Fix BWindow::DecoratorFrame for top-titled windows.

Also take into account the border width for the edge with the title.
Fixes #8614.

Add BWindow::HasShortcut()

Reverse the meaning of BWindow fShowLevel to match BView.

This also matches the client_window_info.show_hide_level field used in Deskbar
and other applications.

While doing this, keep fShowLevel fully in sync between BWindow and app_server,
use one message type for both hiding and showing, and make the decision to show
and hide the window in the app_server.

Lastly make minimize behave as described in the Be Book: hidden windows cannot
be minimized, and minimized windows which get hidden become unminimized.

Sync BWindow fShowLevel with the app_server.

Implementing the window_info.show_hide_level in terms of this solves the
problem of minimized windows also being considered hidden, when really they are
just hidden in the app_server.

window_info.show_hide_level is still defined backwards with a comment making
that clear.

Also removed sending fShowLevel in the minimize request since it is now
maintained in the app_server.

Fixes #4127.

Compilation and 64-bit fixes to libbe.so sources.

Fixed the usual issues - printf format strings, uint32 instead of
addr_t, etc. One thing that isn't so nice is several places where
BList is used to store (u)int32, these require a double cast to addr_t
then void* to silence a warning on x86_64.

Keymap changes from recent activity. No AltGr yet.

Below is a mostly complete summary of the changes in this commit.

* Set the DeadKeys for the US-International Keymap to use the Option map.
* Rename American keymap to US
* Update the US, US-International, and United-Kingdom keymaps to take
out unneeded spaces in the option layer. Also updated the dead keys
and some other keys on the US-International keyboard to use UTF-8
characters rather than there ASCII equivalents when different.
* Make the Option key fall-through when there is no mapping in the Option
table. Option is for special characters, if none, print the regular one.
This is mostly meant for the US keymap which has an empty option map. But
also so that you don't have to repeat the normal, shift, and caps maps in
the option map needlessly. Although the keymaps are still not empty in
some cases that it could be like numpad keys and space.
* Update the /bin/keymap app to use fputs() instead of printf() when there
is no actual formatting taking place. I've gotten into trouble for doing
this before and it is faster to not process the string unnecessarily.
* Also several 80-char limit style fixes and updated comments.
* In Keymap class Reorder the modifier keys to match the keymap files.
Put B_CONTROL_KEY check above B_OPTION_KEY. Neither change has any effect,
they are purely aesthetic.
* Update DumpKeymap() method to use the abbreviated modifier letters so it
will fit in your 80-char wide terminal.
* Tiny style fix in InputServer
* 80-char limit style fix in BWindow and add a comment that the shortcut
gets eaten in the case of Cmd+Q
* Implement IndexForModifier() in KeyboardLayout, although I am not using it.
* Take Caps Lock out of the Modifier keys window because I couldn't get
it to work the way I wanted it to.
* Move key roles to the left column, and the key label on the left. Add column
header labels. Thanks Rimas!
* Add validation and improve marking menu options. Add a 'Disabled' option
to control, option, and command menus to disable the key. Make the key
role text grey if the key roles is disabled. Validation ensures that you
cannot repeat the same key twice in the Modifier keys window since that
won't work. You can't define 2 sets of option keys even if you really want
to. You can disable your control, option, and command keys if you
want, but that is not recommended.
* Rename kUpdateModifiers to kUpdateModifierKeys message to differetiate
it from kUpdateModifier.
* Add shift key to Modifier keys window, use the stop icon instead of the
warning icon to indicate conflicts.
* Allow the Layout system to control the size of the Modifier keys window
again, set the width's of the key role lables to the widest, set the width
of the menu fields to take up the rest of the space minus room for the
conflict views. I didn't like it that the Modifier keys window would change
size based on what options you had selected in the menu fields. Now it
doesn't, but, the layout system still makes it all fit.

Reconnect window and recreate all views.

Change instances of wether in comments to whether. No functional change intended.

Revert part of r43166 that was included accidentally. This resulted in
incorrect behavior while processing certain kinds of messages, and
consequently deadlocks in some apps. Fixes #8101.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@43237 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove extraneous code with respect to telling a menu to close which was
breaking various cases where menus coexisted in a window with a view which
was using event masks:
- If one invoked the menu via, e.g. a BMenuField, and then tried to choose
an item on it, the mouse down would get captured, the menu would be
closed, and the mouse event would be thrown away without ever reaching
the event mask view.

- Furthermore, since the menu was told to terminate early, it would decide
that the user hadn't actually chosen that item (the escape key case),
and never actually invoke it.

Fixes the menu fields in the mouse preflet being broken.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@43166 a95241bf-73f2-0310-859d-f6bbb57e9c96

Tweak the notification window again :
* Spacing of the bprogressbar is now 8pixels on each size
* Remove the useless window tab for now

Also fix DecoratorFrame() again as BORDERED_WINDOW didn't work with it either.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@43034 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fix DecoratorFrame() for kLeftTitledWindowLook windows
* Use it in notification window for better positionning.
Thanks augiedoggie for reporting the problem !


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42588 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch from X512 (ticket #7408): Don't notify input server of focus change if window is not active.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42574 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch from X512 (#7408): only send input method aware messages for active windows


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42557 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Followed Ingo's suggestion, and added a BWindow::Layout() method.
* Changed ShowImage to use that function.
+alpha in case Ingo gives his okay :-)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@41782 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed _CheckSizeLimits() to UpdateSizeLimits() and made it public.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@39495 a95241bf-73f2-0310-859d-f6bbb57e9c96

Big change deriving BLayout from BLayoutItem, and allowing viewless BLayouts.
a few highlights:
* BLayout now derives publicly from BLayoutItem
* Added BAbstractLayout class, which our layouts now derive from
* updated layout builders to avoid creating views when they don't need to
* updated layout classes
* updated AboutSystem to fix a little regression
* more details on #6407
* please tell me about any regressions, I've tried to find them all, but some
may have slipped by.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38207 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Modified PtrScr key behaviour:
- No modifiers: take a screenshot with zero delay and launch the GUI
- Shift-PrtScr: Silent screenshot using the saved GUI settings
- Ctrl-PrtScr: Take a screenshot using the saved GUI settings and copy the
screenshot to the system clipboard
* A few locale related changes (I am not sure how this works when a class is
shared between two applications, I hope I got everything right)



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37252 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rename "ScreenshotApp" to "Screenshot", rename its signature from
"application/x-vnd.haiku-screenshotapp" to "application/x-vnd.haiku-screenshot"
* Rename "Screenshot" to "screenshot", rename its signature from
"application/x-vnd.haiku-screenshot" to "application/x-vnd.haiku-screenshot-cli"
* Move screenshot from /system/apps to /bin


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37059 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Made zooming to full screen center the window when the window cannot be made
full screen (due to size restrictions).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36714 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The window was now always taken to the new workspace with Ctrl-Alt-arrow;
lazy debugging left over that closes ticket #5675 again.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36637 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Unified the Desktop and ServerApp AS_ACTIVATE_WORKSPACE to work in the same
way.
* Use that newly exposed feature of taking the focus window to the new workspace
when using the Ctrl-Alt-Shift-Arrow shortcut.
* This fixes #5675.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36585 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add comment explaining why MenusBeginning() is invoked before handling key events
with the command key pressed.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35567 a95241bf-73f2-0310-859d-f6bbb57e9c96

When a Window is created for holding BViews attached to an offscreen bitmap,
always behave as if within a transaction. Otherwise all drawing calls to such
BViews wait for the server to finish the request. This change gives a tremendous
speed boost for these situations and is compatibly with BeOS, since there you
also had to call Sync() before drawing a bitmap that was painted with attached
views.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35414 a95241bf-73f2-0310-859d-f6bbb57e9c96

Reverting r34335. Wrong location of my "B_NOT_MINIZABLE" check.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34350 a95241bf-73f2-0310-859d-f6bbb57e9c96

Make BWindow honor the B_NOT_MINIMIZABLE flag. Fixes ticket #4337.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34335 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added a BWindow::InViewTransaction() that determines whether or not the
window is currently updating its drawings as suggested by Stippi.
* Add this method to decide whether to call _FontChanged() in
BListView::SetFont().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34109 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BWindow::SendBehind(NULL) will now send the window to back.
* Added shortcuts ctrl-alt-F to get a window to front, and ctrl-alt-b to move it
to the back.
* Updated the user guide with these shortcuts, and also explained the missing
"single click to bring window to front" behaviour.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34108 a95241bf-73f2-0310-859d-f6bbb57e9c96

Spelling

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34081 a95241bf-73f2-0310-859d-f6bbb57e9c96

- this should fix gcc2 build, sorry ;)
- style fixes.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33691 a95241bf-73f2-0310-859d-f6bbb57e9c96

Extend the Ctrl-Alt-arrows shortcuts to also move the current window when shift is held, as on Gnome.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33690 a95241bf-73f2-0310-859d-f6bbb57e9c96

Include the terminating 0 byte when adding as string type. Otherwise someone
reading the message as an actual string will run into problems.
Should fix #4697.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33442 a95241bf-73f2-0310-859d-f6bbb57e9c96

Work based on a patch by Joshua R. Elsasser:
* Some key-sequences cause \0 chars in the "bytes" data which is supposed to
be passed to BView::KeyDown() and BView::KeyUp(). Therefor, one cannot use
string methods for adding/extracting the data to/from the events. For
example, Control-Space now works in the Terminal.

Thanks a lot for the original patch, Joshua!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33338 a95241bf-73f2-0310-859d-f6bbb57e9c96

* IsMinimized() now locks the window. This should fix bug #4274.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32902 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove the forced resizing to preferred size when CenterIn is called, and
instead just check if the size limits need to be set. The code that previously
did this when processing the B_LAYOUT_WINDOW message was extracted into a
private method, which is called from both places.

This may be slightly wasteful in some cases (since the size limits may be set
multiple times), but it is definitely needed because without it both of the
current test apps (DiskProbe and Screenshot) did not have properly centered
windows.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32622 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove useless BRect* version of CenterIn, make remaining version take a const
BRect reference.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32614 a95241bf-73f2-0310-859d-f6bbb57e9c96

Finally implemented BWindow::CenterOnScreen, with associated CenterIn(BRect)
methods as well as Size(). To avoid the problem of centering the window before
it has been resized by the layout system, I force the resizing early. If there
is a better way to do this or some way to avoid doing it repeatedly, let me
know. But I figure the Center* methods should not be called that often.

Updated Screenshot and DiskProbe to use this new method as a test. It certainly
cleaned up DiskProbe. I will update other code over the next few days (if
anyone wants to help, please do :)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32612 a95241bf-73f2-0310-859d-f6bbb57e9c96

removed resolved TODO comment


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32562 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Contrl-~ now switches between application windows. This is handier than the
mechanism used in BeOS, that is to press Control+Option-Tab (which didn't
work on Haiku, though).
* Did not change the window switch logic, though, so it's still not really
nice.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32524 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Our jump-to-next-group never worked, and since no one complained, option-tab
now takes over this role (leaving Alt-Tab) to the application.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32517 a95241bf-73f2-0310-859d-f6bbb57e9c96

Cleanup

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32233 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add TabFrame window property.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32228 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added a basic tool tip API, and implementation.
* The BView API can probably be regarded as good enough; the implementation
might need to be improved over time (also, some things as archivability
aren't fully implemented yet). The ToolTip.h header should get public once
finalized.
* Added new B_MOUSE_IDLE message that is sent to a BView after a certain
time has passed (BToolTipManager::ShowDelay()).
* Added small test app (ToolTipTest) that shows what is already working.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32078 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Don't use B_ZOOM; this does not respect the B_NOT_ZOOMABLE flag, same problem
as with B_MINIMIZE. This fixes bug #4134.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31828 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added AS_GET_WORKSPACE_LAYOUT to get the number of columns/rows of the
workspaces.
* Added shortcuts Ctrl-Alt-{Left|Right|Up|Down} to switch to adjacent
workspaces.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31824 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Don't use the B_MINIMIZE message for the minimize shortcut (ctrl-alt-m), since
that ignores the B_NOT_MINIMIZABLE flag. Instead, there is now a private
_MINIMIZE_ message (naming like those other ugly private BMessage constants in
AppDefs.h) which honours that flag.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31799 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Style cleanups.
* Removed former dead print-screen code.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31797 a95241bf-73f2-0310-859d-f6bbb57e9c96

