Update WLED app links for Android and iOS

thx @blazoncek

.github/FUNDING.yml

@@ -1,2 +1,3 @@
-github: [Aircoookie,blazoncek]
+github: [Aircoookie,blazoncek,DedeHai,lost-hope,willmmiles]
 custom: ['https://paypal.me/Aircoookie','https://paypal.me/blazoncek']
+thanks_dev: u/gh/netmindz

.github/copilot-instructions.md

@@ -0,0 +1,138 @@
+# WLED - ESP32/ESP8266 LED Controller Firmware
+
+WLED is a fast and feature-rich implementation of an ESP32 and ESP8266 webserver to control NeoPixel (WS2812B, WS2811, SK6812) LEDs and SPI-based chipsets. The project consists of C++ firmware for microcontrollers and a modern web interface.
+
+Always reference these instructions first and fallback to search or bash commands only when you encounter unexpected information that does not match the info here.
+
+## Working Effectively
+
+### Initial Setup
+- Install Node.js 20+ (specified in `.nvmrc`): Check your version with `node --version`
+- Install dependencies: `npm ci` (takes ~5 seconds)
+- Install PlatformIO for hardware builds: `pip install -r requirements.txt` (takes ~60 seconds)
+
+### Build and Test Workflow
+- **ALWAYS build web UI first**: `npm run build` -- takes 3 seconds. NEVER CANCEL.
+- **Run tests**: `npm test` -- takes 40 seconds. NEVER CANCEL. Set timeout to 2+ minutes.
+- **Development mode**: `npm run dev` -- monitors file changes and auto-rebuilds web UI
+- **Hardware firmware build**: `pio run -e [environment]` -- takes 15+ minutes. NEVER CANCEL. Set timeout to 30+ minutes.
+
+### Build Process Details
+The build has two main phases:
+1. **Web UI Generation** (`npm run build`):
+   - Processes files in `wled00/data/` (HTML, CSS, JS)
+   - Minifies and compresses web content 
+   - Generates `wled00/html_*.h` files with embedded web content
+   - **CRITICAL**: Must be done before any hardware build
+
+2. **Hardware Compilation** (`pio run`):
+   - Compiles C++ firmware for various ESP32/ESP8266 targets
+   - Common environments: `nodemcuv2`, `esp32dev`, `esp8266_2m`
+   - List all targets: `pio run --list-targets`
+
+## Validation and Testing
+
+### Web UI Testing
+- **ALWAYS validate web UI changes manually**:
+  - Start local server: `cd wled00/data && python3 -m http.server 8080`
+  - Open `http://localhost:8080/index.htm` in browser
+  - Test basic functionality: color picker, effects, settings pages
+- **Check for JavaScript errors** in browser console
+
+### Code Validation
+- **No automated linting configured** - follow existing code style in files you edit
+- **Code style**: Use tabs for web files (.html/.css/.js), spaces (2 per level) for C++ files
+- **C++ formatting available**: `clang-format` is installed but not in CI
+- **Always run tests before finishing**: `npm test`
+
+### Manual Testing Scenarios
+After making changes to web UI, always test:
+- **Load main interface**: Verify index.htm loads without errors
+- **Navigation**: Test switching between main page and settings pages
+- **Color controls**: Verify color picker and brightness controls work
+- **Effects**: Test effect selection and parameter changes
+- **Settings**: Test form submission and validation
+
+## Common Tasks
+
+### Repository Structure
+```
+wled00/                 # Main firmware source (C++)
+  ├── data/            # Web interface files 
+  │   ├── index.htm    # Main UI
+  │   ├── settings*.htm # Settings pages
+  │   └── *.js/*.css   # Frontend resources
+  ├── *.cpp/*.h        # Firmware source files
+  └── html_*.h         # Generated embedded web files (DO NOT EDIT)
+tools/                 # Build tools (Node.js)
+  ├── cdata.js         # Web UI build script
+  └── cdata-test.js    # Test suite
+platformio.ini         # Hardware build configuration
+package.json           # Node.js dependencies and scripts
+.github/workflows/     # CI/CD pipelines
+```
+
+### Key Files and Their Purpose
+- `wled00/data/index.htm` - Main web interface
+- `wled00/data/settings*.htm` - Configuration pages  
+- `tools/cdata.js` - Converts web files to C++ headers
+- `wled00/wled.h` - Main firmware configuration
+- `platformio.ini` - Hardware build targets and settings
+
+### Development Workflow
+1. **For web UI changes**:
+   - Edit files in `wled00/data/`
+   - Run `npm run build` to regenerate headers
+   - Test with local HTTP server
+   - Run `npm test` to validate build system
+
+2. **For firmware changes**:
+   - Edit files in `wled00/` (but NOT `html_*.h` files)
+   - Ensure web UI is built first (`npm run build`)
+   - Build firmware: `pio run -e [target]`
+   - Flash to device: `pio run -e [target] --target upload`
+
+3. **For both web and firmware**:
+   - Always build web UI first
+   - Test web interface manually
+   - Build and test firmware if making firmware changes
+
+## Build Timing and Timeouts
+
+- **Web UI build**: 3 seconds - Set timeout to 30 seconds minimum
+- **Test suite**: 40 seconds - Set timeout to 2 minutes minimum  
+- **Hardware builds**: 15+ minutes - Set timeout to 30+ minutes minimum
+- **NEVER CANCEL long-running builds** - PlatformIO downloads and compilation can take significant time
+
+## Troubleshooting
+
+### Common Issues
+- **Build fails with missing html_*.h**: Run `npm run build` first
+- **Web UI looks broken**: Check browser console for JavaScript errors
+- **PlatformIO network errors**: Try again, downloads can be flaky
+- **Node.js version issues**: Ensure Node.js 20+ is installed (check `.nvmrc`)
+
+### When Things Go Wrong
+- **Clear generated files**: `rm -f wled00/html_*.h` then rebuild
+- **Force web UI rebuild**: `npm run build -- --force` or `npm run build -- -f`
+- **Clean PlatformIO cache**: `pio run --target clean`
+- **Reinstall dependencies**: `rm -rf node_modules && npm install`
+
+## Important Notes
+
+- **DO NOT edit `wled00/html_*.h` files** - they are auto-generated
+- **Always commit both source files AND generated html_*.h files**
+- **Web UI must be built before firmware compilation**
+- **Test web interface manually after any web UI changes**
+- **Use VS Code with PlatformIO extension for best development experience**
+- **Hardware builds require appropriate ESP32/ESP8266 development board**
+
+## CI/CD Pipeline
+The GitHub Actions workflow:
+1. Installs Node.js and Python dependencies
+2. Runs `npm test` to validate build system
+3. Builds web UI with `npm run build` 
+4. Compiles firmware for multiple hardware targets
+5. Uploads build artifacts
+
+Match this workflow in your local development to ensure CI success.

.github/workflows/build.yml

@@ -40,7 +40,10 @@ jobs:
       with:
         node-version-file: '.nvmrc'
         cache: 'npm'
-    - run: npm ci
+    - run: |
+        npm ci
+        VERSION=`date +%y%m%d0`
+        sed -i -r -e "s/define VERSION .+/define VERSION $VERSION/" wled00/wled.h
     - name: Cache PlatformIO
       uses: actions/cache@v4
       with:

.github/workflows/nightly.yml

@@ -27,6 +27,8 @@ jobs:
         with:
           token: ${{ secrets.GITHUB_TOKEN }} 
           sinceTag: v0.15.0
+          # Exclude issues that were closed without resolution from changelog
+          exclude-labels: 'stale,wontfix,duplicate,invalid'
       - name: Update Nightly Release
         uses: andelf/nightly-release@main
         env:

.github/workflows/pr-merge.yaml

@@ -1,12 +1,13 @@
     name: Notify Discord on PR Merge
     on:
       workflow_dispatch:
-      pull_request:
+      pull_request_target:
         types: [closed]
 
     jobs:
       notify:
         runs-on: ubuntu-latest
+        if: github.event.pull_request.merged == true
         steps:
         - name: Get User Permission
           id: checkAccess
@@ -23,11 +24,15 @@
             echo "Current permission level is ${{ steps.checkAccess.outputs.user-permission }}"
             echo "Job originally triggered by ${{ github.actor }}"
             exit 1
-        - name: Checkout code
-          uses: actions/checkout@v3
-          with:
-            ref: ${{  github.event.pull_request.head.sha }} # This is dangerous without the first access check
         - name: Send Discord notification
-          # if: github.event.pull_request.merged == true
+          env:
+            PR_NUMBER: ${{ github.event.pull_request.number }}
+            PR_TITLE: ${{ github.event.pull_request.title }}
+            PR_URL: ${{ github.event.pull_request.html_url }}
+            ACTOR: ${{ github.actor }}
           run: |
-            curl -H "Content-Type: application/json" -d '{"content": "Pull Request ${{ github.event.pull_request.number }} merged by ${{ github.actor }}"}' ${{ secrets.DISCORD_WEBHOOK_BETA_TESTERS }}
+            jq -n \
+              --arg content "Pull Request #${PR_NUMBER} \"${PR_TITLE}\" merged by ${ACTOR}
+            ${PR_URL}. It will be included in the next nightly builds, please test" \
+              '{content: $content}' \
+              | curl -H "Content-Type: application/json" -d @- ${{ secrets.DISCORD_WEBHOOK_BETA_TESTERS }}

.github/workflows/release.yml

@@ -23,7 +23,10 @@ jobs:
       uses: janheinrichmerker/action-github-changelog-generator@v2.3
       with:
           token: ${{ secrets.GITHUB_TOKEN }} 
-          sinceTag: v0.15.0        
+          sinceTag: v0.15.0
+          maxIssues: 500
+          # Exclude issues that were closed without resolution from changelog
+          exclude-labels: 'stale,wontfix,duplicate,invalid'       
     - name: Create draft release
       uses: softprops/action-gh-release@v1
       with:

.github/workflows/usermods.yml

@@ -5,6 +5,9 @@ on:
     paths:
       - usermods/**
       - .github/workflows/usermods.yml
+  pull_request:
+    paths:
+      - usermods/**
     
 jobs:
 

platformio.ini

@@ -10,7 +10,7 @@
 # ------------------------------------------------------------------------------
 
 # CI/release binaries
-default_envs = nodemcuv2, esp8266_2m, esp01_1m_full, nodemcuv2_160, esp8266_2m_160, esp01_1m_full_160, nodemcuv2_compat, esp8266_2m_compat, esp01_1m_full_compat, esp32dev, esp32dev_V4, esp32_eth, lolin_s2_mini, esp32c3dev, esp32s3dev_16MB_opi, esp32s3dev_8MB_opi, esp32s3_4M_qspi, esp32_wrover, usermods
+default_envs = nodemcuv2, esp8266_2m, esp01_1m_full, nodemcuv2_160, esp8266_2m_160, esp01_1m_full_160, nodemcuv2_compat, esp8266_2m_compat, esp01_1m_full_compat, esp32dev, esp32_eth, lolin_s2_mini, esp32c3dev, esp32s3dev_16MB_opi, esp32s3dev_8MB_opi, esp32s3_4M_qspi, esp32_wrover, usermods
 
 src_dir  = ./wled00
 data_dir = ./wled00/data
@@ -142,7 +142,8 @@ lib_deps =
     IRremoteESP8266 @ 2.8.2
     makuna/NeoPixelBus @ 2.8.3
     #https://github.com/makuna/NeoPixelBus.git#CoreShaderBeta
-    https://github.com/Aircoookie/ESPAsyncWebServer.git#v2.4.0
+    https://github.com/Aircoookie/ESPAsyncWebServer.git#v2.4.2
+    marvinroger/AsyncMqttClient @ 0.9.0
   # for I2C interface
     ;Wire
   # ESP-NOW library
@@ -234,25 +235,20 @@ lib_deps_compat =
 
 [esp32_all_variants]
 lib_deps =
-  willmmiles/AsyncTCP @ 1.3.1
+  esp32async/AsyncTCP @ 3.4.7
   bitbank2/AnimatedGIF@^1.4.7
   https://github.com/Aircoookie/GifDecoder#bc3af18
 build_flags =
   -D CONFIG_ASYNC_TCP_USE_WDT=0
+  -D CONFIG_ASYNC_TCP_STACK_SIZE=8192
   -D WLED_ENABLE_GIF
 
 [esp32]
-#platform = https://github.com/tasmota/platform-espressif32/releases/download/v2.0.2.3/platform-espressif32-2.0.2.3.zip
-platform = espressif32@3.5.0
-platform_packages = framework-arduinoespressif32 @ https://github.com/Aircoookie/arduino-esp32.git#1.0.6.4
+platform = ${esp32_idf_V4.platform}
+platform_packages =
 build_unflags = ${common.build_unflags}
-build_flags = -g
-  -DARDUINO_ARCH_ESP32
-  #-DCONFIG_LITTLEFS_FOR_IDF_3_2
-  #use LITTLEFS library by lorol in ESP32 core 1.x.x instead of built-in in 2.x.x
-  -D LOROL_LITTLEFS
-  ; -DARDUINO_USB_CDC_ON_BOOT=0 ;; this flag is mandatory for "classic ESP32" when building with arduino-esp32 >=2.0.3
-  ${esp32_all_variants.build_flags}
+build_flags = ${esp32_idf_V4.build_flags}
+lib_deps = ${esp32_idf_V4.lib_deps}
 
 tiny_partitions = tools/WLED_ESP32_2MB_noOTA.csv
 default_partitions = tools/WLED_ESP32_4MB_1MB_FS.csv
@@ -260,10 +256,7 @@ extended_partitions = tools/WLED_ESP32_4MB_700k_FS.csv
 big_partitions = tools/WLED_ESP32_4MB_256KB_FS.csv     ;; 1.8MB firmware, 256KB filesystem, coredump support
 large_partitions = tools/WLED_ESP32_8MB.csv
 extreme_partitions = tools/WLED_ESP32_16MB_9MB_FS.csv
-lib_deps =
-  https://github.com/lorol/LITTLEFS.git
-  ${esp32_all_variants.lib_deps}
-  ${env.lib_deps}
+
 board_build.partitions = ${esp32.default_partitions}   ;; default partioning for 4MB Flash - can be overridden in build envs
 # additional build flags for audioreactive - must be applied globally
 AR_build_flags = ;; -fsingle-precision-constant ;; forces ArduinoFFT to use float math (2x faster)
@@ -271,8 +264,7 @@ AR_lib_deps =  ;; for pre-usermod-library platformio_override compatibility
 
