* suspend: suppress wake up keypress
Waking the host from suspend is done by pressing any key on the
keyboard, the regular key codes assigned to the keys are not important
and must not be sent - otherwise they usually end up in password prompts
as ghost characters that have to be deleted again. This commit adds
suppression for all keys pressed at the time of wake up. Once a key is
released it functions as a regular key again.
Signed-off-by: Stefan Kerkmann <karlk90@pm.me>
* suspend: update wake up matrix after wake up delay
If USB_SUSPEND_WAKEUP_DELAY is set, the keyboard sleeps during wake up -
which can be up to multiple seconds. To handle key presses and releases
in that time frame we have to handle the following cases:
1. Key not pressed before suspend, and not pressed after wakeup → do
nothing (normal case).
2. Key not pressed before suspend, but pressed after wakeup → set the
wakeup_matrix bit to 1 (so that the press and release events would be
suppressed).
3. Key pressed before suspend, but not pressed after wakeup → do nothing
(the release event will be generated on the first matrix_task() call
after the wakeup).
4. Key pressed before suspend, and still pressed after wakeup → do
nothing (the release event will be generated some time later).
Signed-off-by: Stefan Kerkmann <karlk90@pm.me>
Co-authored-by: Sergey Vlasov <sigprof@gmail.com>
* keyboards: anavi: macropad8: disable snake and rgb_test effects
...to shrink the binary size.
* feat: adds a transmit and receive i2c method
* fix: address the i2c transmit and receive length on u16
* Add AVR/LUFA implementation
Didn't add a progmem version, since that would only apply to receive.
Figured it wasn't worth it, but can add.
* Rearrange order of functions
* Add docs
* Fix doc gen error
* Fix lint issues
* fix more lint issues
* Only perform key and mod remapping in keycode_config() and mod_config() when MAGIC_ENABLE is defined.
* If not set, these functions now return the original keycode or modifier unchanged.
* Reduces firmware size, and unnecessary code when MAGIC_ENABLE is not enabled.
* Removed space saving suggestion with magic functions from squeezing AVR documentation
* Indicate <keyboard>.h LAYOUT() deprecation in understanding_qmk.md
Initial documentation uses <keyboard>.json to define matrix pin definitions, rather than having users #define the LAYOUT() macro in <keyboard>.h - This change brings this docs inline with the [porting_guidelines](https://docs.qmk.fm/porting_your_keyboard_to_qmk), QMK MSYS will also throw an error if this is attempted.
* Update understanding_qmk.md
enclosed '<' using backtick
* style guideline, now builds correctly
prior version indicates directory, new indicates config. still the same idea.
I built three of these and was confused as to why my numpad 0 wasn't
working. Eventually, I consulted this website to see the key matrix for
the Model M
https://sharktastica.co.uk/sims/matrix?kb=enhanced
`kp_0` is in the matrix at [0, 12], not [0, 11]. Simple off by one
error. I fixed this, flashed to all my keyboards, and it worked.
Co-authored-by: Kay Barkbark <kay@laptop>
Stefan Kerkmann