Small style cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31155 a95241bf-73f2-0310-859d-f6bbb57e9c96

Reverted my change to usage of _get_object_token_(), it already means to use
a cached token... doh! Thanks Axel!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31140 a95241bf-73f2-0310-859d-f6bbb57e9c96

Cache a view's server token. Avoids getting it for every
BView method that contacts the server.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31134 a95241bf-73f2-0310-859d-f6bbb57e9c96

* various fixes identified by gcc4, provided by Joe Prostko


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30902 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added DecoratorFrame() method, which returns the outer frame of the window
on screen (ie including the decorator border and tab). Plus the necessary
refactoring as well as some TODO notes about windows with the tab on the left
side.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29820 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Reworked r29180 to make _HandleKeyDown() no longer rely on the active window
but if the message was targeted to the focus view. It will now only handle
shortcuts in this case.
* This also fixes bug #3513.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29776 a95241bf-73f2-0310-859d-f6bbb57e9c96

* it seems we can skip _HandleKeyDown() when not active (keyboard navigation, shortcuts and screenshots).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29180 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Handles shortcuts only when the window is active. This fixes bug #3414 (and hopefully is the R5 behavior).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29179 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Resolve further warnings on GCC4.
* Enable -Werror on GCC4 builds as well (limited to the same selected targets).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29118 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add missing headers (malloc/free, string functions, memcpy, etc. undeclared).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28982 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The move/resize window protocol now uses absolute coordinates rather than
relative ones. This fixes bugs #2658, and #3213; in BWindow::ScreenChanged()
the window does not yet know that it moved on the new screen (when it already
had a position on that workspace).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28908 a95241bf-73f2-0310-859d-f6bbb57e9c96

At some point, there was some hassle about what coordinate system the "where"
BPoint in a mouse moved message is. Turns out it's different (!) according
to the type of message and this is even documented in the BeBook. I don't
really know if we want to copy this, but since there are apps out there that
depend on this, I am implementing it so that it's compatible with BeOS: In
the B_MOUSE_UP/DOWN case, it's the coordinate space of the target view, in
B_MOUSE_MOVED messages, it's that of the window. That explains a few things...
Clicking into Pe's function popup window should now select the right icon.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28742 a95241bf-73f2-0310-859d-f6bbb57e9c96

The previous commit exposed another bug: BWindow::Zoom() incorrectly calculated the max height of the window by only including the border width once. This was previously masked by the fact that the (by default 5) extra pixels were included in its hardcoded default for the height of the tab. Also add the border width to the message returned by GetDecoratorSettings() so Zoom can use the actual value instead of hardcoding it.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28679 a95241bf-73f2-0310-859d-f6bbb57e9c96

Resolved a small todo thanks to r28664:
We now ask the app_server for the window tab's height instead of
hardcoding it.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28678 a95241bf-73f2-0310-859d-f6bbb57e9c96

Minor style fix.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28672 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Replaced some "new" with new(std::nothrow) where appropriate in our base
classes (BView, BWindow, BAlert, BButton - BTextView should be part of this,
too, to make BAlerts work).
* However, it's not that simple, because there is often no way to return an
error. Most of that code obviously assumes to be able to throw exceptions
(it's just not communicated to the caller). Maybe we should just start
documenting exceptions for R1 (and properly use exceptions later on).
* Automatic white space cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28646 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed two more problems with BWindow::FindView(BPoint):
* The function is not supposed to return hidden views.
* After iterating over the child views, the "view" variable was clobbered,
so it didn't work to return the current view if none of it's child views
were hit.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28494 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed numerous problems in BWindow::_FindView:

1) If a view contained the point, but had children, that view would never be returned as a result, even if none of the children matched.
2) When converting coordinates to the child's coordinate space, it was using the bounds rectangle rather than the frame. This in most cases had the result that the coordinate was unchanged, and thus messed up the search completely.

This fixes ticket #3000.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28440 a95241bf-73f2-0310-859d-f6bbb57e9c96

* reintroduce member fOffscreen which got removed in r2917
* check if locking the offscreen window succeeds before calling delete on it
* don't return offscreen windows in CountWindows and WindowAt (works now as on R5)

fixes ticket 1522, 1591, 1946, 2318 and propably more

While creating an BBitmap in BApplication the bitmaps window looper would
be added to the applications gLooperList, thus calling Quit() on that window
and later delete on a stale window pointer in BBitmaps dtor. The Lock() check
would fix the problem, but tests on R5 have shown that BApplication hides the
offscreen window in CountWindows() and WindowAt().



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28236 a95241bf-73f2-0310-859d-f6bbb57e9c96

bonefish + zooey:
* Fixed a general problem with respect to overriding of the reserved
virtual function slots: instead of statically invoking the method
that corresponds to the reserved slot on the class that contains the
slot, we now invoke the virtual Perform() method. Perform() then dispatches
the method invocation to the "proper" class, i.e. the highest class in the
hierarchy that actually implements the requested method.
This fixes a crash in apps that use liblayout's MSlider class and
should fix one or other spurious bug with old apps or libraries, too.
* added new header folder 'binary_compatibility' that contains files that
define the method codes and data structures required by Perform()
* looked for and implemented all used reserved virtual slot functions to
invoke Perform() where necessary or to pass on the method call statically
(for slots that were already maintained by Be)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28124 a95241bf-73f2-0310-859d-f6bbb57e9c96

zooey + axeld + stippi:
Fixed a problem with dispatching B_MOUSE_MOVED messages to views which need
the transit event:
* If the app_server decided to send a B_MOUSE_MOVED to the previous mouse
containing window, it forgot to add the _feed_focus flag, which was a problem
if the message actually happened to contain tokens (views that registered
for events.) On the client side, only those views would receive the message,
while the regular last mouse moved view would not be notified at all.
* On the client side, never change fLastMouseMovedView if the message is not
targeted at the preferred handler. In the above situation, any "registered
for events views" would receive the message first, but viewUnderMouse would
be NULL and then when the regular last mouse moved view received it, it
resulted in the wrong transit (B_OUTSIDE_VIEW instead of B_EXITED_VIEW).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28059 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Refactored a method for getting the transit from a mouse moved message.
* In _StealMouseMessage(), don't maintain fLastMouseMovedView, instead,
prevent B_MOUSE_MOVED message from being stolen that are important for
detecting transit changes. The point is that some apps (like Tracker) are
shooting themselves in the foot because they steal mouse messages via
GetMouse() in one place, but then rely on sane transit values in another
place. The way it works now, the view in question may get notified of the
same mouse moved coordinate twice, once via GetMouse() and once via
MouseMoved().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28002 a95241bf-73f2-0310-859d-f6bbb57e9c96

* use the screenshot app to take a screenshot
* remove libpng and libz dependency in case we build for Haiku



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27801 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed a race condition in BWindow::Show(). If it was the first time Show()
was called, the calling thread could be preempted, or simply be blocked on
calling Lock() after running the window thread. The window thread in turn
could be processing messages. In that case, fShowLevel would still have the
wrong value. For example, MediaPlayer would call IsHidden() on one of it's
views, which would then return true for this case. The result was that the
video view was not hidden and a black rectangle was showing on top of the
controls. This may also have caused other similar problems of course.
* fRunCalled was accessed without holding the lock.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27209 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BWindow::Activate() now also unminimizes a window if necessary.
* Removed superfluous white space.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26757 a95241bf-73f2-0310-859d-f6bbb57e9c96

In the B_MOUSE_MOVED message, "where" is supposed to be in window coordinates
while "be:view_where" is in view coordinates. We made the mistake of having
them both be in view coordinates. This caused a problem for example in Vision.
(#2460)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26491 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added TODO about a problem that the late mouse moved message dropping code
may have.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26353 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Made the mouse moved message dropping code slightly nicer to read.
* Don't drop messages which carry an important transit value. (Thanks, Axel!)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26352 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Introduced a new view event mask flag: B_FULL_POINTER_HISTORY which,
when set, prevents any old mouse moved message discarding.
* BWindow::DispatchMessage(B_MOUSE_MOVED) checks the event time of the
message and discards too old events, but only if there is another event
in the queue and the view does not specify B_FULL_POINTER_HISTORY.
* BView::GetMouse() ignores the checkHistory flag passed to the function
in case the event mask specifies B_NO_POINTER_HISTORY.
B_FULL_POINTER_HISTORY on the other hand prevents the dropping of old
messages.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26341 a95241bf-73f2-0310-859d-f6bbb57e9c96

Use a flag in BWindow to store whether an update to a view (Invalidate()) has
been requested. The first call to a BView::Invalidate() will flush the link
so that app_server is notified as soon as possible. It makes no sense for
further calls to Invalidate() to flush also, since Flush() is not cheap. This
trick makes Invalidate() about 3.2 times faster, making it a cheaper operation.
I could not see any negative effects, I tested with apps that invalidate
multiple different parts inside a window in reaction to something. Thanks go to
Ingo who had the idea.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@26020 a95241bf-73f2-0310-859d-f6bbb57e9c96

gcc4 fix.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25883 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Change the protocol for sending the affected view tokens during an update
session to also include each view's individual update rect (in screen coords).
Should actually not be mush slower than the old version, and hopefully makes
it possible to have smarter BView::Draw() implementations which should make
more than up for any potential speed loss.
* Removed unused version of View::AddTokensForViewsInRegion().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25879 a95241bf-73f2-0310-859d-f6bbb57e9c96

use find_directory() instead of /boot/home/


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25373 a95241bf-73f2-0310-859d-f6bbb57e9c96

Resolve a TODO and fix another ancient bug, #386. Print Screen is now handled
by BWindow, no longer by the app_server. This should stop the "screen freeze"
effect.

This adds a dependency on libpng.so and libz.so to libbe.so. The same
dependencies and the PNGDump code added here can be removed from the
app_server. I am just waiting for a code review of this before doing that.

This implementation still does not give the client a chance to handle it
differently.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@25269 a95241bf-73f2-0310-859d-f6bbb57e9c96

* "where" and "be:view_where" are supposed to contain the same value; dunno
why the latter is then even added, but we do the same thing as BeOS now.
* This fixes bug #2075.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24997 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added a class MenuPrivate to handle access to private BMenu methods.
BMenuItem and BWindow are no longer friends of BMenu, but use this class
instead.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24909 a95241bf-73f2-0310-859d-f6bbb57e9c96

As per discussion on haiku-dev, BWindow now always forces the
B_ASYNCHRONOUS_CONTROLS flag. If anyone does encounter an app that
breaks because of this, please let me know, but a few tests with various
apps seem to show no issues.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24290 a95241bf-73f2-0310-859d-f6bbb57e9c96

I was just going to implement Begin/EndViewTransaction(), but I saw
it is already implemented they way I thought it could be done - nice! Just
some simplifications.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24027 a95241bf-73f2-0310-859d-f6bbb57e9c96

reverted r21961. A menu could be closed, if the user
clicked (with the menu opened) on a view which uses GetMouse() in a loop
(PE, tracker), since it stealed the mousedown message.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@24005 a95241bf-73f2-0310-859d-f6bbb57e9c96

UpdateIfNeeded() no longer keeps the looper's BMessageQueue locked when
calling DispatchMessage().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23870 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Ingo broke binary compatibility of Window.h in r18649, thanks to Stefano
for finding this. This should fix bug #1734.
* Removed unused BWindow members and the temporary PrintToStream() method.
* Indentation cleanup (DirectWindow.h had some spaces instead of tabs).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23863 a95241bf-73f2-0310-859d-f6bbb57e9c96

IsModal() now includes also the kMenuWindowLook. Fixes bug #1269. Please
review.
Added a compile time option to switch off the use of the cached menu
windows. Looks like there is a problem with the focus system, if I keep
the cached menu window around, it "steals" keyboard events from the main
app window after the menu has been opened and closed once.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23343 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed copy&paste bug, thanks Stefano!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23197 a95241bf-73f2-0310-859d-f6bbb57e9c96

* DispatchMessage()'s B_WINDOW_ACTIVATED now checks if there are any pending
activation messages, and always retrieves the information from the latest.
* Reverted r23062, as the above should fix bug #613 as well.
* This should also fix bug #1674.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23196 a95241bf-73f2-0310-859d-f6bbb57e9c96

* removed declarations for methods which are not used/implemented in View.h
* improved naming of some private BView functions and used our underscore
prefix
* added a _DrawAfterChildren method to BView which does the necessary setup
and calls the DrawAfterChildren hook
* call the new _DrawAfterChildren method in the BWindow implementation, this
was easy since the view tokens are already hierachically sorted
* implement support for B_DRAW_ON_CHILDREN in the app_server's ViewLayer, it
simply means that children are ignored for the views clipping region, any
drawing methods will paint over children (therefor the B_DRAW_ON_CHILDREN
flag is even properly named)

With these changes, support for B_DRAW_ON_CHILDREN and the DrawAfterChildren()
hook are implemented and behave exactly as on R5 as far as I can tell, also
for the view background painting.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23154 a95241bf-73f2-0310-859d-f6bbb57e9c96

added a TODO item


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23063 a95241bf-73f2-0310-859d-f6bbb57e9c96

Dispatch also the B_WINDOW_ACTIVATED messages. This fixes bug #613, but
we need to check which other messages are handled by beos here.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@23062 a95241bf-73f2-0310-859d-f6bbb57e9c96

Made the private B_HIDE_APPLICATION message (now a misnomer) hide all applications
that share the same signature - so that this feature is now consistent to what the
Deskbar does.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22304 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added private (as of now) B_HIDE_APPLICATION message that calls
do_minimize_team() (which is what the Deskbar does when you select
"Hide All").
* Added keyboard shortcuts to minimize, and zoom a window, and to hide
the whole application (Cmd+Ctrl+M/Z/H respectively). The former two
are also present in Dano and up.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22302 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed bug #1028 from both sides:
* FindPanel::SetUpAddRemoveButtons() called Window()->FindView() but did not
check if Window() was NULL.
* BWindow now always checks the result of a BAutolock - this is why Tracker
got away with this bug on BeOS; NULL windows cannot be locked...


