x11 modifier handling in events is broken: the modifier state is the state from
before the event, meaning that pressing caps lock will never result in an event where
the modifier is not set in the corresponding event
wayland handles this more sensibly, though it should be detected on key up rather
than key down
fix T5737
The object is not meant to be moved, we only subscribe click events on
it, this isnt really changing the context where the api wants to declare
it const, so for now safe it everywhere as const and register /
unregister the callbacks with casts.
everything used here edje_object_part_swallow_get, which is somehow
wrong, since we dont really know if the everything theme gives the type
swallow, using the plain object_get of edje is the solution here. Its
returning a alias object that is hiding the internal edje object.
the problem here was that in the initial case the function got the
previous state of the tree wrong, so the insertion of a second client
ended up in a unpossible state of the tree, this should not happen
anymore now.
The insertion is now also way more stable, since in a errorcase the
client is not just not placed in the tree but associated with a window
tree, its just not placing the client in the window tree at all.
You can install all updates or select the packages you like.
The update work without asking any password, if it doesn't works check the
org.freedesktop.packagekit.system-update Polkit permission
(work out-of-the-box on Arch Linux, just install packagekit)
this makes modules with a binary helper simpler to build using the
parent module build harness as much as possible. i probably could
simplify this down to a single binary only and it is either setuid or
not... define the deps and flags ... it could be a bit simpler. not
much. i also removed the if's in the build for battery and ifdefs in
src handle it instead (imho simpler to maintain in src). sysinfo still
uses the if's there.
config that is only needed fro modules only needs to be don in the
modules subdir meson.build... i tried doing what modules do but output
files cant have paths.
for now i just added HAVE_WL_DRM but its just the same as
USE_MODULE_WL_DRM, maybe we can replace HAVE_WL_DRM once autotools is
gone, so we have a clear pattern.
a reason for doing that is that you can just pack together targets into
a array and pass them to our helper, and the helper will just handle
them, so even module with eldbus codegen etc is now supported.
This also means that we are just passing the src object directly into
the shared_module call, which means the user of our helper can just pack
everything he needs into the src var and the helper does not need to
know about it.
Summary:
Basically missed this one. CPU_STATES is 6 as it's still supposedly a work-in-progress. But we only need 5 for all
queries (for now).
Reviewers: stephenmhouston, zmike!
Subscribers: cedric
Differential Revision: https://phab.enlightenment.org/D5070
these were added mostly for the purpose of testing the wizard functionality,
but a wizard should only be used when a gadget cannot function without being
configured
ref T5796
many times it's useful to have an event for actual zone geometry change
vs useful geomtry change, so split this out and use the right handler where appropriate
makes gcc warnings happy but otherwise doesn't hurt us as the buffer
should be at the top of the stack (last pushed) and only bytes used
will be allocated.
so i get a new service of type WIRELESS_SERVICE_TYPE_NONE that's going
to suck when accessing arrays by type like wireless does like
array[cs->type] ... so check type value and if its invalid kill off
the cs as we can't do much useful with it. this fixes an actual segv e
gets if you restart the connamn daemon while e runs.
now mdoule build files that fllow one pattern (the most common by far)
all JUSt list their souce files and nothing else. this really cuts
down on build size/complexity.
there are other patterns too (no icons at all) that i'll do next, then
we're down into "weird" land where i'll have to think of some more
interesting ways to deal with this.
It appears that config.geom.x and config.geom.y specify the corner of
an output in global space, but ecore_drm2_output_mode_set's x and y
are offsets into the framebuffer for the corner of the display.
Just pass 0, 0 and everything will be ok.
this module is not loaded by any other (dependency) nor is it loadable
via the gui - no module/desktop there thus will be hidden... so it's
useless/unused... thus remove it as its not usable by users.
now their build fiels are super easy to maintan. we should expand this
kind of ptterning aross the e build and then expand it to handle new
patters where needed like custom binaries (setuid or not), etc etc.
so we';re missing installing desktop files. edj icon files, wizard
data files in the wrong place, and much more. this also cleans up the
module meson build of pretty much all modules and make their build
files cimple and consistent so it's far easier to re-use things from
one module to the next. we should aim for simplicity, consistency
between as much as possible so we can refactor and turn into maybe
functions later. imho that starts with consistency though. until i can
see all the common patterns clearly, i don't want to write functions
yet. it's easier to see if all the files are consistently using the
same vars and formatting etc. etc. etc.
but either way the installation needs fixing so it installs all files
in the right places with the right permissions etc. etc. etc.
this doesn't fix all module build files bt all the ones i found that
were broken installs and they use what i think is a cleaner/simpler
template, BUT there is far too much copy & pastage here... far too
much.i need to find a cleaner way to automate this.
we use these defines with
ifdef xxxx
not #if xxxx
and so they should ne undefed when not there and otherwise value when
defined is irrelevant as long as they are defined.
so use set not set10
also... there isn't realy a need to track the screensaver state...
powersave sleep will drop back to an hour between sleeps if we're in
freeze mode (it could be longer or even be indefinite). it will be
woken up if powersave state changes...
