Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
/*
|
|
|
|
Copyright 2013 Oleg Kostyuk <cub.uanic@gmail.com>
|
|
|
|
Copyright 2017 Erin Call <hello@erincall.com>
|
|
|
|
|
|
|
|
This program is free software: you can redistribute it and/or modify
|
|
|
|
it under the terms of the GNU General Public License as published by
|
|
|
|
the Free Software Foundation, either version 2 of the License, or
|
|
|
|
(at your option) any later version.
|
|
|
|
|
|
|
|
This program is distributed in the hope that it will be useful,
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
GNU General Public License for more details.
|
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
|
|
along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
*/
|
|
|
|
#include <stdint.h>
|
|
|
|
#include <stdbool.h>
|
|
|
|
#include <avr/io.h>
|
|
|
|
#include "wait.h"
|
|
|
|
#include "action_layer.h"
|
|
|
|
#include "print.h"
|
|
|
|
#include "debug.h"
|
|
|
|
#include "util.h"
|
|
|
|
#include "matrix.h"
|
|
|
|
#include "dactyl.h"
|
|
|
|
#include "i2cmaster.h"
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#include "timer.h"
|
|
|
|
|
|
|
|
|
|
|
|
/* Set 0 if debouncing isn't needed */
|
|
|
|
|
|
|
|
#ifndef DEBOUNCING_DELAY
|
|
|
|
# define DEBOUNCING_DELAY 5
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
#endif
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#if (DEBOUNCING_DELAY > 0)
|
|
|
|
static uint16_t debouncing_time;
|
|
|
|
static bool debouncing = false;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifdef MATRIX_MASKED
|
|
|
|
extern const matrix_row_t matrix_mask[];
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#if (DIODE_DIRECTION == ROW2COL) || (DIODE_DIRECTION == COL2ROW)
|
|
|
|
static const uint8_t onboard_row_pins[MATRIX_ROWS] = MATRIX_ONBOARD_ROW_PINS;
|
|
|
|
static const uint8_t onboard_col_pins[MATRIX_COLS] = MATRIX_ONBOARD_COL_PINS;
|
|
|
|
static const bool col_expanded[MATRIX_COLS] = COL_EXPANDED;
|
|
|
|
static const uint8_t expander_row_pins[MATRIX_ROWS] = MATRIX_EXPANDER_ROW_PINS;
|
|
|
|
static const uint8_t expander_col_pins[MATRIX_COLS] = MATRIX_EXPANDER_COL_PINS;
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
#endif
|
|
|
|
|
|
|
|
/* matrix state(1:on, 0:off) */
|
|
|
|
static matrix_row_t matrix[MATRIX_ROWS];
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
static matrix_row_t matrix_debouncing[MATRIX_ROWS];
|
|
|
|
|
|
|
|
#if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
static void init_cols(void);
|
|
|
|
static bool read_cols_on_row(matrix_row_t current_matrix[], uint8_t current_row);
|
|
|
|
static void unselect_rows(void);
|
|
|
|
static void select_row(uint8_t row);
|
|
|
|
static void unselect_row(uint8_t row);
|
|
|
|
#elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
static void init_rows(void);
|
|
|
|
static bool read_rows_on_col(matrix_row_t current_matrix[], uint8_t current_col);
|
|
|
|
static void unselect_cols(void);
|
|
|
|
static void select_col(uint8_t col);
|
|
|
|
static void unselect_col(uint8_t col);
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
static uint8_t expander_reset_loop;
|
|
|
|
uint8_t expander_status;
|
|
|
|
uint8_t expander_input_pin_mask;
|
|
|
|
bool i2c_initialized = false;
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
|
|
|
#ifdef DEBUG_MATRIX_SCAN_RATE
|
|
|
|
uint32_t matrix_timer;
|
|
|
|
uint32_t matrix_scan_count;
|
|
|
|
#endif
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#define ROW_SHIFTER ((matrix_row_t)1)
|
|
|
|
#if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
// bitmask to ensure the row state from the expander only applies to its columns
|
|
|
|
#define EXPANDER_MASK ((matrix_row_t)0b00111111)
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
|
|
|
__attribute__ ((weak))
|
|
|
|
void matrix_init_user(void) {}
|
|
|
|
|
|
|
|
__attribute__ ((weak))
|
|
|
|
void matrix_scan_user(void) {}
|
|
|
|
|
|
|
|
__attribute__ ((weak))
|
|
|
|
void matrix_init_kb(void) {
|
|
|
|