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22102 a95241bf-73f2-0310-859d-f6bbb57e9c96

The client-side implementation of B_NO_POINTER_HISTORY did only work for SetEventMask(),
but not for SetMouseEventMask(). We now track the value of that mask in a dedicated
member variable.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22022 a95241bf-73f2-0310-859d-f6bbb57e9c96

The server is too fast sending the messages; we now filter out old mouse events
client side in case B_NO_POINTER_HISTORY is set.
This fixes bug #1415.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22021 a95241bf-73f2-0310-859d-f6bbb57e9c96

The bug worked around in r21960 also affected Haiku. Now, a menu is only quit if the click
into its "controlling" window targeted the preferred handler - ie. is a standard click that
is not sent because of an event mask.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21961 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed the TODO added by Ingo in r21957: locking the application didn't even make any
sense. Instead, we now lock its app_server connection only. The deadlock as exposed
by starting Icon-O-Matic twice is now gone, at last.
* Fixed the TODO added by Ingo in r21953: moved the thread/handler renaming code in a
dedicated method _SetName() which is now called from _InitData() and SetTitle(); the
"w>" is no longer lost.
* Unlike the BeBook states, BMessageQueue::RemoveMessage() is indeed not supposed to
delete the message it removes.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21959 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added another TODO.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21957 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added TODO.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21953 a95241bf-73f2-0310-859d-f6bbb57e9c96

* this should fix BWindow::FindChild(BPoint)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21537 a95241bf-73f2-0310-859d-f6bbb57e9c96

Modal windows no longer install the 'Q' shortcut anymore. This fixes bug #1256.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21311 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implemented B_CLOSE_ON_ESCAPE as mentioned on the mailing list.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@21308 a95241bf-73f2-0310-859d-f6bbb57e9c96

If there's a opened menu, the window should "eat" the B_MOUSE_DOWN
message. This fixes bug #582 and now menus behave like on beos (or
should, at least)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@20799 a95241bf-73f2-0310-859d-f6bbb57e9c96

Small fix for _StealMouseMessage(): only if "feed focus" is "false", the message
can't be used.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@20737 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added and implemented B_INVALIDATE as available on Dano; to invalidate only a certain
part of the view, you can add a BRect "be:area" to that message - very handy.
* Cleaned up AppDefs.h a bit.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@20683 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implemented direct message passing for local targets; this fixes a deadlock
with PostMessage() in case the message queue is full.
Some notes:
* for synchronous replies, we don't use this mechanism yet, but it could be
extended to do that as well.
* the code looks so complicated because we need a way to access the looper's
queue without locking it (to prevent deadlocks); like Dano's solution, I've
abused BTokenSpace to store a BDirectMessageTarget with a BHandler.
* we also need to decouple the lifetime of a looper's queue from its target,
as we cannot lock the looper, and therefore, can't guarantee it stays valid
as long as we're accessing it outside of BLooper.
* init_clipboard() now needs to be done after the global constructors have
been called - since sending messages now needs gDefaultTokens to be initialized.
Since this is done per image, it shouldn't cause any troubles, though.
* some minor cleanup, removed unused _msg_cache_cleanup_() and friends.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19968 a95241bf-73f2-0310-859d-f6bbb57e9c96

Cleanup:
* Got rid of unused BLooper members
* renamed fTaskID to fThread
* Removed private and deprecated AddLooper()/RemoveLooper()/... stuff; BLooper is now
directly calling BLooperList methods.
* Got rid of extensive and useless comments
* Made a few TODOs more clear
* Merged InitData() and InitData(...) to _InitData(...)
* BLooper::Team() now uses BPrivate::current_team(), sTeamID is gone now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19966 a95241bf-73f2-0310-859d-f6bbb57e9c96

As per Ingo's request, I moved calling BRoster::Private::UpdateActiveApp() into
the app_server (and updated all comments that said otherwise).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19865 a95241bf-73f2-0310-859d-f6bbb57e9c96

Turns out the whole "active app" mechanism in the registrar wasn't used at all; the
Switcher now works as expected.
* Renamed TRoster::ActivateApp() to UpdateActiveApp(), as the app is already activated
at that point (the registrar only keeps track of it).
* BWindow::DispatchMessage() now calls the new BRoster::Private::UpdateActiveApp()
method when it receives a B_WINDOW_ACTIVATED message.
* Added BRoster::_UpdateActiveApp() which calls the new B_REG_UPDATE_ACTIVE_APP.
* Removed now unused B_REG_ACTIVATE_APP.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19857 a95241bf-73f2-0310-859d-f6bbb57e9c96

Messed up logic last time - the cursor should now work as expected again in BTextView.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19856 a95241bf-73f2-0310-859d-f6bbb57e9c96

Turns out our pulse mechanism was broken; we need to honour the rate set by
SetPulseRate() even if it is 0. BView::_Attach() and BView::SetFlags() now
just set the previous pulse rate again (which will start pulsing in case
there is no fPulseRunner yet).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19850 a95241bf-73f2-0310-859d-f6bbb57e9c96

One step closer to the Switcher - still doesn't work, though, but at least the
BWindow code looks okay now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19849 a95241bf-73f2-0310-859d-f6bbb57e9c96

BWindow::IsFront() was implemented incorrectly - we actually need to query the
app_server for this; added a new AS_IS_FRONT_WINDOW command for this.
For example, clicking on the menu bar to bring windows to front in FFM mode
does work now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19725 a95241bf-73f2-0310-859d-f6bbb57e9c96

Notifies input_server about the focus view IM awareness when the window is (de)activated
This fixes the last part of bug #658


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@19166 a95241bf-73f2-0310-859d-f6bbb57e9c96

This fixes bug #851 - GLTeapot unlocks the window before indirectly calling
BWindow::UpdateIfNeeded(); BPopUpMenu::Go() should be independent of this.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18946 a95241bf-73f2-0310-859d-f6bbb57e9c96

We're now using the same priorities for windows as BeOS does - this should improve
the responsiveness of the GUI, and should also fix bug #742.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18821 a95241bf-73f2-0310-859d-f6bbb57e9c96

Merge from layout management branch.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18649 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed a stupid copy&paste bug.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18605 a95241bf-73f2-0310-859d-f6bbb57e9c96

Now this should nail down bug #762 pretty well: BView::GetMouse() no calls the
private BWindow::_StealMouseMessage() which makes sure only messages for the
preferred handler are stolen, and also, that nothing gets lost that shouldn't
get lost.
Also updates the fLastMouseMoved view in case the message is actually removed
due to GetMouse().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18601 a95241bf-73f2-0310-859d-f6bbb57e9c96

Replaced Unlock() with UnlockFully() in BWindow's destructor, fixes bug
406


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18579 a95241bf-73f2-0310-859d-f6bbb57e9c96

We don't add the Alt-w shortcut to modal windows anymore
by default, thanks to Darkwyrm for pointing to this.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18450 a95241bf-73f2-0310-859d-f6bbb57e9c96

Stefano's investigations did the trick: added MenusBeginning()/MenusEnded() in
the shortcut evaluation.
This fixes bug #531.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18338 a95241bf-73f2-0310-859d-f6bbb57e9c96

The BWindow destructor is called with the window locked, but we need to unlock to acquire the menu semaphore. Updated BBitmap to lock he window on quit too. This fixes last problem mentioned in bug 406

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18156 a95241bf-73f2-0310-859d-f6bbb57e9c96

"class" is added by BArchivable::Archive(), no need to duplicate


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@18086 a95241bf-73f2-0310-859d-f6bbb57e9c96

When the B_COMMAND key is pressed, the event will not be forwarded to the target
handler anymore, no matter if a shortcut existed or not.
This fixes bug #498.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17883 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Dropped messages come in a _MESSAGE_DROPPED_ "packet", and were unpacked much too
late, in DispatchMessage(), working around all eventually installed message filters.
This fixes the odd "cannot drop files into file panel" bug mentioned in bug #669.
* They are now handled in _DetermineTarget() and _SanitizeMessage().
* _SanitizeMessage() is now called earlier, so that no user code can see them before;
even though this might give you wrong MouseDown() coordinates in combination with
a redirecting message filter, I think that's exactly what would happen in this
situation under R5 as well.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17839 a95241bf-73f2-0310-859d-f6bbb57e9c96

fix some more bugs in scripting handling, this seems to never end
Having canna input method installed shouldn't crash Haiku anymore


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17810 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed warning.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17807 a95241bf-73f2-0310-859d-f6bbb57e9c96

scripting in BApplication is mostly fixed
BLooper had a be-handler named suite
BWindow is now exposing Active
scripting to BViews isn't working yet


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17797 a95241bf-73f2-0310-859d-f6bbb57e9c96

* fixed some more GetSupportedSuites implementations
* fixed some local variables names
* fixed the case of some message attribute names


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17734 a95241bf-73f2-0310-859d-f6bbb57e9c96

fixed some property_info and value_info for BShelf, BMenu, BWindow and BView
real support is to be implemented for a lot of them


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17728 a95241bf-73f2-0310-859d-f6bbb57e9c96

begin to fix bug #658:
makes use of BWindow::_SetFocus() in BView::MakeFocus()
BWindow::_KeyboardNavigation() now uses BView::MakeFocus()
This is though not enough: _SetFocus isn't called on window activation/deactivation, thus the input server isn't aware of a focus view change in this case.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17723 a95241bf-73f2-0310-859d-f6bbb57e9c96

cleanup of archiving code

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17674 a95241bf-73f2-0310-859d-f6bbb57e9c96

More hacks to BWindow's mousedown handler, to fix some problems with menus. Fixes bug 610 among other things. Note: I don't like this code so much, but apparently BeOS handles it in a similar way. I accept suggestions on how to improve this.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17606 a95241bf-73f2-0310-859d-f6bbb57e9c96

an eventually opened menu is now quit in BWindow::DispatchMessage()
before sending a B_MOUSE_DOWN message to any other view. This fixes bug
594 for real and another bug in BMenuBar. BMenuBar tracking will also
be simplified a bit because of this. Install items to a NULL window on
Show() as does R5 (although I don't know why yet).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17605 a95241bf-73f2-0310-859d-f6bbb57e9c96

sorry, forgot about this change... fixes build

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17604 a95241bf-73f2-0310-859d-f6bbb57e9c96

added a way for BWindow to store and restore arbitrary decor settings,
currently those include only the tab location


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17583 a95241bf-73f2-0310-859d-f6bbb57e9c96

Detect recursions of UpdateIfNeeded. This is necessary as we may call hook functions (FrameResized(), FrameMoved()) that in turn could call UpdateIfNeeded again. This is the case for DarkSite which uses GetMouse() inside FrameResized(). This fixes resizing DarkSite and probably fixes bug 539 too.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17314 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Applied patch by G. Zachrisson to fix the endless loop in _FindNextNavigable().
* Rewrote broken _FindPreviousNavigable() - it now does the exact opposite of
_FindNextNavigable().
* Both methods did not ignore hidden views - they now do.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17287 a95241bf-73f2-0310-859d-f6bbb57e9c96