 
 [esp32_idf_V4]
-;; experimental build environment for ESP32 using ESP-IDF 4.4.x / arduino-esp32 v2.0.5
-;; very similar to the normal ESP32 flags, but omitting Lorol LittleFS, as littlefs is included in the new framework already.
+;; build environment for ESP32 using ESP-IDF 4.4.x / arduino-esp32 v2.0.5
 ;;
 ;; please note that you can NOT update existing ESP32 installs with a "V4" build. Also updating by OTA will not work properly.
 ;; You need to completely erase your device (esptool erase_flash) first, then install the "V4" build from VSCode+platformio.
@@ -283,14 +275,12 @@ build_unflags = ${common.build_unflags}
 build_flags = -g
   -Wshadow=compatible-local ;; emit warning in case a local variable "shadows" another local one
   -DARDUINO_ARCH_ESP32 -DESP32
-  -DARDUINO_USB_CDC_ON_BOOT=0 ;; this flag is mandatory for "classic ESP32" when building with arduino-esp32 >=2.0.3
   ${esp32_all_variants.build_flags}
   -D WLED_ENABLE_DMX_INPUT
 lib_deps =
   ${esp32_all_variants.lib_deps}
   https://github.com/someweisguy/esp_dmx.git#47db25d
   ${env.lib_deps}
-board_build.partitions = ${esp32.default_partitions}   ;; default partioning for 4MB Flash - can be overridden in build envs
 
 [esp32s2]
 ;; generic definitions for all ESP32-S2 boards
@@ -305,10 +295,9 @@ build_flags = -g
   -DARDUINO_USB_MODE=0 ;; this flag is mandatory for ESP32-S2 !
   ;; please make sure that the following flags are properly set (to 0 or 1) by your board.json, or included in your custom platformio_override.ini entry:
   ;; ARDUINO_USB_CDC_ON_BOOT
-  ${esp32_all_variants.build_flags}
+  ${esp32_idf_V4.build_flags}
 lib_deps =
-  ${esp32_all_variants.lib_deps}
-  ${env.lib_deps}
+  ${esp32_idf_V4.lib_deps}
 board_build.partitions = ${esp32.default_partitions}   ;; default partioning for 4MB Flash - can be overridden in build envs
 
 [esp32c3]
@@ -323,10 +312,9 @@ build_flags = -g
   -DARDUINO_USB_MODE=1 ;; this flag is mandatory for ESP32-C3
   ;; please make sure that the following flags are properly set (to 0 or 1) by your board.json, or included in your custom platformio_override.ini entry:
   ;; ARDUINO_USB_CDC_ON_BOOT
-  ${esp32_all_variants.build_flags}
+  ${esp32_idf_V4.build_flags}
 lib_deps =
-  ${esp32_all_variants.lib_deps}
-  ${env.lib_deps}
+  ${esp32_idf_V4.lib_deps}
 board_build.partitions = ${esp32.default_partitions}   ;; default partioning for 4MB Flash - can be overridden in build envs
 board_build.flash_mode = qio
 
@@ -343,10 +331,9 @@ build_flags = -g
   -DCO
   ;; please make sure that the following flags are properly set (to 0 or 1) by your board.json, or included in your custom platformio_override.ini entry:
   ;; ARDUINO_USB_MODE, ARDUINO_USB_CDC_ON_BOOT
-  ${esp32_all_variants.build_flags}
+  ${esp32_idf_V4.build_flags}
 lib_deps =
-  ${esp32_all_variants.lib_deps}
-  ${env.lib_deps}
+  ${esp32_idf_V4.lib_deps}
 board_build.partitions = ${esp32.large_partitions}   ;; default partioning for 8MB flash - can be overridden in build envs
 
 
@@ -441,21 +428,11 @@ custom_usermods = audioreactive
 
 [env:esp32dev]
 board = esp32dev
-platform = ${esp32.platform}
-platform_packages = ${esp32.platform_packages}
-custom_usermods = audioreactive
-build_unflags = ${common.build_unflags}
-build_flags = ${common.build_flags} ${esp32.build_flags} -D WLED_RELEASE_NAME=\"ESP32\" #-D WLED_DISABLE_BROWNOUT_DET
-lib_deps = ${esp32.lib_deps}
-monitor_filters = esp32_exception_decoder
-board_build.partitions = ${esp32.default_partitions}
-
-[env:esp32dev_V4]
-board = esp32dev
 platform = ${esp32_idf_V4.platform}
 build_unflags = ${common.build_unflags}
 custom_usermods = audioreactive
 build_flags = ${common.build_flags} ${esp32_idf_V4.build_flags} -D WLED_RELEASE_NAME=\"ESP32_V4\" #-D WLED_DISABLE_BROWNOUT_DET
+              -DARDUINO_USB_CDC_ON_BOOT=0 ;; this flag is mandatory for "classic ESP32" when building with arduino-esp32 >=2.0.3
 lib_deps = ${esp32_idf_V4.lib_deps}
 monitor_filters = esp32_exception_decoder
 board_build.partitions = ${esp32.default_partitions}
@@ -489,23 +466,9 @@ board_upload.maximum_size = 16777216
 board_build.f_flash = 80000000L
 board_build.flash_mode = dio
 
-;[env:esp32dev_audioreactive]
-;board = esp32dev
-;platform = ${esp32.platform}
-;platform_packages = ${esp32.platform_packages}
-;custom_usermods = audioreactive
-;build_unflags = ${common.build_unflags}
-;build_flags = ${common.build_flags} ${esp32.build_flags} -D WLED_RELEASE_NAME=\"ESP32_audioreactive\" #-D WLED_DISABLE_BROWNOUT_DET
-;lib_deps = ${esp32.lib_deps}
-;monitor_filters = esp32_exception_decoder
-;board_build.partitions = ${esp32.default_partitions}
-;; board_build.f_flash = 80000000L
-;; board_build.flash_mode = dio
-
 [env:esp32_eth]
 board = esp32-poe
-platform = ${esp32.platform}
-platform_packages = ${esp32.platform_packages}
+platform = ${esp32_idf_V4.platform}
 upload_speed = 921600
 custom_usermods = audioreactive
 build_unflags = ${common.build_unflags}
@@ -513,10 +476,10 @@ build_flags = ${common.build_flags} ${esp32.build_flags} -D WLED_RELEASE_NAME=\"
 ;  -D WLED_DISABLE_ESPNOW ;; ESP-NOW requires wifi, may crash with ethernet only
 lib_deps = ${esp32.lib_deps}
 board_build.partitions = ${esp32.default_partitions}
+board_build.flash_mode = dio
 
 [env:esp32_wrover]
 extends = esp32_idf_V4
-platform = ${esp32_idf_V4.platform}
 board = ttgo-t7-v14-mini32
 board_build.f_flash = 80000000L
 board_build.flash_mode = qio

platformio_override.sample.ini

@@ -28,7 +28,6 @@ lib_deps = ${esp8266.lib_deps}
 ;  robtillaart/SHT85@~0.3.3
 ;  ;gmag11/QuickESPNow @ ~0.7.0 # will also load QuickDebug
 ;  https://github.com/blazoncek/QuickESPNow.git#optional-debug  ;; exludes debug library
-;  bitbank2/PNGdec@^1.0.1 ;; used for POV display uncomment following
 ;  ${esp32.AR_lib_deps} ;; needed for USERMOD_AUDIOREACTIVE
 
 build_unflags = ${common.build_unflags}

readme.md

@@ -10,10 +10,12 @@
 
   </p>
 
-# Welcome to my project WLED! ✨
+# Welcome to WLED! ✨
 
 A fast and feature-rich implementation of an ESP32 and ESP8266 webserver to control NeoPixel (WS2812B, WS2811, SK6812) LEDs or also SPI based chipsets like the WS2801 and APA102!
 
+Originally created by [Aircoookie](https://github.com/Aircoookie)
+
 ## ⚙️ Features
 - WS2812FX library with more than 100 special effects  
 - FastLED noise effects and 50 palettes  
@@ -32,7 +34,7 @@ A fast and feature-rich implementation of an ESP32 and ESP8266 webserver to cont
 - Filesystem-based config for easier backup of presets and settings  
 
 ## 💡 Supported light control interfaces
-- WLED app for [Android](https://play.google.com/store/apps/details?id=com.aircoookie.WLED) and [iOS](https://apps.apple.com/us/app/wled/id1475695033)
+- WLED app for [Android](https://play.google.com/store/apps/details?id=ca.cgagnier.wlednativeandroid) and [iOS](https://apps.apple.com/gb/app/wled-native/id6446207239)
 - JSON and HTTP request APIs  
 - MQTT   
 - E1.31, Art-Net, DDP and TPM2.net
@@ -63,6 +65,7 @@ See [here](https://kno.wled.ge/basics/compatible-hardware)!
 
 Licensed under the EUPL v1.2 license  
 Credits [here](https://kno.wled.ge/about/contributors/)!
+CORS proxy by [Corsfix](https://corsfix.com/)
 
 Join the Discord server to discuss everything about WLED!
 

usermods/platformio_override.usermods.ini

@@ -2,7 +2,7 @@
 default_envs = usermods_esp32, usermods_esp32c3, usermods_esp32s2, usermods_esp32s3
 
 [env:usermods_esp32]
-extends = env:esp32dev_V4
+extends = env:esp32dev
 custom_usermods = ${usermods.custom_usermods}
 board_build.partitions = ${esp32.extreme_partitions}  ; We're gonna need a bigger boat
 
@@ -28,4 +28,4 @@ board_build.partitions = ${esp32.extreme_partitions}  ; We're gonna need a bigge
 
 
 [usermods]
-# Added in CI
+# Added in CI

usermods/pov_display/README.md

@@ -0,0 +1,48 @@
+## POV Display usermod
+
+This usermod adds a new effect called “POV Image”.
+
+![the usermod at work](pov_display.gif?raw=true)
+
+###How does it work?
+With proper configuration (see below) the main segment will display a single row of pixels from an image stored on the ESP.
+It displays the image row by row at a high refresh rate.
+If you move the pixel segment at the right speed, you will see the full image floating in the air thanks to the persistence of vision.
+RGB LEDs only (no RGBW), with grouping set to 1 and spacing set to 0.
+Best results with high-density strips (e.g., 144 LEDs/m).
+
+To get it working:
+- Resize your image. The height must match the number of LEDs in your strip/segment.
+- Rotate your image 90° clockwise (height becomes width).
+- Upload a BMP image (24-bit, uncompressed) to the ESP filesystem using the “/edit” URL.
+- Select the “POV Image” effect.
+- Set the segment name to the absolute filesystem path of the image (e.g., “/myimage.bmp”).
+- The path is case-sensitive and must start with “/”.
+- Rotate the pixel strip at approximately 20 RPM.
+- Tune as needed so that one full revolution maps to the image width (if the image appears stretched or compressed, adjust RPM slightly).
+- Enjoy the show!
+
+Notes:
+- Only 24-bit uncompressed BMP files are supported.
+- The image must fit into ~64 KB of RAM (width × height × 3 bytes, plus row padding to a 4-byte boundary).
+- Examples (approximate, excluding row padding):
+  - 128×128 (49,152 bytes) fits.
+  - 160×160 (76,800 bytes) does NOT fit.
+  - 96×192 (55,296 bytes) fits; padding may add a small overhead.
+- If the rendered image appears mirrored or upside‑down, rotate 90° the other way or flip horizontally in your editor and try again.
+- The path must be absolute.
+
+### Requirements
+- 1D rotating LED strip/segment (POV setup). Ensure the segment length equals the number of physical LEDs.
+- BMP image saved as 24‑bit, uncompressed (no alpha, no palette).
+- Sufficient free RAM (~64 KB) for the image buffer.
+
+### Troubleshooting
+- Nothing displays: verify the file exists at the exact absolute path (case‑sensitive) and is a 24‑bit uncompressed BMP.
+- Garbled colors or wrong orientation: re‑export as 24‑bit BMP and retry the rotation/flip guidance above.
+- Image too large: reduce width and/or height until it fits within ~64 KB (see examples).
+- Path issues: confirm you uploaded the file via the “/edit” URL and can see it in the filesystem browser.
+
+### Safety
+- Secure the rotating assembly and keep clear of moving parts.
+- Balance the strip/hub to minimize vibration before running at speed.

usermods/pov_display/bmpimage.cpp

@@ -0,0 +1,146 @@
+#include "bmpimage.h"
+#define BUF_SIZE 64000
+
+byte * _buffer = nullptr;
+
+uint16_t read16(File &f) {
+  uint16_t result;
+  f.read((uint8_t *)&result,2);
+  return result;
+}
+
+uint32_t read32(File &f) {
+  uint32_t result;
+  f.read((uint8_t *)&result,4);
+  return result;
+}
+
+bool BMPimage::init(const char * fn) {
+    File bmpFile;
+    int bmpDepth;
+    //first, check if filename exists
+    if (!WLED_FS.exists(fn)) {
+      return false;
+    }
+    
+    bmpFile = WLED_FS.open(fn);
+    if (!bmpFile) {
+      _valid=false;
+      return false;
+    }
+
+    //so, the file exists and is opened
+    // Parse BMP header
+    uint16_t header = read16(bmpFile);
+    if(header != 0x4D42) { // BMP signature
+      _valid=false;
+      bmpFile.close();
+      return false;
+    }
+
+    //read and ingnore file size
+    read32(bmpFile);
+    (void)read32(bmpFile); // Read & ignore creator bytes
+    _imageOffset = read32(bmpFile); // Start of image data
+    // Read DIB header
+    read32(bmpFile);
+    _width  = read32(bmpFile);
+    _height = read32(bmpFile);
+    if(read16(bmpFile) != 1) { // # planes -- must be '1'
+        _valid=false;
+        bmpFile.close();
+        return false;
+    }
+    bmpDepth = read16(bmpFile); // bits per pixel
+    if((bmpDepth != 24) || (read32(bmpFile) != 0)) { // 0 = uncompressed {
+        _width=0;
+        _valid=false;
+        bmpFile.close();
+        return false;
+    }
+    // If _height is negative, image is in top-down order.
+    // This is not canon but has been observed in the wild.
+    if(_height < 0) {
+        _height = -_height;
+    }
+    //now, we have successfully got all the basics
+    // BMP rows are padded (if needed) to 4-byte boundary
+    _rowSize = (_width * 3 + 3) & ~3;
+    //check image size - if it is too large, it will be unusable
+    if (_rowSize*_height>BUF_SIZE) {
+      _valid=false;
+      bmpFile.close();
+      return false;
+    }
+
+    bmpFile.close();
+    // Ensure filename fits our buffer (segment name length constraint).
+    size_t len = strlen(fn);
+    if (len > WLED_MAX_SEGNAME_LEN) {
+      return false;
+    }
+    strncpy(filename, fn, sizeof(filename));
+    filename[sizeof(filename) - 1] = '\0';
+    _valid = true;
+    return true;
+}
+
+void BMPimage::clear(){
+    strcpy(filename, "");
+    _width=0;
+    _height=0;
+    _rowSize=0;
+    _imageOffset=0;
+    _loaded=false;
+    _valid=false;
+}
+
+bool BMPimage::load(){
+    const size_t size = (size_t)_rowSize * (size_t)_height;
+    if (size > BUF_SIZE) {
+        return false;
+    }
+    File bmpFile = WLED_FS.open(filename);
+    if (!bmpFile) {
+        return false;
+    }
+
+    if (_buffer != nullptr) free(_buffer);
+    _buffer = (byte*)malloc(size);
+    if (_buffer == nullptr) return false;
+
+    bmpFile.seek(_imageOffset);
+    const size_t readBytes = bmpFile.read(_buffer, size);
+    bmpFile.close();
+    if (readBytes != size) {
+        _loaded = false;
+        return false;
+    }
+    _loaded = true;
+    return true;
+}
+
+byte* BMPimage::line(uint16_t n){
+    if (_loaded) {
+        return (_buffer+n*_rowSize);
+    } else {
+        return NULL;
+    }
+}
+
+uint32_t BMPimage::pixelColor(uint16_t x, uint16_t  y){
+    uint32_t pos;
+    byte b,g,r; //colors
+    if (! _loaded) {
+      return 0;
+    }
+    if ( (x>=_width) || (y>=_height) ) {
+      return 0;
+    }
+    pos=y*_rowSize + 3*x;
+    //get colors. Note that in BMP files, they go in BGR order
+    b= _buffer[pos++];
+    g= _buffer[pos++];
+    r= _buffer[pos];
+    return (r<<16|g<<8|b);
+}

