* enable the psram on esp32 and allow big screen
* update CODEOWNERS
* small update
* update CODEOWNERS again.
* Removed the M5STACK because it is a ESP32 device.
* i removed the wrong model
* update the error message.
* Feat: add support for hex color in color component
* Chore: move hex color validator to color component
* Chore: add test
* Chore: fix formatting
* Chore: make linter happy
* Chore: make linter happy
* Fix: parse correct offsets
Co-authored-by: Oxan van Leeuwen <oxan@oxanvanleeuwen.nl>
* Chore: use cv.Invalid
* Fix: remove # because it indicates a comment in yaml
* Fix: only allow hex if no other color value is set
* Fix: tests
* Fix: mutual exclusion of raw and hex colors
* Chore: format file
* Update __init__.py
---------
Co-authored-by: Oxan van Leeuwen <oxan@oxanvanleeuwen.nl>
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Fix missing data array
* Fix incorrect bit offset
* Correct variable types
* Do same conversions as in original library
* Correct clang-format
* Move out float conversion for clarity
* Added check for heater stability
* Correct clang format
* Allow reporting gas resistance when heater is disabled
* Correct clang format
* Better error reporting by @DAVe3283
* Correct signed operation, range switching error was positive all the time
* - Removed cleaning the screen twice.
\Should be handled by `DisplayBuffer::init_internal_();`
- Made ili9341::initalize() protected and renamed
\ it to ili9341::initalize().
- ili9341::initalize() should only init the display
\ and set the width and heigth.
* removed to much
* clang format fixes
* removing trailing underscors for
protected virtual methods
* removed the "override" on display()
* clang fixes
* restored old changes
* Renamed the ili9341 platform to ili9xxx and added
multiple drivers as well. including PR #3848
* fixed most of the clang reported issues
* fixed reported issues
* last fixes
* Setting the right codeowners
* missing changes
* fixed naming Display() method.
* clang again
* clang fix
* fixes reported by @jesserockz & @gpambrozio
* a change to display.py removing an unneeded var
* re-introduce **backlight** option.
* update the ili9488 initialization
* update the ili9488 initialization and fix typo
* fixed typo
* add missing constants
* swap height and width back for the ili9488
* init fixes ili9488
* fixed lint issue
testing the init code
* oeps
* init fixes ili9488
* fixed wrong define
* fixed wrong define again
* removed some spaces
* revert to ili9341
* Remove parts that where used for
the switchplate
* lint fixes and removing unused function
* fix error and introducing 16bit color option
* fix error and introducing 16bit color option
* fix clang issue
* clang fix
* clang issue again
* is this what clang exprect
* clang fix
* clang fix
* try again
* let try again
* and again
* and the last clang fix
* remove the need of wifi
* update dimentions
* update ili8488 init code.
* update dimentions
* allow to change height and width
* dump color mode config
* fix
* fix
* modify logging
* referd back unrelated change
* code formatting commit and moving functions around
* add missing ;
* update code
* update code
* use the correct write_array for sending uint16_t
* fix panic loop
* fix panic loop
* - update the test file
- fixed sending display data
* clang fixes
* clang fixes
* clang fixes again
* remove .gitignore items
* remove .gitignore items
* make sure Update() can can not be called while
called
* clang correction
* adding a test yaml for the ili9341
* Update ili9341 example
* Make test ili9xxx/tests only local
* restore back old ili9341 driver code
* Add a new config for the M5Core
* fix clang request
* reverd to restore of the old ili9341
there is no proper way to say it is depricated.
* Remove the backlight/led pin from the config.
You need to use a proper light platform component
* Ili9488init changes (#88)
Fixed ILI9488 init settings, and adjusted pixel handling code to push pixels in 18 bit format.
This does not change the internal 16-bit representation.
* fixed some leftover clang issues from the merge.
* fixed the slang-tidy request.
* remove `backlight_pin` warning.
---------
Co-authored-by: JD Steffen <jdsteffen81@gmail.com>
* remove unused sensors on pzemdc
* add reset energy action for pzemdc
* fix lint errors
* remove trailing space on pzemdc.h
* make method reset_energy of pzemdc public
* Apply suggestions from code review
---------
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Use PSRam for BLE scan results.
* Format Document
* Use generic define `CONFIG_SPIRAM`.
* Formatting changes.
* Memory allocation is allowed to fail.
* Use mark_failed instead of abort.
---------
Co-authored-by: Your Name <you@example.com>
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Add support for AS7341 spectral color sensor.
* Run clang-format and clang-tidy.
* Post-review changes.
* Apply suggestions from code review
---------
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Add delay between clock changes on sn74hc165
* Increase to 10us
* Add another delay after clock low
* Print input bits every second
* Fix pin order
* Remove log
* Fix for inverted pins
* formatting
* Add initial support for multiple devices
Re-introduce support for multiple I2C devices (it was suppressed in df37a7635f). Devices are identified by their I2C address, and the BME680 can only have the 0x76 or 0x77 address, so this adds support for a maximum of two devices.
* Reintegrate commit ebf13a0b
Reintegrate commit ebf13a0ba0 which was lost in my changes (I were working on old files)
* wrong commit
* wrong commit
* Reintegrate commit ebf13a0b
Reintegrate commit ebf13a0ba0 which was lost due to me working on old files
* Reintroduce newlines at end of files
* Reintroduce newlines at end of files
* Adhere to codebase standards
Obey the "All uses of class members and member functions should be prefixed with this-> to distinguish them from global functions in code review" rule of the Codebase Standards
* Fix formatting according to clang-format
* Perform the BSEC library reinitialization+snapshot only when more than one device is present
* Fix formatting according to clang-format
* Degrade abort message in restore_state_() from warning to verbose
This always happen at initial setup, so it's not a really useful message; when some real problems arise, we'll have a more useful warning from snapshot_state_()
Co-authored-by: Trevor North <trevor@freedisc.co.uk>
* Reduce peak stack usage to avoid bootloops on ESP8266
Achieved mainly by moving the work_buffer needed by the BSEC library to the heap, as a single global work buffer shared by all instances.
::set_config_ has been reverted to a code path similar to the original, as that reduces peak stack usage enough to be OK on ESP8266 even without moving the work_buffer to the heap.
* Fix formatting according to clang-format
* Add support for devices with the same i2c address
Devices are now identified using their index in the BME680BSECComponent::instances member, which became a vector. This allows adding two devices with the same i2c address (which should be placed on different i2c buses). Since a BME680 can only have an address of 0x76 or 0x77, a maximum of 2 devices could be added before this commit. Now there is no theoretical limit on the number of devices which could be added.
* Fix formatting according to clang-format
* Fix formatting according to clang-format
---------
Co-authored-by: Trevor North <trevor@freedisc.co.uk>
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Add support for FS3000 sensor.
* add fs3000 to test yaml
* Clean up code with clang.
* Clean up sensor.py file.
* Update CODEOWNERS file.
* Apply suggestions from code review regarding sensor.py
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Apply suggestions from code review for basic issues regarding C++ code
- removed unnecessary default for FS3000Model
- use "this->" before any sensor update
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
* Move model setup to overall setup function.
* Remove unneeded CONF_ID from sensor.py
* Run clang-format
* Move set_model code to header file now that it is simplified
* Update fs3000.h
---------
Co-authored-by: Jesse Hills <3060199+jesserockz@users.noreply.github.com>
This lets downstream components respond to climate configuration
changes, which take place through ClimateCall objects, without also
being notified every time the state changes, which happens every time
the input sensor announces a new value.
FIXES https://github.com/esphome/feature-requests/issues/2136