Now checks all calls to Lock() - this should fix buggy apps like Globe in bug #499.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17227 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implemented Switcher support in BWindow - doesn't yet work correctly for some
reason, though.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17164 a95241bf-73f2-0310-859d-f6bbb57e9c96

* fix bug 414 by Sync()ing in UpdateIfNeeded() to make
sure all messages that could trigger an update have
actually arrived at the server
* small clean ups here and there, some clarifications in
comments
* check for the return value of Lock() in DisableUpdates()
and EnableUpdates()


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17158 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed starting keyboard navigation (when no view has focus yet in which
case the window receives the keyboard message).
This fixes bug #411.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17156 a95241bf-73f2-0310-859d-f6bbb57e9c96

First steps at getting drag & drop to work properly. Simple drag & drop (draging Tracker items) should work now. Not sure about the negotiated version (with mimetype exchange). Fixed left behind drag bitmaps. Some cleanup.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17058 a95241bf-73f2-0310-859d-f6bbb57e9c96

no need to contact the server for this, transaction means that drawing commands are not immediately flushed, which is handled entirely on client side

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17044 a95241bf-73f2-0310-859d-f6bbb57e9c96

Window's destructor should acquire the menu sem, not delete it. Could help fixing bug 422

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17016 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The BView::FrameResized()/FrameMoved() hooks are now called asynchronously
as in R5, and no longer directly. This fixes bug #301.
* As a side effect, the hooks are now only called when the view is attached
to a window, as on R5.
* Removed dead B_VIEW_RESIZED/MOVED code from BWindow.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@17009 a95241bf-73f2-0310-859d-f6bbb57e9c96

Setting fRunCalled in case of failure is stupid, as this prevents the
window from being freed on Quit(). There still is a memory leak within
StressTest, though - but it's pretty tiny now.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16753 a95241bf-73f2-0310-859d-f6bbb57e9c96

BWindow::Show() will no longer Run() its looper without a valid app_server
connection.
StressTest now detects this case and quits those windows.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16752 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed memory leak in StressTest and BWindow: they both did not delete the
message they passed to a BMessageRunner object.
* Added note about the ownership of the message to the BMessageRunner
documentation.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16751 a95241bf-73f2-0310-859d-f6bbb57e9c96

When the creation of the window failed, all further messages accidently went
to the ServerApp, instead of being dropped - the window link is now updated
correctly in that case.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16750 a95241bf-73f2-0310-859d-f6bbb57e9c96

While the user resizes a window, programmatical resize operations are ignored,
likewise, while the user moves a window around, programmatical moves are ignored
as well.
This fixes bug #264.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16743 a95241bf-73f2-0310-859d-f6bbb57e9c96

* cleaned up fPulseRunner stuff, I have no idea why fPulseEnabled was
useful... so I removed it
* fixed memory leak with SetPulseRate(0), fPulseRunner is now
properly freed in the Window destructor
* fTitle is now freed as well
* fix some potential leaks in BView destructor as well

I wrote a stress test app, which I'm soon going to commit as well...
it shows that not all memory leaks are fixed by this patch, I could
use some help with this...



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16713 a95241bf-73f2-0310-859d-f6bbb57e9c96

implemented input_server notification of IM aware focus changes, not tested


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16690 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed BWindow::Show()/Hide(): fShowLevel was changed and checked without
holding the window lock.
* Run() is now called before creating the window in the app_server when
Show() is called for the first time (which is now checked with fRunCalled
instead of some thread arithmetics).
* Minimize() now sends the show level of a window to the app_server, so that
it can actually determine if minimizing or maximizing the window should
have any effect. This fixes bug #225.
* fShowLevel's meaning is now reversed; when it's above zero, it now means
the window is shown (before, a level less than 1 meant shown). This definitely
better fits its name :)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16536 a95241bf-73f2-0310-859d-f6bbb57e9c96

Accidently broke the build with the previous revision...


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16455 a95241bf-73f2-0310-859d-f6bbb57e9c96

Window has to be locked when it's going down or its children are queried/changed.
This fixed bug #182.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16454 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Shortcut messages now add a boolean "shutdown" field to the message - this
prevents Tracker from being quit by Command+Q.
* Also, it's now possible (as on R5) to remove the Command+Q shortcut.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16411 a95241bf-73f2-0310-859d-f6bbb57e9c96

The server now differentiates between hidden and minimized - that concept somehow
got lost, before.
It might not work 100% correctly yet, but it works good enough to hide the Tracker
status window from the Deskbar, and thus, fixing bug #133.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16314 a95241bf-73f2-0310-859d-f6bbb57e9c96

* fixed window zooming/unzooming behaviour
* fixed FrameResized() and FrameMoved() not
being called anymore in response to ResizeTo()
and MoveTo() since I introduced a check in
DispatchMessage() (fixes bug #123)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16312 a95241bf-73f2-0310-859d-f6bbb57e9c96

This fixes the update problems that I could observe. The client might have been not in sync with the server regarding window position and size, which in turn messed up the ConvertFromScreen() functions and also the view Bounds(). Therefor the current window geometry is also added to the information the client gets when starting an update session. I think the problem was there before my previous changes, just maybe less likely to be observed.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16258 a95241bf-73f2-0310-859d-f6bbb57e9c96

change the way the _UPDATE_ message is used: it is now a mere notification that some views need updating. The BWindow will then pull data from the server which views exactly and the update rect. Therefor, the server can append regions to the current update session even if an _UPDATE_ message has already been sent to the client. If multiple views are invalidated in the client, only one update session will be triggered instead of two with the old implementation. Some drawing defects can be observed, but I know how to reproduce them so I hope to fix them soon.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16245 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Use _FindView() where possible instead of using BTokenSpace::GetToken() directly.
* Fixed some issues in _UnpackMessage():
- The focus view could have been removed before its turn which was ignored
(and could led to a crash, for example when moving the Deskbar around).
It's now tested if it's still valid (can't use _FindView() here, as the
the preferred handler of a window could be any BHandler).
- fLastMouseMovedView could have been NULL in which case there is no
need to let the message be processed (was harmless, though).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16010 a95241bf-73f2-0310-859d-f6bbb57e9c96

The last change accidently put messages to invalid targets back into the game.
_DetermineTarget() is no longer called for these messages, "handler" will stay
NULL and the message will be dropped.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15921 a95241bf-73f2-0310-859d-f6bbb57e9c96

BWindow::_DetermineTarget() now defaults to the window, instead of no target at
all in case the window doesn't have a current focus view.
That allows shortcuts like Command+q/w to work again.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15920 a95241bf-73f2-0310-859d-f6bbb57e9c96

ServerFont:
* fixed weird pointer conversion in SetStyle()
* fixed a potential mix up in operator=() in case the
other ServerFont has fStyle == NULL

ServerWindow:
* the WindowLayer fTopLayer cannot be deleted by
client request, just for safety reasons
* the link is flushed if there is no drawing engine,
but this case is theoretical only
* deleting the ServerWindow object syncs with the
client, so that when BBitmaps are deleted, they
can be sure there are no pending messages (which
would be executed in a nother thread)
* there is no timeout anymore when sending messages
to the client, which made absolutely no sense

AGGTextRenderer:
* renamed fFontManager to fFontCache, because that's
what it really is
* fLastFamilyAndStyle defaulted to the system plain
font and therefor that font was never loaded when
the font never changed meanwhile

DrawingMode:
* I'm not quite sure but I think there was the
potential of a division by zero, at least I
had crashes with "divide error"

HWInterface:
* fix update when the cursor shape changed in
double buffered mode

ViewLayer:
* since the top layer is never really deleted
before its time has come, it is not necessary
to set it to NULL in the ViewLayer destructor

ViewLayer/WindowLayer:
* added a function to collect the view tokens
that are affected by an update session

EventDispatcher:
* use the importance of the message for the timeout
in _SendMessage()
* drop mouse moved events in the server if we're
lagging behind more than 5 ms (Axel, maybe review)

View:
* there were some problems with the locking
of the BWindow looper in RemoveSelf(), since
this is called from the window destructor,
also of BWindows from BBitmaps, which have
never been run (this might need review), at
least I seem to have solved the crashing
problems introduced by actually deleting the
view hirarchy in the BWindow destructor
* fixed _Draw() for being used non-recursively,
temporarily disabled DrawAfterChildren, which
didn't work yet anyways (because views cannot
draw over children in the server yet)

Window:
* small cleanup when deleting shortcuts
* sync with the server when having send
AS_DELETE_WINDOW (see ServerWindow above)
* fixed locking in Begin/EndViewTransaction()
* removed folding of _UPDATE_ messages, since
there is only one ever in the queue
* set the fInTransaction flag during an update,
I plan to use this in BView later to
flush the link when drawing outside of an
update
* BView::_Draw() is now called by view token,
this gives the next leap forward in speed,
the overhead because of drawing clean views
was considerable



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15878 a95241bf-73f2-0310-859d-f6bbb57e9c96

If there was a BMenuItem view with shortcut, it tried to delete the shortcut twice
(as the shortcut list wasn't emptied), courtesy of Stippi.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15866 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BWindow deletes its view hirachry
* BViews set fOwner to NULL recursively when detached
* updated a comment


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15851 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed shutting down apps with file panels - no more hang:
* if file panels shouldn't be quit, we no longer call QuitRequested() for
them and stop quitting windows (which left normal windows open, because
file panels always return "false" in QuitRequested())


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15715 a95241bf-73f2-0310-859d-f6bbb57e9c96

combine _UPDATE_ and B_WINDOW_RESIZED messages, I left the old code commented, because my combining code might need review. I don't understand why the current message is still in the queue and why passing 1 as index to FindMessage does not seem to work, BMessageQueue::RemoveMessage does not delete the message as documented in the BeBook... everywhere else in the code it seems to be taken as a fact

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15621 a95241bf-73f2-0310-859d-f6bbb57e9c96

a first and really naive implementation of drag&drop support, no visuals yet... Axel, no worries I'm going to clean this up and remove stuff from EventDispatcher again...

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15417 a95241bf-73f2-0310-859d-f6bbb57e9c96

The resize code now works correctly, finally - it temporarily did harm ;)
The BViews must be resized directly after every change in the window size - we
cannot wait until B_WINDOW_RESIZED, since subsequent calls have wrong view
sizes, then.
B_WINDOW_RESIZED is only really useful for app_server caused window resizing.
Added TODO to SetLook(): it may change the window size as well.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15385 a95241bf-73f2-0310-859d-f6bbb57e9c96

fixed wrong usage of ConstrainClippingRegion() introduced by myself a couple months ago

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15374 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The client views are now automatically resized on B_WINDOW_RESIZED as they are in the
server.
This saves overhead on both sides, the server doesn't need to build the update message
for the client, and the client doesn't have to unflatten and parse another message.
* This code is actually needed for the new clipping code in the app_server, but shouldn't
do much harm for the old app_server, either.
* Also disabled getting the bounds from the server, as that is just never needed (and would
also break the code).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15353 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BWindow::AddToSubset()/RemoveFromSubset() no longer send their team ID; this
is known by the server, anyway.
* B_MODAL_SUBSET_WINDOW_FEEL now also works as expected.
* Renamed AS_REM_FROM_SUBSET to AS_REMOVE_FROM_SUBSET.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15285 a95241bf-73f2-0310-859d-f6bbb57e9c96

Better app_server communication for SetFeel(), SetLook(), and SetFlags().
SetFeel() gets a status value back as well.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15267 a95241bf-73f2-0310-859d-f6bbb57e9c96

Some cleanup in the communication code.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15197 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Don't trust a message that we got goes to a handler that belongs to us.
* when we terminate gracefully and in the looper's thread, we no longer
unlock ourselves.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15155 a95241bf-73f2-0310-859d-f6bbb57e9c96

Don't trust the app_server that the view token under the mouse belongs to the
current looper.
This needs an investigation in the app_server, as that probably shouldn't happen?
(well, I wrote that part, I should know better)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15154 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed the port where app_server messages are received from "w_rcv_port" to "w<app_server"
to match the name of the application's port for that task.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15142 a95241bf-73f2-0310-859d-f6bbb57e9c96