usermods/pov_display/bmpimage.h

@@ -0,0 +1,50 @@
+#ifndef _BMPIMAGE_H
+#define _BMPIMAGE_H
+#include "Arduino.h"
+#include "wled.h"
+
+/*
+ * This class describes a bitmap image. Each object refers to a bmp file on
+ * filesystem fatfs.
+ * To initialize, call init(), passign to it name of a bitmap file
+ * at the root of fatfs filesystem:
+ *
+ * BMPimage myImage;
+ * myImage.init("logo.bmp");
+ *
+ * For performance reasons, before actually usign the image, you need to load
+ * it from filesystem to RAM:
+ * myImage.load();
+ * All load() operations use the same reserved buffer in RAM, so you can only
+ * have one file loaded at a time. Before loading a new file, always unload the
+ * previous one:
+ * myImage.unload();
+ */
+
+class BMPimage {
+    public:
+        int height()    {return _height; }
+        int width()     {return _width;  }
+        int rowSize()   {return _rowSize;}
+        bool isLoaded() {return _loaded; }
+        bool load();
+        void unload()   {_loaded=false;  }
+        byte * line(uint16_t n);
+        uint32_t pixelColor(uint16_t x,uint16_t  y);
+        bool init(const char* fn);
+        void clear();
+        char * getFilename() {return filename;};
+
+    private:
+        char filename[WLED_MAX_SEGNAME_LEN+1]="";
+        int _width=0;
+        int _height=0;
+        int _rowSize=0;
+        int _imageOffset=0;
+        bool _loaded=false;
+        bool _valid=false;
+};
+
+extern byte * _buffer;
+
+#endif

usermods/pov_display/library.json

@@ -1,7 +1,5 @@
 {
   "name:": "pov_display",
   "build": { "libArchive": false},
-  "dependencies": {
-    "bitbank2/PNGdec":"^1.0.3"
-  }
+  "platforms": ["espressif32"]
 }

usermods/pov_display/pov.cpp

@@ -0,0 +1,47 @@
+#include "pov.h"
+
+POV::POV() {}
+
+void POV::showLine(const byte * line, uint16_t size){
+    uint16_t i, pos;
+    uint8_t r, g, b;
+    if (!line) {
+        // All-black frame on null input
+        for (i = 0; i < SEGLEN; i++) {
+            SEGMENT.setPixelColor(i, CRGB::Black);
+        }
+        strip.show();
+        lastLineUpdate = micros();
+        return;
+    }
+    for (i = 0; i < SEGLEN; i++) {
+        if (i < size) {
+            pos = 3 * i;
+            // using bgr order
+            b = line[pos++];
+            g = line[pos++];
+            r = line[pos];
+            SEGMENT.setPixelColor(i, CRGB(r, g, b));
+        } else {
+            SEGMENT.setPixelColor(i, CRGB::Black);
+        }
+    }
+    strip.show();
+    lastLineUpdate = micros();
+}
+
+bool POV::loadImage(const char * filename){
+  if(!image.init(filename)) return false;
+  if(!image.load()) return false;
+  currentLine=0;
+  return true;
+}
+
+int16_t POV::showNextLine(){
+    if (!image.isLoaded()) return 0;
+    //move to next line
+    showLine(image.line(currentLine), image.width());
+    currentLine++;
+    if (currentLine == image.height()) {currentLine=0;}
+    return currentLine;
+}

usermods/pov_display/pov.h

@@ -0,0 +1,42 @@
+#ifndef _POV_H
+#define _POV_H
+#include "bmpimage.h"
+
+
+class POV {
+    public:
+        POV();
+
+        /* Shows one line. line should be pointer to array which holds  pixel colors
+         * (3 bytes per pixel, in BGR order). Note: 3, not 4!!!
+         *  size should be size of array (number of pixels, not number of bytes)
+         */
+        void showLine(const byte * line, uint16_t size);
+
+        /* Reads from file an image and making it current image */
+        bool loadImage(const char * filename);
+
+        /* Show next line of active image
+           Retunrs the index of next line to be shown (not yet shown!)
+           If it retunrs 0, it means we have completed showing the image and
+            next call will start again
+        */
+        int16_t showNextLine();
+
+        //time since strip was last updated, in micro sec
+        uint32_t timeSinceUpdate() {return (micros()-lastLineUpdate);}
+
+
+        BMPimage * currentImage() {return ℑ}
+
+        char * getFilename() {return image.getFilename();}
+
+    private:
+        BMPimage image;
+        int16_t  currentLine=0;     //next line to be shown
+        uint32_t lastLineUpdate=0; //time in microseconds
+};
+
+
+
+#endif

usermods/pov_display/pov_display.cpp

@@ -1,88 +1,75 @@
 #include "wled.h"
-#include <PNGdec.h>
+#include "pov.h"
 
-void * openFile(const char *filename, int32_t *size) {
-    f = WLED_FS.open(filename);
-    *size = f.size();
-    return &f;
-}
+static const char _data_FX_MODE_POV_IMAGE[] PROGMEM = "POV Image@!;;;;";
 
-void closeFile(void *handle) {
-    if (f) f.close();
-}
-
-int32_t readFile(PNGFILE *pFile, uint8_t *pBuf, int32_t iLen)
-{
-    int32_t iBytesRead;
-    iBytesRead = iLen;
-    File *f = static_cast<File *>(pFile->fHandle);
-    // Note: If you read a file all the way to the last byte, seek() stops working
-    if ((pFile->iSize - pFile->iPos) < iLen)
-	iBytesRead = pFile->iSize - pFile->iPos - 1; // <-- ugly work-around
-    if (iBytesRead <= 0)
-	return 0;
-    iBytesRead = (int32_t)f->read(pBuf, iBytesRead);
-    pFile->iPos = f->position();
-    return iBytesRead;
-}
-
-int32_t seekFile(PNGFILE *pFile, int32_t iPosition)
-{
-    int i = micros();
-    File *f = static_cast<File *>(pFile->fHandle);
-    f->seek(iPosition);
-    pFile->iPos = (int32_t)f->position();
-    i = micros() - i;
-    return pFile->iPos;
-}
-
-void draw(PNGDRAW *pDraw) {
-    uint16_t usPixels[SEGLEN];
-    png.getLineAsRGB565(pDraw, usPixels, PNG_RGB565_LITTLE_ENDIAN, 0xffffffff);
-    for(int x=0; x < SEGLEN; x++) {
-	uint16_t color = usPixels[x];
-	byte r = ((color >> 11) & 0x1F);
-	byte g = ((color >> 5) & 0x3F);
-	byte b = (color & 0x1F);
-	SEGMENT.setPixelColor(x, RGBW32(r,g,b,0));
-    }
-    strip.show();
-}
+static POV s_pov;
 
 uint16_t mode_pov_image(void) {
-    const char * filepath = SEGMENT.name;
-    int rc = png.open(filepath, openFile, closeFile, readFile, seekFile, draw);
-    if (rc == PNG_SUCCESS) {
-	rc = png.decode(NULL, 0);
-	png.close();
-	return FRAMETIME;
-    }
+  Segment& mainseg = strip.getMainSegment();
+  const char* segName = mainseg.name;
+  if (!segName) {
+     return FRAMETIME;
+   }
+  // Only proceed for files ending with .bmp (case-insensitive)
+  size_t segLen = strlen(segName);
+  if (segLen < 4) return FRAMETIME;
+  const char* ext = segName + (segLen - 4);
+  // compare case-insensitive to ".bmp"
+  if (!((ext[0]=='.') &&
+        (ext[1]=='b' || ext[1]=='B') &&
+        (ext[2]=='m' || ext[2]=='M') &&
+        (ext[3]=='p' || ext[3]=='P'))) {
     return FRAMETIME;
+  }
+
+  const char* current = s_pov.getFilename();
+  if (current && strcmp(segName, current) == 0) {
+     s_pov.showNextLine();
+     return FRAMETIME;
+   }
+
+  static unsigned long s_lastLoadAttemptMs = 0;
+  unsigned long nowMs = millis();
+  // Retry at most twice per second if the image is not yet loaded.
+  if (nowMs - s_lastLoadAttemptMs < 500) return FRAMETIME;
+  s_lastLoadAttemptMs = nowMs;
+  s_pov.loadImage(segName);
+  return FRAMETIME;
 }
 
-class PovDisplayUsermod : public Usermod
-{
-  public:
-    static const char _data_FX_MODE_POV_IMAGE[] PROGMEM = "POV Image@!;;;1";
+class PovDisplayUsermod : public Usermod {
+protected:
+  bool enabled = false; //WLEDMM
+  const char *_name; //WLEDMM
+  bool initDone = false; //WLEDMM
+  unsigned long lastTime = 0; //WLEDMM
+public:
 
-    PNG png;
-    File f;
+  PovDisplayUsermod(const char *name, bool enabled)
+    : enabled(enabled) , _name(name) {}
+  
+  void setup() override {
+    strip.addEffect(255, &mode_pov_image, _data_FX_MODE_POV_IMAGE);
+    //initDone removed (unused)
+  }
 
-    void setup() {
-	strip.addEffect(255, &mode_pov_image, _data_FX_MODE_POV_IMAGE);
+
+  void loop() override {
+    // if usermod is disabled or called during strip updating just exit
+    // NOTE: on very long strips strip.isUpdating() may always return true so update accordingly
+    if (!enabled || strip.isUpdating()) return;
+
+    // do your magic here
+    if (millis() - lastTime > 1000) {
+      lastTime = millis();
     }
+  }
 
-    void loop() {
-    }
-
-    uint16_t getId()
-    {
-      return USERMOD_ID_POV_DISPLAY;
-    }
-
-    void connected() {}
+  uint16_t getId() override {
+    return USERMOD_ID_POV_DISPLAY;
+  }
 };
 
-
-static PovDisplayUsermod pov_display;
-REGISTER_USERMOD(pov_display);
+static PovDisplayUsermod pov_display("POV Display", false);
+REGISTER_USERMOD(pov_display);

usermods/udp_name_sync/library.json

@@ -0,0 +1,5 @@
+{
+  "name": "udp_name_sync",
+  "build": { "libArchive": false },
+  "dependencies": {}
+}

usermods/udp_name_sync/udp_name_sync.cpp

@@ -0,0 +1,85 @@
+#include "wled.h"
+
+class UdpNameSync : public Usermod {
+
+  private:
+
+    bool enabled = false;
+    char segmentName[WLED_MAX_SEGNAME_LEN] = {0};
+    static constexpr uint8_t kPacketType = 200; // custom usermod packet type
+    static const char _name[];
+    static const char _enabled[];
+
+  public:
+    /**
+     * Enable/Disable the usermod
+     */
+    inline void enable(bool value) { enabled = value; }
+
+    /**
+     * Get usermod enabled/disabled state
+     */
+    inline bool isEnabled() const { return enabled; }
+
+    void setup() override {
+      // Enabled when this usermod is compiled, set to false if you prefer runtime opt-in
+      enable(true);
+    }
+
+    void loop() override {
+      if (!enabled) return;
+      if (!WLED_CONNECTED) return;
+      if (!udpConnected) return;
+      Segment& mainseg = strip.getMainSegment();
+      if (segmentName[0] == '\0' && !mainseg.name) return; //name was never set, do nothing
+
+      const char* curName = mainseg.name ? mainseg.name : "";
+      if (strncmp(curName, segmentName, sizeof(segmentName)) == 0) return; // same name, do nothing
+
+      IPAddress broadcastIp = uint32_t(Network.localIP()) | ~uint32_t(Network.subnetMask());
+      byte udpOut[WLED_MAX_SEGNAME_LEN + 2];
+      udpOut[0] = kPacketType; // custom usermod packet type (avoid 0..5 used by core protocols)
+
+      if (segmentName[0] != '\0' && !mainseg.name) { // name cleared
+        notifierUdp.beginPacket(broadcastIp, udpPort);
+        segmentName[0] = '\0';
+        DEBUG_PRINTLN(F("UdpNameSync: sending empty name"));
+        udpOut[1] = 0; // explicit empty string
+        notifierUdp.write(udpOut, 2);
+        notifierUdp.endPacket();
+        return;
+      }
+
+      notifierUdp.beginPacket(broadcastIp, udpPort);
+      DEBUG_PRINT(F("UdpNameSync: saving segment name "));
+      DEBUG_PRINTLN(curName);
+      strlcpy(segmentName, curName, sizeof(segmentName));
+      strlcpy((char *)&udpOut[1], segmentName, sizeof(udpOut) - 1); // leave room for header byte
+      size_t nameLen = strnlen((char *)&udpOut[1], sizeof(udpOut) - 1);
+      notifierUdp.write(udpOut, 2 + nameLen);
+      notifierUdp.endPacket();
+      DEBUG_PRINT(F("UdpNameSync: Sent segment name : "));
+      DEBUG_PRINTLN(segmentName);
+      return;
+    }
+
+    bool onUdpPacket(uint8_t * payload, size_t len) override {
+      DEBUG_PRINT(F("UdpNameSync: Received packet"));
+      if (!enabled) return false;
+      if (receiveDirect) return false;
+      if (len < 2) return false;                 // need type + at least 1 byte for name (can be 0)
+      if (payload[0] != kPacketType) return false;
+      Segment& mainseg = strip.getMainSegment();
+      char tmp[WLED_MAX_SEGNAME_LEN] = {0};
+      size_t copyLen = len - 1;
+      if (copyLen > sizeof(tmp) - 1) copyLen = sizeof(tmp) - 1;
+      memcpy(tmp, &payload[1], copyLen);
+      tmp[copyLen] = '\0';
+      mainseg.setName(tmp);
+      DEBUG_PRINT(F("UdpNameSync: set segment name"));
+      return true;
+     }
+};
+
+static UdpNameSync udp_name_sync;
+REGISTER_USERMOD(udp_name_sync);

wled00/FX.cpp

@@ -2328,7 +2328,7 @@ uint16_t mode_colortwinkle() {
       }
 
       if (cur == prev) {  //fix "stuck" pixels
-        color_add(col, col);
+        col = color_add(col, col);
         SEGMENT.setPixelColor(i, col);
       }
       else SEGMENT.setPixelColor(i, col);
@@ -3940,7 +3940,7 @@ uint16_t mode_percent(void) {
 
  	return FRAMETIME;
 }
-static const char _data_FX_MODE_PERCENT[] PROGMEM = "Percent@,% of fill,,,,One color;!,!;!";
+static const char _data_FX_MODE_PERCENT[] PROGMEM = "Percent@!,% of fill,,,,One color;!,!;!";
 
