]> Dogcows Code - chaz/openbox/log
chaz/openbox
17 years agooops, logic error
Dana Jansens [Sat, 17 Mar 2007 20:06:16 +0000 (20:06 +0000)] 
oops, logic error

17 years agomention when the logicalx/y is actually returned
Dana Jansens [Fri, 16 Mar 2007 19:16:07 +0000 (19:16 +0000)] 
mention when the logicalx/y is actually returned

17 years agofix for bug introduced in revision 5643. logical size is supposed to be updated only...
Dana Jansens [Fri, 16 Mar 2007 19:14:53 +0000 (19:14 +0000)] 
fix for bug introduced in revision 5643. logical size is supposed to be updated only when the window is resized.

17 years agoanimate presses on toggled titlebar buttons
Dana Jansens [Fri, 16 Mar 2007 01:33:47 +0000 (01:33 +0000)] 
animate presses on toggled titlebar buttons

17 years ago(no commit message)
Dana Jansens [Thu, 15 Mar 2007 17:46:39 +0000 (17:46 +0000)] 

17 years agoupdate COMPLIANCE
Dana Jansens [Thu, 15 Mar 2007 17:37:30 +0000 (17:37 +0000)] 
update COMPLIANCE

17 years agoonly refocus the omnipresent window if its allowed
Dana Jansens [Thu, 15 Mar 2007 01:11:52 +0000 (01:11 +0000)] 
only refocus the omnipresent window if its allowed

17 years agoclearer engrish
Dana Jansens [Thu, 15 Mar 2007 01:09:10 +0000 (01:09 +0000)] 
clearer engrish

17 years agowhen an omnipresent window has focus and you switch desktops, keep it focused
Dana Jansens [Thu, 15 Mar 2007 01:02:48 +0000 (01:02 +0000)] 
when an omnipresent window has focus and you switch desktops, keep it focused

17 years agodon't fall back focus to omnipresent windows
Dana Jansens [Wed, 14 Mar 2007 21:15:58 +0000 (21:15 +0000)] 
don't fall back focus to omnipresent windows

17 years agomake focus_order into one long list instead of having one per desktop. this actually...
Dana Jansens [Wed, 14 Mar 2007 21:07:30 +0000 (21:07 +0000)] 
make focus_order into one long list instead of having one per desktop. this actually fixes bugs with omnipresent windows and adding desktops. it is just overall a little nicer i think.

also this is not tested. sorry if it breaks everything.

17 years agosyscrash changed his theme up
Dana Jansens [Wed, 14 Mar 2007 20:59:16 +0000 (20:59 +0000)] 
syscrash changed his theme up

17 years agosave the focus_client when falling back, so we don't actually refocus it when we...
Dana Jansens [Wed, 14 Mar 2007 15:56:59 +0000 (15:56 +0000)] 
save the focus_client when falling back, so we don't actually refocus it when we don't want to.

17 years agooh, 2 things in this commit..
Dana Jansens [Wed, 14 Mar 2007 15:51:21 +0000 (15:51 +0000)] 
oh, 2 things in this commit..

1) THIS ONE IS IMPORTANT: don't set client->iconfied if the window is not actually going to be allowed to iconify. heh!

2) changes to focus fallback to avoid crashing and to avoid losing focus all at the same time.

17 years agomove stuff around in client_unmanage to maybe make closing windows a bit nicer
Mikael Magnusson [Wed, 14 Mar 2007 14:17:31 +0000 (14:17 +0000)] 
move stuff around in client_unmanage to maybe make closing windows a bit nicer

17 years agorefix for managing iconic windows, without having the frame map which caused flashing.
Dana Jansens [Wed, 14 Mar 2007 13:58:36 +0000 (13:58 +0000)] 
refix for managing iconic windows, without having the frame map which caused flashing.

instead, keep track of if the frame has been mapped yet. if it hasn't and it is getting hidden, then make sure to unmap the child.