No longer checks if there is already a "be:transit" item in the message;
there should never be one (at least not a valid one).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15125 a95241bf-73f2-0310-859d-f6bbb57e9c96

* _DetermineTarget() will now return fLastMouseMovedView in case there
was no valid "_view_token" in the mouse event.
* _SanitizeMessage() will now only add the "be:transit" field for
B_MOUSE_MOVED messages. It will also now only update fLastMouseMovedView
for B_MOUSE_MOVED messages, and not for all mouse messages.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15124 a95241bf-73f2-0310-859d-f6bbb57e9c96

* applied the logic from BWindow::task_looper() to BLooper as well (messages to
invalid target handlers are dropped).
* B_PREFERRED messages now go to the looper if there is no other preferred handler.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15088 a95241bf-73f2-0310-859d-f6bbb57e9c96

Next big step in the event handling:
* RootLayer's mouse event processing is now at its minimum - the
EventDispatcher handles them now. As a result, a window will now
get only one message per event.
* RootLayer adds "_view_token" to mouse moved messages that specify
the view currently under the cursor.
* There is now a mouse event layer in RootLayer that gets preferred
when it's set - this is now used for the window moving instead of
the previous mechanism.
* changed the previous DistributeMessage() to an UnpackMessage()
method following Adi's suggestion.
* caveat: some things might be functionally broken in RootLayer now
because of removing the mouse notification stuff.
* "be:transit" handling is now done completely client side by
BWindow::_SanitizeMessage(() (similar to what the input_server does).
This should also make the mechanism pretty robust, since every
B_MOUSE_MOVED message can now trigger the view transit (in case a
message is lost). B_WINDOW_ACTIVATED messages should be generated
client side as well.
* renamed AS_LAYER_GET_MOUSE_COORDS to AS_GET_MOUSE as it's not a
layer specific command, and also gets the mouse buttons.
* B_MOUSE_* messages from the up server now contain only a "screen_where"
field; "where" (in window's coordinates) and "be:view_where" are
added in BMessage::_SanitizeMessage().
* messages that don't have a valid target in the looper are now
dropped instead of being sent to the looper - this should be done
in BLooper as well, though.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15087 a95241bf-73f2-0310-859d-f6bbb57e9c96

* changed the way a message is forwarded to the focus view (instead of adding
a suspend focus field to the message, there is now a "feed focus" field in
case the message should be forwarded).
* added a comment to the BPoint version of _FindView() (since it's broken)
* _DistributeMessage() is now called after _DetermineTarget() - so that it
can prevent sending the message twice to the focus view.
* removed BWindow::DoUpdate() as it's no longer used.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15066 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed build for non-Message4 again. Both BMessage::Private classes should be compatible.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15063 a95241bf-73f2-0310-859d-f6bbb57e9c96

Huge cleanup and fixes:
* attachView() is now called _CreateSelf() and creates the app_server
view counterpart for itself, and no longer for a child view.
* removed superfluous deleteView().
* moved drawing from BWindow::DoUpdate() to BView::_Draw().
* made the recursive hook call functions consistent.
* fixed BWindow::DispatchMessage() to send public messages to the intended
target (instead of always handling them itself directly).
* DispatchMessage() no longer eats unmapped key events that were targeted
at the window directly.
* B_KEY_DOWN and B_KEY_UP events are now send to the target view as well
(this couldn't work before as BMessages were broken with B_PREFERRED_TOKEN).
* the default button is now correctly targeted by BWindow::_DetermineTarget()
(previously, the enter key was hacked to get through via _HandleKeyDown()).
* removing a view now also makes sure it won't have focus any longer.
* also, the DetachedFromWindow() hooks are now called first, so that any
changes made there cannot mess up our window anymore.
* removed BView::SetPattern(); _UpdatePattern() now does its job.
* renamend private methods to match our preferred style.
* removed unused methods and variables.
* more consistent naming overall.
* removed _PR3_COMPATIBLE_ stuff - there is definitely no need for us to be
compatible to that one.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15060 a95241bf-73f2-0310-859d-f6bbb57e9c96

* reverted Adi's premature changes to BWindow and restored _DetermineTarget() and
task_looper() again.
* removed BMessenger::fPreferred - whenever you had to specify "usePreferred" separately,
you don't have to do that anymore - use B_PREFERRED_TOKEN instead.
* fixed BTokenSpace::GetToken() semantics: it will no longer touch the "object" argument
in case of failure.
* Introduced a BWindow::_DistributeMessage() that will be part of the event dispatcher
counterpart to the app_server (the other will be _DetermineTarget()).
* Made it easier to use Michael's Message4 implementation: just add the following line
to your UserBuildConfig:
AppendToConfigVar DEFINES : HAIKU_TOP src : USING_MESSAGE4 : global ;
* Introduced ServerWindow::HandlerMessenger() and FocusMessenger() - the first will
target the client handler, while the other will target the preferred handler of the
client looper (usually the view having focus).
* Fixed dano message unflattening in the Message4 code.
* Changed BMessage::PrintToStream() to no longer use macros in the Message4 implementation.
* I hope that's all - it's a huge change, but it's all connected.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15046 a95241bf-73f2-0310-859d-f6bbb57e9c96

Removed BWindow::_DetermineTarget(). BWindow's task_looper() just calls BLooper's one. Took Message4 stuff from BWindow's task_looper() and put it in BLooper::task_looper() - guess this is wanted...

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15037 a95241bf-73f2-0310-859d-f6bbb57e9c96

Following recent changes (Axel's token stuff or Message4 impl) I could not launch a single application/window in app_server. This fixes the problem. Note I always test with DEBUG=1. Am I the only one who had this problem?

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15033 a95241bf-73f2-0310-859d-f6bbb57e9c96

Moved the declaration of roundf() to HaikuBuildCompatibility.h - it's missing from
math.h but exported by libroot.so.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14903 a95241bf-73f2-0310-859d-f6bbb57e9c96

* no longer needs to define roundf() as it's now part of libroot.so (like on BeOS).
* for the libbe_test target, though, I need to declare it, although it's also declared
in BeOS' math.h - Ingo??
* removed comment about B_ASYNCHRONOUS_CONTROLS - it's only used by our control classes
to behave differently depending on that flag (mouse tracking).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14892 a95241bf-73f2-0310-859d-f6bbb57e9c96

Introducing Message4. The changes to the related sources are ifdefed with USING_MESSAGE4 which is defined in Message4.h. To use Message4 the Message4.cpp, Message4.h, MessageUtils4.cpp, MessageUtils4.h and MessagePrivate4.h have to be linked to their counterparts without 4 suffix. Then MessageBody.cpp and MessageField.cpp have to be commented out in the app kit Jamfile and r5_message.cpp has to be added. There remain some bugs to be found. Feel free to change that.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14882 a95241bf-73f2-0310-859d-f6bbb57e9c96

prevent windows from being located at fractional offsets, fixes BAlerts (again)

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14879 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rewrote shortkey handling - it now actually works as expected.
* Some work on _DetermineTarget(), more to come.
* Minor cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14869 a95241bf-73f2-0310-859d-f6bbb57e9c96

Shortcuts now at least somehow work, but the mechanism is still broken.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14867 a95241bf-73f2-0310-859d-f6bbb57e9c96

InitData() no longer calls SetTitle(), but sets the title directly.
SetTitle() now also works when called before the window is shown for the first time
(ie. before the window thread is running).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14852 a95241bf-73f2-0310-859d-f6bbb57e9c96

a quick solution to get some B_EXITED_VIEW transit when the mouse leaves a view, probably gets replaced when Axel is looking into the event handling stuff

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14842 a95241bf-73f2-0310-859d-f6bbb57e9c96

just a change of constants.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14576 a95241bf-73f2-0310-859d-f6bbb57e9c96

cleanup

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14567 a95241bf-73f2-0310-859d-f6bbb57e9c96

the new clipping code uses the same mechanism for updating frame/bounds, so we don't need some things anymore.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14559 a95241bf-73f2-0310-859d-f6bbb57e9c96

small checkin to make the new clipping code work if one desires

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14461 a95241bf-73f2-0310-859d-f6bbb57e9c96

Undid my last checkin

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14406 a95241bf-73f2-0310-859d-f6bbb57e9c96

This change should've been checked in for well over a week, back when I activated the NEW_INPUT_HANDLING define. Sorry about that!

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14393 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed keyboard navigation; it should now behave pretty much like R5 AFAICT:
- moved standard keyboard navigation into the BView::KeyDown() hook
- the window now only handles tab+option/command key
- B_COMMAND_KEY triggers group jumping, not B_CONTROL_KEY (that activates the switcher,
but directly in the app_server)
- fixed broken group navigation: (modifiers & B_COMMAND_KEY & B_SHIFT_KEY) is different
to (modifiers & (B_COMMAND_KEY | B_SHIFT_KEY)) and is just never true with these
constants.
That allows the tab key completion to be used again in Terminal.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14038 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Implemented BRoster::ActivateApp().
* Added the respective case statement in AppServer::DispatchMessage().
The code that actually activates the app is still missing.
* Removed the remnants of the old way of notifying the registrar about
what app got activated (the activated client window did that).



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13820 a95241bf-73f2-0310-859d-f6bbb57e9c96

The window is not supposed to resize the top level view. This removes the
"look-through" area in Terminal, and probably ShowImage as well.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13743 a95241bf-73f2-0310-859d-f6bbb57e9c96

A call to SetSizeLimits() can change the window frame, so it's now updated, too (a separate FrameResized() message is sent anyway).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13742 a95241bf-73f2-0310-859d-f6bbb57e9c96

Some cleanup, some removed typos, some unification, some fixes and some added todos. Most of it related to client-server communication. Apps that rely on BFont stuff should work now (StyledEdit, Fonts, Keymap, Menu, ...). Or should not hang/quit at startup at least.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13625 a95241bf-73f2-0310-859d-f6bbb57e9c96

BWindow::SetDefaultButton() crashed when setting another default button.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13565 a95241bf-73f2-0310-859d-f6bbb57e9c96

usability improvements to scrollbar, sorry had no time to include all of Stefanos drawing code yet, other visual improvements

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13535 a95241bf-73f2-0310-859d-f6bbb57e9c96

We must not delete the server window's port...


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13510 a95241bf-73f2-0310-859d-f6bbb57e9c96

Now drops into the debugger if the window/looper is still locked in task_looper().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13480 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed AS_WINDOW_TITLE to AS_SET_WINDOW_TITLE.
Fixed handling in ServerWindow as stippi's latest commit broke it.
It's now properly done with a separate ServerWindow::SetTitle() method,
that will also take care to rename the window's thread.
Changed naming the window thread in the app_server to "w:<team>:<title>".