When creating the data manager source, passing an id of 1 overwrites
the wl_display's id in the map, causing crashes the next time the
client tries to interact with that object. The client in this case
is Xwayland. Bad things happen.
Instead pass 0 which just chooses an available map slot.
Fix T5738
This cleans up how sysinfo manages object vs thread lifetimes. If thread is still alive dependent on aspects that need to be freed in the gadget removal process, it defers that cleanup from the remove callback to the thread end callback. As for the combination sysinfo gadget, each gadget inside of sysinfo will set a done flag alerting that the cleanup of the combination gadget can happen once all threads are done.
This fixes T5694
See also ac92ff5256.
- eina_strbuf_string_get() returns the internally stored string as
a const char *, and does not free the strbuf itself
- eina_strbuf_string_steal() returns the internal string as a
char *, giving ownership to the caller, and frees the strbuf
itself
- eina_stringshare_add() takes a const char * as input and makes a
copy of the string
As a consequence, ss_add(sb_string_steal()) leaks the internal
string from the strbuf, while ss_add(sb_string_get()) leaks the
strbuf structure.
A one liner here would require either an eina_slstr based API or
an API in stringshare to take ownership of a given string. Both
would be useful APIs :)
this requires we have to force dpms on to reduce power. to avoid
glitches with the pointer staying around in x we need to support
suspending it too so it hides cleanly like the screen dims or undims.
also use the new powersave freeze mode to do this.
note that i've tested this on s3 supporting laptops and non-s3 and it
"works for me". it may require more testing and work. there is more to
power saving than just this as well but for now that's out of scope as
you have to mess with linux device autosuspend timeouts and a bunch
more (wowlan ... blahblah).
i need to find the source of the intermittent wakeups too in e. there
is a long lived timeout (8-ish seconds?) but more specifically e keeps
waking up from fd's and then reading /sys stuff about battery - some
event is causing us to do this... maybe to suspend this or make
battery checking very rare when in freeze mode (or screen off) etc.
so this fixes some glitches as well as supports a new way of sleeping
"alive" when hardware literally doesnt support normal s3 sleep... so
kind of a fix with a feature.
so new laptops now seem to no longer support S3 sleep. sleeping is
done basically by going as idle as possible. you can ask the kernel to
freeze execution BUT this seems to use about the same power as staying
alive in my tests. to support this add 2 things:
1. a FREEZE powersave mode which implies we're alive but want to
really stay as idle as absolutely possible.
2. powersave aware sleep functions that replace the usleeps in threads
so they can switch from being super sleepy when in freeze mode to
normal.
partially reverts 7e05eff3e3
This was causing problems when destroying some xdg v6 clients.
if weston-simple-shm was killed while not on the current desktop
it would remain on deskmirrors.
==20443== Invalid read of size 8
==20443== at 0x28CED526: _bar_exec_new_show (bar.c:980)
==20443== by 0x819D78D: _eo_evas_object_cb (evas_callbacks.c:184)
==20443== by 0xDFB6FED: _event_callback_call (eo_base_class.c:1496)
==20443== by 0xDFB7373: _efl_object_event_callback_legacy_call (eo_base_class.c:1569)
==20443== by 0xDFB743A: efl_event_callback_legacy_call (eo_base_class.c:1572)
==20443== by 0x81DC562: _efl_canvas_object_efl_object_event_callback_legacy_call (evas_object_main.c:993)
==20443== by 0xDFB743A: efl_event_callback_legacy_call (eo_base_class.c:1572)
==20443== by 0x819E1F8: evas_object_event_callback_call (evas_callbacks.c:404)
==20443== by 0x81E6B23: evas_object_inform_call_show (evas_object_inform.c:13)
==20443== by 0x81DECA2: _show (evas_object_main.c:1689)
==20443== by 0x81DF0E7: _efl_canvas_object_efl_gfx_visible_set (evas_object_main.c:1810)
==20443== by 0xDD670B9: efl_gfx_visible_set (efl_gfx.eo.c:21)
==20443== by 0x81DEA93: evas_object_show (evas_object_main.c:1639)
==20443== by 0x483706: _e_comp_intercept_show_helper (e_comp_object.c:1754)
==20443== by 0x483761: _e_comp_intercept_show (e_comp_object.c:1768)
==20443== by 0x81E7536: evas_object_intercept_call_show (evas_object_intercept.c:71)
==20443== by 0x81E7ED2: _evas_object_intercept_call_internal (evas_object_intercept.c:103)
