ivi-shell: harden get_ivi_shell_surface()

Add more sanity checks to get_ivi_shell_surface() just in case.

If the configure hook is set, we must always have non-NULL
configure_private.

Check the ivi_shell_surface matches the surface.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Eugen Friedrich <friedrix@gmail.com>

ivi-shell: add sanity check in ivi_shell_surface_configure

This should not get called unless there is an ivi_shell_surface.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Eugen Friedrich <friedrix@gmail.com>

ivi-shell: add input panel label func

Copied from desktop-shell/input-panel.c, add a label function for the
input panel.

This patch strictly reduces the difference between input-panel.c and
input-panel-ivi.c.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Eugen Friedrich <friedrix@gmail.com>

ivi-shell: add shell surface labels

To be used by the Weston timeline feature for identifying surfaces in a
trace. The 'get_label' functionality can also be used by any debugging
code, too.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Eugen Friedrich <friedrix@gmail.com>

ivi-shell: remove ivi_layout_get_screen_output API

The controller plugins use IVI Layout API with weston outputs.
Therefore, this API is not required.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: remove ivi_layout_get_screen_from_id API

The controller plugins does not use ivi screens for IVI layout APIs.
They use weston outputs directly. Therefore, this API is unnecessary.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: implement get_screen_from_output

It is an internal API, which returns ivi_layout_screen
for a pregiven weston_output.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: use weston_output in public APIs

IVI layout APIs now are called with weston_output pointers,
instead of ivi_layout_screen pointers.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: remove ivi_layout_get_screens API

The compositor data struct already has a list of weston outputs.
Therefore, this API is not required.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: remove ivi_layout_get_screen_resolution API

The controller plugins can get the screen resolution directly from
weston output. Therefore, this API is not required.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: remove ivi_layout_get_id_of_screen API

ivi-screen does not have an id. IVI layout implementation is using
id of weston output. Therefore, this API is unnecessary.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

ivi-shell: remove id_screen

use output id instead

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Acked-by: Wataru Natsume <wnatsume@jp.adit-jv.com>

input: use doubles in the interfaces to notify of input events

This patch is a further step in the wl_fixed_t internal sanitization.
It changes the notify_* functions to take doubles instead of wl_fixed_t
but does not change how these are stored in the various input structs
yet, except for weston_pointer_axis_event.
However this already allows to remove all wl_fixed_t usage in places
like the libinput or the x11 backend.

Reviewed-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

compositor-drm: Renaming of gbm fields

In preparation for follow-on changes to support frame presentation
through EGLDevice+EGLOutput, this change includes the following:
  - Rename drm_backend::format   to gbm_format
  - Rename drm_output::format    to gbm_format
  - Rename drm_output::surface   to gbm_surface
  - Rename drm_output::cursor_bo to gbm_cursor_bo

Signed-off-by: Miguel A Vico Moya <mvicomoya@nvidia.com>
Reviewed-by: Andy Ritger <aritger@nvidia.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
[Pekka: trivial rebase out of the series]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

gl-renderer: Implement & use check_extension

Using strstr(3) for checking for extensions is an error-prone mechanism
as extension names can be prefixes of other extension names (see
https://www.opengl.org/registry/doc/rules.html#using).

This change implements the check_extension() function to properly check
for an extension and replaces all usages of strstr(3).

Signed-off-by: Miguel A Vico Moya <mvicomoya@nvidia.com>
Reviewed-by: Andy Ritger <aritger@nvidia.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
[Pekka: move 'bool' to the same line with 'static']
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

hmi-controller: fix wrong panel width

The width of the first base layer is used for all panels.
Every display has a base layer which is as big as the
display. Therefore, it is wrong to use the width of the
first base layer for all panels, because every display
could have a different resolution.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

hmi-controller: fix wrong background surface size

In current implementation, the size of the first application
layer is used for the background image of a display.

This is wrong because:
1. The background surface should be fullscreen.
2. Each display could have different resolution.

We should use the size of the base layer of each display
for the background image.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

input: Implement wl_seat.release

Avoid a crash because listener is NULL.

Signed-off-by: Quentin Glidic <sardemff7+git@sardemff7.net>
Reviewed-By: David Fort <contact@hardening-consulting.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

compositor-drm: fix memcmp using a bad pointer in drm_outout_choose_initial_mode

current_mode is already the pointer, taking the address of it is wrong.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94562
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
[Pekka: rewrote the patch]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi: Switch from MEM_ALLOC to the now equivalent xzalloc

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>

shared: Print .c file/line number in x* routines

Switches from inline to pre-processor definitions in order to utilize
__FILE__ and __LINE__ from the .c file in order to display the location
of memory allocation failures when failing.

