I found the bug which may occurs enlightenment crash.
There is code to insert E_Gadcon pointer to custom_populate_requests list in e_gadcon_custom_new(). However, there is no code to remove deleted E_Gadcon pointer from custom_populate_requests list in e_gadcon_custom_del(). So, if the e_gadcon_custom_del() is called before calling _e_gadcon_custom_populate_idler(), then enlightenment crash may occurs while e calls _e_gadcon_custom_populate_idler() because e accesses deleted E_Gadcon pointer.
I modified this problem. Please review this patch.
Thanks.
--
Doyoun Kang
SVN revision: 75033
not zone->id. zone->id is for matching u to zinerama/xrandr
screens/outputs. e re-sorts based on highest res screens first, lower
res later.
SVN revision: 67951
the 'need' variable had to be initialized to 0, and a break condition for the loop
also needed to be added otherwise we'd end up in an infinite loop if no gadget
had autoscroll (in which case, the need var was never modified).
SVN revision: 64587
one more todo item being knocked off. but there.
also no proper icon for it yet. mostly code from the old taskbar with
lots of little things fixed.
SVN revision: 64518
gadcon: Fix 'limit' may be used uninitialized.
Fix formatting in some places.
menus: Fix several variables may be used uninitialized warnings.
comp: Add UNUSED where needed.
Fix formatting in some places.
SVN revision: 58467
added e_menu_item_new_relative(E_Menu, E_Menu_Item *rel) which appends
new item to rel or prepends it when rel is NULL.
changed e_gadcon_util_client_menu_append to to take only the gadget menu
and returns the new toplevel menu that should be shown by gcc
depending on the setting this is a new menu with gadget menu as submenu
or the gadet_menu with gadget menu items appended
updated e-modules-extra comes in a minute
SVN revision: 57826
Do not abuse the concept of e_*_init(), make them call-once and those
that needed multiple call are renamed to e_*_update(). To make sure
convert them to EINTERN so the symbols are not exported.
Actually I guess too much is exported as EAPI while they should be
EINTERN, but that would require manual investigation, while this patch
was basically created with sed + grep.
SVN revision: 54795
than the gadget name which is usually just a short version which would
mean less to the user).
Use the Icon from the gadget itself for display in the menu.
SVN revision: 52106
If you have a Gadget and open the menu over it, all gadget specific options are under a separate submenu. The menu items which the Gadget sets theirself can be put on the main menu tree or on the Gadget submenu.
If you don't see a reason for this, check the ibar menu _before_ and _after_. :)
This api change broke all modules but i already fixed them all. Hope everything works... I also shortend text strings, etc, etc.
Hope you like it! :)
SVN revision: 52041
* Remove vim modelines:
find . -name '*.[chx]' -exec sed -i '/\/\*$/ {N;N;/ \* vim:ts/d}' \{\} \;
find . -name '*.[chx]' -exec sed -i '/\/[\*\/] *vim:/d' \{\} \;
* Remove leading blank lines:
find . -name '*.[cxh]' -exec sed -i '/./,$!d'
If you use vim, use this in your .vimrc:
set ts=8 sw=3 sts=8 expandtab cino=>5n-3f0^-2{2(0W1st0
SVN revision: 50816
anyway nice that i didnt had to rewrite gadcon layout to fix this issue,
but at least starting the rewrite made me realize the above mentioned
difference. moving and resizing of gcc in non-shrinked mode works
now nicely again.
SVN revision: 42637
- fix dnd of gadcon clients between gadcons
- apply fix from victor for gadcon client movements
- make layout algorithm orientation agnostic
SVN revision: 42539
Hi guys.
Here is my solution for #113 ticket request. This allows to transfer a
gadget to different locations. In short,
- any module can register arbitrary number of locations by invoking
e_gadcon_location_register().
- module is responsible to provide 2 functions, one for removing
particular gadget and one to add particular gadget class.
- module responsible to equip its gadcon(s) with reference to registered
location. If a gadcon contains non-NULL location reference, each gadget
contained by this gadcon can have submenu on its context menu. Template
function for constructing such submenu is provided
e_gadcon_client_add_location_menu. Gadgets can be moved from and to this
location by invoking menu items. Gadcons with NULL location reference
can accept transferred gadgets by module logic but can't send contained
gadgets to another module.
- module responsible to unregister location when gadcon is no more
available.
- menu items for context submenu is sorted by E_GADCON_SITE_XXX assigned
to that. Module can assign and change an icon for context menu item
corresponding to its location(s).
- Currently only shelves and gadman (desktop) used in gadget transfer
operations. I believe it is no trouble to ivolve any other
gadcon-capable module into the process. If you want to have that, please
just let me know.
Here the patch in attachmet. It even works, at least sometimes and at
least for me. Hope you'll find it useful.
SVN revision: 42121
Formatting.
Whitespace Removal.
Variable cleanup.
Add missing 'static' declaration on private function.
Move EAPI functions to be together in the source before local functions.