17 years agoa) fix resizerelative moving windows when they reach their minimum size do this by...
Dana Jansens [Wed, 14 Mar 2007 03:45:56 +0000 (03:45 +0000)] 
a) fix resizerelative moving windows when they reach their minimum size do this by...
b) adding a function to let you see what size/position a window will be given when you actually try move/resize it to some given values. (client_try_configure)

17 years agoadd support for 8bit TrueColor visuals. Who knew these existed. Fixes vnc's 8bit...
Dana Jansens [Wed, 14 Mar 2007 03:05:47 +0000 (03:05 +0000)] 
add support for 8bit TrueColor visuals. Who knew these existed. Fixes vnc's 8bit mode. (bug 2878)

17 years agoupdate german translation
Mikael Magnusson [Wed, 14 Mar 2007 02:27:46 +0000 (02:27 +0000)] 
update german translation

17 years agomake frame and client window in a consistant mapped state when the window is grabbed
Dana Jansens [Wed, 14 Mar 2007 02:23:50 +0000 (02:23 +0000)] 
make frame and client window in a consistant mapped state when the window is grabbed

17 years agoless wasted updates of wm_state
Dana Jansens [Wed, 14 Mar 2007 02:07:30 +0000 (02:07 +0000)] 
less wasted updates of wm_state

17 years agofocus fallback
Mikael Magnusson [Wed, 14 Mar 2007 01:17:39 +0000 (01:17 +0000)] 
focus fallback

17 years agocomparing strings to random locations in memory is probably not what we wanted to do
Mikael Magnusson [Tue, 13 Mar 2007 12:34:58 +0000 (12:34 +0000)] 
comparing strings to random locations in memory is probably not what we wanted to do

17 years agoalways unfocus the window when it disappears, no special cases.. (except unmanage)
Dana Jansens [Tue, 13 Mar 2007 04:15:25 +0000 (04:15 +0000)] 
always unfocus the window when it disappears, no special cases.. (except unmanage)

17 years agoset IconicState on windows that are not visible always. this means windows on other...
Dana Jansens [Tue, 13 Mar 2007 04:04:13 +0000 (04:04 +0000)] 
set IconicState on windows that are not visible always. this means windows on other desktops too.

as well, this turned out to be a nice cleanup of the wmstate code.

regards to the random dude in #openbox complaining about the ICCCM to make me think to look this up and see if openbox was compliant.

17 years agosave and restore pre-maximized/fullscreen sizes properly
Dana Jansens [Tue, 13 Mar 2007 02:29:08 +0000 (02:29 +0000)] 
save and restore pre-maximized/fullscreen sizes properly

much cleanup on how per-app settings are applied to new mapping windows. maybe i broke it? but it'll be much quicker now, and it's consistant with the rest of the code, so hooray for that. pls let me know if i broke any of it :( hee..

17 years agoremove event_lastttime. use CurrentTime when handling events without a timestamp...
Dana Jansens [Mon, 12 Mar 2007 23:23:39 +0000 (23:23 +0000)] 
remove event_lastttime. use CurrentTime when handling events without a timestamp in them

17 years ago?
Dana Jansens [Mon, 12 Mar 2007 23:22:13 +0000 (23:22 +0000)] 
?

17 years agofallback is still getting x errors. instead, fallback when the client iconifies,...
Dana Jansens [Mon, 12 Mar 2007 17:31:11 +0000 (17:31 +0000)] 
fallback is still getting x errors. instead, fallback when the client iconifies, and don't fallback when nothing has focus, send it to where we can control things though.

17 years agofocusins can occur on windows that aren't clients when there's latency involved....
Dana Jansens [Mon, 12 Mar 2007 06:09:02 +0000 (06:09 +0000)] 
focusins can occur on windows that aren't clients when there's latency involved. wm_torture proved this point. fallback in that situation. also fallback when nothing is focused.

17 years agodon't validate inside client_focus. instead, validate before you call it!
Dana Jansens [Mon, 12 Mar 2007 06:04:00 +0000 (06:04 +0000)] 
don't validate inside client_focus. instead, validate before you call it!

17 years agovalidate clients before trying to focus them in fallback
Dana Jansens [Mon, 12 Mar 2007 05:57:30 +0000 (05:57 +0000)] 
validate clients before trying to focus them in fallback

also validate before trying to set the focus on a window

this should fix those crashes, and should just be better over all

17 years agoupdate focus_hilite appropriately for when changing desktops
Dana Jansens [Mon, 12 Mar 2007 05:44:16 +0000 (05:44 +0000)] 
update focus_hilite appropriately for when changing desktops

17 years agoi rewrote handling of focus events. this is pretty much based on blackbox's current...
Dana Jansens [Mon, 12 Mar 2007 05:25:34 +0000 (05:25 +0000)] 
i rewrote handling of focus events. this is pretty much based on blackbox's current form, as well as reading the xlib programming manual at:
http://tronche.com/gui/x/xlib/events/input-focus/normal-and-grabbed.html

this may break for people. that'd be nice to hear about, so it can be fixed. but hopefully this is more robust. it sure is a lot more simple.

