* UART and BLE/GATT transports for a protobuf encoded RPC
request/response protocol.
* Custom framing protocol is used to frame a give message.
* Requests/responses are divided into major "subsystems" which
handle requests and create response messages.
* Notification support, including mapping local events to RPC
notifications by a given subsystem.
* Meta responses for "no response" and "unlock needed".
* Initial basic lock state support in a new core section, and allow specifying
if a given RPC callback requires unlocked state or not.
* Add behavior subsystem with full metadata support and examples of
using callback to serialize a repeated field without extra stack space needed.
Co-authored-by: Cem Aksoylar <caksoylar@users.noreply.github.com>
* Studio will use stable layer IDs to refer to layers, so that
layer reordering doesn't affect behavior bindings, so update to match.
* Fix a few layer metadata entries that missed being refactored.
* Because settings load is delayed now, enable external
power on init, and let it be disabled on settings load
later, to ensure power is on early for
other peripheral initialization.
Fixes: #2361
* Add bindings to allow creating multiple physical layouts that specify
their key's physical attributes, and the matching matrix transform
and dependant kscan to use.
* Synthesize a basic physical layout if none specified, for backwards
compatibility.
* Update matrix transform API to explicitly pass in the selected transform
to the API instead of using a fixed chosen transform.
* Move kscan subscription and handling into the physical layout code, so
that selecting a different physical layout at runtime can also use the
correct kscan instance.
* Add `physical_layouts.dtsi` file to include so you can use the
pre-configured `&key_physical_attrs` for adding you layout keys.
* Avoid doing duplicate calls to setings_load_subtree, which iterates
NVS fully each time under the hood, and instead use on settings_load
later in the lifecycle.
* Add a new feature for tracking a given behavior by a new concept
of a "behavior local ID" which is a stable 16-bit identifier for
a given behavior, that is resilient to new behaviors being added
and requires no additional work on the part of the behavior
authors.
* Add implementations for either settings lookup table, or CRC16
hashing of behavior device names for generating behavior local
IDs.
This adds a new config value `ZMK_HID_SEPARATE_MOD_RELEASE_REPORT`
where, if enabled, the report for a key release is sent separately to
the accompanying modifier release signals, which are then sent in a
second report.
This fixes an issue where certain applications are unable to work with
implicitly modified keys (e.g. colon) due to them registering the
modifier release prior to the actual key release.
Have tested this on my personal keyboard and `wev` now shows the signals
in the correct order.
=> **Previously:** ```LSHIFT (pressed) -> colon (pressed) -> LSHIFT
(released) -> **semi**colon (released)```
=> **Now:** ```LSHIFT (pressed) -> colon (pressed) -> colon (released)
-> LSHIFT (released)```
(This time without accidental files)
* For upcoming ZMK studio work, make a set of rich metadata available
to provide a friendly name for a behavior, and allow super flexible
descriptions of the parameters the behaviors take.
* Add ability to validate a zmk_behavior_binding against
the behavior metadata available.
* In order to be sure the rest of the system is fully ready before
intializing, because init may result in immediate events being
triggered when used with toggle direct kscan inner devices.
* Add dedicated init priority for the sideband kscan.
* Refactor sideband code for clarity.
* Tweaks to direct kscan for clarity.
* Make sideband behavior row optional for brevity.
* Allow overriding ZMK Uno sideband behaviors.
* Instead of gpio key behavior trigger, add new kscan driver that
decorates/wraps a given kscan driver and will invoke basic system
behavior assigned to a given row + column, without the need for keymap
mapping in the matrix transform, bypassing keymaps entirely.
* Better naming for gpio-key behavior triggers.
* Tweaks to scanned behavior trigger to avoid bad semaphore use,
and reduce chance of issues with slowly scanned matrixes.
* Various code cleanups of style issues.
Initial work on a soft on/off support for ZMK. Triggering soft off
puts the device into deep sleep with only a specific GPIO pin
configured to wake the device, avoiding waking from other key
presses in the matrix like the normal deep sleep.
Co-authored-by: Cem Aksoylar <caksoylar@users.noreply.github.com>