 
 /*
@@ -4766,30 +4766,17 @@ class AuroraWave {
 };
 
 uint16_t mode_aurora(void) {
-  //aux1 = Wavecount
-  //aux2 = Intensity in last loop
-
   AuroraWave* waves;
+  SEGENV.aux1 = map(SEGMENT.intensity, 0, 255, 2, W_MAX_COUNT); // aux1 = Wavecount
+  if(!SEGENV.allocateData(sizeof(AuroraWave) * SEGENV.aux1)) {  // 20 on ESP32, 9 on ESP8266
+    return mode_static(); //allocation failed
+  }
+  waves = reinterpret_cast<AuroraWave*>(SEGENV.data);
 
-//TODO: I am not sure this is a correct way of handling memory allocation since if it fails on 1st run
-// it will display static effect but on second run it may crash ESP since data will be nullptr
-
-  if(SEGENV.aux0 != SEGMENT.intensity || SEGENV.call == 0) {
-    //Intensity slider changed or first call
-    SEGENV.aux1 = map(SEGMENT.intensity, 0, 255, 2, W_MAX_COUNT);
-    SEGENV.aux0 = SEGMENT.intensity;
-
-    if(!SEGENV.allocateData(sizeof(AuroraWave) * SEGENV.aux1)) { // 26 on 32 segment ESP32, 9 on 16 segment ESP8266
-      return mode_static(); //allocation failed
-    }
-
-    waves = reinterpret_cast<AuroraWave*>(SEGENV.data);
-
+  if(SEGENV.call == 0) {
     for (int i = 0; i < SEGENV.aux1; i++) {
       waves[i].init(SEGLEN, CRGB(SEGMENT.color_from_palette(hw_random8(), false, false, hw_random8(0, 3))));
     }
-  } else {
-    waves = reinterpret_cast<AuroraWave*>(SEGENV.data);
   }
 
   for (int i = 0; i < SEGENV.aux1; i++) {
@@ -7541,9 +7528,9 @@ uint16_t mode_2Ddistortionwaves() {
       byte valueG = gdistort + ((a2-( ((xoffs - cx1) * (xoffs - cx1) + (yoffs - cy1) * (yoffs - cy1))>>7 ))<<1);
       byte valueB = bdistort + ((a3-( ((xoffs - cx2) * (xoffs - cx2) + (yoffs - cy2) * (yoffs - cy2))>>7 ))<<1);
 
-      valueR = gamma8(cos8_t(valueR));
-      valueG = gamma8(cos8_t(valueG));
-      valueB = gamma8(cos8_t(valueB));
+      valueR = cos8_t(valueR);
+      valueG = cos8_t(valueG);
+      valueB = cos8_t(valueB);
 
       if(SEGMENT.palette == 0) {
         // use RGB values (original color mode)
@@ -8495,7 +8482,6 @@ static const char _data_FX_MODE_PARTICLEPERLIN[] PROGMEM = "PS Fuzzy Noise@Speed
 #define NUMBEROFSOURCES 8
 uint16_t mode_particleimpact(void) {
   ParticleSystem2D *PartSys = nullptr;
-  uint32_t i = 0;
   uint32_t numMeteors;
   PSsettings2D meteorsettings;
   meteorsettings.asByte = 0b00101000; // PS settings for meteors: bounceY and gravity enabled
@@ -8508,7 +8494,7 @@ uint16_t mode_particleimpact(void) {
     PartSys->setBounceY(true); // always use ground bounce
     PartSys->setWallRoughness(220); // high roughness
     numMeteors = min(PartSys->numSources, (uint32_t)NUMBEROFSOURCES);
-    for (i = 0; i < numMeteors; i++) {
+    for (uint32_t i = 0; i < numMeteors; i++) {
       PartSys->sources[i].source.ttl = hw_random16(10 * i); // set initial delay for meteors
       PartSys->sources[i].source.vy = 10; // at positive speeds, no particles are emitted and if particle dies, it will be relaunched
     }
@@ -8530,7 +8516,7 @@ uint16_t mode_particleimpact(void) {
   numMeteors = min(PartSys->numSources, (uint32_t)NUMBEROFSOURCES);
   uint32_t emitparticles; // number of particles to emit for each rocket's state
 
-  for (i = 0; i < numMeteors; i++) {
+  for (uint32_t i = 0; i < numMeteors; i++) {
     // determine meteor state by its speed:
     if ( PartSys->sources[i].source.vy < 0) // moving down, emit sparks
       emitparticles = 1;
@@ -8546,7 +8532,7 @@ uint16_t mode_particleimpact(void) {
   }
 
   // update the meteors, set the speed state
-  for (i = 0; i < numMeteors; i++) {
+  for (uint32_t i = 0; i < numMeteors; i++) {
     if (PartSys->sources[i].source.ttl) {
       PartSys->sources[i].source.ttl--; // note: this saves an if statement, but moving down particles age twice
       if (PartSys->sources[i].source.vy < 0) { // move down
@@ -8837,7 +8823,7 @@ uint16_t mode_particleGEQ(void) {
         //set particle properties TODO: could also use the spray...
         PartSys->particles[i].ttl = 20 + map(SEGMENT.intensity, 0,255, emitspeed>>1, emitspeed + hw_random16(emitspeed)) ; // set particle alive, particle lifespan is in number of frames
         PartSys->particles[i].x = xposition + hw_random16(binwidth) - (binwidth>>1); // position randomly, deviating half a bin width
-        PartSys->particles[i].y = PS_P_RADIUS; // start at the bottom (PS_P_RADIUS is minimum position a particle is fully in frame)
+        PartSys->particles[i].y = 0; // start at the bottom
         PartSys->particles[i].vx = hw_random16(SEGMENT.custom1>>1)-(SEGMENT.custom1>>2) ; //x-speed variation: +/- custom1/4
         PartSys->particles[i].vy = emitspeed;
         PartSys->particles[i].hue = (bin<<4) + hw_random16(17) - 8; // color from palette according to bin

wled00/FX.h

@@ -686,6 +686,7 @@ class Segment {
 
     // 1D strip
     uint16_t virtualLength() const;
+    uint16_t maxMappingLength() const;
     [[gnu::hot]] void setPixelColor(int n, uint32_t c) const; // set relative pixel within segment with color
     inline void setPixelColor(unsigned n, uint32_t c) const                    { setPixelColor(int(n), c); }
     inline void setPixelColor(int n, byte r, byte g, byte b, byte w = 0) const { setPixelColor(n, RGBW32(r,g,b,w)); }
@@ -808,6 +809,8 @@ class Segment {
     inline void wu_pixel(uint32_t x, uint32_t y, CRGB c) {}
   #endif
   friend class WS2812FX;
+  friend class ParticleSystem2D;
+  friend class ParticleSystem1D;
 };
 
 // main "strip" class (108 bytes)

wled00/FX_fcn.cpp

@@ -66,13 +66,15 @@ Segment::Segment(const Segment &orig) {
   _dataLen = 0;
   pixels = nullptr;
   if (!stop) return;  // nothing to do if segment is inactive/invalid
-  if (orig.name) { name = static_cast<char*>(d_malloc(strlen(orig.name)+1)); if (name) strcpy(name, orig.name); }
-  if (orig.data) { if (allocateData(orig._dataLen)) memcpy(data, orig.data, orig._dataLen); }
   if (orig.pixels) {
+    // allocate pixel buffer: prefer IRAM/PSRAM
     pixels = static_cast<uint32_t*>(d_malloc(sizeof(uint32_t) * orig.length()));
-    if (pixels) memcpy(pixels, orig.pixels, sizeof(uint32_t) * orig.length());
-    else {
-      DEBUG_PRINTLN(F("!!! Not enough RAM for pixel buffer !!!"));
+    if (pixels) {
+      memcpy(pixels, orig.pixels, sizeof(uint32_t) * orig.length());
+      if (orig.name) { name = static_cast<char*>(d_malloc(strlen(orig.name)+1)); if (name) strcpy(name, orig.name); }
+      if (orig.data) { if (allocateData(orig._dataLen)) memcpy(data, orig.data, orig._dataLen); }
+    } else {
+      DEBUGFX_PRINTLN(F("!!! Not enough RAM for pixel buffer !!!"));
       errorFlag = ERR_NORAM_PX;
       stop = 0; // mark segment as inactive/invalid
     }
@@ -107,12 +109,14 @@ Segment& Segment::operator= (const Segment &orig) {
     pixels = nullptr;
     if (!stop) return *this;  // nothing to do if segment is inactive/invalid
     // copy source data
-    if (orig.name) { name = static_cast<char*>(d_malloc(strlen(orig.name)+1)); if (name) strcpy(name, orig.name); }
-    if (orig.data) { if (allocateData(orig._dataLen)) memcpy(data, orig.data, orig._dataLen); }
     if (orig.pixels) {
+      // allocate pixel buffer: prefer IRAM/PSRAM
       pixels = static_cast<uint32_t*>(d_malloc(sizeof(uint32_t) * orig.length()));
-      if (pixels) memcpy(pixels, orig.pixels, sizeof(uint32_t) * orig.length());
-      else {
+      if (pixels) {
+        memcpy(pixels, orig.pixels, sizeof(uint32_t) * orig.length());
+        if (orig.name) { name = static_cast<char*>(d_malloc(strlen(orig.name)+1)); if (name) strcpy(name, orig.name); }
+        if (orig.data) { if (allocateData(orig._dataLen)) memcpy(data, orig.data, orig._dataLen); }
+      } else {
         DEBUG_PRINTLN(F("!!! Not enough RAM for pixel buffer !!!"));
         errorFlag = ERR_NORAM_PX;
         stop = 0; // mark segment as inactive/invalid
@@ -159,8 +163,16 @@ bool Segment::allocateData(size_t len) {
     return false;
   }
   // prefer DRAM over SPI RAM on ESP32 since it is slow
-  if (data) data = (byte*)d_realloc(data, len);
-  else      data = (byte*)d_malloc(len);
+  if (data) {
+    data = (byte*)d_realloc_malloc(data, len); // realloc with malloc fallback
+    if (!data) {
+      data = nullptr;
+      Segment::addUsedSegmentData(-_dataLen); // subtract original buffer size
+      _dataLen = 0;   // reset data length
+    }
+  }
+  else data = (byte*)d_malloc(len);
+
   if (data) {
     memset(data, 0, len);  // erase buffer
     Segment::addUsedSegmentData(len - _dataLen);
@@ -170,7 +182,6 @@ bool Segment::allocateData(size_t len) {
   }
   // allocation failed
   DEBUG_PRINTLN(F("!!! Allocation failed. !!!"));
-  Segment::addUsedSegmentData(-_dataLen); // subtract original buffer size
   errorFlag = ERR_NORAM;
   return false;
 }
@@ -271,11 +282,13 @@ void Segment::startTransition(uint16_t dur, bool segmentCopy) {
       _t->_oldSegment = new(std::nothrow) Segment(*this); // store/copy current segment settings
       _t->_start = millis();                              // restart countdown
       _t->_dur   = dur;
+      _t->_prevPaletteBlends = 0;
       if (_t->_oldSegment) {
         _t->_oldSegment->palette = _t->_palette;          // restore original palette and colors (from start of transition)
         for (unsigned i = 0; i < NUM_COLORS; i++) _t->_oldSegment->colors[i] = _t->_colors[i];
+        DEBUGFX_PRINTF_P(PSTR("-- Updated transition with segment copy: S=%p T(%p) O[%p] OP[%p]\n"), this, _t, _t->_oldSegment, _t->_oldSegment->pixels);
+        if (!_t->_oldSegment->isActive()) stopTransition();
       }
-      DEBUG_PRINTF_P(PSTR("-- Updated transition with segment copy: S=%p T(%p) O[%p] OP[%p]\n"), this, _t, _t->_oldSegment, _t->_oldSegment->pixels);
     }
     return;
   }
@@ -291,13 +304,12 @@ void Segment::startTransition(uint16_t dur, bool segmentCopy) {
     #endif
     for (int i=0; i<NUM_COLORS; i++) _t->_colors[i] = colors[i];
     if (segmentCopy) _t->_oldSegment = new(std::nothrow) Segment(*this); // store/copy current segment settings
-    #ifdef WLED_DEBUG
     if (_t->_oldSegment) {
-      DEBUG_PRINTF_P(PSTR("-- Started transition: S=%p T(%p) O[%p] OP[%p]\n"), this, _t, _t->_oldSegment, _t->_oldSegment->pixels);
+      DEBUGFX_PRINTF_P(PSTR("-- Started transition: S=%p T(%p) O[%p] OP[%p]\n"), this, _t, _t->_oldSegment, _t->_oldSegment->pixels);
+      if (!_t->_oldSegment->isActive()) stopTransition();
     } else {
-      DEBUG_PRINTF_P(PSTR("-- Started transition without old segment: S=%p T(%p)\n"), this, _t);
+      DEBUGFX_PRINTF_P(PSTR("-- Started transition without old segment: S=%p T(%p)\n"), this, _t);
     }
-    #endif
   };
 }
 