Now xmalloc, et al calls will produce:

    [weston-info] clients/weston-info.c:714: out of memory (1024)

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

clients & tests: Unify multiple definitions of x*alloc and related functions

Direct fail_on_null calls now produce output like:

    [weston-info] clients/weston-info.c:714: out of memory

xmalloc, et al produce output on failure like:

    [weston-info] out of memory (-1)

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: simplify ivi_layout_screen_set_render_order

It is not necessary check the ivi-id of pregiven objects (layer or
surface). Traversing the list of all objects is sure to find the exact
same pointer we start with, bugs aside.

The controller modules are responsible for providing valid pointers. We
cannot protect from invalid pointers anyway.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
[Pekka: fix subject, add commit message]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: simplify ivi_layout_screen_add_layer

It is not necessary check the ivi-id of pregiven objects (layer or
surface). Traversing the list of all objects is sure to find the exact
same pointer we start with, bugs aside.

The controller modules are responsible for providing valid pointers. We
cannot protect from invalid pointers anyway.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
[Pekka: commit message]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: simplify ivi_layout_set_render_order

It is not necessary check the ivi-id of pregiven objects (layer or
surface). Traversing the list of all objects is sure to find the exact
same pointer we start with, bugs aside.

The controller modules are responsible for providing valid pointers. We
cannot protect from invalid pointers anyway.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
[Pekka: commit message]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: simplify ivi_layout_layer_add_surface

It is not necessary check the ivi-id of pregiven objects (layer or
surface). Traversing the list of all objects is sure to find the exact
same pointer we start with, bugs aside.

The controller modules are responsible for providing valid pointers. We
cannot protect from invalid pointers anyway.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
[Pekka: commit message]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: simplify ivi_layout_remove_surface

It is not necessary check the ivi-id of pregiven objects (layer or
surface). Traversing the list of all objects is sure to find the exact
same pointer we start with, bugs aside.

The controller modules are responsible for providing valid pointers. We
cannot protect from invalid pointers anyway.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
[Pekka: commit message]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: remove add_configured_listener left-over

In all my rebases, this got accidentally left behind. The implementation
was removed in 4a7503976b5d300a4474b149d877cf579c2a8c05 but
32ca791df8f855898e007512f3b996a66ffcb09c reintroduced it.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: introduce ivi-layout-shell.h

This new header encapsulates the API that ivi-layout offers to
ivi-shell.c to call.

ivi-shell.c no longer uses ivi-layout-private.h. This limits the
ivi-layout internal structures to just ivi-layout code.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emre Ucan <eucan@de.adit-jv.com>

ivi-shell: remove configured signal from ivi-layout

Now that ivi-layout calls directly into ivi-shell.c, this signal is no
longer used. Remove it.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emre Ucan <eucan@de.adit-jv.com>

ivi-shell: call shell_surface_send_configure() directly

For some reason, it seems that ivi-layout.c has tried hard to avoid
calling directly into ivi-shell.c. This means there is a jump through
hoops just to get the configure event sent to the clients. Ivi-shell
registers a listener for a ivi-layout signal for sending the event.

Instead, let ivi-layout.c call directly into ivi-shell.c, and expose a
function to send out the configure events. This reduces some confusion
on who calls what.

The main idea though is that this makes ivi-shell.c not depend on struct
ivi_layout_surface fields directly anymore. In following patches,
ivi_layout_surface can be made opaque for ivi-shell.c.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Emre Ucan <eucan@de.adit-jv.com>

clients: fix incorrect format handling in simple-shm

the `shm_format` function seems to assume the `wl_shm_format`
enum has bit-exclusive enumerations which is not true.

Signed-off-by: Murray Calavera <murray.calavera@gmail.com>
[Pekka: fix whitespace with an 'if'.]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: add include guards on ivi-shell.h

It's our standard practice. This file will get used a bit more in the
future.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: include config.h in ivi-layout-transition.c

Every .c file must include config.h as the first thing.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

simple-egl: use roundtrip after get_registry request

Sometimes weston-simple-egl aborts in create_surface
under some conditions. It is because wl_display_dispatch()
may not be enough to make sure we have all requried objects.
Can be modeled by wldbg:

  $ wldbg -i weston-simple-egl
  (wldbg) b re get_registry
  (wldbg) c
  (wldbg) c

After these steps the weston-simple-egl aborts, because
it has not got shell neither ivi-shell objects

Signed-off-by: Marek Chalupa <mchqwerty@gmail.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: remove content_observer notification

The content observer notification struct and its
set API (ivi_layout_surface_set_content_observer)
are removed. Because they are unused.