17 years agoanother silly bug. i rule the school.
Dana Jansens [Mon, 12 Mar 2007 03:09:09 +0000 (03:09 +0000)] 
another silly bug. i rule the school.

17 years agothat was a silly mistake. fullscreen windows amongst others will not crash now :D
Dana Jansens [Mon, 12 Mar 2007 03:07:00 +0000 (03:07 +0000)] 
that was a silly mistake. fullscreen windows amongst others will not crash now :D

17 years agoroll back focus fallback change, so it behaves well with new stacking
Dana Jansens [Mon, 12 Mar 2007 02:41:34 +0000 (02:41 +0000)] 
roll back focus fallback change, so it behaves well with new stacking

17 years agook this is from prevous changes but has a bunch of debug stuff in it
Dana Jansens [Mon, 12 Mar 2007 02:34:34 +0000 (02:34 +0000)] 
ok this is from prevous changes but has a bunch of debug stuff in it

17 years agoremove the group option for the restacking commands in rc.xml, cuz groups are clever now
Dana Jansens [Mon, 12 Mar 2007 02:27:58 +0000 (02:27 +0000)] 
remove the group option for the restacking commands in rc.xml, cuz groups are clever now

17 years agoRewrite the stacking code. It's a lot faster now, I should think. It's def a more...
Dana Jansens [Mon, 12 Mar 2007 02:24:40 +0000 (02:24 +0000)] 
Rewrite the stacking code. It's a lot faster now, I should think. It's def a more clever algorithm. It deals with group transients much better.

On that note, utility and menu and toolbar window types are now treated as group transients in terms of stacking and focus and such.

17 years agono separator at the start of the menu
Mikael Magnusson [Mon, 12 Mar 2007 02:12:51 +0000 (02:12 +0000)] 
no separator at the start of the menu

17 years agomake the combined client list menu not suck quite as bad
Mikael Magnusson [Mon, 12 Mar 2007 01:59:49 +0000 (01:59 +0000)] 
make the combined client list menu not suck quite as bad

17 years agoupdate copyright notice in --version output
Mikael Magnusson [Mon, 12 Mar 2007 01:12:31 +0000 (01:12 +0000)] 
update copyright notice in --version output

17 years agoWe were grabbing SubstructureNotifyMask on the root window. I don't know why.. It...
Dana Jansens [Sun, 11 Mar 2007 22:16:42 +0000 (22:16 +0000)] 
We were grabbing SubstructureNotifyMask on the root window. I don't know why.. It meant that we got unmap notifies for windows we weren't even managing and stuff. It ended up breaking VLC when it mapped-unmapped-mapped a window. We grab StructureNotify for all the clients, so I am not sure why this was ever there.

17 years agoextra braces.. why does the world hate me so?
Dana Jansens [Sun, 11 Mar 2007 19:27:44 +0000 (19:27 +0000)] 
extra braces.. why does the world hate me so?

17 years ago.. i am pretending to be mika :D :D ok this time for real
Dana Jansens [Sun, 11 Mar 2007 19:27:16 +0000 (19:27 +0000)] 
.. i am pretending to be mika :D :D ok this time for real

17 years agooh.. wasted verbosity.. *strums guitar*
Dana Jansens [Sun, 11 Mar 2007 19:26:28 +0000 (19:26 +0000)] 
oh.. wasted verbosity.. *strums guitar*

17 years agorewrite the focus steal prevention code to clean it up hopefully a bit. also don...
Dana Jansens [Sun, 11 Mar 2007 19:25:57 +0000 (19:25 +0000)] 
rewrite the focus steal prevention code to clean it up hopefully a bit. also don't steal focus from globally active focus model clients

17 years agoHonor transient_for=Root, even thought this is crap and breaks some KDE stuff. Also...
Dana Jansens [Sun, 11 Mar 2007 19:15:28 +0000 (19:15 +0000)] 
Honor transient_for=Root, even thought this is crap and breaks some KDE stuff. Also include a big comment as to why this is suck.

