Remove deprecate/clutter-frame-source.hAdam Jackson2019-02-155-324/+0
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/445
* clutter: Remove deprecated/clutter-timeout-pool.hAdam Jackson2019-02-154-570/+0
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/445
* clutter: Remove deprecated/clutter-backend.hAdam Jackson2019-02-154-247/+0
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/445
* clutter: Remove deprecated/clutter-util.hAdam Jackson2019-02-155-66/+0
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/445
* cogl: Remove CoglBool, use gboolean insteadNiels De Graef2019-02-15228-1197/+1185
| | | | | | | | | | | | This basically reverts commit 54735dec, which tried to avoid the GLib-defined types in favor the standard C ones. One exception to this is the bool type, for which the commit introduces a new type CoglBool. Let's just get rid of this type in favor of having consistency with the GLib types. Note by the way that neither CoglBool nor gboolean (which has a size of `int`) are completely compatible with bool (size `char`). https://gitlab.gnome.org/GNOME/mutter/merge_requests/321
* wayland/xdg-shell: Do not maximize if not possibleOlivier Fourdan2019-02-152-0/+6
| | | | | | | | | | | | | | | | | On X11, if a window cannot be maximized because its minimum size is already larger than the output size, a request to maximize will be ignored. On Wayland, however, we would still honor the maximize request and switch the window state to maximized, without actually moving the window which leads to weird visual effects, as the window end up being maximized in-place. To avoid this, make sure the window has the maximize functionality available prior to change its state in xdg-shell `set_maximized` request. Closes: https://gitlab.gnome.org/GNOME/mutter/issues/463
* gpu: Remove unimplemented ::get_kms_file_path()Niels De Graef2019-02-151-3/+0
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/444
* gitlab-ci.yml: Add check for issue or MR URLJonas Ådahl2019-02-142-0/+45
| | | | | | | | | | | This adds a pipeline stage for merge requests that checks that the commit message contains an URL to either a issue or a merge request. This means that for merge requests without corresponding issues will always fail initially, as the merge request URL is not known until after it is created. This is still arguably better than accidentally merging merge requests without URLs. https://gitlab.gnome.org/GNOME/mutter/merge_requests/440
* README: Add contribution sectionJonas Ådahl2019-02-141-0/+14
| | | | | | | Short and incomplete blurb about coding style and commit message guidelines. https://gitlab.gnome.org/GNOME/mutter/merge_requests/440
* window-actor: Remove public declaration for get_x_windowMarco Trevisan (Treviño)2019-02-141-5/+0
| | | | | | | meta_window_actor_get_x_window has been removed in commit 422648e2 but has not been removed from the header for ages. https://gitlab.gnome.org/GNOME/mutter/merge_requests/439
* window: Return focusable ability looking at properties onlyMarco Trevisan (Treviño)2019-02-142-3/+3
| | | | | | | | | | | As per commit 43633d6b, we mark an unmanaging window as not focusable, while this is true, it might cause not resetting the current focused window when unmanaging it causing a crash. Also this wouldn't allow to check if a window can be focused when unmanaging it, so let's revert the previous behavior. Fixes: https://gitlab.gnome.org/GNOME/mutter/issues/462
* wayland: Reset text-input state after commitCarlos Garnacho2019-02-141-4/+6
| | | | | | | | | This was wrongly done just before enable, which is not right as per the protocol. A side effect was that input purpose/hints were eagerly reset before being applied, thus not properly honored, noticed in the doing of emoji/numeric OSK panels. https://gitlab.gnome.org/GNOME/mutter/merge_requests/410
* Updated Danish translationAlan Mortensen2019-02-131-42/+60
|
* Update Romanian translationDaniel Șerbănescu2019-02-131-44/+53
|
* window: Move can_ping to a function and implement in X11Marco Trevisan (Treviño)2019-02-138-7/+44
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/421
* window-x11: Move delete_window to MetaWindowX11PrivateMarco Trevisan (Treviño)2019-02-136-5/+20
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/421
* window-x11: Move take_focus to MetaWindowX11PrivateMarco Trevisan (Treviño)2019-02-136-11/+30
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/421
* window: Add is_focusable class methodMarco Trevisan (Treviño)2019-02-135-8/+30
| | | | | | | Implement is_focusable for both x11 and wayland and just use this check so that we can abstract things more and be less dependent on window backend. https://gitlab.gnome.org/GNOME/mutter/merge_requests/421
* Updated Czech translationMarek Cernocky2019-02-121-41/+48
|
* Updated Slovenian translationMatej Urbančič2019-02-121-42/+51
|
* meta-launcher: g_propagate_*_error takes ownershipNiels De Graef2019-02-121-2/+6
| | | | | | | | | | | | | | | | This means we need to make sure we don't accidentally free the provided source GError (which automatically happens with `g_autoptr`), so use `g_steal_pointer()`. This fixes an issue where, when launched in a bubblewrap environment (such as the one provided by Buildstream), mutter would give the following warning message: ``` mutter-WARNING **: 8:31:35:069: Can't initialize KMS backend: (null) ``` ... which isn't that useful when trying to debug the actual issue.
