* X11: General cleanup
This is almost entirely internal changes, and as usual, doesn't actually
fix any problems people have complained about.
- `XSetInputFocus` can't be called before the window is visible. This
was previously handled by looping (with a sleep) and querying for the
window's state until it was visible. Now we use `XIfEvent`, which blocks
until we receive `VisibilityNotify`. Note that this can't be replaced
with an `XSync` (I tried).
- We now call `XSync` at the end of window creation and check for
errors, assuring that broken windows are never returned. When creating
invisible windows, this is the only time the output buffer is flushed
during the entire window creation process (AFAIK). For visible windows,
`XIfEvent` will generally flush, but window creation has overall been
reduced to the minimum number of flushes.
- `check_errors().expect()` has been a common pattern throughout the
backend, but it seems that people (myself included) didn't make a
distinction between using it after synchronous requests and asynchronous
requests. Now we only use it after async requests if we flush first,
though this still isn't correct (since the request likely hasn't been
processed yet). The only real solution (besides forcing a sync *every
time*) is to handle asynchronous errors *asynchronously*. For future
work, I plan on adding logging, though I don't plan on actually
*handling* those errors; that's more of something to hope for in the
hypothetical async/await XCB paradise.
- We now flush whenever it makes sense to. `util::Flusher` was added to
force contributors to be aware of the output buffer.
- `Window::get_position`, `Window::get_inner_position`,
`Window::get_inner_size`, and `Window::get_outer_size` previously all
required *several* round-trips. On my machine, it took an average of
around 80µs. They've now been reduced to one round-trip each, which
reduces my measurement to 16µs. This was accomplished simply by caching
the frame extents, which are expensive to calculate (due to various
queries and heuristics), but change infrequently and predictably. I
still recommend that application developers use these methods sparingly
and generally prefer storing the values from `Resized`/`Moved`, as
that's zero overhead.
- The above change enabled me to change the `Moved` event to supply
window positions, rather than client area positions. Additionally, we no
longer generate `Moved` for real (as in, not synthetic)
`ConfigureNotify` events. Real `ConfigureNotify` events contain
positions relative to the parent window, which are typically constant
and useless. Since that position would be completely different from the
root-relative positions supplied by synthetic `ConfigureNotify` events
(which are the vast majority of them), that meant real `ConfigureNotify`
events would *always* be detected as the position having changed, so the
resultant `Moved` was multiple levels of misleading. In practice, this
meant a garbage `Moved` would be sent every time the window was resized;
now a resize has to actually change the window's position to be
accompanied by `Moved`.
- Every time we processed an `XI_Enter` event, we would leak 4 bytes via
`util::query_pointer` (`XIQueryPointer`). `XIButtonState` contains a
dynamically-allocated mask field which we weren't freeing. As this event
occurs with fairly high frequency, long-running applications could
easily accumulate substantial leaks. `util::PointerState::drop` now
takes care of this.
- The `util` module has been split up into several sub-modules, as it
was getting rather lengthy. This accounts for a significant part of this
diff, unfortunately.
- Atoms are now cached. Xlib caches them too, so `XInternAtom` wouldn't
typically be a round-trip anyway, but the added complexity is
negligible.
- Switched from `std::sync::Mutex` to `parking_lot::Mutex` (within this
backend). There appears to be no downside to this, but if anyone finds
one, this would be easy to revert.
- The WM name and supported hints are now global to the application, and
are updated upon `ReparentNotify`, which should detect when the WM was
replaced (assuming a reparenting WM was involved, that is). Previously,
these values were per-window and would never update, meaning replacing
the WM could potentially lead to (admittedly very minor) problems.
- The result of `Window2::create_empty_cursor` will now only be used if
it actually succeeds.
- `Window2::load_cursor` no longer re-allocates the cursor name.
- `util::lookup_utf8` previously allocated a 16-byte buffer on the heap.
Now it allocates a 1024-byte buffer on the stack, and falls back to
dynamic allocation if the buffer is too small. This base buffer size is
admittedly gratuitous, but less so if you're using IME.
- `with_c_str` was finally removed.
- Added `util::Format` enum to help prevent goofs when dealing with
format arguments.
- `util::get_property`, something I added way back in my first winit PR,
only calculated offsets correctly for `util::Format::Char`. This was
concealed by the accomodating buffer size, as it would be very rare for
the offset to be needed; however, testing with a buffer size of 1,
`util::Format::Long` would read from the same offset multiple times, and
`util::Format::Short` would miss data. This function now works correctly
for all formats, relying on the simple fact that the offset increases by
the buffer size on each iteration. We also account for the extra byte
that `XGetWindowProperty` allocates at the end of the buffer, and copy
data from the buffer instead of moving it and taking ownership of the
pointer.
- Drag and drop now reliably works in release mode. This is presumably
related to the `util::get_property` changes.
- `util::change_property` now exists, which should make it easier to add
features in the future.
- The `EventsLoop` device map is no longer in a mutex.
- `XConnection` now implements `Debug`.
- Valgrind no longer complains about anything related to winit (with
either the system allocator or jemalloc, though "not having valgrind
complain about jemalloc" isn't something to strive for).