SVN revision: 41246
1. Gadcon sities:
-- 'E_Gadcon_Site' enum to list all available sities,
-- added 'is_site' callback for distinguish gadcon sities,
-- 'e_gadcon_site_is_*' helpers written to use in gadget modules,
-- gadcon config dialogs modified to show only gadgets that appropriate
to selected gadcon site,
-- all gadget modules updated for using new callback.
When callback is not provided in module, then gadget can be used in
any gadcon site.
2. Filemanager:
-- added error icon to unexisting path dialog,
-- added callback 'func.abort' to 'E_Fm2_Op_Registry_Entry' to specify
abort operation sequence (if not provided, then operation is not
cancelable),
-- added parameter to 'e_fm2_op_registry_entry_add' to specify abort
method on construct 'E_Fm2_Op_Registry_Entry' or make non-cancelable
operation if this function is not provided,
-- added 'e_fm2_operation_abort' call to cancel efm2 operation,
-- added method '_e_fm2_operation_abort_internal' that represent efm2
default handler of operation cancelling,
-- 'done' and 'total' fields of 'E_Fm2_Op_Registry_Entry' changed to
off_t type, also all message handlers modified to work with it
(to show progress for large files correctly),
-- improved file operation progress processing,
-- added cancel button to operation progress gadget in filemanager window,
-- added delete file glyph to operation progress gadget,
-- improved status message for efm2 operation progress.
3. Filemanager slave:
-- fixed critical bug in message handling via stdin/stdout,
-- fixed critical bug in removing task from list,
-- improved 'E_FM_OP_ABORT' handling,
-- fix rollback counting on moving files.
4. Fileman_opinfo module:
-- a new theme writed: status string and gadget for view operation status
in summary and detail modes,
-- summary mode: operation glyph, state message, progressbar, abort button,
-- detail mode: all for summary + from filename, to filename,
-- summary/detailed button works as trigger,
-- added source of module theme and used images.
4. Pathbar module:
-- non-critical fix, just for safety.
Also patch contains several minor updates to serve main task.
SVN revision: 41224
The old code that query the zone based on x,y was causing me
segfaults. When battery module was in a hidden shelf, the position was
outside any zone and then segmentation fault.
As far as I understand, gcc->gadcon->zone should always exist and
always be correct, but I don't have xinerama setup now to confirm
that.
If you know this is wrong, let me know!
SVN revision: 41147
if e_modapi_init() called e_gadcon_provider_register() the module
gc_init() would be called immediately, becore e_modapi_init() finished
and thus mod->data was unset (it's set by E when that function
returns).
most modules did not show this problem since they keep lots of globals
around and use them, but not mixer module. Mixer module did check for
global mixer_mod to be set and also mixer_mod->data, thus exposed this
problem, with "mixer is not able to show on desktop/gadget" but did in
shelves, since it would refuse to load immediately but would be loaded
later at e_shelf_config_new()->e_gadcon_populate().
Now gadcons are populated from an idler, avoiding all these problems
and probably will impact user's perceived load time since it will do
less work before getting to main loop and being able to process
events.
SVN revision: 40714
e_util_menu_item_fdo_icon_set to e_util_menu_item_theme_icon_set
As it looks for icons in e theme AND in fdo icon themes.
* Use the new function everywere instead of e_util_menu_item_edje_icon_set,
that is now marked as DEPRECATED
You should not see any difference with this commit. With the next few commit I'm
going to rename all the icons in the theme following fdo standard. I will not change
any icons, just rename them to match fdo ones.
SVN revision: 39370
* etk,
* PROTO/exalt,
* E-MODULES-EXTRA/diskio,
* E-MODULES-EXTRA/drawer,
* E-MODULES-EXTRA/penguins,
* E-MODULES-EXTRA/slideshow,
* E-MODULES-EXTRA/mail,
* E-MODULES-EXTRA/forecasts,
* E-MODULES-EXTRA/iiirk,
* E-MODULES-EXTRA/places,
* e,
* ewl,
* ecore,
* elitaire,
* entrance,
* e_dbus,
* efreet: Here we go, move from Ecore_List to Eina_List.
NOTE: This patch is huge, I did test it a lot, and I hope nothing is
broken. But if you think something change after this commit, please
contact me ASAP.
SVN revision: 39200
vanishing, like with e_shelf_locked_set. gadcon providers set their own lock
function with e_gadcon_util_lock_func_set, which is called by
e_gadcon_locked_set.
SVN revision: 38759
according to raster e_gadcon_canvas_zone_geometry_get() was supposed
to return zone-relative coordinates, but it was implemented
incorrectly and everything now depends on absolute coordinates, so
let's keep it absolute.
SVN revision: 38589
Some cases like not having o_base and o_frame are impossible, but I
added the cheap "else return;" just to keep it quiet. But some cases
were really bugs and were fixed.
SVN revision: 38556
e_menu had this behaviour that submenus were not deleted
automatically, relying on the parent menu to hook to post_deactivate
and then delete them.
This was good because maybe you don't want to delete these children,
you might want to reuse them sometime.
