Remove the effectively unused `initial_suspend_` variable. It wasn't
doing anything useful, because it was set to false on the first round
through the event loop. Also remove the unused reference to the internal
AVR USB core variable `_usbSuspendState`.
Saves 18 bytes of flash on AVR, and 32 on GD32.
Signed-off-by: Taylor Yu <tlyu@mit.edu>
Fix a race condition in `sketch.mk` that tends to pop up when doing
`make -j smoke-sketches`. Ignore failures from `install -d`, because
they seem to only occur due to a TOCTOU condition in `install`.
Signed-off-by: Taylor Yu <tlyu@mit.edu>
The difference between the AVR and the GD32 implementation was the way we
checked if the device is suspended, otherwise the logic was exactly the same,
duplicated.
To remove the need for duplicating the same code for every architecture we
support, lift out the suspension check instead. This way the core logic becomes
shared between all of them.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
detecting the device port at evaluation time rather than execution time.
This meant that we were doing the "where is the board" check for any
compilation target, even if we'd never flash.
Arduino's board probing is somewhat heavyweight and can take a couple of
seconds.
We move that logic into a shell expression executed at runtime.
On my laptop, this shaves 10 seconds off make -j 9 simulator tests,
which is pretty nice since that used to take about 30 seconds.
But on a plain `make simulator-tests`, it shaved a full minute from the
2 minute and 30 second runtime.
When flushing the on-tap action, we need to use `handleKeyEvent`, because our
address may not be valid, and `handleKeyswitchEvent` will not perform the action
if the address is invalid.
This addresses keyboardio/Chrysalis#1055.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
This reverts commit 1d8eb6d2e2.
Rainbow effects on the Model 100 had some harsh transitions,
especially from red/orange to yellow.
@obra reported that the gamma implementation was disabled on the
Model 100 earlier because it broke the firmware so badly that
the keyboard was unusable for typing. This no longer seems to be
the case, possibly due to changes in the Arduino core and/or
<pgmspace.h> compatibility.
Signed-off-by: Taylor Yu <tlyu@mit.edu>
Updates for the FocusSerial documentation, showing the new patterns, and
documenting the new methods.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
With the new patterns, `Focus.inputMatchesCommand(command, cmd)` felt wrong, so
this patch renames the `command` argument of `onFocusEvent()` to `input`, to
better match what it really is.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
In the past, we were using `Focus.handleHelp()` to see if we're handling a
`help` command, and print the commands supported by the handler. We also used
`strcmp_P` directly to compare (parts of) our input against command supported by
the handler. This approach works, but had multiple major disadvantages: it
duplicated strings between `handleHelp` and the `strcmp_P` calls, and it relied
on fragile substring pointers to save space.
To replace all that, this patch implements a different approach. Help handling
is split between a check (`Focus.inputMatchesHelp()`) and a
reply (`Focus.printHelp()`), the latter of which takes a list of `PSTR()`
strings, rather than one single string. This allows us to reuse the same
strings for comparing against the handler's input.
The new approach no longer uses the fragile substring pointers, nor does it use
`strcmp_P` directly, but goes through a wrapper (`Focus.inputMatchesCommand()`)
instead.
These changes lead to a more readable pattern. While we do use longer strings as
a result, there is less duplication, and the new patterns also require less
code, so we end up with saving space, at least on AVR devices.
The old methods are still available and usable, but they're deprecated.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
`Serial.read()` is unbuffered, and will return `-1` (which we cast to an
unsigned char, so 255) immediately if there is no data in the incoming buffer.
This is unlike every other kind of read we do, which use `parseInt()`, and are
thus buffered reads with a timeout.
The problem with returning `-1` immediately is that Chrysalis sends data in
chunks, so if we end up trying to read a char at a chunk boundary, we'll end up
reading -1s, which we treat as valid data, and cast it to an unsigned char,
completely throwing off the protocol in the process.
By using `readBytes()`, we have a one second window during which more data can
arrive, and as such, is consistent with the rest of our reads.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
The new plugin enables configuring some aspects of SpaceCadet through Focus: the
current mode, and the global timeout. This is makes it possible to ship firmware
with SpaceCadet included, disabled by default, but still allow one to enable it
without having to map and tap the enable key.
The settings are also persisted into storage.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
The example in the documentation was referring to a function that does not
exist. Correct it to use the one that does.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
Fall back to `dash` on macOS, because `bash` randomly drops serial
input, causing the tool to hang.
Flush the command buffer before sending the requested command. A
failed upload session can cause characters to remain in the command
buffer.
Redirect stdin instead of using a separate file descriptor. Also do
this before running `stty`. This allows the `stty` settings to
actually take effect on macOS, which seems to reset the termios
state of serial devices upon the last close of the device.
Tested on macOS 10.15 and Ubuntu 20.04.
Signed-off-by: Taylor Yu <tlyu@mit.edu>
Previously, a newline that wasn't read with the rest of the command in
a single event would get appended to the command buffer instead of
remaining in the "peek" buffer, so the command dispatch would never
fire.
Stash a space delimiter instead of storing it in the buffer, only to
overwrite it later.
Fix an off-by-one error that could cause a buffered command to not be
null-terminated. This probably didn't cause a problem in practice,
because Focus plugins mostly did a `strcmp` against a fixed string that
is smaller than the command buffer size.
Signed-off-by: Taylor Yu <tlyu@mit.edu>
The new plugin provides a Focus-based interface for setting custom layer names.
The layer names aren't used internally, they're purely for use by host-side
applications.
This addresses the Kaleidoscope-side of keyboardio/Chrysalis#3.
Signed-off-by: Gergely Nagy <algernon@keyboard.io>
Alpha Chen
Gergely Nagy
Norwin
Jesse Vincent
Taylor Yu
Gergely Nagy
Jesse Vincent