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13461 a95241bf-73f2-0310-859d-f6bbb57e9c96

MenuField layouts the menu bar better with respect to fDivider, it aligns better with other controls. fDivider in TextControl is an integer number now, small fix and small cleanup in Menu, Window::InitData takes an optional BBitmap token to construct an offscreen window, fixed PrivateScreen IndexForColor, View prevents being located at fractional coordinates as in R5, BBitmap unlocks its offscreen window since it is never Show()n and needs manual unlocking, fixed Slider offscreen window mode and improved triange thumb drawing, ScrollView would not crash when passing a NULL target just for kicks, the private MenuBar class now implements Draw to draw itself a little differently inside the BMenuField (dark right and bottom side) - though how it currently sets the clipping region prevents the text controls to draw in Playground, needs fixing

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13450 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed resizing/moving using the new clipping code. Stephan (I know I wrote about these changes in Layer.cpp) I saw the changes you did at how B_VIEW_RESIZE/MOVED are handled. This might break compatibility with R5. I think you guys should discuss this.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13396 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed SetLook() (is not implemented server-side, though).
Finally got the thread rename code right under R5 8-)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13391 a95241bf-73f2-0310-859d-f6bbb57e9c96

Some refactoring and cleanup:
- renamed some members and methods to fit our style guide
- moved removeSelf() to RemoveSelf() (as those two are essentially
the same), and fixed it on the way: the state of the child views
is now also updated by the new _UpdateStateForRemove() method
- Moved BWindow::sendPulse() and activateView() to BView::_Pulse()
and BView::_Activate()
- some minor stuff

This also fixes the broken previous commit - I forgot to update
View.h; this update contains the previous changes as well. Sorry
for the inconvenience.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13334 a95241bf-73f2-0310-859d-f6bbb57e9c96

Rewrote the push/pop state and ViewAttr stuff:
- it now seems to work finally correctly
- renamed ViewAttr to ViewState and put it into the BPrivate namespace
- some refactoring (moved some private BView methods to ViewState)
- renamed AS_LAYER_MOVETO/RESIZETO to *_TO (note the underscore)
- exchanged BView::originX/Y with fParentOffset (BPoint)
- divided AS_LAYER_GET_COLORS into separate ones for high/low/view color
- BView::SetPattern() now actually works as expected (ie. updates
only if necessary)
- exchanged the ViewAttr::flags with ViewState::valid_flags which inverses
the previous logic (which wasn't even used consistently)
- fState was initialized twice (incorrectly by the ViewAttr constructor,
and then again correctly by initCachedState()) - now the ViewState
constructor does the job alone, but correctly
- BView::PushState() no longer resets the state (it did so only locally
anyway...)
- cleanup


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13325 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Radically culled the list of BApplication friends and introduced a
BApplication::Private class for accessing relevant parts.
* Pulled the app server connection and IK initialization out of
InitData() into a new method _InitGUIContext() and introduced a private
constructor that allows to avoid this initialization. This will be used
for servers that don't have GUI respectively want to init the app server
connection later.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13312 a95241bf-73f2-0310-859d-f6bbb57e9c96

AS_CREATE_WINDOW now also gets the actual frame width and size limits of
the window on server side - ie. if the app_server could not create a
window of the size requested, BWindow::fFrame will still be correct
(and the size limits will mirror actual decorator limits).
Renamed fMinWindHeight and friends to fMinHeight and so on.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13302 a95241bf-73f2-0310-859d-f6bbb57e9c96

B_WINDOW_RESIZED sends int32 values, not floats.
Made message parsing more robust for a lot of messages.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13301 a95241bf-73f2-0310-859d-f6bbb57e9c96

The thread renaming was broken under R5. Cleaned it up a little.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13289 a95241bf-73f2-0310-859d-f6bbb57e9c96

fMenuSem was not initialized, and therefore, the window tried to delete some
arbitrary semaphore on destruction.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13250 a95241bf-73f2-0310-859d-f6bbb57e9c96

No longer draws hidden views (the app_server still handles them wrong, though,
when they are hidden while being attached to the window).
DrawAfterChildren() is now called at the appropriate place.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13224 a95241bf-73f2-0310-859d-f6bbb57e9c96

a new graphics state is pushed before calling BView::Draw() and popped after it. So any state changes you do in your Draw() function will be forgotton when you're done, as on R5. I have not done a lot of testing for bugs in the state stack within the app_server though...

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13174 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed BAppServerLink to AppServerLink, BPortLink to PortLink, LinkMsgReader
to LinkReceiver, LinkMsgSender to LinkSender, and put everything into the
BPrivate namespace.
Made AppServerLink a cheap object - it will use the applications receiver/sender
and not create its own buffers.
Fixed broken communication stuff here and there (mostly Font.cpp).
Put the newly introduced set|get_system_colors() into the BPrivate namespace -
please don't introduce private functions into the public namespace!!!
Also fixed their broken communication use, as Darkwyrm obviously forgot about
it again: the sequence Flush(); GetNextMessage() without error checking is
purely wrong and can make the app hang and/or crash! :-)
Other minor cleanup.
The input_server used some test mode with the haiku build target which is
probably wrong.
Hopefully I did not forget anything this time.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13128 a95241bf-73f2-0310-859d-f6bbb57e9c96

DispatchMessage() is called with the looper already locked, so we call fLink->Flush() directly, which does not lock it again. Changed some TODO to be more informative, removed one.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13076 a95241bf-73f2-0310-859d-f6bbb57e9c96

No need to Lock()/Unlock() as Flush() does it already. Sync() wouldn't have unlocked the window in case FlushWithReply() wouldn't succeed.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13061 a95241bf-73f2-0310-859d-f6bbb57e9c96

Sometimes I have seen crashes in BWindow::determine_target(), and this could very well fix it.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13060 a95241bf-73f2-0310-859d-f6bbb57e9c96

Now we flush the command buffer after handling a B_PULSE message. This fixes the PulseTest (and probably some other things).

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@13059 a95241bf-73f2-0310-859d-f6bbb57e9c96

BPortLink now has a FlushWithReply() method itself.
BPortLink::AttachString() now accepts a length argument, and will no longer
send a terminating null byte; LinkMsgReader::ReadString(), however, will
make sure the string read is null terminated.
Changed client communication code to use FlushWithReply() instead of Flush()
and GetNextReply() - there were many bugs and shortcomings in the code, I
hope I've fixed them all.
Converted ClientFontList.cpp to our coding style (but not completely, the
class members are missing).
Some more cleanup - I hope Adi will adopt our coding style one day!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12998 a95241bf-73f2-0310-859d-f6bbb57e9c96

fixed some pretty bad bugs that prevented menus from working (they now do): The BMessage* in AddShortcut is taken in responsibility by the BWindow, so we cannot directly use the message from BMenuItem, the be_app was locked in InitData but never unlocked, it fixes BMenus only working once, and who knows what else it fixes. A little cleanup with _BCmdKey usage, also note that it is inefficient to RemoveItem()s from a list in the destructor of any class using a BList as data container! Simply delete the items and be done with it. The BList destructor will take care of the rest and free its storage in one go.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12982 a95241bf-73f2-0310-859d-f6bbb57e9c96

The window was removing the app_server connection too early, and thus hang in Quit().
Removed the stopConnection() method.
Cleaned up header a bit.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12959 a95241bf-73f2-0310-859d-f6bbb57e9c96

huge cleanup, support for SetSizeLimits, support for truncating strings, numerous decorator bug fixes, Layer does not draw when view color is B_TRANSPARENT_COLOR, cleaner dispatching of mouse events to the WinBorder in RootLayer, commented the char map selection in the font server, as it seems glyph lookup by unicode index works much better with the default map

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12945 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed build with debug enabled. Renamed a static variable.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12938 a95241bf-73f2-0310-859d-f6bbb57e9c96

BBitmap now uses the correct (private) BWindow constructor. Partially implemented that constructor. Fixed another typo in BView, verified and removed some TODOs

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12884 a95241bf-73f2-0310-859d-f6bbb57e9c96

Now we delete the menu semaphore in the window's destructor

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12785 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed BWindow::UpdateIfNeeded() by moving the code from BView::GetMouse() over.
GetMouse() will now just call UpdateIfNeeded().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12736 a95241bf-73f2-0310-859d-f6bbb57e9c96

Converted Window.cpp to our coding style guide and (unfortunately) fixed a
myriad of small bugs here and there, as well as some very odd coding techniques:
- the "Minimize" property was not handled
- the "Minimize" property was not settable
- the property/scripting stuff was horrible, anyway
- there are more and better error codes than B_ERROR (could someone tell the
app_server guys? :))
- added some ToDo items here and there
- and much more I don't remember


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12723 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implemented private function BWindow::DequeueAll() which reads all pending
messages from the port.
BView::GetMouse() now calls this function if it was called from the window's
thread and thus blocking the message loop.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12716 a95241bf-73f2-0310-859d-f6bbb57e9c96

Work in Progress. The server keeps the client window up to date on layer movement/resizing. This fixes quite a few problems and brings support for FrameMoved and Resized hooks. But implementing it this way has its own set of problem, most importantly: When a BView calles Window()->CurrentMessage() in its FrameMoved/Resized hooks, it will see something very different from what it would see in R5. This needs to be fixed, but I have not had a good idea how to do this other than faking the current message in BWindow, which I didn't look into.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12689 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixes to my fixes of BView::MakeFocus(), the previous focus BView needs to be unfocused of course so the derived classes implementation gets called. Simplified BWindow::setFocus(). BView calls FrameResized() and FrameMoved() if it has no parent, I don't know how R5 handles it, but I added a TODO...

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12683 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added handler for _MENUS_DONE_

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12583 a95241bf-73f2-0310-859d-f6bbb57e9c96

added some safety checks, cosmetic changes - please note that it is safe to call delete with a NULL pointer, unlike free(), so please don't clutter the code

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12564 a95241bf-73f2-0310-859d-f6bbb57e9c96

work around for layers not being moved that have not yet been added to the tree. See WinBorder::MoveTo for the explaination. The same problem should be at other places and most likely for normal views as well. This fixes BWindow::MoveXX() when it is not yet Show()n

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12500 a95241bf-73f2-0310-859d-f6bbb57e9c96

fixed BWindow::MoveXX(), BWindow::ResizeXX() and BView::ConvertToScreen()

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12491 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added some TODO comments in Window.cpp, improved coordinate conversion in View.cpp. I'm not sure if this really works correctly, I need to write more tests. The scrolling test actually does more of what it is supposed to. At least the scrolling properly follows the mouse now. Missing are the movement of child views, the rebuilding of clipping, and I think that the scrolling needs to be synced to update requests having been fullfilled, or else some parts of the View will be CopyBits()ed which have not been drawn after having been scrolled into view by previous calls to ScrollBy(). Any insights appreciated.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12485 a95241bf-73f2-0310-859d-f6bbb57e9c96

small changes and fixes here and there

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12480 a95241bf-73f2-0310-859d-f6bbb57e9c96

Adjusted SetFeel() to current server implementation

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12457 a95241bf-73f2-0310-859d-f6bbb57e9c96

Forward message to the handler set inside task_looper() no to fFocus

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12395 a95241bf-73f2-0310-859d-f6bbb57e9c96

No more B_MESSAGE_NOT_UNDERSTOOD on B_UNMAPPED_KEY_DOWN, B_UNMAPPED_KEY_UP, B_MODIFIERS_CHANGED events

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12394 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added server support for Workspaces() and SetWorkspaces()

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12286 a95241bf-73f2-0310-859d-f6bbb57e9c96

Forgot a warning message

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12196 a95241bf-73f2-0310-859d-f6bbb57e9c96

Send only one AS_BEGIN/END_UPDATE message, not for every view

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12195 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove debug leftover.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12083 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implement the BWindow::task_loop again. I know that this is code duplication, but look at
the TODO item for an explanation. We could do this as a hack in Looper.cpp though.
Also implement the determine_target function (incomplete).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12082 a95241bf-73f2-0310-859d-f6bbb57e9c96

BWindow's frame events hooks work now. Frame gets updated on window move/resize.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12073 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix KeyDown/KeyUp calls so that we don't always get numBytes = 0.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@12062 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implemented BView::SetEventMask()

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@11945 a95241bf-73f2-0310-859d-f6bbb57e9c96

This fixes a problem with window's contents not being redrawn after a Show(), Hide and Show() again

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@11876 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed some weird includes (no need or even support to specify "posix" as part of the include path).


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11575 a95241bf-73f2-0310-859d-f6bbb57e9c96

mouse & keyboard messages now use the full power of application_kit


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11281 a95241bf-73f2-0310-859d-f6bbb57e9c96

BView::MouseUp/KeyDown/KeyUp() hook methods are called for the BView for which the action takes place


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11241 a95241bf-73f2-0310-859d-f6bbb57e9c96

BView::MouseDown() finally gets called


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11239 a95241bf-73f2-0310-859d-f6bbb57e9c96

Keyboard messages are dispached to BWindow's handler if no focus view


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@10931 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added some changes for the new mouse wheel support


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9813 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed a stupid mistake in the message to create a window


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8619 a95241bf-73f2-0310-859d-f6bbb57e9c96

Client-server messaging fixes


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8543 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix 2 instances of mismatched locking


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8532 a95241bf-73f2-0310-859d-f6bbb57e9c96

Pahtz's changes from PortLink/BSession/PortMessage/PortQueue to BPortLink


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8519 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added implementation of _set_menu_sem_() to Window.cpp (should it go to a different place ?), thus fixing the build (at least here, I hope I didn't forget anything else).
Thanks to Bill Hayden for reporting and sorry again for the trouble.
Changed a comment in PopUpMenu.cpp


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8394 a95241bf-73f2-0310-859d-f6bbb57e9c96

Implemented a method for the update process. Rearanged some code.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8321 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added missing private AddShortcut variant.
Minor style update.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8194 a95241bf-73f2-0310-859d-f6bbb57e9c96

chnaged a mesage code


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8062 a95241bf-73f2-0310-859d-f6bbb57e9c96

::MoveBy() fixed.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8054 a95241bf-73f2-0310-859d-f6bbb57e9c96

hidden state was wrongly reported


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8040 a95241bf-73f2-0310-859d-f6bbb57e9c96

Changed internal protocol. Ha ha. Just deleted a line. :-)))


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8034 a95241bf-73f2-0310-859d-f6bbb57e9c96

Major code cleanup


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8001 a95241bf-73f2-0310-859d-f6bbb57e9c96

modified Resize methods


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@7598 a95241bf-73f2-0310-859d-f6bbb57e9c96

modified the way top_view sends its data to app_server. It now send _all_ is data


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@7144 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Modified a little the protocol for AS_LAYER_CREATE_ROOT message. Replaced PortLink with BSession. Was
necessary!