But it ended that nobody was using this feature, and worse: most use
cases were failing to delete these children, causing memory leak.
This commit changes the default behaviour and thus remove all the
existing code that was replicating such work. If one want the old
behavior, just call e_menu_item_submenu_set(mi, NULL) and it will be
unbounded from the parent.
If you experience any problems with that, try adding some printf()
before _e_menu_free() and _e_menu_item_free() and print some relevant
information like m->category and m->header.title or mi->label in order
to figure out the problematic menu.
SVN revision: 38528
But is so big i fear i could have broken some piece of code. So report any wrong
behaviour to me (cedric on #edevelop).
So moving e17 and efreet to eina_hash. With a little efreet API break so they
must come together.
SVN revision: 38185
Version 3 add the *client_class param to icon(),label(),id_new(), id_del()
and the *orient param to orient().
All the modules (with a gadcon_client) need to be fixed...just add the new params
to follow the EXTRA modules update.
SVN revision: 37450
Some people is using it for some time now without problems, so I'm
adding it to SVN to get some broader use. Remember to recompile ALL
libraries that depend on Evas as it will change the
EVAS_SMART_CLASS_VERSION and old classes will fail to load.
This will also change Edje so it will postpone _edje_recalc() to
render time, calculate() callback, however some methods will force
early recalculation.
SVN revision: 35860
1. configure/build changes to allow cross-compiling painlessly
2. pager module namespace changes - this was still dirty afdter the namespace
cleanup, so clean it up
3. add a powersave subsystem - doesnt have an "automatic" way to turn on and
off right now, this i think is best provided by modules (that do things like
monitor acpi status's (eg close lid of laptop), AC power status etc. etc.
this allows e to nicely defer "power" expensive actions to avoid disk
spinups etc.
4. move to use the new ecore poller system - discussed long ago as part of
power management/saving issues. now it exists
5. add a canvas idle flush call that helsp cope with the new shm greedy
software x11 engine stuff
6. use the new powersave subsystem where appropriate
7. fix non-zeroed/initted memory access in e_fm_main
8. fix mem leak for e menus
9. remove ipc handlers for changed/removed config values
10. use animaotr not timer for menu scrolls - then menu scrolls obey the fps
config
11. fix up timer/poll happienss of cursor idle stuff
12. remove avoid damage from popups for now - causing problems
13. change battery and temp readouts to b e shorter so they fit
14. pager can emit signals on focus change for mini-windows now
15. temperature module now uses a slave process and uses stdin/out to talk to
it and get output - this makes e smoother as in my expereicne i found getting
the temp on my laptop actually took like 200ms so e "hang" for 200ms while
reading the acpi files - so now the subprocess does it and just writesa back
to e when it gets it.
ecore:
1. add ecore_pollers. see the documentation on them in doxygen comments :)
2. fix timers to only go off when they have to - bug there that made e's
select time out a LOT more than it needed to. defensive coding hid the
problem. now fixed. e should be much more power friendly now.
3. formatting/niceness in ecore_exe stuff
4. some comments on comments with SIGIO ideas vs. select
5. add call to be able to add an idle enterer at the start of the list of
them, not just the end (as has been the default)
6. fix ecore_evas to support auto evas idler calls after 0.5 secs of idle in
all canvases - and to do it right
7. if argb destination - set the shape EVENT shape (to mask out events in
transparent regions much like shape does withotu translucency)
8. in ecore_x add support for the event shape
evas:
1. fix cache to work properly and not just always fill up (as it seemed to
like to think cahce useage dropped below 0 when it didnt and thus just
over-fill)
2. software x11 engine now ONLY uses shm segments - no ximages over the
socket. this ximage hack was there to avoid the 2 round trips involved in
setting up an shm image - now i mitigated that wih an shm image cache pool.
it keeps shm images around and repurposes them for new update regions if
appropriate. this means many fewer shm creates (about 1/100th the number) and
since we recycle the memory less 0 memory page filling by the kernel - in the
end, i recorded about a 10-20% speedup over the old software x11 engine.
simple tests i have seen up to 120% speedups. idle flush now does something -
it frees all the cached shm segments. it has a hard-coded limit of 4mb worth
of shm segments (or 32 segments - whichever comes first) to keep around. once
can never complain much about speedups methinks :). also evas will defer sync
until the NEXT frame is written - this means evas can calculate the next
frame of data while x dma's/copies the images to the screen at the same time
(if you hve a dual core or multi-cpu machnike or your xserver is able to use
DMA to copy image data to the screen/video ram then this should see a decent
speedup).
SVN revision: 33448
Mike Blumenkrantz
Carsten Haitzler
Vincent Torri
Cedric BAIL
Youness Alaoui
Hannes Janetzek
Christopher Michael
Peter van de Werken
Gustavo Sverzut Barbieri
Sebastian Dransfeld
Miculcy Brian
Lucas De Marchi
Michael BOUCHAUD
victor
semernin
Viktor Kojouharov
Davide Andreoli
Chidambar Zinnoury
toma