matrix_init_user();
|
|
|
|
}
|
|
|
|
|
|
|
|
__attribute__ ((weak))
|
|
|
|
void matrix_scan_kb(void) {
|
|
|
|
matrix_scan_user();
|
|
|
|
}
|
|
|
|
|
|
|
|
inline
|
|
|
|
uint8_t matrix_rows(void)
|
|
|
|
{
|
|
|
|
return MATRIX_ROWS;
|
|
|
|
}
|
|
|
|
|
|
|
|
inline
|
|
|
|
uint8_t matrix_cols(void)
|
|
|
|
{
|
|
|
|
return MATRIX_COLS;
|
|
|
|
}
|
|
|
|
|
|
|
|
void matrix_init(void)
|
|
|
|
{
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
init_expander();
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#if (DIODE_DIRECTION == COL2ROW)
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
unselect_rows();
|
|
|
|
init_cols();
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
unselect_cols();
|
|
|
|
init_rows();
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
|
|
|
// initialize matrix state: all keys off
|
|
|
|
for (uint8_t i=0; i < MATRIX_ROWS; i++) {
|
|
|
|
matrix[i] = 0;
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
matrix_debouncing[i] = 0;
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef DEBUG_MATRIX_SCAN_RATE
|
|
|
|
matrix_timer = timer_read32();
|
|
|
|
matrix_scan_count = 0;
|
|
|
|
#endif
|
|
|
|
|
|
|
|
matrix_init_quantum();
|
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
void init_expander(void) {
|
|
|
|
if (! i2c_initialized) {
|
|
|
|
i2c_init();
|
|
|
|
wait_us(1000000);
|
|
|
|
}
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
if (! expander_input_pin_mask) {
|
|
|
|
#if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
for (int col = 0; col < MATRIX_COLS; col++) {
|
|
|
|
if (col_expanded[col]) {
|
|
|
|
expander_input_pin_mask |= (1 << expander_col_pins[col]);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
for (int row = 0; row < MATRIX_ROWS; row++) {
|
|
|
|
expander_input_pin_mask |= (1 << expander_row_pins[row]);
|
|
|
|
}
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
expander_status = i2c_start(I2C_ADDR_WRITE); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(IODIRA); if (expander_status) goto out;
|
|
|
|
|
|
|
|
/*
|
|
|
|
Pin direction and pull-up depends on both the diode direction
|
|
|
|
and on whether the column register is 0 ("A") or 1 ("B"):
|
|
|
|
+-------+---------------+---------------+
|
|
|
|
| | ROW2COL | COL2ROW |
|
|
|
|
+-------+---------------+---------------+
|
|
|
|
| Reg 0 | input, output | output, input |
|
|
|
|
+-------+---------------+---------------+
|
|
|
|
| Reg 1 | output, input | input, output |
|
|
|
|
+-------+---------------+---------------+
|
|
|
|
*/
|
|
|
|
|
|
|
|
#if (EXPANDER_COLUMN_REGISTER == 0)
|
|
|
|
# if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
# elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
# endif
|
|
|
|
#elif (EXPANDER_COLUMN_REGISTER == 1)
|
|
|
|
# if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
# elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
# endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
#endif
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
i2c_stop();
|
|
|
|
|
|
|
|
// set pull-up
|
|
|
|
// - unused : off : 0
|
|
|
|
// - input : on : 1
|
|
|
|
// - driving : off : 0
|
|
|
|
expander_status = i2c_start(I2C_ADDR_WRITE); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(GPPUA); if (expander_status) goto out;
|
|
|
|
#if (EXPANDER_COLUMN_REGISTER == 0)
|
|
|
|
# if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
# elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
# endif
|
|
|
|
#elif (EXPANDER_COLUMN_REGISTER == 1)
|
|
|
|
# if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
# elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
expander_status = i2c_write(expander_input_pin_mask); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(0); if (expander_status) goto out;
|
|
|
|
# endif
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
out:
|
|
|
|
i2c_stop();
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
uint8_t matrix_scan(void)
|
|
|
|
{
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
