Document changes and release 0.7.1-1~pureos1 into byzantium

See merge request Librem5/debs/pkg-phoc!1

phoc Debian release 0.7.1-1

Update to phoc 0.7.1

See merge request DebianOnMobile-team/phoc!5

Latest supported is 0.12.0.

Update to upstream version '0.7.1'
with Debian dir daacacf251980b290f8d1669ba3a1bc3f52e6bed

Guido Gunther authored 3 years ago and Sebastian Krzyszkowiak committed 3 years ago

Until we allow to bind all keys let's allow for these as well since
this block includes e.g. PrintScreen for screenshots.

Passing an invalid keybinding currently doesn't end well:

 ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000 (pc 0xaaaaadbff580 bp 0xffffc1f94d00 sp 0xffffc1f94d00 T0)
 The signal is caused by a READ memory access.
 Hint: address points to the zero page.
    #0 0xaaaaadbff580 in phosh_private_keyboard_event_accelerator_is_registered ../src/phosh.c:131
    #1 0xaaaaadbff7d8 in phosh_private_accelerator_already_subscribed ../src/phosh.c:149
    #2 0xaaaaadbffa68 in phosh_private_keyboard_event_grab_accelerator_request ../src/phosh.c:184
    #3 0xffffafb47044  (/lib/aarch64-linux-gnu/libffi.so.7+0x6044)
    #4 0xffffafb4676c  (/lib/aarch64-linux-gnu/libffi.so.7+0x576c)
    #5 0xffffb0031a1c in wl_closure_invoke ../src/connection.c:1018
    #6 0xffffb002cd7c in wl_client_connection_data ../src/wayland-server.c:432
    #7 0xffffb002fa64 in wl_event_loop_dispatch ../src/event-loop.c:1027
    #8 0xaaaaadbdce38 in wayland_event_source_dispatch ../src/server.c:42
    #9 0xffffb0273bc4 in g_main_dispatch ../../../glib/gmain.c:3325
    #10 0xffffb0273bc4 in g_main_context_dispatch ../../../glib/gmain.c:4043
    #11 0xffffb0273e58 in g_main_context_iterate ../../../glib/gmain.c:4119
    #12 0xffffb02741ac in g_main_loop_run ../../../glib/gmain.c:4317
    #13 0xaaaaadbdc7f0 in main ../src/main.c:137
    #14 0xffffafd8c214 in __libc_start_main ../csu/libc-start.c:308
    #15 0xaaaaadbdb610  (/usr/bin/phoc+0x1b610)

Always set and check both keybinding results. This tests that
we don't invoke the wrong callback accidentally.

Currently one needs to look up what's tested in an array but each most
of the keys are only used a single time. So drop that array and use a
define for the single key that is used multiple times so it's clear we
just add an accelerator.

The current accelerator tests try to encode two boolean values
into two variables making it a bit hard to read. Use a single
enum instead.

Guido Gunther authored 4 years ago and Sebastian Krzyszkowiak committed 4 years ago

Don't try to go on without a cursor (which can only happen when calloc
fails). wlroots also only returns NULL in case of memory problems.

This allows to remove the NULL checks elsewhere

Update to upstream version '0.7.0'
with Debian dir 68d1dac2d0627e82ce0cf052b2c6ed4115885b57

louib authored 4 years ago and Sebastian Krzyszkowiak committed 4 years ago

This commit adds the `build-dep` one-liner to install the dependencies
by leveraging the debian control file. The instructions are from
the same section in phosh's README. Also fixes a typo in the building
section, which seems to be from phosh.

See https://github.com/swaywm/wlroots/pull/1919

Move the cursor under the finger until touch points receive their own
cursors.

Otherwise the old cursor stays there until the pointer gets moved.

This also removes the destroy signal listener of the pending surface.

There's no need to redraw unactivated clients since they're
hidden anyway, so that just wastes resources. It also looks
bad in phosh activity chooser.

Sebastian Krzyszkowiak authored 4 years ago and Guido Gunther committed 4 years ago

The actual window geometry should stay in the same position
regardless of surface geometry being changed.

Sebastian Krzyszkowiak authored 4 years ago and Guido Gunther committed 4 years ago

We're interested in restoring the actual window into the same
position, regardless of whether the shadows around it happened
to change their size meanwhile.

Sebastian Krzyszkowiak authored 4 years ago and Guido Gunther committed 4 years ago

This makes it consistent with PHOC_EDGE_SNAP_THRESHOLD and actually makes
more sense.

This follows the convention used by other projects like phosh.