@@ -357,6 +369,7 @@ void Segment::beginDraw(uint16_t prog) {
     // minimum blend time is 100ms maximum is 65535ms
     #ifndef WLED_SAVE_RAM
     unsigned noOfBlends = ((255U * prog) / 0xFFFFU) - _t->_prevPaletteBlends;
+    if(noOfBlends > 255) noOfBlends = 255; // safety check
     for (unsigned i = 0; i < noOfBlends; i++, _t->_prevPaletteBlends++) nblendPaletteTowardPalette(_t->_palT, Segment::_currentPalette, 48);
     Segment::_currentPalette = _t->_palT; // copy transitioning/temporary palette
     #else
@@ -418,14 +431,15 @@ void Segment::setGeometry(uint16_t i1, uint16_t i2, uint8_t grp, uint8_t spc, ui
 
   unsigned oldLength = length();
 
-  DEBUG_PRINTF_P(PSTR("Segment geometry: %d,%d -> %d,%d [%d,%d]\n"), (int)i1, (int)i2, (int)i1Y, (int)i2Y, (int)grp, (int)spc);
+  DEBUGFX_PRINTF_P(PSTR("Segment geometry: %d,%d -> %d,%d [%d,%d]\n"), (int)i1, (int)i2, (int)i1Y, (int)i2Y, (int)grp, (int)spc);
   markForReset();
-  startTransition(strip.getTransition()); // start transition prior to change (if segment is deactivated (start>stop) no transition will happen)
-  stateChanged = true; // send UDP/WS broadcast
+  if (_t) stopTransition(); // we can't use transition if segment dimensions changed
+  stateChanged = true;      // send UDP/WS broadcast
 
   // apply change immediately
   if (i2 <= i1) { //disable segment
-    d_free(pixels);
+    deallocateData();
+    p_free(pixels);
     pixels = nullptr;
     stop = 0;
     return;
@@ -442,21 +456,25 @@ void Segment::setGeometry(uint16_t i1, uint16_t i2, uint8_t grp, uint8_t spc, ui
   #endif
   // safety check
   if (start >= stop || startY >= stopY) {
-    d_free(pixels);
+    deallocateData();
+    p_free(pixels);
     pixels = nullptr;
     stop = 0;
     return;
   }
-  // re-allocate FX render buffer
+  // allocate FX render buffer
   if (length() != oldLength) {
-    if (pixels) pixels = static_cast<uint32_t*>(d_realloc(pixels, sizeof(uint32_t) * length()));
-    else        pixels = static_cast<uint32_t*>(d_malloc(sizeof(uint32_t) * length()));
+    // allocate render buffer (always entire segment), prefer IRAM/PSRAM. Note: impact on FPS with PSRAM buffer is low (<2% with QSPI PSRAM) on S2/S3
+    p_free(pixels);
+    pixels = static_cast<uint32_t*>(d_malloc(sizeof(uint32_t) * length()));
     if (!pixels) {
-      DEBUG_PRINTLN(F("!!! Not enough RAM for pixel buffer !!!"));
+      DEBUGFX_PRINTLN(F("!!! Not enough RAM for pixel buffer !!!"));
+      deallocateData();
       errorFlag = ERR_NORAM_PX;
       stop = 0;
       return;
     }
+
   }
   refreshLightCapabilities();
 }
@@ -563,10 +581,10 @@ Segment &Segment::setName(const char *newName) {
   if (newName) {
     const int newLen = min(strlen(newName), (size_t)WLED_MAX_SEGNAME_LEN);
     if (newLen) {
-      if (name) name = static_cast<char*>(d_realloc(name, newLen+1));
-      else      name = static_cast<char*>(d_malloc(newLen+1));
+      if (name) d_free(name); // free old name
+      name = static_cast<char*>(d_malloc(newLen+1));
+      if (mode == FX_MODE_2DSCROLLTEXT) startTransition(strip.getTransition(), true); // if the name changes in scrolling text mode, we need to copy the segment for blending
       if (name) strlcpy(name, newName, newLen+1);
-      name[newLen] = 0;
       return *this;
     }
   }
@@ -645,6 +663,14 @@ uint16_t Segment::virtualLength() const {
   return vLength;
 }
 
+#ifndef WLED_DISABLE_2D
+// maximum length of a mapped 1D segment, used in PS for buffer allocation
+uint16_t Segment::maxMappingLength() const {
+  uint32_t vW = virtualWidth();
+  uint32_t vH = virtualHeight();
+  return max(sqrt32_bw(vH*vH + vW*vW), (uint32_t)getPinwheelLength(vW, vH)); // use diagonal
+}
+#endif
 // pixel is clipped if it falls outside clipping range
 // if clipping start > stop the clipping range is inverted
 bool IRAM_ATTR_YN Segment::isPixelClipped(int i) const {
@@ -729,8 +755,8 @@ void IRAM_ATTR_YN Segment::setPixelColor(int i, uint32_t col) const
         }
         break;
       case M12_pCorner:
-        for (int x = 0; x <= i; x++) setPixelColorRaw(XY(x, i), col);
-        for (int y = 0; y <  i; y++) setPixelColorRaw(XY(i, y), col);
+        for (int x = 0; x <= i; x++) setPixelColorXY(x, i, col); // note: <= to include i=0. Relies on overflow check in sPC()
+        for (int y = 0; y <  i; y++) setPixelColorXY(i, y, col);
         break;
       case M12_sPinwheel: {
         // Uses Bresenham's algorithm to place coordinates of two lines in arrays then draws between them
@@ -1061,27 +1087,14 @@ void Segment::blur(uint8_t blur_amount, bool smear) const {
 /*
  * Put a value 0 to 255 in to get a color value.
  * The colours are a transition r -> g -> b -> back to r
- * Inspired by the Adafruit examples.
+ * Rotates the color in HSV space, where pos is H. (0=0deg, 256=360deg)
  */
 uint32_t Segment::color_wheel(uint8_t pos) const {
-  if (palette) return color_from_palette(pos, false, false, 0); // never wrap palette
+  if (palette) return color_from_palette(pos, false, false, 0); // only wrap if "always wrap" is set
   uint8_t w = W(getCurrentColor(0));
-  pos = 255 - pos;
-  if (useRainbowWheel) {
-    CRGB rgb;
-    hsv2rgb_rainbow(CHSV(pos, 255, 255), rgb);
-    return RGBW32(rgb.r, rgb.g, rgb.b, w);
-  } else {
-    if (pos < 85) {
-      return RGBW32((255 - pos * 3), 0, (pos * 3), w);
-    } else if (pos < 170) {
-      pos -= 85;
-      return RGBW32(0, (pos * 3), (255 - pos * 3), w);
-    } else {
-      pos -= 170;
-      return RGBW32((pos * 3), (255 - pos * 3), 0, w);
-    }
-  }
+  uint32_t rgb;
+  hsv2rgb(CHSV32(static_cast<uint16_t>(pos << 8), 255, 255), rgb);
+  return rgb | (w << 24); // add white channel
 }
 
 /*
@@ -1139,21 +1152,27 @@ void WS2812FX::finalizeInit() {
   #if defined(ARDUINO_ARCH_ESP32) && !defined(CONFIG_IDF_TARGET_ESP32C3)
   // determine if it is sensible to use parallel I2S outputs on ESP32 (i.e. more than 5 outputs = 1 I2S + 4 RMT)
   unsigned maxLedsOnBus = 0;
+  unsigned busType = 0;
   for (const auto &bus : busConfigs) {
     if (Bus::isDigital(bus.type) && !Bus::is2Pin(bus.type)) {
       digitalCount++;
+      if (busType == 0) busType = bus.type; // remember first bus type
+      if (busType != bus.type) {
+        DEBUG_PRINTF_P(PSTR("Mixed digital bus types detected! Forcing single I2S output.\n"));
+        useParallelI2S = false; // mixed bus types, no parallel I2S
+      }
       if (bus.count > maxLedsOnBus) maxLedsOnBus = bus.count;
     }
   }
   DEBUG_PRINTF_P(PSTR("Maximum LEDs on a bus: %u\nDigital buses: %u\n"), maxLedsOnBus, digitalCount);
-  // we may remove 300 LEDs per bus limit when NeoPixelBus is updated beyond 2.9.0
-  if (maxLedsOnBus <= 300 && useParallelI2S) BusManager::useParallelOutput(); // must call before creating buses
+  // we may remove 600 LEDs per bus limit when NeoPixelBus is updated beyond 2.8.3
+  if (maxLedsOnBus <= 600 && useParallelI2S) BusManager::useParallelOutput(); // must call before creating buses
   else useParallelI2S = false; // enforce single I2S
+  digitalCount = 0;
   #endif
 
   // create buses/outputs
   unsigned mem = 0;
-  digitalCount = 0;
   for (const auto &bus : busConfigs) {
     mem += bus.memUsage(Bus::isDigital(bus.type) && !Bus::is2Pin(bus.type) ? digitalCount++ : 0); // includes global buffer
     if (mem <= MAX_LED_MEMORY) {
@@ -1175,8 +1194,9 @@ void WS2812FX::finalizeInit() {
     if (busEnd > _length) _length = busEnd;
     // This must be done after all buses have been created, as some kinds (parallel I2S) interact
     bus->begin();
-    bus->setBrightness(bri);
+    bus->setBrightness(scaledBri(bri));
   }
+  BusManager::initializeABL(); // init brightness limiter
   DEBUG_PRINTF_P(PSTR("Heap after buses: %d\n"), ESP.getFreeHeap());
 
   Segment::maxWidth  = _length;
@@ -1189,10 +1209,9 @@ void WS2812FX::finalizeInit() {
   deserializeMap();     // (re)load default ledmap (will also setUpMatrix() if ledmap does not exist)
 
   // allocate frame buffer after matrix has been set up (gaps!)
-  if (_pixels) _pixels = static_cast<uint32_t*>(d_realloc(_pixels, getLengthTotal() * sizeof(uint32_t)));
-  else         _pixels = static_cast<uint32_t*>(d_malloc(getLengthTotal() * sizeof(uint32_t)));
+  d_free(_pixels); // using realloc on large buffers can cause additional fragmentation instead of reducing it
+  _pixels = static_cast<uint32_t*>(d_malloc(getLengthTotal() * sizeof(uint32_t)));
   DEBUG_PRINTF_P(PSTR("strip buffer size: %uB\n"), getLengthTotal() * sizeof(uint32_t));
-
   DEBUG_PRINTF_P(PSTR("Heap after strip init: %uB\n"), ESP.getFreeHeap());
 }
 
@@ -1237,7 +1256,8 @@ void WS2812FX::service() {
         // if segment is in transition and no old segment exists we don't need to run the old mode
         // (blendSegments() takes care of On/Off transitions and clipping)
         Segment *segO = seg.getOldSegment();
-        if (segO && (seg.mode != segO->mode || blendingStyle != BLEND_STYLE_FADE)) {
+        if (segO && segO->isActive() && (seg.mode != segO->mode || blendingStyle != BLEND_STYLE_FADE ||
+            (segO->name != seg.name && segO->name && seg.name && strncmp(segO->name, seg.name, WLED_MAX_SEGNAME_LEN) != 0))) {
           Segment::modeBlend(true);         // set semaphore for beginDraw() to blend colors and palette
           segO->beginDraw(prog);            // set up palette & colors (also sets draw dimensions), parent segment has transition progress
           _currentSegment = segO;           // set current segment
@@ -1278,7 +1298,7 @@ static uint8_t _add       (uint8_t a, uint8_t b) { unsigned t = a + b; return t
 static uint8_t _subtract  (uint8_t a, uint8_t b) { return b > a ? (b - a) : 0; }
 static uint8_t _difference(uint8_t a, uint8_t b) { return b > a ? (b - a) : (a - b); }
 static uint8_t _average   (uint8_t a, uint8_t b) { return (a + b) >> 1; }
-#ifdef CONFIG_IDF_TARGET_ESP32C3
+#if defined(ESP8266) || defined(CONFIG_IDF_TARGET_ESP32C3)
 static uint8_t _multiply  (uint8_t a, uint8_t b) { return ((a * b) + 255) >> 8; } // faster than division on C3 but slightly less accurate
 #else
 static uint8_t _multiply  (uint8_t a, uint8_t b) { return (a * b) / 255; } // origianl uses a & b in range [0,1]
@@ -1289,10 +1309,10 @@ static uint8_t _darken    (uint8_t a, uint8_t b) { return a < b ? a : b; }
 static uint8_t _screen    (uint8_t a, uint8_t b) { return 255 - _multiply(~a,~b); } // 255 - (255-a)*(255-b)/255
 static uint8_t _overlay   (uint8_t a, uint8_t b) { return b < 128 ? 2 * _multiply(a,b) : (255 - 2 * _multiply(~a,~b)); }
 static uint8_t _hardlight (uint8_t a, uint8_t b) { return a < 128 ? 2 * _multiply(a,b) : (255 - 2 * _multiply(~a,~b)); }
-#ifdef CONFIG_IDF_TARGET_ESP32C3
-static uint8_t _softlight (uint8_t a, uint8_t b) { return (((b * b * (255 - 2 * a) + 255) >> 8) + 2 * a * b + 255) >> 8; } // Pegtop's formula (1 - 2a)b^2 + 2ab
+#if defined(ESP8266) || defined(CONFIG_IDF_TARGET_ESP32C3)
+static uint8_t _softlight (uint8_t a, uint8_t b) { return (((b * b * (255 - 2 * a))) + ((2 * a * b + 256) << 8)) >> 16; } // Pegtop's formula (1 - 2a)b^2
 #else
-static uint8_t _softlight (uint8_t a, uint8_t b) { return (b * b * (255 - 2 * a) / 255 + 2 * a * b) / 255; } // Pegtop's formula (1 - 2a)b^2 + 2ab
+static uint8_t _softlight (uint8_t a, uint8_t b) { return (b * b * (255 - 2 * a) + 255 * 2 * a * b) / (255 * 255); } // Pegtop's formula (1 - 2a)b^2 + 2ab
 #endif
 static uint8_t _dodge     (uint8_t a, uint8_t b) { return _divide(~a,b); }
 static uint8_t _burn      (uint8_t a, uint8_t b) { return ~_divide(a,~b); }
@@ -1435,8 +1455,10 @@ void WS2812FX::blendSegment(const Segment &topSegment) const {
       }
       uint32_t c_a = BLACK;
       if (x < vCols && y < vRows) c_a = seg->getPixelColorRaw(x + y*vCols); // will get clipped pixel from old segment or unclipped pixel from new segment
-      if (segO && blendingStyle == BLEND_STYLE_FADE && topSegment.mode != segO->mode && x < oCols && y < oRows) {
-        // we need to blend old segment using fade as pixels ae not clipped
+      if (segO && blendingStyle == BLEND_STYLE_FADE
+        && (topSegment.mode != segO->mode || (segO->name != topSegment.name && segO->name && topSegment.name && strncmp(segO->name, topSegment.name, WLED_MAX_SEGNAME_LEN) != 0))
+        && x < oCols && y < oRows) {
+        // we need to blend old segment using fade as pixels are not clipped
         c_a = color_blend16(c_a, segO->getPixelColorRaw(x + y*oCols), progInv);
       } else if (blendingStyle != BLEND_STYLE_FADE) {
         // workaround for On/Off transition
@@ -1529,67 +1551,9 @@ void WS2812FX::blendSegment(const Segment &topSegment) const {
   Segment::setClippingRect(0, 0);             // disable clipping for overlays
 }
 
