AgeCommit message (Collapse)AuthorFilesLines
2010-12-24Allow plugins to initiate scale when initiated to change the matchSam Spilsbury1-7/+8
2010-12-24Don't clear terminate state until action is actually finished clearing up.Sam Spilsbury1-3/+3
Clearing the terminate state before this will confuse other plugins which check activity states
2010-12-23Add signals for viewport change begin/endSam Spilsbury1-0/+6
2010-12-19Other mouse buttons should be allowed to change the selected window.Sam Spilsbury1-4/+7
But only button 1 should be able to select windows and terminate scale mode. Fixes incorrectly selected window when the mouse has not been moved after scale mode has been initiated and the mouse is hovering over an unselected window, and then Button2, or Button3 eg is pressed. This broke scaleaddon's pull window and close window bindings. Fixes LP #691087
2010-12-13CleanupSam Spilsbury1-2/+1
2010-12-13Send window id in decoration notificationSam Spilsbury1-1/+5
2010-12-13Fix unintialized variable usageSam Spilsbury2-6/+8
2010-12-13Merge branch 'master' of git+ssh:// Spilsbury2-1/+4
2010-12-13Also beautifySam Spilsbury1-8/+8
2010-12-13Break after updating after reparentSam Spilsbury1-0/+1
2010-12-13Add a compiz event for when we have been decoratedSam Spilsbury1-0/+3
2010-12-13package mode is release debug for pluginsDidier Roche2-1/+4
also ship the slipped gconf cmake helper
2010-12-13Fix buffer overflow on some platforms (Non NULL passed to XOpenDisplay)Sam Spilsbury1-1/+3
2010-12-11Ensure decoration is updated on shade. Fixes artifacts.Sam Spilsbury2-4/+33
Previously we'd rely upon a change in winDecorAtom to tell us when to update the decoration, however now that we are reparenting we can't do that anymore so hook windowNotify to watch for shade and unshade and then force the decoration to be updated on resizeNotify if this is the case.
2010-12-09Fix some typoes and handle the case where the window is already under a dockSam Spilsbury1-9/+19
2010-12-09Be a little bit more clever about constraining resizes to workareas.Sam Spilsbury2-4/+147
Basically we want to constrain resizes to workareas, but not in the case where two outputs touch. In the case that two outputs touch they must be directly adjacent and there must not be a panel or dock in the way
2010-12-07Don't stack focus-denied windows in a disallowed positionSam Spilsbury1-6/+11
When a window is focus-stealing-prevented, we should only stack it below the active window if that is possible. If it's not (the window has the "Above" state, for example), then we should stack it at the normal location. Forward port 984bca5801d6179b9f8081c4b6c466063975d940 to master
2010-12-07Only reflect the decoration size in MapRequest gravity adjustment, notSam Spilsbury1-5/+5
the whole window size. Forward port a86c461684f122fd286884c4ff28f6ea673fe669
2010-12-07Also update desktop keysSam Spilsbury1-6/+6
2010-12-07Update path names in gconf keybindingsSam Spilsbury1-13/+13
2010-12-07Cleanup key binding list from redundant and unneeded keys and add scaleSam Spilsbury1-11/+3
binding. Forward port c9993a7739e9bf6b595e5efa62605027241c3d7f to master
2010-12-07Also handle windows that have a server border set in pointer placement mode.Sam Spilsbury1-2/+5
Also fix typo. Forward port 1d5f80cd52274484ecb59a1c8f7e7b6cea3b1844 to master
2010-12-07Correctly reflect that the switcher window is managed.Sam Spilsbury2-1/+17
Forward port 46371f9172089a3e60648b834611d5694df2e80b to master
2010-12-07Make short descriptions in opacity a little more verboseSam Spilsbury1-12/+12
Forward port f089351c2d6d62302e59f52a810b722808a4a574 to master
2010-12-07Move window position validation to validateWindowResizeRequest function.Sam Spilsbury1-55/+57
Forward port 2a37340f9610e7f80a66ed6ab48190c181784f6b to master
2010-12-07Don't make minimized windows unmanaged.Sam Spilsbury1-1/+1
