194 lines
7.4 KiB
Markdown
194 lines
7.4 KiB
Markdown
# Overview
|
|
|
|
This is my personal userspace file. Most of my code exists here, as it's heavily shared.
|
|
|
|
## Custom userspace handlers
|
|
|
|
Specifically QMK works by using customized handlers for everything. This allows for multiple levels of customization.
|
|
|
|
`matrix_scan` calls `matrix_scan_quantum`, which alls `matrix_scan_kb`, which calls `matrix_scan_user`.
|
|
`process_record` calls a bunch of stuff, but eventually calls `process_record_kb` which calls `process_record_user`
|
|
The same goes for `matrix_init`, `layer_state_set`, `led_set`, and a few other functions.
|
|
|
|
All (most) `_user` functions are handled here instead. To allow keyboard specific configuration, I've created `_keymap` functions that can be called by the keymap.c files instead.
|
|
|
|
This allows for keyboard specific configuration while maintaining the ability to customize the board.
|
|
|
|
My [Ergodox EZ Keymap](https://github.com/qmk/qmk_firmware/blob/master/layouts/community/ergodox/drashna/keymap.c#L297) is a good example of this, as it uses the LEDs as modifier indicators.
|
|
|
|
## Keyboard Layout Templates
|
|
|
|
This borrows from @jola5's "Not quite neo" code. This allows me to maintain blocks of keymaps in the userspace, so that I can modify the userspace, and this is reflected in all of the keyboards that use it, at once.
|
|
|
|
This makes adding tap/hold mods, or other special keycodes or functions to all keyboards super easy, as it's done to all of them at once.
|
|
|
|
The caveat here is that the keymap needs a processor/wrapper, as it doesn't like the substitutions. However, this is as simple as just pushing it through a define. For instance:
|
|
|
|
`#define LAYOUT_ergodox_wrapper(...) LAYOUT_ergodox(__VA_ARGS__)`
|
|
|
|
Once that's been done and you've switched the keymaps to use the "wrapper", it will read the substitution blocks just fine.
|
|
|
|
Credit goes to @jola5 for first implementing this awesome idea.
|
|
|
|
|
|
## Custom Keycodes
|
|
|
|
Keycodes are defined in the drashna.h file and need to be included in the keymap.c files, so that they can be used there.
|
|
|
|
A bunch of macros are present and are only included on boards that are not the Ergodox EZ or Orthodox, as they are not needed for those boards.
|
|
|
|
Included is a custom macro for compiling my keyboards. This includes the bootloader target (`:teensy`, `:avrdude`, or `:dfu`), and keeps RGBLIGHT, AUDIO and/or FAUXCLICKY enabled, if it previously was (regardless of the rules file).
|
|
|
|
This also includes a modified RESET keycode as well, that sets the underglow to red.
|
|
|
|
## Layer Indication
|
|
|
|
This uses the `layer_state_set_*` command to change the layer color, to indicate which layer it is on. This includes the default keymap, as well.
|
|
|
|
Since this is done via userspace, it is the same between all systems.
|
|
|
|
Additionally, there is a custom keycode to toggle layer indication. And all RGB keycodes disable layer indication by default, as well. This way, I can leave special effects doing when I want.
|
|
|
|
Also. I use `rgblight_sethsv` since it works with animation modes (that support it).
|
|
|
|
## Diablo Layer
|
|
|
|
This layer has some special handling.
|
|
|
|
When Tap Dances are enabled, this layer has the ability to "spam" keypresses.
|
|
|
|
For instance, tapping the TD "1" twice causes the layer to hit "1" ever 1 second (appoximately). This is useful for auto-hotkeying skills (such as bone armor or devour).
|
|
|
|
Tappind once disables this, and switching layers temporarily disables this, until you switch back to the layer.
|
|
|
|
For critics that think this is cheating, search "diablo 3 num lock auto cast". This is just a simpler method, since I no longer own a normal (non QMK) numpad.
|
|
|
|
## Secret Macros
|
|
|
|
With help from gitter and Colinta, this adds the ability to add hidden macros from other users.
|
|
|
|
First, I have several files that are hidden/excluded from Git/GitHub. These contain everything needed for the macros. To hide these files, open `.git/info/exclude` and add `secrets.c` and `secrets.h` to that file, below the comments.
|
|
|
|
And this requires `KC_SECRET_1` through `KC_SECRET_5` to be defined in your `<name>.h` file to define the keycodes for the new macros.