if (expander_status) { // if there was an error
|
|
|
|
if (++expander_reset_loop == 0) {
|
|
|
|
// since expander_reset_loop is 8 bit - we'll try to reset once in 255 matrix scans
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
// this will be approx bit more frequent than once per second
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
print("trying to reset expander\n");
|
|
|
|
init_expander();
|
|
|
|
if (expander_status) {
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
print("left side not responding\n");
|
|
|
|
} else {
|
|
|
|
print("left side attached\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef DEBUG_MATRIX_SCAN_RATE
|
|
|
|
matrix_scan_count++;
|
|
|
|
|
|
|
|
uint32_t timer_now = timer_read32();
|
|
|
|
if (TIMER_DIFF_32(timer_now, matrix_timer)>1000) {
|
|
|
|
print("matrix scan frequency: ");
|
|
|
|
pdec(matrix_scan_count);
|
|
|
|
print("\n");
|
|
|
|
|
|
|
|
matrix_timer = timer_now;
|
|
|
|
matrix_scan_count = 0;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#if (DIODE_DIRECTION == COL2ROW)
|
|
|
|
for (uint8_t current_row = 0; current_row < MATRIX_ROWS; current_row++) {
|
|
|
|
# if (DEBOUNCING_DELAY > 0)
|
|
|
|
bool matrix_changed = read_cols_on_row(matrix_debouncing, current_row);
|
|
|
|
|
|
|
|
if (matrix_changed) {
|
|
|
|
debouncing = true;
|
|
|
|
debouncing_time = timer_read();
|
|
|
|
}
|
|
|
|
# else
|
|
|
|
read_cols_on_row(matrix, current_row);
|
|
|
|
# endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
for (uint8_t current_col = 0; current_col < MATRIX_COLS; current_col++) {
|
|
|
|
# if (DEBOUNCING_DELAY > 0)
|
|
|
|
bool matrix_changed = read_rows_on_col(matrix_debouncing, current_col);
|
|
|
|
|
|
|
|
if (matrix_changed) {
|
|
|
|
debouncing = true;
|
|
|
|
debouncing_time = timer_read();
|
|
|
|
}
|
|
|
|
# else
|
|
|
|
read_rows_on_col(matrix, current_col);
|
|
|
|
# endif
|
|
|
|
|
|
|
|
}
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
# if (DEBOUNCING_DELAY > 0)
|
|
|
|
if (debouncing && (timer_elapsed(debouncing_time) > DEBOUNCING_DELAY)) {
|
|
|
|
for (uint8_t i = 0; i < MATRIX_ROWS; i++) {
|
|
|
|
matrix[i] = matrix_debouncing[i];
|
|
|
|
}
|
|
|
|
debouncing = false;
|
|
|
|
}
|
|
|
|
# endif
|
|
|
|
|
|
|
|
matrix_scan_quantum();
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool matrix_is_modified(void) // deprecated and evidently not called.
|
|
|
|
{
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#if (DEBOUNCING_DELAY > 0)
|
|
|
|
if (debouncing) return false;
|
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
inline
|
|
|
|
bool matrix_is_on(uint8_t row, uint8_t col)
|
|
|
|
{
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
return (matrix[row] & (ROW_SHIFTER << col));
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
inline
|
|
|
|
matrix_row_t matrix_get_row(uint8_t row)
|
|
|
|
{
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#ifdef MATRIX_MASKED
|
|
|
|
return matrix[row] & matrix_mask[row];
|
|
|
|
#else
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
return matrix[row];
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#endif
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
void matrix_print(void)
|
|
|
|
{
|
|
|
|
print("\nr/c 0123456789ABCDEF\n");
|
|
|
|
for (uint8_t row = 0; row < MATRIX_ROWS; row++) {
|
|
|
|
phex(row); print(": ");
|
|
|
|
pbin_reverse16(matrix_get_row(row));
|
|
|
|
print("\n");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
uint8_t matrix_key_count(void)
|
|
|
|
{
|
|
|
|
uint8_t count = 0;
|
|
|
|
for (uint8_t i = 0; i < MATRIX_ROWS; i++) {
|
|
|
|
count += bitpop16(matrix[i]);
|
|
|
|
}
|
|
|
|
return count;
|
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#if (DIODE_DIRECTION == COL2ROW)
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
static void init_cols(void) {
|
|
|
|
for (uint8_t x = 0; x < MATRIX_COLS; x++) {
|
|
|
|
if (! col_expanded[x]) {
|
|
|
|
uint8_t pin = onboard_col_pins[x];
|
|
|
|
_SFR_IO8((pin >> 4) + 1) &= ~_BV(pin & 0xF); // IN