* X11: Add better diagnostics when initialization fails
* X11: Handle XIQueryDevice failure
* X11: Use correct types in error handler
* Add get_inner_position for windows, prototypes for other platforms
* Fix linux builds
* Implement get_inner_position for osx
* Add get_inner_pos implementations for other platforms
* Fixed get_inner_position on macOS
* Corrected set_position on macOS
* Added CHANGELOG entry
* Add min/max size setting for win32 and wayland backends
* Implement dynamic min/max size on macos
* Add min/max size setting for x11
* Add empty functions for remaining platforms
* Improved min/max size setting for x11
* Added CHANGELOG entry for new min/max methods
* Added documentation for new min/max methods
* On win32, bound window size to min/max dimensions on window creation
* On win32, force re-check of window size when changing min/max dimensions
* Fix freeze when setting min and max size
This has been stubbed on all platforms other than X11. The X11 implementation has also been
revised to toggle correctly, as it was previously only able to remove decorations.
* Implement public API for high-DPI #105
* Recover get_inner_size_points and get_inner_size_pixels and change their implementation assuming get_inner_size() returns size in pixels
* Update changelog for high-DPI changes
* wayland: don't create a second event_queue
As each EventsLoop has its own context, this is no longer necessary.
* wayland: buffer events rather than direct dispatch
Changes the behavior of the event loop to first internally
buffer the events generated by the wayland handlers, and then
dispatch them to the client's closure.
- It simplifies the event loop logic
- It makes it possible for the user to call window methods such as
`set_title()` or `set_inner_size()` without causing a deadlock
* wayland: add is_ready() & fix protocol errors
Adds a `is_ready()` method to the windows to advertize
when it is legal to start drawing, and fix a few wayland
protocol mishandling in the process.
* Fix X11 screen resolution change using XrandR
The previous XF86 resolution switching was broken and everything
seems to have moved on to xrandr. Use that instead while cleaning
up the code a bit as well.
* Use XRandR for actual multiscreen support in X11
* Use actual monitor names in X11
* Get rid of ptr::read usage in X11
* Use a bog standard Vec instead of VecDeque
* Get rid of the XRandR mode switching stuff
Wayland has made the decision that apps shouldn't change screen
resolutions and just take the screens as they've been setup. In the
modern world where GPU scaling is cheap and LCD panels are scaling
anyway it makes no sense to make "physical" resolution changes when
software should be taking care of it. This massively simplifies the
code and makes it easier to extend to more niche setups like MST and
videowalls.
* Rename fullscreen options to match new semantics
* Implement XRandR 1.5 support
* Get rid of the FullScreen enum
Moving to just having two states None and Some(MonitorId) and then
being able to set full screen in the current monitor with something
like:
window.set_fullscreen(Some(window.current_monitor()));
* Implement Window::get_current_monitor()
Do it by iterating over the available monitors and finding which
has the biggest overlap with the window. For this MonitorId needs
a new get_position() that needs to be implemented for all platforms.
* Add unimplemented get_position() to all MonitorId
* Make get_current_monitor() platform specific
* Add unimplemented get_current_monitor() to all
* Implement proper primary monitor selection in X11
* Shut up some warnings
* Remove libxxf86vm package from travis
Since we're no longer using XF86 there's no need to keep the package
around for CI.
* Don't use new struct syntax
* Fix indentation
* Adjust Android/iOS fullscreen/maximized
On Android and iOS we can assume single screen apps that are already
fullscreen and maximized so there are a few methods that are implemented
by just returning a fixed value or not doing anything.
* Mark OSX/Win fullscreen/maximized unimplemented()!
These would be safe as no-ops but we should make it explicit so
there is more of an incentive to actually implement them.
* Don't use UNIX_BACKEND in Window2::new
* Move get_available_monitors and get_primary_monitor to EventsLoop
* Remove UNIX_BACKEND
* Restore choosing the Linux backend
* Return a XNotSupported for new_x11()
* Fix fullscreen example
* Rework MonitorId::get_native_identifier
* Try fix compilation
* Returns the monitor ID on wayland as well
* Try fix compilation
* Fix iOS compilation
Use the enum to make a single fullscreen API that's much more
consistent. Both set_fullscreen() and with_fullscreen() take the
same enum and support all the variations so you can build the window
however you want and switch between the modes at runtime.
There are two kinds of fullscreen. One where you take over the whole
output the other where you just set the window size to the screen
size and get rid of decorations. The first one already existed,
implement the second which is more common for normal desktop apps.
Use an enum to consolidate all the fullscreen states.
This removes the need for the EventsLoop::interrupt method by inroducing
a ControlFlow type. This new type is to be returned by the user's
callback and indicates whether the `EventsLoop` should continue waiting
for events or break from the loop.
Only the wayland, x11 and api_transition backends have been updated so
far, and only the wayland backend has actually been tested.
X11 and Wayland implementations are now half implemented, however both
still do not correctly break from the inner blocking event dispatch
functions when `wakeup` is called, which they should do.
This expands input events to represent sub-pixel mouse positions, devices responsible for generating events, and raw
device-oriented events. The X11 back end is refactored to make full use of the new expressiveness. Other backends have
had new functionality minimally stubbed out, save for the macos backend which already supports sub-pixel mouse
positions.