17 years agoset micro_version correctly in configure.ac
Mikael Magnusson [Sun, 11 Mar 2007 18:42:16 +0000 (18:42 +0000)] 
set micro_version correctly in configure.ac

17 years agomake smart placement more smart with multiple monitors. hopefully?
Dana Jansens [Sun, 11 Mar 2007 18:39:34 +0000 (18:39 +0000)] 
make smart placement more smart with multiple monitors. hopefully?

17 years agochange version to 3.4, i think that will be next release number? and reset the librar...
Dana Jansens [Sun, 11 Mar 2007 17:36:27 +0000 (17:36 +0000)] 
change version to 3.4, i think that will be next release number? and reset the libraries binary ages stuff, cuz they aren't backwards compatible with the last release.

17 years agoclient_configure_full is trying to move a window, it shouldn't determine the window...
Dana Jansens [Sun, 11 Mar 2007 17:34:07 +0000 (17:34 +0000)] 
client_configure_full is trying to move a window, it shouldn't determine the window's monitor from its current position, but from the new position instead. move client_monitor()'s logic into screen, more generically, and let client_configure_full use that

17 years agofallback to transient parents properly
Dana Jansens [Sun, 11 Mar 2007 17:05:07 +0000 (17:05 +0000)] 
fallback to transient parents properly

17 years agostacking has been made more reliable with groups and group transients.
Dana Jansens [Sun, 11 Mar 2007 16:58:19 +0000 (16:58 +0000)] 
stacking has been made more reliable with groups and group transients.

this was a pretty invasive change in client.c though, so it may break things?

it did expose some bugginess in client_calc_layer, which is now better than ever, hopefully there isn't more to be found.

17 years agonote things in CHANGELOG and mark xml theme entry as maybe
Mikael Magnusson [Sun, 11 Mar 2007 15:52:52 +0000 (15:52 +0000)] 
note things in CHANGELOG and mark xml theme entry as maybe

17 years agoallow setting startupnotify per execute action, default to disabled. use like this...
Mikael Magnusson [Sun, 11 Mar 2007 15:41:47 +0000 (15:41 +0000)] 
allow setting startupnotify per execute action, default to disabled. use like this: <action name="execute"><execute>program</execute><startupnotify><enabled>true</enabled><name>something descriptive</name><icon>name of an icon in the icon theme to show</icon></startupnotify</action>

17 years agolook for more things to cancel startup notification on, from the wmclass. if the...
Dana Jansens [Sun, 11 Mar 2007 14:31:11 +0000 (14:31 +0000)] 
look for more things to cancel startup notification on, from the wmclass. if the wmclass is related to the name or the binary name, then it will cancel it also. this is so that oldschool stuff doesn't thing its opening forever.

17 years agomake _NET_ACTIVE_WINDOW requests look at the timestamp when the user didn't
Dana Jansens [Sun, 11 Mar 2007 14:17:14 +0000 (14:17 +0000)] 
make _NET_ACTIVE_WINDOW requests look at the timestamp when the user didn't
request it

17 years agofix non instrusive stacking adds, they didnt account for layers at all
Dana Jansens [Sun, 11 Mar 2007 14:16:24 +0000 (14:16 +0000)] 
fix non instrusive stacking adds, they didnt account for layers at all

17 years agoadd _NET_WM_FULL_PLACEMENT, cuz openbox does do placement, even tho it can be sucky...
Dana Jansens [Sun, 11 Mar 2007 05:00:56 +0000 (05:00 +0000)] 
add _NET_WM_FULL_PLACEMENT, cuz openbox does do placement, even tho it can be sucky right now, lol. but it's good for dialogs and stuff, which is what this is talking about.

also add _NET_WM_MOVERESIZE_CANCEL.

17 years agowow... this is a big commit...
Dana Jansens [Sun, 11 Mar 2007 04:44:15 +0000 (04:44 +0000)] 
wow... this is a big commit...

all related to _NET_WM_USER_TIME and focus stealing prevention

a) add launcher startup notification. this means when you run something from
   the openbox menu or a key/mouse binding, that startup notification will go
   on in openbox and other applications like your panel or something
b) add the _NET_WM_USER_TIME property for windows
c) use the _NET_WM_USER_TIME data and startup notification to prevent focus
   stealing.
d) cookie party !! ! all are invited.
e) oh yeah, and pass around timestamps for a lot more things. like, when you
   run an action, send the timestamp for the event that is running the action.
   this is important for startup notification. this also affects menus.
f) yes.. cookies..

would it be a good idea to disable focus stealing prevention if a window takes
too long to load? i mean.. maybe after a certain length of time, a user can't be
expected to not do anything in any other windows, but would they still want the
new application to focus then? HMM. open question i guess..