|
|
|
|
_SFR_IO8((pin >> 4) + 2) |= _BV(pin & 0xF); // HI
|
|
|
|
}
|
|
|
|
}
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
static bool read_cols_on_row(matrix_row_t current_matrix[], uint8_t current_row) {
|
|
|
|
// Store last value of row prior to reading
|
|
|
|
matrix_row_t last_row_value = current_matrix[current_row];
|
|
|
|
|
|
|
|
// Clear data in matrix row
|
|
|
|
current_matrix[current_row] = 0;
|
|
|
|
|
|
|
|
// Select row and wait for row selection to stabilize
|
|
|
|
select_row(current_row);
|
|
|
|
wait_us(30);
|
|
|
|
|
|
|
|
// Read columns from expander, unless it's in an error state
|
|
|
|
if (! expander_status) {
|
|
|
|
expander_status = i2c_start(I2C_ADDR_WRITE); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(GPIOA); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_start(I2C_ADDR_READ); if (expander_status) goto out;
|
|
|
|
|
|
|
|
current_matrix[current_row] |= (~i2c_readNak()) & EXPANDER_MASK;
|
|
|
|
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
out:
|
|
|
|
i2c_stop();
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
// Read columns from onboard pins
|
|
|
|
for (uint8_t col_index = 0; col_index < MATRIX_COLS; col_index++) {
|
|
|
|
if (! col_expanded[col_index]) {
|
|
|
|
uint8_t pin = onboard_col_pins[col_index];
|
|
|
|
uint8_t pin_state = (_SFR_IO8(pin >> 4) & _BV(pin & 0xF));
|
|
|
|
current_matrix[current_row] |= pin_state ? 0 : (ROW_SHIFTER << col_index);
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
}
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
|
|
|
|
unselect_row(current_row);
|
|
|
|
|
|
|
|
return (last_row_value != current_matrix[current_row]);
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
static void select_row(uint8_t row) {
|
|
|
|
// select on expander, unless it's in an error state
|
|
|
|
if (! expander_status) {
|
|
|
|
// set active row low : 0
|
|
|
|
// set other rows hi-Z : 1
|
|
|
|
expander_status = i2c_start(I2C_ADDR_WRITE); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(GPIOB); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(0xFF & ~(1<<row)); if (expander_status) goto out;
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
out:
|
|
|
|
i2c_stop();
|
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
// select on teensy
|
|
|
|
uint8_t pin = onboard_row_pins[row];
|
|
|
|
_SFR_IO8((pin >> 4) + 1) |= _BV(pin & 0xF); // OUT
|
|
|
|
_SFR_IO8((pin >> 4) + 2) &= ~_BV(pin & 0xF); // LOW
|
|
|
|
}
|
|
|
|
|
|
|
|
static void unselect_row(uint8_t row)
|
|
|
|
{
|
|
|
|
// No need to explicitly unselect expander pins--their I/O state is
|
|
|
|
// set simultaneously, with a single bitmask sent to i2c_write. When
|
|
|
|
// select_row selects a single pin, it implicitly unselects all the
|
|
|
|
// other ones.
|
|
|
|
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
// unselect on teensy
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
uint8_t pin = onboard_row_pins[row];
|
|
|
|
_SFR_IO8((pin >> 4) + 1) &= ~_BV(pin & 0xF); // OUT
|
|
|
|
_SFR_IO8((pin >> 4) + 2) |= _BV(pin & 0xF); // LOW
|
|
|
|
}
|
|
|
|
|
|
|
|
static void unselect_rows(void) {
|
|
|
|
for (uint8_t x = 0; x < MATRIX_ROWS; x++) {
|
|
|
|
unselect_row(x);
|
|
|
|
}
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
#elif (DIODE_DIRECTION == ROW2COL)
|
|
|
|
|
|
|
|
static void init_rows(void)
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
{
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
for (uint8_t x = 0; x < MATRIX_ROWS; x++) {
|
|
|
|
uint8_t pin = onboard_row_pins[x];
|
|
|
|
_SFR_IO8((pin >> 4) + 1) &= ~_BV(pin & 0xF); // IN
|
|
|
|
_SFR_IO8((pin >> 4) + 2) |= _BV(pin & 0xF); // HI
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool read_rows_on_col(matrix_row_t current_matrix[], uint8_t current_col)
|
|
|
|
{
|
|
|
|
bool matrix_changed = false;
|
|
|
|
|
|
|
|
uint8_t column_state = 0;
|
|
|
|
|
|
|
|
//select col and wait for selection to stabilize
|
|
|
|
select_col(current_col);
|
|
|
|