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@7078 a95241bf-73f2-0310-859d-f6bbb57e9c96

added a member in a message


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@6116 a95241bf-73f2-0310-859d-f6bbb57e9c96

added a memeber to a server message


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@6107 a95241bf-73f2-0310-859d-f6bbb57e9c96

changed and if statement in AddToSubset


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@6067 a95241bf-73f2-0310-859d-f6bbb57e9c96

modified so use BSession again
various fixes especialy regarding app_server connection


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@6048 a95241bf-73f2-0310-859d-f6bbb57e9c96

Converted a lot of BSession use to PortLink where streamed messaging is not appropriate


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@5622 a95241bf-73f2-0310-859d-f6bbb57e9c96

Removed use of BSession in constructor - conversion to using a BAppServerLink.
BSession streams are not meant for this kind of use


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@5600 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fix line array bugs, missing breaks, gcc3 incompatibilities, and a few typos


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@5301 a95241bf-73f2-0310-859d-f6bbb57e9c96

*modified ReadRawFromPort() and ConvertToMessage() methods to use BSession class.
*modified/added B_VIEW_(MOVED/RESIZED) handlers in ConvertToMessage()/DispatchMessage()


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4802 a95241bf-73f2-0310-859d-f6bbb57e9c96

* made fLastViewToken member equal with top_view's token. This avoids an unnecessary message to be sent to app_server.
* added some debugging code?
* other changes...


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4685 a95241bf-73f2-0310-859d-f6bbb57e9c96

some appserver connection code


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4574 a95241bf-73f2-0310-859d-f6bbb57e9c96

changed to avoid some compiler warnings


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4488 a95241bf-73f2-0310-859d-f6bbb57e9c96

Modified to use the new BSession class


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4431 a95241bf-73f2-0310-859d-f6bbb57e9c96

WindowLookToInteger and WindowFeelToInteger return the values specified in R5's Window.h


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@4063 a95241bf-73f2-0310-859d-f6bbb57e9c96

Updates for changes in PortLink & PortMessage


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3958 a95241bf-73f2-0310-859d-f6bbb57e9c96

gcc 3 fixes


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3845 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed a crasher in setFocus() and some dead code in ResizeTo()


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3666 a95241bf-73f2-0310-859d-f6bbb57e9c96

Removed some bugs to allow the server to run. We can now display (but not play with) a real BWindow!


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3634 a95241bf-73f2-0310-859d-f6bbb57e9c96

Janitorial work and gcc3-related fixes


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3408 a95241bf-73f2-0310-859d-f6bbb57e9c96

Checkin for Adrian Oanca


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3069 a95241bf-73f2-0310-859d-f6bbb57e9c96

Removed a linker error caused by the restructuring of the BMessage sources


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3046 a95241bf-73f2-0310-859d-f6bbb57e9c96

A few bugfixes to allow for simple app_server testing


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@3002 a95241bf-73f2-0310-859d-f6bbb57e9c96

_get_object_token_ has found its permanent home in AppMisc.h, which the
other files were modified to include.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2957 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added BHandler tokens to app_server communications setup


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2947 a95241bf-73f2-0310-859d-f6bbb57e9c96

Check-in for Adrian Oanca - some serious code coming in!!


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2918 a95241bf-73f2-0310-859d-f6bbb57e9c96

More tweaks to talk to server betted in the constructor
Fleshed out a couple message-handling routines to better be able to debug server


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2905 a95241bf-73f2-0310-859d-f6bbb57e9c96

Window now talks nice-nice to app_server


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2903 a95241bf-73f2-0310-859d-f6bbb57e9c96

Modified to utilize member PortLink instead of creating new ones


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2891 a95241bf-73f2-0310-859d-f6bbb57e9c96

Updated sources to reflect changes in <ServerProtocol.h>


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2832 a95241bf-73f2-0310-859d-f6bbb57e9c96

#Include tweaks


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2078 a95241bf-73f2-0310-859d-f6bbb57e9c96

Adding initial BWindow implementation by John Hedditch


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@2076 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add BWindow::HasShortcut()

Reverse the meaning of BWindow fShowLevel to match BView.

This also matches the client_window_info.show_hide_level field used in Deskbar
and other applications.

While doing this, keep fShowLevel fully in sync between BWindow and app_server,
use one message type for both hiding and showing, and make the decision to show
and hide the window in the app_server.

Lastly make minimize behave as described in the Be Book: hidden windows cannot
be minimized, and minimized windows which get hidden become unminimized.

Sync BWindow fShowLevel with the app_server.

Implementing the window_info.show_hide_level in terms of this solves the
problem of minimized windows also being considered hidden, when really they are
just hidden in the app_server.

window_info.show_hide_level is still defined backwards with a comment making
that clear.

Also removed sending fShowLevel in the minimize request since it is now
maintained in the app_server.

Fixes #4127.

Compilation and 64-bit fixes to libbe.so sources.

Fixed the usual issues - printf format strings, uint32 instead of
addr_t, etc. One thing that isn't so nice is several places where
BList is used to store (u)int32, these require a double cast to addr_t
then void* to silence a warning on x86_64.

Keymap changes from recent activity. No AltGr yet.

Below is a mostly complete summary of the changes in this commit.

* Set the DeadKeys for the US-International Keymap to use the Option map.
* Rename American keymap to US
* Update the US, US-International, and United-Kingdom keymaps to take
out unneeded spaces in the option layer. Also updated the dead keys
and some other keys on the US-International keyboard to use UTF-8
characters rather than there ASCII equivalents when different.
* Make the Option key fall-through when there is no mapping in the Option
table. Option is for special characters, if none, print the regular one.
This is mostly meant for the US keymap which has an empty option map. But
also so that you don't have to repeat the normal, shift, and caps maps in
the option map needlessly. Although the keymaps are still not empty in
some cases that it could be like numpad keys and space.
* Update the /bin/keymap app to use fputs() instead of printf() when there
is no actual formatting taking place. I've gotten into trouble for doing
this before and it is faster to not process the string unnecessarily.
* Also several 80-char limit style fixes and updated comments.
* In Keymap class Reorder the modifier keys to match the keymap files.
Put B_CONTROL_KEY check above B_OPTION_KEY. Neither change has any effect,
they are purely aesthetic.
* Update DumpKeymap() method to use the abbreviated modifier letters so it
will fit in your 80-char wide terminal.
* Tiny style fix in InputServer
* 80-char limit style fix in BWindow and add a comment that the shortcut
gets eaten in the case of Cmd+Q
* Implement IndexForModifier() in KeyboardLayout, although I am not using it.
* Take Caps Lock out of the Modifier keys window because I couldn't get
it to work the way I wanted it to.
* Move key roles to the left column, and the key label on the left. Add column
header labels. Thanks Rimas!
* Add validation and improve marking menu options. Add a 'Disabled' option
to control, option, and command menus to disable the key. Make the key
role text grey if the key roles is disabled. Validation ensures that you
cannot repeat the same key twice in the Modifier keys window since that
won't work. You can't define 2 sets of option keys even if you really want
to. You can disable your control, option, and command keys if you
want, but that is not recommended.
* Rename kUpdateModifiers to kUpdateModifierKeys message to differetiate
it from kUpdateModifier.
* Add shift key to Modifier keys window, use the stop icon instead of the
warning icon to indicate conflicts.
* Allow the Layout system to control the size of the Modifier keys window
again, set the width's of the key role lables to the widest, set the width
of the menu fields to take up the rest of the space minus room for the
conflict views. I didn't like it that the Modifier keys window would change
size based on what options you had selected in the menu fields. Now it
doesn't, but, the layout system still makes it all fit.

Reconnect window and recreate all views.

Change instances of wether in comments to whether. No functional change intended.

Revert part of r43166 that was included accidentally. This resulted in
incorrect behavior while processing certain kinds of messages, and
consequently deadlocks in some apps. Fixes #8101.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@43237 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove extraneous code with respect to telling a menu to close which was
breaking various cases where menus coexisted in a window with a view which
was using event masks:
- If one invoked the menu via, e.g. a BMenuField, and then tried to choose
an item on it, the mouse down would get captured, the menu would be
closed, and the mouse event would be thrown away without ever reaching
the event mask view.

- Furthermore, since the menu was told to terminate early, it would decide
that the user hadn't actually chosen that item (the escape key case),
and never actually invoke it.

Fixes the menu fields in the mouse preflet being broken.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@43166 a95241bf-73f2-0310-859d-f6bbb57e9c96

Tweak the notification window again :
* Spacing of the bprogressbar is now 8pixels on each size
* Remove the useless window tab for now

Also fix DecoratorFrame() again as BORDERED_WINDOW didn't work with it either.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@43034 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fix DecoratorFrame() for kLeftTitledWindowLook windows
* Use it in notification window for better positionning.
Thanks augiedoggie for reporting the problem !


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42588 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch from X512 (ticket #7408): Don't notify input server of focus change if window is not active.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42574 a95241bf-73f2-0310-859d-f6bbb57e9c96

Patch from X512 (#7408): only send input method aware messages for active windows


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@42557 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Followed Ingo's suggestion, and added a BWindow::Layout() method.
* Changed ShowImage to use that function.
+alpha in case Ingo gives his okay :-)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@41782 a95241bf-73f2-0310-859d-f6bbb57e9c96

Renamed _CheckSizeLimits() to UpdateSizeLimits() and made it public.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@39495 a95241bf-73f2-0310-859d-f6bbb57e9c96

Big change deriving BLayout from BLayoutItem, and allowing viewless BLayouts.
a few highlights:
* BLayout now derives publicly from BLayoutItem
* Added BAbstractLayout class, which our layouts now derive from
* updated layout builders to avoid creating views when they don't need to
* updated layout classes
* updated AboutSystem to fix a little regression
* more details on #6407
* please tell me about any regressions, I've tried to find them all, but some
may have slipped by.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@38207 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Modified PtrScr key behaviour:
- No modifiers: take a screenshot with zero delay and launch the GUI
- Shift-PrtScr: Silent screenshot using the saved GUI settings
- Ctrl-PrtScr: Take a screenshot using the saved GUI settings and copy the
screenshot to the system clipboard
* A few locale related changes (I am not sure how this works when a class is
shared between two applications, I hope I got everything right)



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37252 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Rename "ScreenshotApp" to "Screenshot", rename its signature from
"application/x-vnd.haiku-screenshotapp" to "application/x-vnd.haiku-screenshot"
* Rename "Screenshot" to "screenshot", rename its signature from
"application/x-vnd.haiku-screenshot" to "application/x-vnd.haiku-screenshot-cli"
* Move screenshot from /system/apps to /bin


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@37059 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Made zooming to full screen center the window when the window cannot be made
full screen (due to size restrictions).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36714 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The window was now always taken to the new workspace with Ctrl-Alt-arrow;
lazy debugging left over that closes ticket #5675 again.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36637 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Unified the Desktop and ServerApp AS_ACTIVATE_WORKSPACE to work in the same
way.
* Use that newly exposed feature of taking the focus window to the new workspace
when using the Ctrl-Alt-Shift-Arrow shortcut.
* This fixes #5675.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36585 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add comment explaining why MenusBeginning() is invoked before handling key events
with the command key pressed.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35567 a95241bf-73f2-0310-859d-f6bbb57e9c96

When a Window is created for holding BViews attached to an offscreen bitmap,
always behave as if within a transaction. Otherwise all drawing calls to such
BViews wait for the server to finish the request. This change gives a tremendous
speed boost for these situations and is compatibly with BeOS, since there you
also had to call Sync() before drawing a bitmap that was painted with attached
views.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@35414 a95241bf-73f2-0310-859d-f6bbb57e9c96

Reverting r34335. Wrong location of my "B_NOT_MINIZABLE" check.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34350 a95241bf-73f2-0310-859d-f6bbb57e9c96

Make BWindow honor the B_NOT_MINIMIZABLE flag. Fixes ticket #4337.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34335 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added a BWindow::InViewTransaction() that determines whether or not the
window is currently updating its drawings as suggested by Stippi.
* Add this method to decide whether to call _FontChanged() in
BListView::SetFont().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34109 a95241bf-73f2-0310-859d-f6bbb57e9c96

* BWindow::SendBehind(NULL) will now send the window to back.
* Added shortcuts ctrl-alt-F to get a window to front, and ctrl-alt-b to move it
to the back.
* Updated the user guide with these shortcuts, and also explained the missing
"single click to bring window to front" behaviour.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34108 a95241bf-73f2-0310-859d-f6bbb57e9c96

Spelling

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@34081 a95241bf-73f2-0310-859d-f6bbb57e9c96

- this should fix gcc2 build, sorry ;)
- style fixes.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33691 a95241bf-73f2-0310-859d-f6bbb57e9c96

Extend the Ctrl-Alt-arrows shortcuts to also move the current window when shift is held, as on Gnome.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33690 a95241bf-73f2-0310-859d-f6bbb57e9c96

Include the terminating 0 byte when adding as string type. Otherwise someone
reading the message as an actual string will run into problems.
Should fix #4697.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33442 a95241bf-73f2-0310-859d-f6bbb57e9c96

Work based on a patch by Joshua R. Elsasser:
* Some key-sequences cause \0 chars in the "bytes" data which is supposed to
be passed to BView::KeyDown() and BView::KeyUp(). Therefor, one cannot use
string methods for adding/extracting the data to/from the events. For
example, Control-Space now works in the Terminal.