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

ivi-shell: Remove all surface transitions when it is being removed.

If surface transitions are not removed when surface is being removed, it
can lead to crash later when transition will finish, as it will try to
reference already freed memory.
This change exposes function that can remove all existing transitions
for given surface and it is being called during surface cleanup.

Signed-off-by: Mateusz Polrola <mateuszx.potrola@intel.com>
Reviewed-by: Emre Ucan <eucan@de.adit-jv.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

desktop-shell: make panel clock configurable

This patch enhances the panel clock by adding a config file
option which can be used to either disable the clock or make
it also show seconds in the current clock format.

v2: Implement suggestions from Pekka:
    - Include Signed-off-by
    - Coding style fixes
    - Implement clock widget allocation by using
      width from cairo_text_extents
    - Highlight config option values in man page
v3: Implement suggestions from Pekka and Bryce:
    - Use CLOCK_FORMAT_* instead of FORMAT_* in the enum
    - Switch to using fixed clock widget size instead
      of one returned from cairo_text_extents
    - Fixes to config option highlighting in the man page
v4: Implement more suggestions from Pekka and Bryce:
    - Improve patch changelog
    - Move the check for CLOCK_FORMAT_NONE into the
      caller function
    - Fix a memory leak in panel_create introduced by
      previous revision of this patch

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=57583
Signed-off-by: Armin Krezović <armin.krezovic@fet.ba>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

clients: Use zalloc

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>

platform: explicitly cast the return value of weston_platform_get_egl_proc_address

This allows the header to be consumed by C++ compilers, because C++ does
away with C's implicit cast from (void*).
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

option-parser: Handle short double-arg options

weston allows both short and long style options to take arguments.  In
the case of short options, allow an optional space between the option
name and value.  E.g., previously you could launch weston this way:

  weston -i2 -cmyconfig.ini

now you can also launch it like this:

  weston -i 2 -c myconfig.ini

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>

configure: Make jpeglib an optional dependency.

It doesn’t make sense to fail the entire build when jpeglib isn’t
present, so this commit makes it optional just like libwebp in the
previous one, disabled with --without-jpeg and forced with --with-jpeg.

Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

protocol: migrate to stable presentation-time.xml

Remove the unstable presentation_timing.xml file, and use
presentation-time.xml from wayland-protocols instead to generate all the
Presentation extension bindings.

The following renames are done according to the XML changes:
- generated header includes
- enum constants and macros prefixed with WP_
- interface symbol names prefixed with wp_
- protocol API calls prefixed with wp_

Clients use wp_presentation_interface.name rather than hardcoding the
global interface name: presentation-shm, weston-info, presentation-test.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
[Pekka: updated wayland-protocols dependency to 1.2]

ivi-shell: remove ivi_layout_layer_set_dimension API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_set_dimension API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_layer_set_position API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_set_position API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_layer_get_orientation API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_get_orientation API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_layer_get_dimension API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_get_dimension API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_layer_get_position API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_get_position API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_layer_get_opacity API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_get_opacity API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_layer_get_visibility API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

ivi-shell: remove ivi_layout_surface_get_visibility API

Signed-off-by: Emre Ucan <eucan@de.adit-jv.com>
Acked-by: wataru_natsume <wataru_natsume@xddp.denso.co.jp>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

hmi-controller: fix leak of ivi surface list

get_surfaces_on_layer() allocates memory and stores the pointer to
'ivisurfs'. But it was not freed.

Signed-off-by: Wataru Natsume <WATARU_NATSUME@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

remove warning of ivisurface reassign on the ivilayer

The warning of ivisurface reassign can be removed. It is ok to reassign
a surface to a layer it is already on.

The warning started to show up during normal operations since patch
"hmi-controller: remove duplicate commit_changes in random mode".

Signed-off-by: Wataru Natsume <WATARU_NATSUME@xddp.denso.co.jp>
[Pekka: rewrote commit message, removed unneeded comments.]
Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

hmi-controller: remove duplicate commit_changes in random mode

Previous code cleaned up surfaces in layer once and then added
surfaces to a layer in random. In this flow, two commitchanges are
required.

Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
[WATARU_NATSUME@xddp.denso.co.jp: Removes unnecessary check]
Signed-off-by: Wataru Natsume <WATARU_NATSUME@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

libinput-device: Remove unnecessary function call

When we handle keyboard key events, we already retrieve the key state
at the top of this function, so there is no real need to call the same
libinput function again as we can just reuse the 'key_state' variable
that we have above.

Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

tests: cleanup test runner script.

Cleaned up test runner script to unify sections launching weston.
This makes the sections more legible and differences easier to spot.

Signed-off-by: Jon A. Cruz <jonc@osg.samsung.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

Makefile: handle also stable wayland-protocols

Fix the protostability function to handle stable protocol files
correctly. Stable protocol XML file names do not have 'stable' in their
name, nor do we want to write that in the prerequisite lists in the
Makefile.

Function 'protoname' does not need fixing, because for stable protocol
prerequisites, the sed pattern will not match, and it passes stem
through as is, which is correct.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

Makefile: move presentation_timing source out of toytoolkit

Toytoolkit sources don't actually use the presentation_timing client
protocol bindings for anything. Apparently they were there only because
that's how they end up in BUILT_SOURCES.

Move them from toytoolkit sources to BUILT_SOURCES where also other such
things are.

Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

config-parser: Check malloc and strdup returns

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Marek Chalupa <mchqwerty@gmail.com>

configure: Make WebP support togglable, and improve its error message.

The current way was enabling WebP support whenever libwebp was found,
giving no way to the user to disable it if they had the library
installed but didn’t want to link against it.  This adds a
--without-webp configure option to never link against it, and a
--with-webp one to fail the build if it isn’t found, the default being
to use it if it is present.

Additionally, we now tell the user when WebP support has been disabled
and they try to load a WebP file.

Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

libinput-device: Remove unnecessary function call

When we handle pointer button events, we already retrieve the button
state at the top of this function, so there is no real need to call
the same function again as we can just reuse the 'button_state'
variable that we have above.

Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

build: Require dbus for systemd-login

systemd-login support requires dbus (see "dbus.h" header in
"launcher-logind.c") but the configure script was only
checking libsystemd-login availability to define the
HAVE_SYSTEMD_LOGIN macro, which results in undefined
symbols in launcher-unit.

Put the systemd-login checks after the dbus ones, and only
run the checks if it is present. Also mention dbus in the
error message if "--enable-systemd-login" was forced.

Signed-off-by: Manuel Bachmann <manuel.bachmann@iot.bzh>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

shell: Don't send extra configure events on click

The click_to_activate handler fires on every mouse click for a surface
so let's be a little quicker to early return if you're clicking on the
surface that already has activation.

This prevents (among other side effects) the sending of two xdg_configure
events for every mouse click.

This should also make having two seats with keyboards behave in the same
way as a single seat.  Previously the second seat could have a keyboard
focus on the surface and prevent some of the extra processing (including
the extra configure events) from taking place.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

configure.ac: bump to version 1.10.90 for open development

Master is open for new feature development again.

configure.ac: bump to version 1.10.0 for the official release

releasing.txt: Script now figures out release name on its own

configure.ac: bump to version 1.9.93 for the RC1 release

compositor-drm: the cursor plane only supports ARGB

Keep XRGB apps out of the cursor plane, only ARGB is supported.

This prevents programs like weston-simple-shm from landing in the cursor
plane and being misrendered.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

compositor-drm: update internal cursor plane location when disabling cursor

When the cursor plane is disabled the kernel can lose its location.
If we don't update our internal idea of where the plane is at that time,
the next time we set a cursor it can show up at 0,0.

This can show up when an application is put in the cursor plane, removed
from the plane, then put back at the same location.  It might show up at
0,0 when it's reinstated.

We now use INT32_MIN as a location for disabled cursors so enabling the
plane will always cause an update.

Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Signed-off-by: Derek Foreman <derekf@osg.samsung.com>

compositor-x11: send pointer frame after axis event

clients that implement pointer interface of version 5
wait for the frame event, so without it the scrolling
does not work (GTK+ clients do not scroll now for example).
Xcb axis events are discrete, so it's fine to send
frame after every single axis event

Signed-off-by: Marek Chalupa <mchqwerty@gmail.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>

xwm: Don't clear the selection if it has no text type available

weston maintains a copy of the most recently selected "thing" - it picks
the first available type when it copies, and saves that one only.