17 years agoadd the _NET_WM_USER_TIME property support. When focus_new is enabled, don't focus...
Dana Jansens [Sun, 11 Mar 2007 02:06:34 +0000 (02:06 +0000)] 
add the _NET_WM_USER_TIME property support. When focus_new is enabled, don't focus new windows if the user is doing something in another window since it launched. If we can tell when it launched, either from the _NET_WM_USER_TIME or from startup notification.

17 years agoremove the kde icon hint, they use the EWMH icon now
Dana Jansens [Sun, 11 Mar 2007 00:59:13 +0000 (00:59 +0000)] 
remove the kde icon hint, they use the EWMH icon now

17 years agoremove the kde frame extents property, add the _NET_FRAME_EXTENTS property in its...
Dana Jansens [Sun, 11 Mar 2007 00:12:37 +0000 (00:12 +0000)] 
remove the kde frame extents property, add the _NET_FRAME_EXTENTS property in its stead

17 years ago>:| I put old code back somehow.. thanks mika. hooray peer review :)
Dana Jansens [Sun, 11 Mar 2007 00:00:05 +0000 (00:00 +0000)] 
>:| I put old code back somehow.. thanks mika. hooray peer review :)

17 years ago1) Remove support for the Urgent hint. This will no longer do anything within Openbox
Dana Jansens [Sat, 10 Mar 2007 23:57:33 +0000 (23:57 +0000)] 
1) Remove support for the Urgent hint. This will no longer do anything within Openbox
2) Add support for _NET_WM_STATE_DEMANDS_ATTENTION. Yay!

17 years agoskip windows in alt-tab that have SKIP_PAGER set. perhaps this is a better metric...
Dana Jansens [Sat, 10 Mar 2007 23:09:27 +0000 (23:09 +0000)] 
skip windows in alt-tab that have SKIP_PAGER set. perhaps this is a better metric.. also this is for mikachu <3

17 years agopeople have started using SKIP_TASKBAR, and it's not for stuff you don't want to...
Dana Jansens [Sat, 10 Mar 2007 23:03:08 +0000 (23:03 +0000)] 
people have started using SKIP_TASKBAR, and it's not for stuff you don't want to be able to alt-tab to. so don't put that in here anymore.

17 years agomake startup notification a little more robust. obconf is just broken, between
Dana Jansens [Sat, 10 Mar 2007 22:52:50 +0000 (22:52 +0000)] 
make startup notification a little more robust. obconf is just broken, between
the application and it's .desktop.

17 years agolet windows move themselves off the screen somewhat, but don't let them resize themse...
Dana Jansens [Sat, 10 Mar 2007 22:24:47 +0000 (22:24 +0000)] 
let windows move themselves off the screen somewhat, but don't let them resize themselves off the screen (unless they're moving too)

17 years agono longer honor transient_for = None or Root. it violates the ICCCM and is causing...
Dana Jansens [Sat, 10 Mar 2007 20:52:04 +0000 (20:52 +0000)] 
no longer honor transient_for = None or Root. it violates the ICCCM and is causing problems. there are proper ways to make group transient windows with the EWMH.

17 years agoadd grouptran test, which will actually not make valid transient windows
Dana Jansens [Sat, 10 Mar 2007 20:50:49 +0000 (20:50 +0000)] 
add grouptran test, which will actually not make valid transient windows

add missing include to title.c

17 years agoUpdate with all important changes
Mikael Magnusson [Sat, 10 Mar 2007 10:26:50 +0000 (10:26 +0000)] 
Update with all important changes

17 years agosome more stuff
Mikael Magnusson [Sat, 10 Mar 2007 10:22:30 +0000 (10:22 +0000)] 
some more stuff

17 years agoat least the CHANGELOG file gets some updates
Mikael Magnusson [Sat, 10 Mar 2007 10:21:17 +0000 (10:21 +0000)] 
at least the CHANGELOG file gets some updates