wait_us(30);
|
|
|
|
|
|
|
|
if (current_col < 6) {
|
|
|
|
// read rows from expander
|
|
|
|
if (expander_status) {
|
|
|
|
// it's already in an error state; nothing we can do
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
expander_status = i2c_start(I2C_ADDR_WRITE); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(GPIOB); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_start(I2C_ADDR_READ); if (expander_status) goto out;
|
|
|
|
column_state = i2c_readNak();
|
|
|
|
|
|
|
|
out:
|
|
|
|
i2c_stop();
|
|
|
|
|
|
|
|
column_state = ~column_state;
|
|
|
|
} else {
|
|
|
|
for (uint8_t current_row = 0; current_row < MATRIX_ROWS; current_row++) {
|
|
|
|
if ((_SFR_IO8(onboard_row_pins[current_row] >> 4) & _BV(onboard_row_pins[current_row] & 0xF)) == 0) {
|
|
|
|
column_state |= (1 << current_row);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
for (uint8_t current_row = 0; current_row < MATRIX_ROWS; current_row++) {
|
|
|
|
// Store last value of row prior to reading
|
|
|
|
matrix_row_t last_row_value = current_matrix[current_row];
|
|
|
|
|
|
|
|
if (column_state & (1 << current_row)) {
|
|
|
|
// key closed; set state bit in matrix
|
|
|
|
current_matrix[current_row] |= (ROW_SHIFTER << current_col);
|
|
|
|
} else {
|
|
|
|
// key open; clear state bit in matrix
|
|
|
|
current_matrix[current_row] &= ~(ROW_SHIFTER << current_col);
|
|
|
|
}
|
|
|
|
|
|
|
|
// Determine whether the matrix changed state
|
|
|
|
if ((last_row_value != current_matrix[current_row]) && !(matrix_changed))
|
|
|
|
{
|
|
|
|
matrix_changed = true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
unselect_col(current_col);
|
|
|
|
|
|
|
|
return matrix_changed;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void select_col(uint8_t col)
|
|
|
|
{
|
|
|
|
if (col_expanded[col]) {
|
|
|
|
// select on expander
|
|
|
|
if (expander_status) { // if there was an error
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
// do nothing
|
|
|
|
} else {
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
// set active col low : 0
|
|
|
|
// set other cols hi-Z : 1
|
|
|
|
expander_status = i2c_start(I2C_ADDR_WRITE); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(GPIOA); if (expander_status) goto out;
|
|
|
|
expander_status = i2c_write(0xFF & ~(1<<col)); if (expander_status) goto out;
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
out:
|
|
|
|
i2c_stop();
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// select on teensy
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
uint8_t pin = onboard_col_pins[col];
|
|
|
|
_SFR_IO8((pin >> 4) + 1) |= _BV(pin & 0xF); // OUT
|
|
|
|
_SFR_IO8((pin >> 4) + 2) &= ~_BV(pin & 0xF); // LOW
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void unselect_col(uint8_t col)
|
|
|
|
{
|
|
|
|
if (col_expanded[col]) {
|
|
|
|
// No need to explicitly unselect expander pins--their I/O state is
|
|
|
|
// set simultaneously, with a single bitmask sent to i2c_write. When
|
|
|
|
// select_col selects a single pin, it implicitly unselects all the
|
|
|
|
// other ones.
|
|
|
|
} else {
|
|
|
|
// unselect on teensy
|
|
|
|
uint8_t pin = onboard_col_pins[col];
|
|
|
|
_SFR_IO8((pin >> 4) + 1) &= ~_BV(pin & 0xF); // IN
|
|
|
|
_SFR_IO8((pin >> 4) + 2) |= _BV(pin & 0xF); // HI
|
Handwired/Dactyl keyboard (#2058)
* Copy the ergodox_ez code to handwired/dactyl
Differences from the Ergodox:
* Use QMK_SUBPROJECT_H instead of QMK_KEYBOARD_H, since it's under
handwired
* Omitted several keymaps. They'll eventually be broken (since the
Dactyl has fewer keys), and I don't want to try to fix them.
* Omitted the keymap images for the default layout, since they depict
a different keyboard.
* Everything that said Ergodox now says Dactyl, naturally.
* [whitespace] Delete trailing whitespace
My editor does this automatically so it's just gonna keep cropping up...
* Cut the dactyl down to the right number of columns
(Remember, throughout matrix.c, everything called "row" is really a
column, and vice-versa).