==20443== by 0x81E88B0: _evas_object_intercept_call_evas (evas_object_intercept.c:236)
==20443== by 0x81DF0CA: _efl_canvas_object_efl_gfx_visible_set (evas_object_main.c:1807)
==20443== by 0xDD670B9: efl_gfx_visible_set (efl_gfx.eo.c:21)
==20443== by 0x81DEA93: evas_object_show (evas_object_main.c:1639)
==20443== by 0x4A6793: _e_desk_show_begin (e_desk.c:821)
==20443== by 0x4A4E39: e_desk_show (e_desk.c:312)
==20443== by 0x537C2E: _e_int_menus_clients_item_cb (e_int_menus.c:1624)
==20443== by 0x548D3F: _e_menu_active_call (e_menu.c:2056)
==20443== by 0x54ABFB: _e_menu_cb_mouse_up (e_menu.c:2789)
==20443== by 0xC636B66: _ecore_call_handler_cb (ecore_private.h:325)
==20443== by 0xC637B3F: _ecore_event_call (ecore_events.c:518)
==20443== by 0xC641158: _ecore_main_loop_iterate_internal (ecore_main.c:2397)
==20443== by 0xC63EC7E: ecore_main_loop_begin (ecore_main.c:1299)
==20443== by 0x43DE81: main (e_main.c:1081)
==20443== Address 0x20 is not stack'd, malloc'd or (recently) free'd
I missed this in my last commit - we probably shouldn't be calling
e_comp_render_queue or e_comp_shape_queue_block() after hiding the
ecore_evas anyway - and by removing the e_comp_shape_queue_block()
in the activation callback I made things asymmetrical. Ungood.
The code intended to force evas to redraw when we switch back from
another virtual console is failing to do so. Remove it and replace
it with simpler code that successfully forces a redraw.
key presses during desklock should only be received by the lock implementation
and not by any other handler. this ensures that nothing unexpected can happen
with focus and simplifies overall key handling
All three of these gadgets are ports of the existing modules of the same name and are contained within those directories. Once the move from shelves -> bryce and gadcon->gadgets is complete, backlight and mixer will likely need to go into the sysinfo gadget.
when plugging a screen in and out, there is the case that a zone has a
usefull geometry of 0x0, which means all clients on this zone are
resized to 0x0. Which leads to a CRIT message in the compositor, which
leads (ref commit 5d875e6a3d) to a abort()
which is really really annoying. Give here a short ERR about that case
so we are leaving out the compositor, since this really only strikes on
tiling, since in normal mode the client just keeps its size.
in the event that a client has not yet committed the changes from the
most recent resize event, it's legal for a client to have acked the previous
configure, ack this one, and then do nothing
this ensures that the last resize event(s) sizes are applied by the client
calling `pulseaudio` starts a new daemon in the background. this is incorrect
behavior when a daemon already exists, so use --start. tracking the exe of
this process has no effect other than to determine when the fork()ing parent exits,
which is usually immediately
ref 35bb87529f
==21266== 3,488 (96 direct, 3,392 indirect) bytes in 2 blocks are definitely lost in loss record 10,417 of 10,680
==21266== at 0xE1E5D49: _eina_chained_mempool_alloc_in (eina_chained_mempool.c:212)
==21266== by 0xE1E5FDC: eina_chained_mempool_malloc (eina_chained_mempool.c:324)
==21266== by 0xE1A016E: eina_mempool_malloc (eina_inline_mempool.x:90)
==21266== by 0xE1A03C2: _eina_list_mempool_list_new (eina_list.c:222)
==21266== by 0xE1A11C5: eina_list_append (eina_list.c:578)
==21266== by 0x2910B667: _bar_fill (bar.c:1565)
==21266== by 0x2910D1A5: _bar_recalculate_job (bar.c:2047)
==21266== by 0xC602C2C: _ecore_job_event_handler (ecore_job.c:98)
==21266== by 0xC5FBBCE: _ecore_call_handler_cb (ecore_private.h:317)
==21266== by 0xC5FCB5D: _ecore_event_call (ecore_events.c:518)
==21266== by 0xC605EEB: _ecore_main_loop_iterate_internal (ecore_main.c:2381)
==21266== by 0xC603C99: ecore_main_loop_begin (ecore_main.c:1289)
==21266== by 0x43DD0D: main (e_main.c:1089)
there is no event that indicates that the mouse went to a other zone. To
solve this we simply update the current split type each time when
changing or using the type.
If someone starts to drag a client arround, then the client will shrink
into a icon, that now is always at the position of the mouse cursor, if
the drag ends, the client will be placed in the client currently below
it. The client will be placed in a place where the mouse cursor was
currently closer to.
Stephen 'Okra' Houston
Carsten Haitzler
Marcel Hollerbach
Mike Blumenkrantz
Davide Andreoli
Al Poole
Cedric BAIL
Derek Foreman
Jean-Philippe Andre
Stephen Houston
Andy Williams