17 years agoyet another pointless commit to test post-commit hook
Mikael Magnusson [Sat, 10 Mar 2007 10:19:15 +0000 (10:19 +0000)] 
yet another pointless commit to test post-commit hook

17 years agoanother pointless commit to test post-commit hook
Mikael Magnusson [Sat, 10 Mar 2007 10:16:32 +0000 (10:16 +0000)] 
another pointless commit to test post-commit hook

17 years agopointless commit to test post-commit hook
Mikael Magnusson [Sat, 10 Mar 2007 10:10:20 +0000 (10:10 +0000)] 
pointless commit to test post-commit hook

17 years agomore cleverness for window placing.
Dana Jansens [Sat, 10 Mar 2007 01:08:19 +0000 (01:08 +0000)] 
more cleverness for window placing.

make sure its 10% on the screen in some direction always..

but if the application is placing itself, make sure its on the screen entirely,
and also put it entirely on one monitor if you have xinerama.

if it's bigger than the monitor's space though, it won't do anything with it..

17 years agolet you put windows with the titlebar off screen.. for mikachu.
Dana Jansens [Sat, 10 Mar 2007 00:44:43 +0000 (00:44 +0000)] 
let you put windows with the titlebar off screen.. for mikachu.

but also don't let applications put themselves under the struts!

17 years agoone more fix.. oops :)
Dana Jansens [Thu, 8 Mar 2007 03:32:29 +0000 (03:32 +0000)] 
one more fix.. oops :)

17 years agofix rendering of the new top corner resize handles
Dana Jansens [Thu, 8 Mar 2007 03:31:42 +0000 (03:31 +0000)] 
fix rendering of the new top corner resize handles

17 years agoa) remove the top_grip_height business from the theme
Dana Jansens [Thu, 8 Mar 2007 01:26:03 +0000 (01:26 +0000)] 
a) remove the top_grip_height business from the theme
b) make the focus indicator based on border width and padding, not the handle height
c) add 2 more windows for the top corners for resizing, much easier to find them now. also base their width on padding, not the handle height or other weird things

17 years agowas using paddingx for y-things, oops
Dana Jansens [Thu, 8 Mar 2007 01:08:44 +0000 (01:08 +0000)] 
was using paddingx for y-things, oops

17 years agoFix possible crashingness, from last commit.
Dana Jansens [Thu, 8 Mar 2007 01:03:13 +0000 (01:03 +0000)] 
Fix possible crashingness, from last commit.

17 years agoSplit RrPaint to RrPaint and RrPaintPixmap, so you can paint things other than window...
Dana Jansens [Wed, 7 Mar 2007 22:57:10 +0000 (22:57 +0000)] 
Split RrPaint to RrPaint and RrPaintPixmap, so you can paint things other than windows. If you don't free the returned pixmap though, you will be leaking memory in the server!

17 years agoBe more lenient with errors in xml file. If there are errors, try to keep loading...
Dana Jansens [Wed, 7 Mar 2007 22:56:37 +0000 (22:56 +0000)] 
Be more lenient with errors in xml file. If there are errors, try to keep loading it anyways.

17 years agodon't install the old themeupdate.py file
Dana Jansens [Tue, 6 Mar 2007 02:58:24 +0000 (02:58 +0000)] 
don't install the old themeupdate.py file

17 years agoupdate my theme
Mikael Magnusson [Mon, 5 Mar 2007 16:55:21 +0000 (16:55 +0000)] 
update my theme

17 years agoput artists original comments back into the themes after regenerating them
Dana Jansens [Mon, 5 Mar 2007 16:48:49 +0000 (16:48 +0000)] 
put artists original comments back into the themes after regenerating them

17 years agomissed updating this one
Dana Jansens [Mon, 5 Mar 2007 16:46:51 +0000 (16:46 +0000)] 
missed updating this one

17 years agoproper shadow tint values
Dana Jansens [Mon, 5 Mar 2007 16:33:22 +0000 (16:33 +0000)] 
proper shadow tint values

17 years agoscale the shadows up, they were /100 before, now they are /255
Dana Jansens [Mon, 5 Mar 2007 16:32:50 +0000 (16:32 +0000)] 
scale the shadows up, they were /100 before, now they are /255

17 years agotheme converter had a bug! menu item font is parsed now.
Dana Jansens [Mon, 5 Mar 2007 16:28:04 +0000 (16:28 +0000)] 
theme converter had a bug! menu item font is parsed now.