-// To disable brightness limiter we either set output max current to 0 or single LED current to 0
-static uint8_t estimateCurrentAndLimitBri(uint8_t brightness, uint32_t *pixels) {
-  unsigned milliAmpsMax = BusManager::ablMilliampsMax();
-  if (milliAmpsMax > 0) {
-    unsigned milliAmpsTotal = 0;
-    unsigned avgMilliAmpsPerLED = 0;
-    unsigned lengthDigital = 0;
-    bool useWackyWS2815PowerModel = false;
-
-    for (size_t i = 0; i < BusManager::getNumBusses(); i++) {
-      const Bus *bus = BusManager::getBus(i);
-      if (!(bus && bus->isDigital() && bus->isOk())) continue;
-      unsigned maPL = bus->getLEDCurrent();
-      if (maPL == 0 || bus->getMaxCurrent() > 0) continue; // skip buses with 0 mA per LED or max current per bus defined (PP-ABL)
-      if (maPL == 255) {
-        useWackyWS2815PowerModel = true;
-        maPL = 12; // WS2815 uses 12mA per channel
-      }
-      avgMilliAmpsPerLED += maPL * bus->getLength();
-      lengthDigital += bus->getLength();
-      // sum up the usage of each LED on digital bus
-      uint32_t busPowerSum = 0;
-      for (unsigned j = 0; j < bus->getLength(); j++) {
-        uint32_t c = pixels[j + bus->getStart()];
-        byte r = R(c), g = G(c), b = B(c), w = W(c);
-        if (useWackyWS2815PowerModel) { //ignore white component on WS2815 power calculation
-          busPowerSum += (max(max(r,g),b)) * 3;
-        } else {
-          busPowerSum += (r + g + b + w);
-        }
-      }
-      // RGBW led total output with white LEDs enabled is still 50mA, so each channel uses less
-      if (bus->hasWhite()) {
-        busPowerSum *= 3;
-        busPowerSum >>= 2; //same as /= 4
-      }
-      // powerSum has all the values of channels summed (max would be getLength()*765 as white is excluded) so convert to milliAmps
-      milliAmpsTotal += (busPowerSum * maPL * brightness) / (765*255);
-    }
-    if (lengthDigital > 0) {
-      avgMilliAmpsPerLED /= lengthDigital;
-
-      if (milliAmpsMax > MA_FOR_ESP && avgMilliAmpsPerLED > 0) { //0 mA per LED and too low numbers turn off calculation
-        unsigned powerBudget = (milliAmpsMax - MA_FOR_ESP); //80/120mA for ESP power
-        if (powerBudget > lengthDigital) { //each LED uses about 1mA in standby, exclude that from power budget
-          powerBudget -= lengthDigital;
-        } else {
-          powerBudget = 0;
-        }
-        if (milliAmpsTotal > powerBudget) {
-          //scale brightness down to stay in current limit
-          unsigned scaleB = powerBudget * 255 / milliAmpsTotal;
-          brightness = ((brightness * scaleB) >> 8) + 1;
-        }
-      }
-    }
-  }
-  return brightness;
-}
-
 void WS2812FX::show() {
+  if (!_pixels) return; // no pixels allocated, nothing to show
+
   unsigned long showNow = millis();
   size_t diff = showNow - _lastShow;
 
@@ -1614,10 +1578,6 @@ void WS2812FX::show() {
   show_callback callback = _callback;
   if (callback) callback(); // will call setPixelColor or setRealtimePixelColor
 
-  // determine ABL brightness
-  uint8_t newBri = estimateCurrentAndLimitBri(_brightness, _pixels);
-  if (newBri != _brightness) BusManager::setBrightness(newBri);
-
   // paint actual pixels
   int oldCCT = Bus::getCCT(); // store original CCT value (since it is global)
   // when cctFromRgb is true we implicitly calculate WW and CW from RGB values (cct==-1)
@@ -1628,7 +1588,11 @@ void WS2812FX::show() {
     if (_pixelCCT) { // cctFromRgb already exluded at allocation
       if (i == 0 || _pixelCCT[i-1] != _pixelCCT[i]) BusManager::setSegmentCCT(_pixelCCT[i], correctWB);
     }
-    BusManager::setPixelColor(getMappedPixelIndex(i), realtimeMode && arlsDisableGammaCorrection ? _pixels[i] : gamma32(_pixels[i]));
+
+    uint32_t c = _pixels[i]; // need a copy, do not modify _pixels directly (no byte access allowed on ESP32)
+    if(c > 0 && !(realtimeMode && arlsDisableGammaCorrection))
+        c = gamma32(c); // apply gamma correction if enabled note: applying gamma after brightness has too much color loss
+    BusManager::setPixelColor(getMappedPixelIndex(i), c);
   }
   Bus::setCCT(oldCCT);  // restore old CCT for ABL adjustments
 
@@ -1640,9 +1604,6 @@ void WS2812FX::show() {
   // See https://github.com/Makuna/NeoPixelBus/wiki/ESP32-NeoMethods#neoesp32rmt-methods
   BusManager::show();
 
-  // restore brightness for next frame
-  if (newBri != _brightness) BusManager::setBrightness(_brightness);
-
   if (diff > 0) { // skip calculation if no time has passed
     size_t fpsCurr = (1000 << FPS_CALC_SHIFT) / diff; // fixed point math
     _cumulativeFps = (FPS_CALC_AVG * _cumulativeFps + fpsCurr + FPS_CALC_AVG / 2) / (FPS_CALC_AVG + 1);   // "+FPS_CALC_AVG/2" for proper rounding
@@ -1677,7 +1638,7 @@ void WS2812FX::setTransitionMode(bool t) {
 
 // wait until frame is over (service() has finished or time for 1 frame has passed; yield() crashes on 8266)
 void WS2812FX::waitForIt() {
-  unsigned long maxWait = millis() + getFrameTime();
+  unsigned long maxWait = millis() + getFrameTime() + 100; // TODO: this needs a proper fix for timeout!
   while (isServicing() && maxWait > millis()) delay(1);
   #ifdef WLED_DEBUG
   if (millis() >= maxWait) DEBUG_PRINTLN(F("Waited for strip to finish servicing."));
@@ -1707,7 +1668,7 @@ void WS2812FX::setBrightness(uint8_t b, bool direct) {
   if (_brightness == 0) { //unfreeze all segments on power off
     for (const Segment &seg : _segments) seg.freeze = false; // freeze is mutable
   }
-  BusManager::setBrightness(b);
+  BusManager::setBrightness(scaledBri(b));
   if (!direct) {
     unsigned long t = millis();
     if (_segments[0].next_time > t + 22 && t - _lastShow > MIN_SHOW_DELAY) trigger(); //apply brightness change immediately if no refresh soon
@@ -1862,7 +1823,7 @@ void WS2812FX::makeAutoSegments(bool forceReset) {
     for (size_t i = 1; i < s; i++) {
       _segments.emplace_back(segStarts[i], segStops[i]);
     }
-    DEBUG_PRINTF_P(PSTR("%d auto segments created.\n"), _segments.size());
+    DEBUGFX_PRINTF_P(PSTR("%d auto segments created.\n"), _segments.size());
 
   } else {
 
@@ -1986,7 +1947,7 @@ bool WS2812FX::deserializeMap(unsigned n) {
   }
 
   d_free(customMappingTable);
-  customMappingTable = static_cast<uint16_t*>(d_malloc(sizeof(uint16_t)*getLengthTotal())); // do not use SPI RAM
+  customMappingTable = static_cast<uint16_t*>(d_malloc(sizeof(uint16_t)*getLengthTotal())); // prefer DRAM for speed
 
   if (customMappingTable) {
     DEBUG_PRINTF_P(PSTR("ledmap allocated: %uB\n"), sizeof(uint16_t)*getLengthTotal());

wled00/FXparticleSystem.cpp

@@ -17,9 +17,8 @@
 // local shared functions (used both in 1D and 2D system)
 static int32_t calcForce_dv(const int8_t force, uint8_t &counter);
 static bool checkBoundsAndWrap(int32_t &position, const int32_t max, const int32_t particleradius, const bool wrap); // returns false if out of bounds by more than particleradius
-static void fast_color_add(CRGB &c1, const CRGB &c2, uint8_t scale = 255); // fast and accurate color adding with scaling (scales c2 before adding)
-static void fast_color_scale(CRGB &c, const uint8_t scale); // fast scaling function using 32bit variable and pointer. note: keep 'scale' within 0-255
-//static CRGB *allocateCRGBbuffer(uint32_t length);
+static uint32_t fast_color_add(CRGBW c1, const CRGBW c2, uint8_t scale = 255); // fast and accurate color adding with scaling (scales c2 before adding)
+static uint32_t fast_color_scale(CRGBW c, const uint8_t scale); // fast scaling function using 32bit variable and pointer. note: keep 'scale' within 0-255
 #endif
 
 #ifndef WLED_DISABLE_PARTICLESYSTEM2D
@@ -558,7 +557,11 @@ void ParticleSystem2D::pointAttractor(const uint32_t particleindex, PSparticle &
 // warning: do not render out of bounds particles or system will crash! rendering does not check if particle is out of bounds
 // firemode is only used for PS Fire FX
 void ParticleSystem2D::render() {
-  CRGB baseRGB;
+  if(framebuffer == nullptr) {
+    PSPRINTLN(F("PS render: no framebuffer!"));
+    return;
+  }
+  CRGBW baseRGB;
   uint32_t brightness; // particle brightness, fades if dying
   TBlendType blend = LINEARBLEND; // default color rendering: wrap palette
   if (particlesettings.colorByAge) {
@@ -569,13 +572,13 @@ void ParticleSystem2D::render() {
     for (int32_t y = 0; y <= maxYpixel; y++) {
       int index = y * (maxXpixel + 1);
       for (int32_t x = 0; x <= maxXpixel; x++) {
-        fast_color_scale(framebuffer[index], motionBlur); // note: could skip if only globalsmear is active but usually they are both active and scaling is fast enough
+        framebuffer[index] = fast_color_scale(framebuffer[index], motionBlur); // note: could skip if only globalsmear is active but usually they are both active and scaling is fast enough
         index++;
       }
     }
   }
   else { // no blurring: clear buffer
-    memset(framebuffer, 0, (maxXpixel+1) * (maxYpixel+1) * sizeof(CRGB));
+    memset(framebuffer, 0, (maxXpixel+1) * (maxYpixel+1) * sizeof(CRGBW));
   }
 
   // go over particles and render them to the buffer
@@ -593,14 +596,12 @@ void ParticleSystem2D::render() {
       baseRGB = ColorFromPaletteWLED(SEGPALETTE, particles[i].hue, 255, blend);
       if (particles[i].sat < 255) {
         CHSV32 baseHSV;
-        rgb2hsv((uint32_t((byte(baseRGB.r) << 16) | (byte(baseRGB.g) << 8) | (byte(baseRGB.b)))), baseHSV); // convert to HSV
+        rgb2hsv(baseRGB.color32, baseHSV); // convert to HSV
         baseHSV.s = min(baseHSV.s, particles[i].sat); // set the saturation but don't increase it
-        uint32_t tempcolor;
-        hsv2rgb(baseHSV, tempcolor); // convert back to RGB
-        baseRGB = (CRGB)tempcolor;
+        hsv2rgb(baseHSV, baseRGB.color32); // convert back to RGB
       }
     }
-    brightness = gamma8(brightness); // apply gamma correction, used for gamma-inverted brightness distribution
+    if(gammaCorrectCol) brightness = gamma8(brightness); // apply gamma correction, used for gamma-inverted brightness distribution
     renderParticle(i, brightness, baseRGB, particlesettings.wrapX, particlesettings.wrapY);
   }
 
@@ -621,18 +622,10 @@ void ParticleSystem2D::render() {
   if (smearBlur) {
     blur2D(framebuffer, maxXpixel + 1, maxYpixel + 1, smearBlur, smearBlur);
   }
-
-  // transfer the framebuffer to the segment
-  for (int y = 0; y <= maxYpixel; y++) {
-    int index = y * (maxXpixel + 1); // current row index for 1D buffer
-    for (int x = 0; x <= maxXpixel; x++) {
-      SEGMENT.setPixelColorXY(x, y, framebuffer[index++]);
-    }
-  }
 }
 