|
|
|
|
|
|
### .git/info/exclude
|
|
|
|
```
|
|
# git ls-files --others --exclude-from=.git/info/exclude
|
|
# Lines that start with '#' are comments.
|
|
# For a project mostly in C, the following would be a good set of
|
|
# exclude patterns (uncomment them if you want to use them):
|
|
# *.[oa]
|
|
# *~
|
|
/users/drashna/secrets.c
|
|
/users/drashna/secrets.h
|
|
```
|
|
|
|
Then you can create these files:
|
|
|
|
### secrets.c
|
|
|
|
```c
|
|
#include "drashna.h" // replace with your keymap's "h" file, or whatever file stores the keycodes
|
|
|
|
#if (__has_include("secrets.h") && !defined(NO_SECRETS))
|
|
#include "secrets.h"
|
|
#else
|
|
// `PROGMEM const char secret[][x]` may work better, but it takes up more space in the firmware
|
|
// And I'm not familiar enough to know which is better or why...
|
|
static const char * const secret[] = {
|
|
"test1",
|
|
"test2",
|
|
"test3",
|
|
"test4",
|
|
"test5"
|
|
};
|
|
#endif
|
|
|
|
bool process_record_secrets(uint16_t keycode, keyrecord_t *record) {
|
|
switch (keycode) {
|
|
case KC_SECRET_1 ... KC_SECRET_5: // Secrets! Externally defined strings, not stored in repo
|
|
if (!record->event.pressed) {
|
|
clear_oneshot_layer_state(ONESHOT_OTHER_KEY_PRESSED);
|
|
send_string_with_delay(secret[keycode - KC_SECRET_1], MACRO_TIMER);
|
|
}
|
|
return false;
|
|
break;
|
|
}
|
|
return true;
|
|
}
|
|
```
|
|
|
|
### secrets.h
|
|
|
|
```c
|
|
static const char * const secrets[] = {
|
|
"secret1",
|
|
"secret2",
|
|
"secret3",
|
|
"secret4",
|
|
"secret5"
|
|
};
|
|
```
|
|
|
|
Replacing the strings with the codes that you need.
|
|
|
|
### name.c
|
|
|
|
In the `<name>.c` file, you will want to add this to the top:
|
|
|
|
```c
|
|
__attribute__ ((weak))
|
|
bool process_record_secrets(uint16_t keycode, keyrecord_t *record) {
|
|
return true;
|
|
}
|
|
```
|
|
|
|
This is so that the function can be called here, and replaced in the `secrets.c` file, and so it won't error out if it doesn't exist.
|
|
|
|
And then, in the `process_record_user` function, assuming you have `return process_record_keymap(keycode, record)` here, you'll want to replace the "final" return with the following. Otherwise, you want to replace the `return true;` with `return process_record_secrets(keycode, record);`
|
|
|
|
```c
|
|
return process_record_keymap(keycode, record) && process_record_secrets(keycode, record);
|
|
}
|
|
```
|
|
|
|
### rules.mk
|
|
|
|
Here, you want your `/users/<name>/rules.mk` file to "detect" the existence of the `secrets.c` file, and only add it if the file exists. To do so, add this block:
|
|
|
|
```make
|
|
ifneq ("$(wildcard $(USER_PATH)/secrets.c)","")
|
|
SRC += secrets.c
|
|
endif
|
|
```
|
|
|
|
Additionally, if you want to make sure that you can disable the function without messing with the file, you need to add this to your `/users/<name>/rules.mk`, so that it catches the flag:
|
|
|
|
```make
|
|
ifeq ($(strip $(NO_SECRETS)), yes)
|
|
OPT_DEFS += -DNO_SECRETS
|
|
endif
|
|
```
|
|
|
|
Then, if you run `make keyboard:name NO_SECRETS=yes`, it will default to the test strings in your `<name>.c` file, rather than reading from your file.
|
|
|
|
## Pro Micro Hacking
|
|
|
|
Well, you can get the QMK DFU bootloader working on the ProMicro. But you need to change fuses.
|
|
|
|
What worked to get into the firmware properly was:
|
|
|
|
```
|
|
Low: 0x5E High: 0xD9 Extended: 0xC3 Lock: 0x3F
|
|
```
|
|
|
|
The reason that there was some issues before, is that JTAG was still enabled, and using some of the pins that the keyboard uses. Disabling JTAG (either by fuse, or modifying the matrix code for splits fixes the issue).
|
|
|
|
And for reference, if you want to go back to caterina, the default fuse settings I believe were:
|
|
|
|
```
|
|
Low: 0xFF High: 0xD8 Extended: 0xC3 Lock: 0x3F
|
|
```
|