When an application quits weston will make the saved selection active.

When xwm sees the selection set it will check if any of the offered types
are text.  If no text type is offered it will clear the selection.

weston then interprets this in the same way as an application exiting and
causing the selection to be unset, and we get caught in a live lock with
both weston and xwayland consuming as much cpu as they can.

The simple fix is to just remove the test for text presence.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Carlos Garnacho <carlosg@gnome.org>

shell: Check more thoroughly for undefined surface resource

The surface can have an undefined resource in certain situations (such
as with xwayland).  So, since NULL is a valid state for this parameter,
and since the wl_resource_*, etc. calls require their parameters to be
non-NULL, make a practice of always checking the surface resource before
making wayland calls.

update v2:
  * Fix some c/p errors for pointer names
  * Drop null ptr check in add_popup_grab; probably redundant now

Signed-off-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

ivi-layout: apply opacity to weston_view correctly

update_opacity is only called when a ivi-surface is visible. But the
previous code also checks event masks redundantly. However if the event
happens when ivi-surface is invisible, opacity is not calculated. This
patch removes this redundant check to fix potential bug.

Signed-off-by: Nobuhiko Tanibata <NOBUHIKO_TANIBATA@xddp.denso.co.jp>
Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>

configure.ac: bump to version 1.9.92 for the beta release

input: Fix crash when touchscreen generates out of screen coordinate

With change 61ed7b6b, global touch coordinates are being passed to the
touch grab.  However, touch->grab is undefined in certain circumstances
such as when the touch screen raises an axis X value larger than the
maximum expected.  Move the check for this condition earlier, before our
first use of the pointer.

Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=92736

Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>

data-device: Check harder for selection/non-wayland sources

We're not always dealing with weston_data_sources that have a
wl_resource, or data_sources that belong to drag-and-drop. Check
harder for these on the drag-and-drop code paths triggered from
common code.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>

xwayland: zalloc the x11_data_sources

The wrapped weston_data_source struct has new fields which were left
uninitialized, so its access is unreliable.

The data source in xwayland/dnd.c should be eventually setting the
drag-and-drop actions, but it is a lot more incomplete than that
(read: completely), so falls out of the scope of this patch.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>

clipboard: zalloc the clipboard_source

The wrapped weston_data_source struct has new fields which were left
uninitialized, so its access is unreliable.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com
Tested-by: Derek Foreman <derekf@osg.samsung.com>

input: Don't try to send pointer frames if no focus_client

Prevents a segfault when mousing into clients that don't get_pointer
like weston-simple-shm and weston-simple-damage.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Marek Chalupa <mchqwerty@gmail.com>

releasing: register_release now inserts the email announcement urls

configure.ac: bump to version  for the  release

dnd: Turn into a full blown example

In order to keep things simple, weston-dnd made a few choices that
turn out to be unrealistic, a few tweaks have been done to make it
less of a playground demo:

- It now caters for copy/move operations, instead of just move,
  which still remains the default nonetheless.
- As "move" operations are no longer assumed, the item isn't removed
  on start_drag, instead it is made translucent until the drag
  operation finishes (and we know whether the item is to be
  removed after transfer or left as is)
- For the same reasons, "Drop nowhere to delete item" no longer
  happens. Drag-and-drop is a failable operation and must not result
  in data loss.
- As multiple actions are now allowed, we set the pointer icon
  surface accordingly to the current operation.

This makes weston-dnd a better example of what applications usually
want to do here.

Changes since v2:
  - Updated to behave alright-ish with version < 3.

Changes since v1:
  - Remove unneeded include. Remove extra newlines. Other minor
    code fixes.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

data-device: Implement compositor-chosen actions

Set up a keyboard grab during drag-and-drop, so we can translate
modifiers into preferred actions. The compositor chosen action
is stored in the current weston_data_source in order to make it
accessible to the source/offer at the time of calculating the new
action, but would conceptually be part of weston_drag.

The mapping has been made similar to what GTK+/QT usually do, the
shift key defaults to "move" and ctrl defaults to "copy".

Changes since v2:
  - Use enum types and values for the compositor action. Fix
    code formatting issues.

