Prior to this change, the virtual mouse update was restricted to
updating once every `mouse_delay` microseconds, and the allowed values
were a small fixed set. Some users reported that even at the highest
speed, the speed was slower than they were used to (with a high DPI
mouse).
This change modifies the speed calculation algorithm as follows. It
keeps the slowest speed to refresh the mouse every 70 ms. As the speed
increases, the refresh rate drops by 5 ms for every increment in speed,
until the refresh rate caps at once every 10 ms. Beyond that, a
multiplicative factor begins to take effect, with each speed increase
adding 0.25 to the factor. That is, speed 13 would multiply the axis
components by 1.25 _and_ refresh every 10 ms. Speed 14 would bump the
factor to 1.50, speed 15 to 1.75, and so on, until the factor tops out
at 6.0.
Prior to this change, the mouse delays were between 50 ms to 250 ms,
with a difference of 50 ms between steps. Unfortunately, this was too
slow at lower speeds, therefore, the delays have been changed to vary
from 30 ms to 70 ms with a difference of 10 ms between steps. This gives
a much smoother mouse response.
`runstatedir` is only available in Autoconf 2.70, but unless the
distribution is a bleeding edge system, it most likely uses Autoconf
2.69. That said, several major distributions have backported runstatedir
support to the older versions, hiding the issue. See #35.
This change replaces all references to runstatedir to use
$localstatedir/run instead, which is what is recommended by the autoconf
manual.
This also updates the build instructions to add --localstatedir and
--sysconfdir. This is because the lack of the options would have them
default to `$(prefix)/var` and `$(prefix)/etc` respectively, and with
prefix set to `/usr`, these would be the bogus directories `/usr/var`
and `/usr/etc`.
Prior to this change, we were treating ERROR_NO_DEVICE as if the hidapi
library itself would return such a code. However, that is not the case,
and we should be treating any error condition as a critical issue and
treat it as if the device was disconnected. The worst case scenario is
that it would have to re-enumerate the HID device list and reopen the
joystick.
Prior to this change, the button change events were only happening on
periodic intervals corresponding to the change in the mouse REL_X and
REL_Y values. However, this has the issue that it tends to miss a few
events, especially those related to the scroll wheel.
This change reports button and wheel events immediately when receiving
the report, but it leaves the motion to be updated by the thread.
This change adds a separate thread to initialize and read reports from
the supported X52 device. This will then process and raise input events
for a virtual device.
Prior to this change, the build would fail on macOS systems because the
evdev sources were only included on Linux systems, and macOS does not
have evdev/libevdev. By separating out the configuration and update
threads, this should build on macOS, but the configuration would be
ignored.
This change adds the configuration and build related changes for
supporting the virtual mouse. Subsequent commits will add support for
reading the IO interface and translating it to mouse commands.
Prior to this change, the device check was sending a vendor specific
command with wIndex and wValue both set to 0 every 50 ms. On some
systems, this was causing issues with the joystick flapping the state,
and reporting weird values from the stick, and generally sluggish
response.
This change uses the updated libx52 library which uses the hotplug
notification to determine if the device is connected, and should resolve
the issues seen.
Fixes#33
Prior to this commit, the libx52_is_connected API was simply checking if
the device handle was non-NULL. However, this was insufficient, since
the device disconnection would not reset the handle, and was relying on
the daemon to manually disconnect.
The libusb hotplug API provides functionality to register a callback on
device insertion/removal. libx52 only registers for removal, and will
automatically disconnect the device on receiving the callback. This also
modifies libx52_is_connected to fallback to checking if the kernel
driver is active if the linked libusb does not support hotplug (unlikely).
Finally, this commit adds support for the new hotplug related functions
to the libusbx52 preload library. While the preload library doesn't
actually support hotplug, it is sufficient to pretend that it does.
Prior to this change, if the clock thread is disabled, then
disconnecting and reconnecting the X52 device would cause the daemon to
not detect the transition. As a result, the daemon would stay in a state
where it thinks the device is still connected, and therefore, not
actually apply any of the saved configuration, until it received a
SIGHUP to refresh the configuration.
This change adds a routine that sends a dummy vendor command. This
vendor command does nothing on my X52 Pro (VID 06a3, PID 0762), but
serves as a check to see if the daemon can send vendor commands to the
device. If the device is indeed disconnected, then that is a sufficient
indicator to disable the update thread and re-enable the acquisition
thread.
POSIX.1-2004 requires that localtime() is required to behave as if
tzset() was called, but there is no such requirement imposed upon
localtime_r(). Therefore, we need to call tzset ourselves to ensure that
the timezone fields are updated.
Prior to this change, the user needed to install inih as a dependency,
either from the distribution repositories, or from source. On some
platforms (notably macOS), inih is not available prepackaged, and must
be installed by the user. This tends to cause needless friction.
This change imports the ini.c and ini.h files from the upstream inih
repository into the X52 source tree. This will allow us to build the
repository on any system with the original set of dependencies, and not
have to force the user to install packages themselves.
0256807 Use strrchr only if the compiler supports __builtin_strrchr
204aade Use __builtin_strrchr to get file base name
aa0a554 Remove the use of config.h
8983bf8 Ignore test and benchmark programs
git-subtree-dir: lib/pinelog
git-subtree-split: 02568074170f0725196de4e52450db024cc39895
A missing comma at the end of the STRINGIFY line was causing builds with
clang to have segfaults with the default configuration. It turned out
that due to the missing comma, the N_("Unknown LED state %d") and
N_("off") parameters were getting merged into a single parameter by
clang, but interestingly, not by GCC.
As a result, when building with clang, the array is "on", "red",
"amber", "green" - note the missing "off" at the beginning of the array.
This causes clang generated builds to segfault when attemping to log a
trace message when configuring LED A (which defaults to green, and the
index of LIBX52_LED_STATE_GREEN exceeds the array bounds).
AC_CHECK_HEADERS and AC_SEARCH_LIBS do not abort when the header or
library respectively are not found. Since inih is a hard requirement,
make sure that the action-if-not-found fields call AC_MSG_ERROR.
Fixes#32
This change changes the parser element in the CFG macros to be a "type"
element instead. This is handled in the config parser source, where the
macro definition appends `_parser` to the type field. This allows us to
(in the future) add a `_dump` function to dump the configuration to a
file.
nirenjan