 // calculate pixel positions and brightness distribution and render the particle to local buffer or global buffer
-__attribute__((optimize("O2"))) void ParticleSystem2D::renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGB& color, const bool wrapX, const bool wrapY) {
+__attribute__((optimize("O2"))) void ParticleSystem2D::renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGBW& color, const bool wrapX, const bool wrapY) {
   uint32_t size = particlesize;
   if (advPartProps && advPartProps[particleindex].size > 0) // use advanced size properties (0 means use global size including single pixel rendering)
     size = advPartProps[particleindex].size;
@@ -641,7 +634,8 @@ __attribute__((optimize("O2"))) void ParticleSystem2D::renderParticle(const uint
     uint32_t x = particles[particleindex].x >> PS_P_RADIUS_SHIFT;
     uint32_t y = particles[particleindex].y >> PS_P_RADIUS_SHIFT;
     if (x <= (uint32_t)maxXpixel && y <= (uint32_t)maxYpixel) {
-      fast_color_add(framebuffer[x + (maxYpixel - y) * (maxXpixel + 1)], color, brightness);
+      uint32_t index = x + (maxYpixel - y) * (maxXpixel + 1); // flip y coordinate (0,0 is bottom left in PS but top left in framebuffer)
+      framebuffer[index] = fast_color_add(framebuffer[index], color, brightness);
     }
     return;
   }
@@ -681,20 +675,22 @@ __attribute__((optimize("O2"))) void ParticleSystem2D::renderParticle(const uint
   // - scale brigthness with gamma correction (done in render())
   // - apply inverse gamma correction to brightness values
   // - gamma is applied again in show() -> the resulting brightness distribution is linear but gamma corrected in total
-  pxlbrightness[0] = gamma8inv(pxlbrightness[0]); // use look-up-table for invers gamma
-  pxlbrightness[1] = gamma8inv(pxlbrightness[1]);
-  pxlbrightness[2] = gamma8inv(pxlbrightness[2]);
-  pxlbrightness[3] = gamma8inv(pxlbrightness[3]);
+  if(gammaCorrectCol) {
+    pxlbrightness[0] = gamma8inv(pxlbrightness[0]); // use look-up-table for invers gamma
+    pxlbrightness[1] = gamma8inv(pxlbrightness[1]);
+    pxlbrightness[2] = gamma8inv(pxlbrightness[2]);
+    pxlbrightness[3] = gamma8inv(pxlbrightness[3]);
+  }
 
   if (advPartProps && advPartProps[particleindex].size > 1) { //render particle to a bigger size
-    CRGB renderbuffer[100]; // 10x10 pixel buffer
+    uint32_t renderbuffer[100]; // 10x10 pixel buffer
     memset(renderbuffer, 0, sizeof(renderbuffer)); // clear buffer
     //particle size to pixels: < 64 is 4x4, < 128 is 6x6, < 192 is 8x8, bigger is 10x10
     //first, render the pixel to the center of the renderbuffer, then apply 2D blurring
-    fast_color_add(renderbuffer[4 + (4 * 10)], color, pxlbrightness[0]); // oCrder is: bottom left, bottom right, top right, top left
-    fast_color_add(renderbuffer[5 + (4 * 10)], color, pxlbrightness[1]);
-    fast_color_add(renderbuffer[5 + (5 * 10)], color, pxlbrightness[2]);
-    fast_color_add(renderbuffer[4 + (5 * 10)], color, pxlbrightness[3]);
+    renderbuffer[4 + (4 * 10)] = fast_color_add(renderbuffer[4 + (4 * 10)], color, pxlbrightness[0]); // order is: bottom left, bottom right, top right, top left
+    renderbuffer[5 + (4 * 10)] = fast_color_add(renderbuffer[5 + (4 * 10)], color, pxlbrightness[1]);
+    renderbuffer[5 + (5 * 10)] = fast_color_add(renderbuffer[5 + (5 * 10)], color, pxlbrightness[2]);
+    renderbuffer[4 + (5 * 10)] = fast_color_add(renderbuffer[4 + (5 * 10)], color, pxlbrightness[3]);
     uint32_t rendersize = 2; // initialize render size, minimum is 4x4 pixels, it is incremented int he loop below to start with 4
     uint32_t offset = 4; // offset to zero coordinate to write/read data in renderbuffer (actually needs to be 3, is decremented in the loop below)
     uint32_t maxsize = advPartProps[particleindex].size;
@@ -751,7 +747,8 @@ __attribute__((optimize("O2"))) void ParticleSystem2D::renderParticle(const uint
           else
           continue;
         }
-        fast_color_add(framebuffer[xfb + (maxYpixel - yfb) * (maxXpixel + 1)], renderbuffer[xrb + yrb * 10]);
+        uint32_t idx = xfb + (maxYpixel - yfb) * (maxXpixel + 1); // flip y coordinate (0,0 is bottom left in PS but top left in framebuffer)
+        framebuffer[idx] = fast_color_add(framebuffer[idx], renderbuffer[xrb + yrb * 10]);
       }
     }
     } else { // standard rendering (2x2 pixels)
@@ -786,8 +783,10 @@ __attribute__((optimize("O2"))) void ParticleSystem2D::renderParticle(const uint
       }
     }
     for (uint32_t i = 0; i < 4; i++) {
-      if (pixelvalid[i])
-        fast_color_add(framebuffer[pixco[i].x + (maxYpixel - pixco[i].y) * (maxXpixel + 1)], color, pxlbrightness[i]); // order is: bottom left, bottom right, top right, top left
+      if (pixelvalid[i]) {
+        uint32_t idx = pixco[i].x + (maxYpixel - pixco[i].y) * (maxXpixel + 1); // flip y coordinate (0,0 is bottom left in PS but top left in framebuffer)
+        framebuffer[idx] = fast_color_add(framebuffer[idx], color, pxlbrightness[i]); // order is: bottom left, bottom right, top right, top left
+      }
     }
   }
 }
@@ -970,17 +969,17 @@ __attribute__((optimize("O2"))) void ParticleSystem2D::collideParticles(PSpartic
 // update size and pointers (memory location and size can change dynamically)
 // note: do not access the PS class in FX befor running this function (or it messes up SEGENV.data)
 void ParticleSystem2D::updateSystem(void) {
-  PSPRINTLN("updateSystem2D");
+  //PSPRINTLN("updateSystem2D");
   setMatrixSize(SEGMENT.vWidth(), SEGMENT.vHeight());
   updatePSpointers(advPartProps != nullptr, advPartSize != nullptr); // update pointers to PS data, also updates availableParticles
-  PSPRINTLN("\n END update System2D, running FX...");
+  //PSPRINTLN("\n END update System2D, running FX...");
 }
 
 // set the pointers for the class (this only has to be done once and not on every FX call, only the class pointer needs to be reassigned to SEGENV.data every time)
 // function returns the pointer to the next byte available for the FX (if it assigned more memory for other stuff using the above allocate function)
 // FX handles the PSsources, need to tell this function how many there are
 void ParticleSystem2D::updatePSpointers(bool isadvanced, bool sizecontrol) {
-  PSPRINTLN("updatePSpointers");
+  //PSPRINTLN("updatePSpointers");
   // Note on memory alignment:
   // a pointer MUST be 4 byte aligned. sizeof() in a struct/class is always aligned to the largest element. if it contains a 32bit, it will be padded to 4 bytes, 16bit is padded to 2byte alignment.
   // The PS is aligned to 4 bytes, a PSparticle is aligned to 2 and a struct containing only byte sized variables is not aligned at all and may need to be padded when dividing the memoryblock.
@@ -988,11 +987,8 @@ void ParticleSystem2D::updatePSpointers(bool isadvanced, bool sizecontrol) {
   particles = reinterpret_cast<PSparticle *>(this + 1); // pointer to particles
   particleFlags = reinterpret_cast<PSparticleFlags *>(particles + numParticles); // pointer to particle flags
   sources = reinterpret_cast<PSsource *>(particleFlags + numParticles); // pointer to source(s) at data+sizeof(ParticleSystem2D)
-  framebuffer = reinterpret_cast<CRGB *>(sources + numSources); // pointer to framebuffer
-  // align pointer after framebuffer
-  uintptr_t p = reinterpret_cast<uintptr_t>(framebuffer + (maxXpixel+1)*(maxYpixel+1));
-  p = (p + 3) & ~0x03; // align to 4-byte boundary
-  PSdataEnd = reinterpret_cast<uint8_t *>(p); // pointer to first available byte after the PS for FX additional data
+  framebuffer = SEGMENT.getPixels(); // pointer to framebuffer
+  PSdataEnd = reinterpret_cast<uint8_t *>(sources + numSources); // pointer to first available byte after the PS for FX additional data (already aligned to 4 byte boundary)
   if (isadvanced) {
     advPartProps = reinterpret_cast<PSadvancedParticle *>(PSdataEnd);
     PSdataEnd = reinterpret_cast<uint8_t *>(advPartProps + numParticles);
@@ -1015,8 +1011,8 @@ void ParticleSystem2D::updatePSpointers(bool isadvanced, bool sizecontrol) {
 // for speed, 1D array and 32bit variables are used, make sure to limit them to 8bit (0-255) or result is undefined
 // to blur a subset of the buffer, change the xsize/ysize and set xstart/ystart to the desired starting coordinates (default start is 0/0)
 // subset blurring only works on 10x10 buffer (single particle rendering), if other sizes are needed, buffer width must be passed as parameter
-void blur2D(CRGB *colorbuffer, uint32_t xsize, uint32_t ysize, uint32_t xblur, uint32_t yblur, uint32_t xstart, uint32_t ystart, bool isparticle) {
-  CRGB seeppart, carryover;
+void blur2D(uint32_t *colorbuffer, uint32_t xsize, uint32_t ysize, uint32_t xblur, uint32_t yblur, uint32_t xstart, uint32_t ystart, bool isparticle) {
+  CRGBW seeppart, carryover;
   uint32_t seep = xblur >> 1;
   uint32_t width = xsize; // width of the buffer, used to calculate the index of the pixel
 
@@ -1030,12 +1026,11 @@ void blur2D(CRGB *colorbuffer, uint32_t xsize, uint32_t ysize, uint32_t xblur, u
     carryover =  BLACK;
     uint32_t indexXY = xstart + y * width;
     for (uint32_t x = xstart; x < xstart + xsize; x++) {
-      seeppart = colorbuffer[indexXY]; // create copy of current color
-      fast_color_scale(seeppart, seep); // scale it and seep to neighbours
+      seeppart = fast_color_scale(colorbuffer[indexXY], seep); // scale it and seep to neighbours
       if (x > 0) {
-        fast_color_add(colorbuffer[indexXY - 1], seeppart);
-        if (carryover) // note: check adds overhead but is faster on average
-          fast_color_add(colorbuffer[indexXY], carryover);
+        colorbuffer[indexXY - 1] = fast_color_add(colorbuffer[indexXY - 1], seeppart);
+        if (carryover.color32) // note: check adds overhead but is faster on average
+          colorbuffer[indexXY] = fast_color_add(colorbuffer[indexXY], carryover);
       }
       carryover = seeppart;
       indexXY++; // next pixel in x direction
@@ -1052,12 +1047,11 @@ void blur2D(CRGB *colorbuffer, uint32_t xsize, uint32_t ysize, uint32_t xblur, u
     carryover = BLACK;
     uint32_t indexXY = x + ystart * width;
     for (uint32_t y = ystart; y < ystart + ysize; y++) {
-      seeppart = colorbuffer[indexXY]; // create copy of current color
-      fast_color_scale(seeppart, seep); // scale it and seep to neighbours
+      seeppart = fast_color_scale(colorbuffer[indexXY], seep); // scale it and seep to neighbours
       if (y > 0) {
-        fast_color_add(colorbuffer[indexXY - width], seeppart);
-        if (carryover) // note: check adds overhead but is faster on average
-          fast_color_add(colorbuffer[indexXY], carryover);
+        colorbuffer[indexXY - width] = fast_color_add(colorbuffer[indexXY - width], seeppart);
+        if (carryover.color32) // note: check adds overhead but is faster on average
+          colorbuffer[indexXY] = fast_color_add(colorbuffer[indexXY], carryover);
       }
       carryover = seeppart;
       indexXY += width; // next pixel in y direction
@@ -1068,13 +1062,7 @@ void blur2D(CRGB *colorbuffer, uint32_t xsize, uint32_t ysize, uint32_t xblur, u
 //non class functions to use for initialization
 uint32_t calculateNumberOfParticles2D(uint32_t const pixels, const bool isadvanced, const bool sizecontrol) {
   uint32_t numberofParticles = pixels;  // 1 particle per pixel (for example 512 particles on 32x16)
-#ifdef ESP8266
-  uint32_t particlelimit = ESP8266_MAXPARTICLES; // maximum number of paticles allowed (based on one segment of 16x16 and 4k effect ram)
-#elif ARDUINO_ARCH_ESP32S2
-  uint32_t particlelimit = ESP32S2_MAXPARTICLES; // maximum number of paticles allowed (based on one segment of 32x32 and 24k effect ram)
-#else
-  uint32_t particlelimit = ESP32_MAXPARTICLES; // maximum number of paticles allowed (based on two segments of 32x32 and 40k effect ram)
-#endif
+  uint32_t particlelimit = MAXPARTICLES_2D; // maximum number of paticles allowed
   numberofParticles = max((uint32_t)4, min(numberofParticles, particlelimit)); // limit to 4 - particlelimit
   if (isadvanced) // advanced property array needs ram, reduce number of particles to use the same amount
     numberofParticles = (numberofParticles * sizeof(PSparticle)) / (sizeof(PSparticle) + sizeof(PSadvancedParticle));
@@ -1087,16 +1075,8 @@ uint32_t calculateNumberOfParticles2D(uint32_t const pixels, const bool isadvanc
 }
 
 uint32_t calculateNumberOfSources2D(uint32_t pixels, uint32_t requestedsources) {
-#ifdef ESP8266
-  int numberofSources = min((pixels) / 8, (uint32_t)requestedsources);
-  numberofSources = max(1, min(numberofSources, ESP8266_MAXSOURCES)); // limit
-#elif ARDUINO_ARCH_ESP32S2
-  int numberofSources = min((pixels) / 6, (uint32_t)requestedsources);
-  numberofSources = max(1, min(numberofSources, ESP32S2_MAXSOURCES)); // limit
-#else
-  int numberofSources = min((pixels) / 4, (uint32_t)requestedsources);
-  numberofSources = max(1, min(numberofSources, ESP32_MAXSOURCES)); // limit
-#endif
+  int numberofSources = min((pixels) / SOURCEREDUCTIONFACTOR, (uint32_t)requestedsources);
+  numberofSources = max(1, min(numberofSources, MAXSOURCES_2D)); // limit
   // make sure it is a multiple of 4 for proper memory alignment
   numberofSources = (numberofSources+3) & ~0x03;
   return numberofSources;
@@ -1115,10 +1095,7 @@ bool allocateParticleSystemMemory2D(uint32_t numparticles, uint32_t numsources,
   if (sizecontrol)
     requiredmemory += sizeof(PSsizeControl) * numparticles;
   requiredmemory += sizeof(PSsource) * numsources;
-  requiredmemory += sizeof(CRGB) * SEGMENT.virtualLength(); // virtualLength is witdh * height
-  requiredmemory += additionalbytes + 3; // add 3 to ensure there is room for stuffing bytes
-  //requiredmemory = (requiredmemory + 3) & ~0x03; // align memory block to next 4-byte boundary
-  PSPRINTLN("mem alloc: " + String(requiredmemory));
+  requiredmemory += additionalbytes;
   return(SEGMENT.allocateData(requiredmemory));
 }
 
