Instead of running the tests right after compiling them, compile them all first,
then run them one after another. This way the test output is kept much closer
together, which is easier to glance over.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
The workaround was put in place as an attempt to get virtual builds going on
macOS, natively. It wasn't enough, and it doesn't work, so lets drop it.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
We do a number of things here:
- Bump the distro version from Trusty to Bionic, because we need newer tools.
- Install cmake.
- Unset CC, because it interferes with our virtual builds.
- Run the googletest-based suite before building the examples
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
This adds a googletest-based test harness (with googletest pulled into
`testing/googletest`, so we have a fixed state of it), and a few test
cases that demonstrate its use.
Original work by Eric Paniagua in #898, with minor cleanups by Gergely
Nagy.
Signed-off-by: Eric Paniagua <epaniagua@google.com>
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
We do not need `getShortName()` anymore: setting the shortname is now done via
device properties instead. Doing it with `getShortName` only results in
duplicated symbols.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
We can pass arguments to the entrypoint from the `docker run` commandline, so we
do not need to do that via an environment variable. This way, we're an
environment variable and an `eval` shorter.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
To make it easier to reproduce things, and to help build in a clean environment,
this adds a thin Dockerfile that has Arduino and arduino-cli pre-installed, and
- along with the `bin/run-docker` script - is set up so that one can easily run
arbitrary commands in the context of the current bundle and Kaleidoscope.
The first run will take a while, because docker will build the image. Subsequent
runs will use the cache.
To use: `bin/run-docker make`, for example. Any argument passed to the
`bin/run-docker` will be eval-ed within the container, and will run there.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
Previously, rolling over from one System Control key to another would
cause the second one to be released as soon as the first one was
released, because the empty release report would be sent
unconditionally on release of any System Control key.
This change stores the value of the last System Control key
pressed. When a System Control key is released, it first checks to see
that the released key's keycode matches the last one pressed before
sending the empty report.
Signed-off-by: Michael Richters <gedankenexperimenter@gmail.com>
Since the PR was made for the ButterStick and FaunchPad, ATMega32U4Keyboard has
had a breaking API update. Follow up on that now.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
Key masking was a bandaid, and we have better ways to achieve the same thing
now. All current users have been switched over to different methods now, so lets
deprecate the masking.
We only put the `DEPRECATED` label on the `maskKey` method, because the rest are
used internally too, and we do not want to emit warnings for those.
Fixes#884.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
We want to remove the use of key masking, so instead of masking the key when
escaping a OneShot, map it to `NoKey` instead, and continue doing so until
released. Which is effectively what masking did, but localized and simpler.
Doing this will make our cache have `NoKey` for the key until release, and we'll
avoid sending unintended Escape keycodes, without having to use the global
masking functions.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
We introduced the masking to avoid sending extra keys when the mapped key
changes prior to release - but since the introduction of the caching mechanism,
we no longer need to do this.
However, for the caching to work the way we want it to, we need to map the key
to `NoKey` once, upon interrupting.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
squash! TapDance: Do not mask interrupting keys anymore
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
To make their purpose clearer, rearrange our state: we now have the row-based
array on the top level, instead of every member being an array on its own. The
name of the state variable was changed to `matrix_state_`, to reflect its
purpose. This also allowed us to have its members be named `current`,
`previous`, `debouncer` and `masks`.
All devices using these APIs, and the documentation were updated accordingly.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
We want `readCols` as a separate function, so we can tell the compiler to apply
different optimizations to it.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
Jesse Vincent
Gergely Nagy
Eric Paniagua
Michael Richters
Eric Paniagua
Gergely Nagy
Romain Heller
tokencoder