summaryrefslogtreecommitdiff
path: root/include
AgeCommit message (Collapse)AuthorFilesLines
2011-08-03add matrix invert method (from mesa)Travis Watkins1-0/+2
2011-07-08Merge in lp:~smspillaz/compiz-core/compiz-core.dist-buildsystem and turnSam Spilsbury2-5/+4
CTest on by default for everything
2011-07-07Add a proper set of a release targets in the CMake buildsystem for all compizSam Spilsbury1-5/+1
projects to use make release-prep: generates NEWS ChangeLog and AUTHORS file make distcheck: generates a tarball make release-signoff: signs the tarball using the key in RELEASE_KEY and if in a git repository, creates tags and branches for that
2011-07-07Merge in lp:~unity-team/compiz-core/compiz-core.fix_706918 and hookSam Spilsbury6-59/+184
PluginClassHandler tests into make test
2011-07-07Initialize index even if allocating the plugin class failed due to a buggySam Spilsbury1-0/+2
base class
2011-07-05Break the dependency between CompTimeoutSource and CompTimerSam Spilsbury1-3/+15
2011-07-05Use a private implementation for CompTimerSam Spilsbury1-7/+2
2011-07-05Further break the dependency of timers on screenSam Spilsbury1-0/+5
2011-07-04Break CompTimer's dependency on CompScreenSam Spilsbury3-4/+60
2011-06-25Added PluginClassHandler testSam Spilsbury1-1/+6
2011-06-25Abstract out CompPrivate storage from CompScreen so we can make it testableSam Spilsbury5-13/+97
2011-06-17Refactor the code somewhatSam Spilsbury1-109/+70
2011-06-17Make ::get () return a new class instance and initialize the indexSam Spilsbury2-13/+86
if it hasn't been done already for this particular kind of class. Fixes LP#706918
2011-05-08Remove DECOR_NORMAL_ATOM_NAME, the API is now to put all default decorationsSam Spilsbury1-1/+0
in a list in one property
2011-05-07Update API to allow multiple frames in one property (frame caching,Sam Spilsbury1-8/+60
multiple default frame styles) and also to allow decorators to specify in the frame property which window types, states and actions the frames should apply to
2011-04-30Track windows on CreateNotify again, don't rely on trying to beat the raceSam Spilsbury1-0/+1
condition by putting that on MapRequest, instead check at reparent time if the window has already been reparented server-side and don't reparent it again.
2011-03-23Bump core ABISam Spilsbury1-1/+1
2011-03-16CompOption::Class had virtual functions without a virtual destructor. That's ↵Rob Taylor1-0/+1
usually bad.
2011-03-16Region type in x is typedef struct _XRegion *Region. That means returning a ↵Rob Taylor1-1/+1
const Regsion makes no sense - that's setting the returned pointer to const.
2011-03-15Reflect difference between actual input frame geometry and border extentsSam Spilsbury1-2/+6
in the core API
2011-02-24Specify plugin name when loading images to find the correctSam Spilsbury2-1/+2
image loading dir for plugins which don't specify absolute paths to their images
2011-02-01Merge branch 'master' of ../../../coreSam Spilsbury4-9/+17
Conflicts: include/core/core.h src/window.cpp
2011-02-01Allow semantic difference between frame window geometry and frame size.Sam Spilsbury3-5/+10
Rationale: some window decorators or themes might want to specify a border input area that is slightly larger than the visible borders on screen (eg for enhanced resize handles, a11y purposes, etc) or the theme might want to specify some additional decoration which should affect placement but should not be interacted with. In this case we have a priv->border and priv->input window property. priv->border specifies the actual semantic size of the window decoration that should affect placement and priv->input specifies the extents of the frame window past the client. Fixes LP: #710271
2011-01-27Don't track window geometries or properties until they are actually managed.Sam Spilsbury1-2/+3
The window might get some ConfigureRequests which won't be reflected in PrivateWindow at MapRequest time. Fixes LP #707853
2011-01-26Fix windows jumping around when clicking on parts that trigger moveSam Spilsbury3-3/+4
Since we no longer have an unconditional passive grab on Button1 for active and raised windows, we need to check if a client sent us a _NET_WM_MOVERESIZE ClientMessage and update the pointer buffer on that. Also clean up the pointer buffer handling code again - reduce the number of roundtrips required. Remove debug message.
2011-01-20Merge branch 'master' of git://git.compiz.org/compiz/coreSam Spilsbury2-2/+5
2011-01-13Update core ABISam Spilsbury1-1/+1
2011-01-13Don't create CompWindow and initialize plugins until MapRequest.Sam Spilsbury2-2/+5
Having all plugins track windows on CreateNotify doesn't make any sense because they are not visible, and the ones that they would need to track would be override-redirect anyways. Also tracking windows in this way broke reparenting in such a way that a number of buggy applications using xembed (eg gnome-panel applets and Adobe Flash would consistently crash).