Changes since v1:
  - Handle the keyboard grab being cancelled. Initialize new
    wl_data_source fields.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

client: Add DnD cursors to the managed cursors list

That way we'll be able to set the corresponding pointer surface to
a current DnD operation.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

data-device: Implement DnD actions

The policy in weston in order to determine the chosen DnD action is
deliberately simple, and is probably the minimals that any compositor
should be doing here.

Besides honoring the set_actions requests on both wl_data_source and
wl_data_offer, weston now will emit the newly added "action" events
notifying both source and dest of the chosen action.

The "dnd" client has been updated too (although minimally), so it
notifies the compositor of a "move" action on both sides.

Changes since v8:
  - Add back wl_data_offer.source_actions emission, gone during last
    code shuffling. Fix nits found in review.

Changes since v7:
  - Fixes spotted during review. Add client-side version checks.
    Implement .action emission as specified in protocol patch v11.

Changes since v6:
  - Emit errors as defined in DnD actions patch v10.

Changes since v5:
  - Use enum types and values for not-a-bitfield stored values.
    handle errors when finding unexpected dnd_actions values.

Changes since v4:
  - Added compositor-side version checks. Spaces vs tabs fixes.
    Fixed resource versioning. Initialized new weston_data_source/offer
    fields.

Changes since v3:
  - Put data_source.action to use in the dnd client, now updates
    the dnd surface like data_source.target events do.

Changes since v2:
  - Split from DnD progress notification changes.

Changes since v1:
  - Updated to v2 of DnD actions protocol changes, implement
    wl_data_offer.source_actions.
  - Fixed coding style issues.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Michael Catanzaro <mcatanzaro@igalia.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

data-device: Implement DnD progress notification

Weston now sends wl_data_source.dnd_drop_performed and .dnd_finished in
order to notify about the different phases of DnD.

wl_data_source.cancelled is also used as mentioned in the docs, being
emitted also on DnD when the operation is meant to fail (eg. source
and dest didn't agree on a mimetype).

The dnd demo is also fixed so the struct dnd_drag isn't leaked.

https://bugs.freedesktop.org/show_bug.cgi?id=91943
https://bugs.freedesktop.org/show_bug.cgi?id=91944

Changes since v6:
  - Add client-side version checks. Minor code shuffling.

Changes since v5:
  - Dissociate source and offer after cancel. Updated to
    apply on top of c9f8f8a7f.

Changes since v4:
  - Make wl_data_offer.finish with the wrong state an error.

Changes since v3:
  - Fixed wl_data_source.dnd_finished vs cancelled emission on
    when interoperating with version < 3 drag destinations.

Changes since v2:
  - Handle wl_data_offer.finish. Fixed commit log inconsistencies.
    Added version checks. Spaces vs tabs fixes. Fixed resource
    versioning.

Changes since v1:
  - Updated to protocol v2.

Signed-off-by: Carlos Garnacho <carlosg@gnome.org>
Reviewed-by: Michael Catanzaro <mcatanzaro@igalia.com>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

Support axis source, axis discrete, frame and axis stop events

[jonas: only send focus wl_pointer.frame if resource supports it]

Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>
Signed-off-by: Jonas Ådahl <jadahl@gmail.com>

Add a weston_pointer_axis_event

Use an event struct to pass axis events around. This helps dealing with the
upcoming axis discrete changes.

Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

compositor-rdp: update link to wheel delta explanation

The previous one is a 404 now.

Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Reviewed-by: Jonas Ådahl <jadahl@gmail.com>

clients: Fix compositor version check for WL_SURFACE_DAMAGE_BUFFER_SINCE_VERSION

Signed-off-by: Chris Michael <cpmichael@osg.samsung.com>
Reviewed-By: David Fort <contact@hardening-consulting.com>

clients: fix build on Linux

Cannot find out why stropts.h is needed and Linux doesn't support
streams anyway, so there is no stropts.h.

Signed-off-by: Olivier Fourdan <ofourdan@redhat.com>
Reviewed-by: Derek Foreman <derekf@osg.samsung.com>

gl-renderer: Fix comment typo

compositor: create_data_source(): Fix potential crash on OOM

Noticed this while working on primary selection, in the event we run out
of memory when trying to create a new data source, there's a chance
we'll fail on wl_resource_create() and crash from source->resource being
set to NULL.

Signed-off-by: Lyude <cpaul@redhat.com>
Reviewed-by: Marek Chalupa <mchqwerty@gmail.com>
Reviewed-by: Bryce Harrington <bryce@osg.samsung.com>

compositor: Make types consistent for bitshifting unsigned ints