Forward port 8dcfbb256322565228cbcda5230de95343887fac
2010-12-07Fix window offset at the start of close animation.Sam Spilsbury2-21/+22
Forward port 0a28ca484a8a72060b580e0ccc75ea20f3ddab1a
2010-12-07resize: Remove constrain_y, limit to work area instead of screen edges.Sam Spilsbury2-60/+36
Forward port 3075fa6eac1f4c7faa804b319f916899589702cc
2010-12-07resize: Fix typoSam Spilsbury1-1/+1
2010-12-07resize: Allow resizing outside screen edges when initiated with key.Sam Spilsbury2-4/+8
Forward port 6e281003bbcb01ac028aa963ab1bee0349c41376 to master
2010-12-07place: Fix typo (off by one error)Sam Spilsbury1-2/+2
Forward port 4832c514b0c460febc90720a61c11d6a0f64ec84 to master
2010-12-07Also set constrainY to FALSE if move is not loaded.Sam Spilsbury1-1/+1
Forward port 615d823f5edc26eaa12e83c97099c7ab0f017968 to master
2010-12-07Constrain resizes to screen edges and fix constraining.Sam Spilsbury1-5/+39
Forward port 6c12e5a60297143a72c4e4bb104b0c35102ff3f4 to master
2010-12-07resize: Obey move's constrain_y optionSam Spilsbury2-1/+35
Forward port 0d23e499774769d295509f189fbf4b7c21763ef6 to master
2010-12-07Fix typoSam Spilsbury1-2/+2
Forward port 466184685e570b6736adce7800253815d72e1c02 to master
2010-12-07Fix assertion if core is specified as plugin on the command line.Sam Spilsbury1-14/+21
Forward port 2a0087732f23cd87284366ec0dc224cbd5fd68ec to master
2010-12-07scale: Properly remove closing window.Sam Spilsbury2-8/+25
Forward port 96e59168920650f49c5e79b342261c863694fb23 to master
2010-12-07Switcher: prevent simulataneous plugin uses that don't make sense.Sam Spilsbury2-7/+7
Forward port 1974f9b5bfbab24023a0a647b606eedccbb83d04 to master
2010-12-07Avoid adding core twice to plugin listSam Spilsbury1-0/+2
Forward port 20cdf69d096616104cd26fc30c8efd7e4e4e164d to master
2010-12-07Fix sibling checking logic.Sam Spilsbury1-2/+2
Forward port f4af2061d93057843efd437b30b8199dc89defdd to master
2010-12-07Make sure findValidStackSibling below doesn't return a sibling thatSam Spilsbury1-0/+15
we're not allowed to stack under. Forward port 1baf514fefb9177764af99cf508596686583244e to master
2010-12-07Fix up coding style, add some comments.Sam Spilsbury1-0/+6
Forward port 24dd71f04892ddd031ac2bf7b166e10b74aa1bbd to master
2010-12-07Never unload plugins passed to the cmdlineSam Spilsbury3-20/+80
Forward port 27430deae222b4380c0e91da849f930a297afb73 to master
2010-12-07gnomecompat: Default to gnome-terminal for terminal keybinding.Sam Spilsbury1-1/+1
Forward port 060a99697cde9105c85623cb464a36b46da3bf97 to master
2010-12-07Revert gravity adjustment made at MapRequest time at UnmapNotify timeSam Spilsbury3-16/+38
Forward port a0be58f5e9f5e419af4f5f15505e9a9f6fe47bdd to master
2010-12-07Don't take screenshot if Escape was pressed.Sam Spilsbury1-0/+3
Forward port c9239e92fe0b2d6bfc674b523313617017ca062f to master
2010-12-06Revert "Fix windows being forever treated as "not managed" in focus () and ↵Sam Spilsbury2-1/+1
not entering into scale mode, etc" This reverts commit 51e01727d309f39e3f42ecf6b53b4ebb31a385dc. There is a better way to do this
2010-12-06Fix startup order.Sam Spilsbury5-93/+186
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
2010-12-05Fix windows being forever treated as "not managed" in focus () and not ↵Sam Spilsbury2-1/+1
entering into scale mode, etc We need to re-set unmanaging too once managed is set again
2010-12-04Fix windows consistently moving by gravity * window decoration extents.Sam Spilsbury1-1/+15
The behaviour from 0.8 was ported to update decoration windows on the window destructor, but because we are moving windows on a timer in order to handle reparenting issues when decorating them, this also causes the time to be destroyed when the CompWindow is, so the windows never actually get moved back into place correctly. FIXME: Timers should have a referencing system, it does not make sense that we have to preserve their scope