17 years agowas looking for menu items font in the wrong place
Dana Jansens [Mon, 5 Mar 2007 16:27:02 +0000 (16:27 +0000)] 
was looking for menu items font in the wrong place

17 years agofollow new xsd
Dana Jansens [Mon, 5 Mar 2007 16:24:35 +0000 (16:24 +0000)] 
follow new xsd

17 years agopoint and padding became redundant.. do remove padding type from xsd and use point...
Dana Jansens [Mon, 5 Mar 2007 16:23:13 +0000 (16:23 +0000)] 
point and padding became redundant.. do remove padding type from xsd and use point type. also fix bug in theme parser where it was looking for wrong one.

17 years agodon't use handle_height for focus indicators, use top_grip_height. is this better?
Dana Jansens [Mon, 5 Mar 2007 16:10:32 +0000 (16:10 +0000)] 
don't use handle_height for focus indicators, use top_grip_height. is this better?

17 years agoscary commit..but here goes.
Dana Jansens [Mon, 5 Mar 2007 15:44:17 +0000 (15:44 +0000)] 
scary commit..but here goes.

YOUR THEMES ARE NOW OFFICIALLY BROKEN.

Openbox has just moved it's theme format to an XML based one. The details of
this format can be found in data/themerc.xsd (and http://openbox.org/themerc.xsd

ALSO! This is very good and important and stuff! In the tools directory you
will find THEMETOXML ! This tool takes a themerc on stdin, and spits out
the same theme in theme.xml format. So this is all you need to do to update
your themes.

PLEASE NOTE: This themetoxml does _not_ install itself anywhere. It simply
builds and then lives out in its tools/themetoxml directory, and that's it. So
if you want to use it, that is where to find it.

In moving to the new XML format, a number of additions/changes to the theme
engine have been made. Themetoxml takes these into account and will set all
the new things appropriately to make your theme look the same as it always has.

New additions include..
* padding now has an horizontal and vertical component, instead of being one number
* menus can have different borders than windows (color and size)
* menu offset can now be negative. it's a little weird, but someone will want it no doubt
* fonts are no longer controled by the theme at all, however font shadowing is, and on that note..
* font shadows are now any color you want, not just black and white
* you can now set the shadow anywhere you can set the text's color, so you have more control, i.e. you can set shadow on active menu items but not inactive, or disabled, etc.
* every color now has an alpha channel. at the moment they don't do anything, besides the font shadow one, but it leaves room for future explorations. it is REALLY HIGHLY RECOMMENDED that you set the alpha to 255 all the time, until such time as it could be useful. otherwise one day your theme may turn awful for people.
* font colors are in the range 0-255, in case you were wondering, and they have to be specified in decimal
* if you'd like to change you font's you can do so in your configuration file. this is how it is going to stay. changing the font in the theme assumes too much about peoples eye sight and locality and stuff. it doesn't belong there, sorry. the system-wide default rc.xml includes the new font settings for your viewing pleasure, and ill drop an example of it below.
* shadows can now be positioned in any direction, they have both an x and a y offset which can be negative and positive. and offset of 0,0 will disable the shadow

This isn't a release or anything. If someone had some good ideas about the xml theme format, I'd like to hear them. But I don't think it will be changing much right now beyond where it is. I don't even know how the new functionality will play out for themers, so we'll see.

Whew.. I guess that's it. I'm not sure if I mentioned every little change or not, but oh well. Mileage may vary.. Please send any feedback.

Here's the font configuration example. Hopefully ObConf will let you set this real soon.

<theme>
...
  <font place="ActiveWindow">
    <name>arial,sans</name>
    <size>7</size>
    <weight>bold</weight>
    <slant>normal</slant>
  </font>
  <font place="InactiveWindow">
    <name>arial,sans</name>
    <size>7</size>
    <weight>bold</weight>
    <slant>normal</slant>
  </font>
  <font place="MenuTitle">
    <name>arial,sans</name>
    <size>8</size>
    <weight>bold</weight>
    <slant>normal</slant>
  </font>
  <font place="MenuItem">
    <name>arial,sans</name>
    <size>8</size>
    <weight>bold</weight>
    <slant>normal</slant>
  </font>
</theme>

This page took 0.054981 seconds and 4 git commands to generate.