* Prep-work for datetime entities
* Fix some includes and remove some restrictions on printing time on displays
* format
* format
* More formatting
* Move function contents
* Ignore clang-tidy
* fixing rectangle's `shrink` and `extend`
* fixed the rect::shrink and rect::inside methods
and added rect:equal() method
* fixed internal clang issue again. When would is
this going to be fixed :(
* fixed internal clang issue again. When would is
this going to be fixed :(
* remove trailing space
* adding Clipping support to the displaybuffer
- add rect structure
* removed unused define
* add missing property for storing the clipped areas
* include log header
* Move Rect method's code to cpp file
- removed obsolete remarks
* fixed reported issues
* make Rect class methods public
* clang fix
* Remove commented code
* Renaming clipping methods
* Multiple changes:
- replaced 32766 with VALUE_NO_SET
- fixed the way *_clipping(left, top, right, bottom) is stored
- add `is_clipping();`
- make sure that all clipped region are closed after `do_update_()`
- rename de parameters for `Rect::expand();`
* remove unneeded space
* replace define with static const uint8_t
* correcting my copy paste mistake
---------
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
Currently, in each loop during DisplayBuffer::update_() the display is
cleared by calling DisplayBuffer::clear().
This prevents more efficient display usages that do not render the
screen in each loop, but only if necessary. This can be helpful, for
example, if images are rendered. This would cause the loop time to be
exceeded frequently.
This change adds a new optional flag "auto_clear" that can be used to
control the clearing behavior. If unset, the DisplayBuffer defaults to
enabled auto clearing, the current behavior and thus backward compatible.
This flag applies to displays that use DisplayBuffer.
Example excerpt:
globals:
- id: state
type: bool
restore_value: no
initial_value: "false"
- id: state_processed
type: bool
restore_value: no
initial_value: "false"
switch:
- platform: template
name: "State"
id: state_switch
lambda: |-
return id(state);
turn_on_action:
- globals.set:
id: state
value: "true"
- globals.set:
id: state_processed
value: "false"
turn_off_action:
- globals.set:
id: state
value: "false"
- globals.set:
id: state_processed
value: "false"
display:
- platform: ili9341
# ...
auto_clear_enabled: false
lambda: |-
if (!id(state_processed)) {
it.fill(COLOR_WHITE);
if (id(state)) {
it.image(80, 20, id(image1));
} else {
it.image(80, 20, id(image2));
}
id(state_processed) = true;
}
Co-authored-by: Tim Niemueller <timdn@google.com>
* Migrate ESPColor to Color
* color.h constructor fix
* Updated componets to use Color
Added a using for ESPColor
* Lint fixes
* Fixed value error
* Update display components to use colorutil
* Updated to latest PR comments
* Fixed COLOR_WHITE
* Moved esp_scale to color_utils
* Rename color_utils to display_color_utils
* SSD1325 grayscale support implemented
* Linted
* Fix garbage on display at power-up
* Rebase
* Linted
* Add brightness control, CS fixes
* Fix up SSD1327 init
* Add turn_on() and turn_off() methods
* Set brightness later in setup(), logging tweak
* Added is_on() method
* Added grayscale image support
* Updated to use Color, 1327 GS fix
* Move stop/is_running implementation to Action base class
Try to fix issue #1105. Until now if an UpdateComponentAction, a
LambdaAction, or any action that can contain those inside their
"else" or "then" action lists, resulted in a call to script.stop on
the script that contains them, the script would continue running,
because they didn't implement a stop() method. Basically only
the asynchronous ones did: DelayAction, WaitUntilAction and
ScriptWaitAction.
With this change num_running_ in Action replaces
DelayAction::num_running_ and WaitUntilAction::triggered_ to provide
the same is_running logic to other actions.
* Make some Action methods protected
Apparently play()/stop() etc. are not meant to be called directly by
users of the class and if they're called directly that would not give
the expected result for the classes that have an empty play().
Make all methods except play_complex, stop_comples and is_running
protected. While there also make RemoteTransmitterActionBase::encode
protected.
* lint
* format
Co-authored-by: Guillermo Ruffino <glm.net@gmail.com>
Jesse Hills