Correct issues with usermods not being linked.
- Explicitly set libArchive: false in usermod library.json files
- Fix up symlink path generation on Windows
- Add validation script to report usermod linkage in resulting binary
Neither the info panel nor the settings dialog can be trusted to
accurately report the usermod list:
- Not all usermods necessarily add to the info panel
- Not all usermods necessarily add to the config page
- #4609 is required for the config page to be correct
Add a short list to the info object that lists the loaded usermod IDs.
This is not displayed via the UI, but can be queried with curl or web
debug tools.
To be removed when usermod loading is working well.
* add support for up to 10 ESPNow remotes
* removed debug line
* changed todo comment
* fixed some issues, shortened html/java code
- reverting name to `linked_remote`
- ESPNow remote list is now hidden if unchecked
- shortened java script function names and variables to save flash
- removed now obsolete settings in xml.cpp
- correct checking of valid hex string for remote list in java script
* fixed indentation, using emplace_back instead of push_back, using JsonVariant, replaced buttons with +/-
* shortened java code
* updated java code, fixed bug
- element is now properly removed
- `+` button is hidden if list is full
- user needs to remove a remote, then reload the page to add it (workaround for edge case that needs more code to handle otherwise)
* add limit
* clearer usage description
- fixed inconsitencies in size rendering
- fixed palette being wrapped in color by position and color by age modes
- Fixed bug in memory layout: for some unknown reason, if flags come before particles, last flag is sometimes overwritten, changing memory laout seems to fix that
- New color modes in PS Fireworks 1D:
- custom3 slider < 16: lower saturation (check1: single color or multi-color explosions)
- custom3 slider <= 23: full saturation (check1: single color or multi-color explosions)
- custom3 slider > 23: color by speed (check 1 has not effect here)
- custom slider = max: color by age or color by position (depends on check1)
- Photoshop-style segment/layer blending
- return of strip ABL
- remove global LED buffer in favour of segment-local buffer
- new effect blending modes/transitions
- custom palettes moved out of WS2812FX class
- increased limits (matrix size, LED RAM)
- added "rainbow"-mode colorwheel
- replaced palettes with gamma unmodified ones
- move gamma adjustment to last step before sending to LEDs
- Segment & WS2812FX class reorganisation (mutable members, reordered members, protected members)
* Removed memory manager from PS
- reverted all changes related to memory manager
- moved local buffer into effect data memory
- some RAM issues may occur on larger setups: tested on S3 it works fine up to 32x32 but runs into memory issues at 64x32
* fixed ifdef, improved readability, add optimize "o2" flags to improve speed
- added struct for x and y coordinates, thx to @blazoncek
* cleanup and minor improvements
- removed local buffer for ESP8266 in 1D system to save on RAM
- increased particle brightness in PS Impact
- minor tweak in collision binning (might improve speed)
- removed comments and some other unused stuff
- fixed a few compiler wranings
* fixed init sequence bug
- since Update object is accessible even with ArduinoOTA disabled it should possible to use HTTP OTA regardless
- this saves about 12kB while still allowing OTA updates
- HTTP OTA updates can be blocked using PIN or OTA lock
* added Sonic Boom AR FX, some tweaks to Sonic Stream
* added white color option to Sonic Stream
* improvements to collisions (speed look-ahead)
* code prettified
* added "playful" mode to PS Chase plus some minor speed optimizations
* Adding new FX: PS Springy with many config options
Rather than reading the file off disk, have the json/cfg endpoint
return the live config from system state data. This can improve
UI behaviour as it can never be out of date or include values that
do not apply to the current firmware install.
Break the actual JSON assembly apart from the file writing code. This
permits calling it in other contexts, allowing us to pull the live
config data even if the filesystem is out of date.
Part of the ongoing quest to migrate macro definitions to typed
language constructs. This actually yields a small improvement in
code size, likely from the byte->int conversion.