2011-01-11Merge branch 'glibmm-experimental'Sam Spilsbury4-20/+30
Conflicts: src/screen.cpp
2011-01-11Merge git://git.compiz.org/compiz/core into glibmm-experimentalSam Spilsbury1-0/+1
2011-01-08Fix grabs activating when we don't need them to, fixing some applicationsSam Spilsbury1-0/+1
receiving wrong events
2011-01-05Use clock_gettime and add a TIMESPECDIFF macroSam Spilsbury1-0/+7
2011-01-04Re-write timers to be a regular Glib::SourceSam Spilsbury2-12/+4
We previously used Glib::TimeoutSource but this one is broken for a number of usecases that we need, so create our own timeout source which is similar to the stock one, except that we only insert one source into the glib stack with grouped timeout times for the rest of our timers. Allows the use of setTimes () and setCallback () without glib becoming a disaster-zone. Also fixes a number of crashes related to the buggy glibmm implementation.
2010-12-08Merge git://git.compiz.org/compiz/core into glibmm-experimentalSam Spilsbury1-2/+4
Conflicts: include/core/screen.h
2010-12-06Fix startup order.Sam Spilsbury1-0/+3
Previously we were doing something like this: init -> screen init -> init core plugin screen -> create windows -> eventLoop -> updatePlugins -> initPlugin (initScreen (initWindow))). This order has a number of problems with it - namely that config plugins will be loaded on one full pass of the event loop and then core options are initialized, before which we were doing certain startup procedures on the default options, which resulted in strange race condition bugs. Also, if we are compositing we have a brief (visible) switch from 2D mode to 3D mode - in reality we want to be in 3D mode ASAP. The new order is init -> updatePlugins (load plugins specified on command line) -> screen init -> load command line plugins -> update plugins (load new plugins specified by config plugins on the command line) -> init screens -> create window (init plugins for window) -> eventLoop. This means that it takes longer for us to get to the event loop, but at least we can zip through the initialization phase quite quickly
2010-12-03Merge branch 'master' of git://git.compiz.org/compiz/core into ↵Sam Spilsbury1-3/+3
glibmm-experimental
2010-12-03Revert "Ensure that we can change the input focus, even if we requested a ↵Sam Spilsbury1-2/+0
client to take focus and it failed to do it before it was unmapped" This reverts commit 1084f0be5c200d46ad3b4358a477963d31040736. Danny suggested a better way to do this from a commit to 0.8
2010-12-03Ensure that we can change the input focus, even if we requested a client to ↵Sam Spilsbury1-0/+2
take focus and it failed to do it before it was unmapped
2010-12-02Make w->managed () wrapableSam Spilsbury1-3/+3
2010-12-02Merge branch 'glibmm-experimental' of ↵Unity1-1/+1
git://git.compiz.org/users/dbo/compiz-with-glib-mainloop into glibmm-experimental
2010-12-02Remove doPoll () and rewrite CompWatchFd to use Glib::IOSourceSam Spilsbury1-0/+2
This changes compiz to monitor file descriptors directly in the glib mainloop which fixes things like not exiting correctly with session managers and the like.
2010-11-27No need to reset the source, it happens for us on destructorSam Spilsbury1-1/+1
2010-11-25Return false once we're done with the sourceSam Spilsbury1-1/+1
2010-11-20Port C based glib to C++ glibmmSam Spilsbury2-12/+18
2010-11-12Merge branch 'master' of git://git.compiz.org/compiz/coreSam Spilsbury1-2/+17
2010-11-11Listen for FocusIn/FocusOut on NotifyGrab NotifyUngrab and set an internal ↵Sam Spilsbury1-2/+17
bit that plugins can query
2010-11-11Bump core ABISam Spilsbury1-1/+1
2010-11-09[PATCH] Be a little bit more clever when handing option construction in ↵Danny Baumann1-3/+5
handleActionEvent. Previously we would have a static CompAction::Vector which we would resize to lower sizes (so that triggerFooBindings doesn't die) but this causes us to be reconstruction CompOptions all the time which is expensive. Instead just reset the option to make it "dead" (eg keep it in memory for longer) but don't create and recreate them all the time Also copy the active state of a CompAction on copy-construct
2010-11-09Don't increase core action register reference counts if the same action was ↵Sam Spilsbury1-1/+5
already registered
2010-11-09[PATCH] Be a little bit more clever when handing option construction in ↵Danny Baumann1-3/+5
handleActionEvent. Previously we would have a static CompAction::Vector which we would resize to lower sizes (so that triggerFooBindings doesn't die) but this causes us to be reconstruction CompOptions all the time which is expensive. Instead just reset the option to make it "dead" (eg keep it in memory for longer) but don't create and recreate them all the time Also copy the active state of a CompAction on copy-construct