Thanks a lot for the original patch, Joshua!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33338 a95241bf-73f2-0310-859d-f6bbb57e9c96

* IsMinimized() now locks the window. This should fix bug #4274.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32902 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove the forced resizing to preferred size when CenterIn is called, and
instead just check if the size limits need to be set. The code that previously
did this when processing the B_LAYOUT_WINDOW message was extracted into a
private method, which is called from both places.

This may be slightly wasteful in some cases (since the size limits may be set
multiple times), but it is definitely needed because without it both of the
current test apps (DiskProbe and Screenshot) did not have properly centered
windows.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32622 a95241bf-73f2-0310-859d-f6bbb57e9c96

Remove useless BRect* version of CenterIn, make remaining version take a const
BRect reference.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32614 a95241bf-73f2-0310-859d-f6bbb57e9c96

Finally implemented BWindow::CenterOnScreen, with associated CenterIn(BRect)
methods as well as Size(). To avoid the problem of centering the window before
it has been resized by the layout system, I force the resizing early. If there
is a better way to do this or some way to avoid doing it repeatedly, let me
know. But I figure the Center* methods should not be called that often.

Updated Screenshot and DiskProbe to use this new method as a test. It certainly
cleaned up DiskProbe. I will update other code over the next few days (if
anyone wants to help, please do :)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32612 a95241bf-73f2-0310-859d-f6bbb57e9c96

removed resolved TODO comment


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32562 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Contrl-~ now switches between application windows. This is handier than the
mechanism used in BeOS, that is to press Control+Option-Tab (which didn't
work on Haiku, though).
* Did not change the window switch logic, though, so it's still not really
nice.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32524 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Our jump-to-next-group never worked, and since no one complained, option-tab
now takes over this role (leaving Alt-Tab) to the application.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32517 a95241bf-73f2-0310-859d-f6bbb57e9c96

Cleanup

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32233 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add TabFrame window property.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32228 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added a basic tool tip API, and implementation.
* The BView API can probably be regarded as good enough; the implementation
might need to be improved over time (also, some things as archivability
aren't fully implemented yet). The ToolTip.h header should get public once
finalized.
* Added new B_MOUSE_IDLE message that is sent to a BView after a certain
time has passed (BToolTipManager::ShowDelay()).
* Added small test app (ToolTipTest) that shows what is already working.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@32078 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Don't use B_ZOOM; this does not respect the B_NOT_ZOOMABLE flag, same problem
as with B_MINIMIZE. This fixes bug #4134.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31828 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Added AS_GET_WORKSPACE_LAYOUT to get the number of columns/rows of the
workspaces.
* Added shortcuts Ctrl-Alt-{Left|Right|Up|Down} to switch to adjacent
workspaces.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31824 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Don't use the B_MINIMIZE message for the minimize shortcut (ctrl-alt-m), since
that ignores the B_NOT_MINIMIZABLE flag. Instead, there is now a private
_MINIMIZE_ message (naming like those other ugly private BMessage constants in
AppDefs.h) which honours that flag.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31799 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Style cleanups.
* Removed former dead print-screen code.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31797 a95241bf-73f2-0310-859d-f6bbb57e9c96

Small style cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31155 a95241bf-73f2-0310-859d-f6bbb57e9c96

Reverted my change to usage of _get_object_token_(), it already means to use
a cached token... doh! Thanks Axel!


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31140 a95241bf-73f2-0310-859d-f6bbb57e9c96

Cache a view's server token. Avoids getting it for every
BView method that contacts the server.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31134 a95241bf-73f2-0310-859d-f6bbb57e9c96

* various fixes identified by gcc4, provided by Joe Prostko


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@30902 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added DecoratorFrame() method, which returns the outer frame of the window
on screen (ie including the decorator border and tab). Plus the necessary
refactoring as well as some TODO notes about windows with the tab on the left
side.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29820 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Reworked r29180 to make _HandleKeyDown() no longer rely on the active window
but if the message was targeted to the focus view. It will now only handle
shortcuts in this case.
* This also fixes bug #3513.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29776 a95241bf-73f2-0310-859d-f6bbb57e9c96

* it seems we can skip _HandleKeyDown() when not active (keyboard navigation, shortcuts and screenshots).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29180 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Handles shortcuts only when the window is active. This fixes bug #3414 (and hopefully is the R5 behavior).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29179 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Resolve further warnings on GCC4.
* Enable -Werror on GCC4 builds as well (limited to the same selected targets).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@29118 a95241bf-73f2-0310-859d-f6bbb57e9c96

Add missing headers (malloc/free, string functions, memcpy, etc. undeclared).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28982 a95241bf-73f2-0310-859d-f6bbb57e9c96

* The move/resize window protocol now uses absolute coordinates rather than
relative ones. This fixes bugs #2658, and #3213; in BWindow::ScreenChanged()
the window does not yet know that it moved on the new screen (when it already
had a position on that workspace).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28908 a95241bf-73f2-0310-859d-f6bbb57e9c96

At some point, there was some hassle about what coordinate system the "where"
BPoint in a mouse moved message is. Turns out it's different (!) according
to the type of message and this is even documented in the BeBook. I don't
really know if we want to copy this, but since there are apps out there that
depend on this, I am implementing it so that it's compatible with BeOS: In
the B_MOUSE_UP/DOWN case, it's the coordinate space of the target view, in
B_MOUSE_MOVED messages, it's that of the window. That explains a few things...
Clicking into Pe's function popup window should now select the right icon.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28742 a95241bf-73f2-0310-859d-f6bbb57e9c96

The previous commit exposed another bug: BWindow::Zoom() incorrectly calculated the max height of the window by only including the border width once. This was previously masked by the fact that the (by default 5) extra pixels were included in its hardcoded default for the height of the tab. Also add the border width to the message returned by GetDecoratorSettings() so Zoom can use the actual value instead of hardcoding it.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28679 a95241bf-73f2-0310-859d-f6bbb57e9c96

Resolved a small todo thanks to r28664:
We now ask the app_server for the window tab's height instead of
hardcoding it.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28678 a95241bf-73f2-0310-859d-f6bbb57e9c96

Minor style fix.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28672 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Replaced some "new" with new(std::nothrow) where appropriate in our base
classes (BView, BWindow, BAlert, BButton - BTextView should be part of this,
too, to make BAlerts work).
* However, it's not that simple, because there is often no way to return an
error. Most of that code obviously assumes to be able to throw exceptions
(it's just not communicated to the caller). Maybe we should just start
documenting exceptions for R1 (and properly use exceptions later on).
* Automatic white space cleanup.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28646 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed two more problems with BWindow::FindView(BPoint):
* The function is not supposed to return hidden views.
* After iterating over the child views, the "view" variable was clobbered,
so it didn't work to return the current view if none of it's child views
were hit.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28494 a95241bf-73f2-0310-859d-f6bbb57e9c96

Fixed numerous problems in BWindow::_FindView:

1) If a view contained the point, but had children, that view would never be returned as a result, even if none of the children matched.
2) When converting coordinates to the child's coordinate space, it was using the bounds rectangle rather than the frame. This in most cases had the result that the coordinate was unchanged, and thus messed up the search completely.

This fixes ticket #3000.



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28440 a95241bf-73f2-0310-859d-f6bbb57e9c96

* reintroduce member fOffscreen which got removed in r2917
* check if locking the offscreen window succeeds before calling delete on it
* don't return offscreen windows in CountWindows and WindowAt (works now as on R5)

fixes ticket 1522, 1591, 1946, 2318 and propably more

While creating an BBitmap in BApplication the bitmaps window looper would
be added to the applications gLooperList, thus calling Quit() on that window
and later delete on a stale window pointer in BBitmaps dtor. The Lock() check
would fix the problem, but tests on R5 have shown that BApplication hides the
offscreen window in CountWindows() and WindowAt().



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28236 a95241bf-73f2-0310-859d-f6bbb57e9c96

bonefish + zooey:
* Fixed a general problem with respect to overriding of the reserved
virtual function slots: instead of statically invoking the method
that corresponds to the reserved slot on the class that contains the
slot, we now invoke the virtual Perform() method. Perform() then dispatches
the method invocation to the "proper" class, i.e. the highest class in the
hierarchy that actually implements the requested method.
This fixes a crash in apps that use liblayout's MSlider class and
should fix one or other spurious bug with old apps or libraries, too.
* added new header folder 'binary_compatibility' that contains files that
define the method codes and data structures required by Perform()
* looked for and implemented all used reserved virtual slot functions to
invoke Perform() where necessary or to pass on the method call statically
(for slots that were already maintained by Be)


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28124 a95241bf-73f2-0310-859d-f6bbb57e9c96

zooey + axeld + stippi:
Fixed a problem with dispatching B_MOUSE_MOVED messages to views which need
the transit event:
* If the app_server decided to send a B_MOUSE_MOVED to the previous mouse
containing window, it forgot to add the _feed_focus flag, which was a problem
if the message actually happened to contain tokens (views that registered
for events.) On the client side, only those views would receive the message,
while the regular last mouse moved view would not be notified at all.
* On the client side, never change fLastMouseMovedView if the message is not
targeted at the preferred handler. In the above situation, any "registered
for events views" would receive the message first, but viewUnderMouse would
be NULL and then when the regular last mouse moved view received it, it
resulted in the wrong transit (B_OUTSIDE_VIEW instead of B_EXITED_VIEW).


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28059 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Refactored a method for getting the transit from a mouse moved message.
* In _StealMouseMessage(), don't maintain fLastMouseMovedView, instead,
prevent B_MOUSE_MOVED message from being stolen that are important for
detecting transit changes. The point is that some apps (like Tracker) are
shooting themselves in the foot because they steal mouse messages via
GetMouse() in one place, but then rely on sane transit values in another
place. The way it works now, the view in question may get notified of the
same mouse moved coordinate twice, once via GetMouse() and once via
MouseMoved().


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@28002 a95241bf-73f2-0310-859d-f6bbb57e9c96

* use the screenshot app to take a screenshot
* remove libpng and libz dependency in case we build for Haiku



git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27801 a95241bf-73f2-0310-859d-f6bbb57e9c96

* Fixed a race condition in BWindow::Show(). If it was the first time Show()
was called, the calling thread could be preempted, or simply be blocked on
calling Lock() after running the window thread. The window thread in turn
could be processing messages. In that case, fShowLevel would still have the
wrong value. For example, MediaPlayer would call IsHidden() on one of it's
views, which would then return true for this case. The result was that the
video view was not hidden and a black rectangle was showing on top of the
controls. This may also have caused other similar problems of course.
* fRunCalled was accessed without holding the lock.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@27209 a95241bf-73f2-0310-859d-f6bbb57e9c96