* gitlab-ci: Don't upload test artifactsMarco Trevisan (Treviño)2019-02-111-3/+0
| | | | | | | There's no need to upload test artifacts since they don't produce anything that we care about so far, while this phase slows down the pipeline execution. https://gitlab.gnome.org/GNOME/mutter/merge_requests/436
* input-mapper: Use g_auto to free a string arrayMarco Trevisan (Treviño)2019-02-111-3/+1
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/435
* input-mapper: Don't loop infinitely in EDID matchingMarco Trevisan (Treviño)2019-02-111-2/+2
| | | | | | | Iterate over all the monitor product words to check for a partial matching on EDID, otherwise we would hang inside an infinite while loop. Fixes https://gitlab.gnome.org/GNOME/mutter/issues/459
* Update Swedish translationAnders Jonsson2019-02-091-43/+50
|
* Updated Lithuanian translationAurimas Černius2019-02-091-54/+51
|
* Update Turkish translationEmin Tufan Çetin2019-02-091-55/+52
|
* Update Galician translationFran Dieguez2019-02-091-54/+51
|
* Updated French translationClaude Paroz2019-02-091-57/+49
|
* Update French translationCharles Monzat2019-02-091-17/+18
| | | | Cherry-pick 28d2d54189b from gnome-3-30 branch.
* Update Finnish translationJiri Grönroos2019-02-091-141/+176
|
* Bump version to 3.31.903.31.90Florian Müllner2019-02-072-1/+26
| | | | Update NEWS.
* Update Polish translationPiotr Drąg2019-02-061-43/+50
|
* Update Catalan translationJordi Mas2019-02-061-51/+66
|
* Update Indonesian translationKukuh Syafaat2019-02-061-42/+49
|
* wayland/surface: Add support for wp_viewporterRobert Mader2019-02-067-20/+407
| | | | | | | | | | | | | | | | This adds the required bits to wayland surfaces and ties them up to the compositor parts. It is based on and very similar in nature to buffer transforms. From the specification: > The global interface exposing surface cropping and scaling > capabilities is used to instantiate an interface extension for a > wl_surface object. This extended interface will then allow cropping > and scaling the surface contents, effectively disconnecting the > direct relationship between the buffer and the surface size. https://gitlab.gnome.org/GNOME/mutter/merge_requests/323
* shaped-texture: Add support for viewportsRobert Mader2019-02-064-27/+262
| | | | | | | | | | | This implements the viewporter protocol which offers a cropping and scaling capabilities to wayland clients. There are several use cases for this, for example video players and games, both as a convenience function and as potential performance optimization when paired with hardware overlays etc. https://gitlab.gnome.org/GNOME/mutter/merge_requests/323
* region-utils: Add API to crop and scale an integer regionRobert Mader2019-02-062-0/+53
| | | | https://gitlab.gnome.org/GNOME/mutter/merge_requests/323
View Lorry Controller Status