* Generate api docs from source code
* Add a bunch of doxygen comments
* more doxygen comments
* Add the in-progress api docs
* script to generate docs from travis
* Add doc generation to the travis job
* make travis_docs.sh commit the work it does
* make sure the docs script exits cleanly
This is an inelegant hack for #2522 but makes things work. Basically we give `action.c` a chance to handle the hold event early so that we can swap the keyboard for later keys. Later, to allow the hold to happen again quickly we nuke the key record so that tapping is reset. I tried to find a cleaner way, honestly.
* add breathing to bananasplit
* backlight breathing overhaul
* fix the backlight_tick thing.
* fix for vision_division backlight
* fix a few keymaps and probably break breathing for some weirdly set-up boards.
* remove BL_x keycodes because they made unreasonable assumptions
* some fixes for BL keycodes
* integer cie lightness scaling
* use cie lightness for non-breathing backlight and make breathing able to reach true max brightness
* add RETRO_TAP: tap anyway, even after TAP_TERM, if no interruption
* consistent variable name
* add option doc
* change name for consistency
* make RETRO_TAPPING default to off
* Add 80ms delay for KC_CAPS when used as a tap key
Workaround for the macOS caps lock delay
* Revert "Increase TAPPING_TERM for the Clueboard"
This reverts commit a74e69e9fa.
The oneshot cancellation code do not depend on the
action_tapping_process and since process_record get called via the
action_tapping_process logic moved the oneshot cancellation code into
the action_exec function just before the action_tapping_process call
After setting a ONESHOT_TIMEOUT value, the oneshot layer state would
not expire without an event being triggered (key pressed). The reason
was that in the process_record function we would return priort to
execute the process_action function if it detected a NOEVENT cycle. The
process_action contained the codes to timeout the oneshot layer state.
The codes to clear the oneshot layer state have been move just in
front of where we check for the NOEVENT cycle in the process_record
function.
This adds an action, `ACTION_SWAP_HANDS`, that swaps the the keys on the keyboard across a keymap-defined hemisphere in order to support one-hand typing without requiring a separate one-handed layer. See updated `doc/keymap.md` for more information.
* implements leader key for planck experimental
* allows override of leader timeout
* adds ability to use the leader key in seq
* fixes leader keycode
* adds chording prototype
* fixes keycode detection
* moves music mode to quantum.c
* disables chording by default
* adds music sequencer functionality
* implements audio/music functions in quantum.c
* splits up process_action to allow independent processing of actions
* merging?
This commit is mostly a cherry-pick from `ahtn` at
https://github.com/tmk/tmk_keyboard/pull/255.
These are the changes:
* Adds ACTION_LAYER_ONESHOT
* Adds ONESHOT_TAP_TOGGLE
* Mentions sticky keys in the docs on oneshot.
on mod keys, register LGUI, LSFT etc. as normal mods
instead of weak mods:
- they won't be cleared when pressing another key (#188)
- they won't be cleared by layer switching
- LSFT(KC_LGUI) will now have the same behavior as LGUI(KC_LSFT)
If one wants to temporarily disable the key cache (for example because
it interferes with a macro), `disable_action_cache` must be set to
`true`. `process_action_nocache` is a simple wrapper doing just that for
a single call.
- new macro_mods bit field for mods applied by macros
- weak_mods now only used for ACT_{L,R}MODS (i.e. LSFT, RSFT, LCTL etc.)
- clear the _weak_ mods on every key *pressed* such that LSFT etc.
can no more interfere with the next key