* Remove LED-related code
* Tighten up the Dactyl's build options
* Whitespace cleanup in twimaster.c
* Hardtabs -> spaces
* No more trailing whitespace
* Typo fix
* Correct the CPU frequency units
The Teensy's CPU definitely doesn't run at 16 petahertz...
* Restore access to ONEHAND_ENABLE
I turned it off in 26d47cb42622d990a7c3335e7fcc151aa3edfbf0 while
desperately debugging; I just wanted to ensure it wasn't causing the
problem I was seeing. It was not, in fact, causing the problem, so it's
back.
Also fixed the swap matrix in dactyl.c, since it still referred to
columns that exist in the Ergodox but not the Dactyl.
* Clearer phrasing about TWI's effect on scan rate
* Fix up the Dactyl's firmware-loading instructions
Sadly, the Dactyl has no hole for the onboard reset button.
* Dvorak keymap for the Dactyl
* The Erincalling Layout
* Erincalling layout: Add a := key
I've been working in Go, which uses := a lot, and it's awkward to type
in this layout.
* Dactyl README: link to the dactyl-keyboard repo
* Add a missing copyright line
I don't know how much this matters? Honestly, it's enough for me that my
name is on the git commit. But hey, let's be consistent until there's a
specific reason not to be, right?
* Dactyl: remove commented-out code
I hate it I hate it I hate it
There's not even any information about what it was trying to do!!!! >:(
* Add a note about the row/column ridiculousness
* [whitespace] realign some constants
* Don't claim B4 is tied to VCC
It doesn't matter at all? I honestly don't know what the reason ever
was. It looks like it dates back to the original ErgoDox and I've never
seen one sentence about the purpose.
I've been skipping that wire for some time, and I promise it works fine.
* Dactyl keymaps: Send RALT for right-hand alt key
Not terribly important but I just like things tidy OK
* typo fix
* Refer to "dactyl.h" explicitly
QMK_SUBPROJECT_H has been working locally, but fails in CI. Strange!
* Dactyl: Don't use QMK_SUBPROJECT_H at all
It's still breaking in CI, even though it was a never a problem locally.
2017-11-26 08:18:59 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Bring dactyl/matrix.c in line with quantum/matrix.c (#2613)
* Use the new debounce algorithm in dactyl/matrix.c [#2065]
This incorporates the fixed/optimized debounce code added to
quantum/matrix.c in:
* 508eddf8ba8548d3f71e1c09a404839beb49f45c
* 4c6960835c0a6e29670dabdc27117d7d3c7f99f5
* 32f88c07173b795c6981c779057dceba00aeb1cb
* f4030289744fc6dc82dd85c955070c0845813cc5
* a06115df19a74d39b08758472b221e630c3680d3
* Fix the row/column swap in dactyl [#2065]
With a column-driven keyboard, reading from the mcp23081 returns a
column-state, which takes some extra work to translate into the
row-state used in the actual matrix. The ergodox_ez code sidestepped
that problem by calling rows "columns" and columns "rows." With this
change, the dactyl now calls rows "rows" and columns "columns."
* Cleanup: variable names, documentation [#2065]
* Support MATRIX_MASKED in dactyl/matrix.c [#2065]
* Only unselect one col in unselect_col [#2065]
Bonus: saves one i2c transaction per matrix_scan!
* Implement COL2ROW in dactyl/matrix.c [#2065]
* Fix a typo in dactyl/matrix.c
This entirely doesn't matter. The PORT values are set during
init_keyboard and never change. They're repeatedly set to the same
thing. These PORT lines shouldn't even exist, but since they do, they
should at least look right.
* Implement COL_PINS/ROW_PINS for dactyl [#2065]
* Rename "mcp23018" to "expander" [#2065]
I honestly don't know whether/how well this code works with other I/O
expanders, but at least in theory, it should be generic enough to work
with others. Given that, the variable names shouldn't refer to a
specific model of expander.
* Remove matrix_power_up from dactyl/matrix.c [#2065]
It's commented out in quantum/matrix.c, and the dactyl has no power
up/down behavior beyond being unplugged (which goes to matrix_init), so
there's no sense keeping it around.
* Only initialize expander_input_mask once [#2065]
...and rename input_mask to expander_input_mask, since now that it isn't
scoped to init_expander it isn't clear that it's only for the expander.
2018-04-27 20:47:41 +02:00
|
|
|
static void unselect_cols(void)
|
|
|
|
{
|
|
|
|
for(uint8_t x = 0; x < MATRIX_COLS; x++) {
|
|
|
|
unselect_col(x);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|