@@ -1132,17 +1109,26 @@ bool initParticleSystem2D(ParticleSystem2D *&PartSys, uint32_t requestedsources,
 
   uint32_t numparticles = calculateNumberOfParticles2D(pixels, advanced, sizecontrol);
   PSPRINT(" segmentsize:" + String(cols) + " x " + String(rows));
-  PSPRINT(" request numparticles:" + String(numparticles));
+  PSPRINTLN(" request numparticles:" + String(numparticles));
   uint32_t numsources = calculateNumberOfSources2D(pixels, requestedsources);
-  if (!allocateParticleSystemMemory2D(numparticles, numsources, advanced, sizecontrol, additionalbytes))
-  {
-    DEBUG_PRINT(F("PS init failed: memory depleted"));
-    return false;
+  bool allocsuccess = false;
+  while(numparticles >= 4) { // make sure we have at least 4 particles or quit
+    if (allocateParticleSystemMemory2D(numparticles, numsources, advanced, sizecontrol, additionalbytes)) {
+      PSPRINTLN(F("PS 2D alloc succeeded"));
+      allocsuccess = true;
+      break; // allocation succeeded
+    }
+    numparticles = ((numparticles / 2) + 3) & ~0x03; // cut number of particles in half and try again, must be 4 byte aligned
+    PSPRINTLN(F("PS 2D alloc failed, trying with less particles..."));
+  }
+  if (!allocsuccess) {
+    PSPRINTLN(F("PS 2D alloc failed, not enough memory!"));
+    return false; // allocation failed
   }
 
   PartSys = new (SEGENV.data) ParticleSystem2D(cols, rows, numparticles, numsources, advanced, sizecontrol); // particle system constructor
 
-  PSPRINTLN("******init done, pointers:");
+  PSPRINTLN(F("2D PS init done"));
   return true;
 }
 
@@ -1435,28 +1421,26 @@ void ParticleSystem1D::applyFriction(int32_t coefficient) {
 // if wrap is set, particles half out of bounds are rendered to the other side of the matrix
 // warning: do not render out of bounds particles or system will crash! rendering does not check if particle is out of bounds
 void ParticleSystem1D::render() {
-  CRGB baseRGB;
+  if(framebuffer == nullptr) {
+    PSPRINTLN(F("PS render: no framebuffer!"));
+    return;
+  }
+  CRGBW baseRGB;
   uint32_t brightness; // particle brightness, fades if dying
   TBlendType blend = LINEARBLEND; // default color rendering: wrap palette
   if (particlesettings.colorByAge || particlesettings.colorByPosition) {
     blend = LINEARBLEND_NOWRAP;
   }
 
-  #ifdef ESP8266 // no local buffer on ESP8266
-  if (motionBlur)
-    SEGMENT.fadeToBlackBy(255 - motionBlur);
-  else
-    SEGMENT.fill(BLACK); // clear the buffer before rendering to it
-  #else
   if (motionBlur) { // blurring active
     for (int32_t x = 0; x <= maxXpixel; x++) {
-      fast_color_scale(framebuffer[x], motionBlur);
+      framebuffer[x] = fast_color_scale(framebuffer[x], motionBlur);
     }
   }
   else { // no blurring: clear buffer
-    memset(framebuffer, 0, (maxXpixel+1) * sizeof(CRGB));
+    memset(framebuffer, 0, (maxXpixel+1) * sizeof(CRGBW));
   }
-  #endif
+
   // go over particles and render them to the buffer
   for (uint32_t i = 0; i < usedParticles; i++) {
     if ( particles[i].ttl == 0 || particleFlags[i].outofbounds)
@@ -1469,48 +1453,39 @@ void ParticleSystem1D::render() {
     if (advPartProps) { //saturation is advanced property in 1D system
       if (advPartProps[i].sat < 255) {
         CHSV32 baseHSV;
-        rgb2hsv((uint32_t((byte(baseRGB.r) << 16) | (byte(baseRGB.g) << 8) | (byte(baseRGB.b)))), baseHSV); // convert to HSV
+        rgb2hsv(baseRGB.color32, baseHSV); // convert to HSV
         baseHSV.s = min(baseHSV.s, advPartProps[i].sat); // set the saturation but don't increase it
-        uint32_t tempcolor;
-        hsv2rgb(baseHSV, tempcolor); // convert back to RGB
-        baseRGB = (CRGB)tempcolor;
+        hsv2rgb(baseHSV, baseRGB.color32); // convert back to RGB
       }
     }
-    brightness = gamma8(brightness); // apply gamma correction, used for gamma-inverted brightness distribution
+    if(gammaCorrectCol) brightness = gamma8(brightness); // apply gamma correction, used for gamma-inverted brightness distribution
     renderParticle(i, brightness, baseRGB, particlesettings.wrap);
   }
   // apply smear-blur to rendered frame
   if (smearBlur) {
-    #ifdef ESP8266
-    SEGMENT.blur(smearBlur, true); // no local buffer on ESP8266
-    #else
     blur1D(framebuffer, maxXpixel + 1, smearBlur, 0);
-    #endif
   }
 
   // add background color
-  uint32_t bg_color = SEGCOLOR(1);
+  CRGBW bg_color = SEGCOLOR(1);
   if (bg_color > 0) { //if not black
-    CRGB bg_color_crgb = bg_color; // convert to CRGB
     for (int32_t i = 0; i <= maxXpixel; i++) {
-      #ifdef ESP8266 // no local buffer on ESP8266
-      SEGMENT.addPixelColor(i, bg_color, true);
-      #else
-      fast_color_add(framebuffer[i], bg_color_crgb);
-      #endif
+      framebuffer[i] = fast_color_add(framebuffer[i], bg_color);
     }
   }
-
-  #ifndef ESP8266
-  // transfer the frame-buffer to segment
-  for (int x = 0; x <= maxXpixel; x++) {
-    SEGMENT.setPixelColor(x, framebuffer[x]);
+#ifndef WLED_DISABLE_2D
+  // transfer local buffer to segment if using 1D->2D mapping
+  if(SEGMENT.is2D() && SEGMENT.map1D2D) {
+    for (int x = 0; x <= maxXpixel; x++) {
+    //for (int x = 0; x < SEGMENT.vLength(); x++) {
+      SEGMENT.setPixelColor(x, framebuffer[x]); // this applies the mapping
+    }
   }
-  #endif
+#endif
 }
 
 // calculate pixel positions and brightness distribution and render the particle to local buffer or global buffer
-__attribute__((optimize("O2"))) void ParticleSystem1D::renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGB &color, const bool wrap) {
+__attribute__((optimize("O2"))) void ParticleSystem1D::renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGBW &color, const bool wrap) {
   uint32_t size = particlesize;
   if (advPartProps) // use advanced size properties (1D system has no large size global rendering TODO: add large global rendering?)
     size = advPartProps[particleindex].size;
@@ -1518,11 +1493,7 @@ __attribute__((optimize("O2"))) void ParticleSystem1D::renderParticle(const uint
   if (size == 0) { //single pixel particle, can be out of bounds as oob checking is made for 2-pixel particles (and updating it uses more code)
     uint32_t x =  particles[particleindex].x >> PS_P_RADIUS_SHIFT_1D;
     if (x <= (uint32_t)maxXpixel) { //by making x unsigned there is no need to check < 0 as it will overflow
-      #ifdef ESP8266 // no local buffer on ESP8266
-      SEGMENT.addPixelColor(x, color.scale8(brightness), true);
-      #else
-      fast_color_add(framebuffer[x], color, brightness);
-      #endif
+      framebuffer[x] = fast_color_add(framebuffer[x], color, brightness);
     }
     return;
   }
@@ -1548,18 +1519,19 @@ __attribute__((optimize("O2"))) void ParticleSystem1D::renderParticle(const uint
   // - scale brigthness with gamma correction (done in render())
   // - apply inverse gamma correction to brightness values
   // - gamma is applied again in show() -> the resulting brightness distribution is linear but gamma corrected in total
-  pxlbrightness[0] = gamma8inv(pxlbrightness[0]); // use look-up-table for invers gamma
-  pxlbrightness[1] = gamma8inv(pxlbrightness[1]);
-
+  if(gammaCorrectCol) {
+    pxlbrightness[0] = gamma8inv(pxlbrightness[0]); // use look-up-table for invers gamma
+    pxlbrightness[1] = gamma8inv(pxlbrightness[1]);
+  }
   // check if particle has advanced size properties and buffer is available
   if (advPartProps && advPartProps[particleindex].size > 1) {
-    CRGB renderbuffer[10]; // 10 pixel buffer
+    uint32_t renderbuffer[10]; // 10 pixel buffer
     memset(renderbuffer, 0, sizeof(renderbuffer)); // clear buffer
     //render particle to a bigger size
     //particle size to pixels: 2 - 63 is 4 pixels, < 128 is 6pixels, < 192 is 8 pixels, bigger is 10 pixels
     //first, render the pixel to the center of the renderbuffer, then apply 1D blurring
-    fast_color_add(renderbuffer[4], color, pxlbrightness[0]);
-    fast_color_add(renderbuffer[5], color, pxlbrightness[1]);
+    renderbuffer[4] = fast_color_add(renderbuffer[4], color, pxlbrightness[0]);
+    renderbuffer[5] = fast_color_add(renderbuffer[5], color, pxlbrightness[1]);
     uint32_t rendersize = 2; // initialize render size, minimum is 4 pixels, it is incremented int he loop below to start with 4
     uint32_t offset = 4; // offset to zero coordinate to write/read data in renderbuffer (actually needs to be 3, is decremented in the loop below)
     uint32_t blurpasses = size/64 + 1; // number of blur passes depends on size, four passes max
@@ -1593,7 +1565,7 @@ __attribute__((optimize("O2"))) void ParticleSystem1D::renderParticle(const uint
       #ifdef ESP8266 // no local buffer on ESP8266
       SEGMENT.addPixelColor(xfb, renderbuffer[xrb], true);
       #else
-      fast_color_add(framebuffer[xfb], renderbuffer[xrb]);
+      framebuffer[xfb] = fast_color_add(framebuffer[xfb], renderbuffer[xrb]);
       #endif
     }
   }
@@ -1613,11 +1585,7 @@ __attribute__((optimize("O2"))) void ParticleSystem1D::renderParticle(const uint
     }
     for (uint32_t i = 0; i < 2; i++) {
       if (pxlisinframe[i]) {
-        #ifdef ESP8266 // no local buffer on ESP8266
-        SEGMENT.addPixelColor(pixco[i], color.scale8((uint8_t)pxlbrightness[i]), true);
-        #else
-        fast_color_add(framebuffer[pixco[i]], color, pxlbrightness[i]);
-        #endif
+        framebuffer[pixco[i]] = fast_color_add(framebuffer[pixco[i]], color, pxlbrightness[i]);
       }
     }
   }
@@ -1753,7 +1721,7 @@ __attribute__((optimize("O2"))) void ParticleSystem1D::collideParticles(PSpartic
 // update size and pointers (memory location and size can change dynamically)
 // note: do not access the PS class in FX befor running this function (or it messes up SEGENV.data)
 void ParticleSystem1D::updateSystem(void) {
-  setSize(SEGMENT.virtualLength()); // update size
+  setSize(SEGMENT.vLength()); // update size
   updatePSpointers(advPartProps != nullptr);
 }
 
@@ -1768,15 +1736,16 @@ void ParticleSystem1D::updatePSpointers(bool isadvanced) {
   particles = reinterpret_cast<PSparticle1D *>(this + 1); // pointer to particles
   particleFlags = reinterpret_cast<PSparticleFlags1D *>(particles + numParticles); // pointer to particle flags
   sources = reinterpret_cast<PSsource1D *>(particleFlags + numParticles); // pointer to source(s)
-  #ifdef ESP8266 // no local buffer on ESP8266
-  PSdataEnd = reinterpret_cast<uint8_t *>(sources + numSources);
-  #else
-  framebuffer = reinterpret_cast<CRGB *>(sources + numSources); // pointer to framebuffer
-  // align pointer after framebuffer to 4bytes
-  uintptr_t p = reinterpret_cast<uintptr_t>(framebuffer + (maxXpixel+1)); // maxXpixel is SEGMENT.virtualLength() - 1
-  p = (p + 3) & ~0x03; // align to 4-byte boundary
-  PSdataEnd = reinterpret_cast<uint8_t *>(p); // pointer to first available byte after the PS for FX additional data
-  #endif
+  PSdataEnd = reinterpret_cast<uint8_t *>(sources + numSources);   // pointer to first available byte after the PS for FX additional data (already aligned to 4 byte boundary)
+#ifndef WLED_DISABLE_2D
+  if(SEGMENT.is2D() && SEGMENT.map1D2D) {
+    framebuffer = reinterpret_cast<uint32_t *>(sources + numSources); // use local framebuffer for 1D->2D mapping
+    PSdataEnd = reinterpret_cast<uint8_t *>(framebuffer + SEGMENT.maxMappingLength()); // pointer to first available byte after the PS for FX additional data (still aligned to 4 byte boundary)
+  }
+  else
+#endif
+    framebuffer = SEGMENT.getPixels();  // use segment buffer for standard 1D rendering
+
   if (isadvanced) {
     advPartProps = reinterpret_cast<PSadvancedParticle1D *>(PSdataEnd);
     PSdataEnd = reinterpret_cast<uint8_t *>(advPartProps + numParticles); // since numParticles is a multiple of 4, this is always aligned to 4 bytes. No need to add padding bytes here
@@ -1797,32 +1766,20 @@ void ParticleSystem1D::updatePSpointers(bool isadvanced) {
 //non class functions to use for initialization, fraction is uint8_t: 255 means 100%
 uint32_t calculateNumberOfParticles1D(const uint32_t fraction, const bool isadvanced) {
   uint32_t numberofParticles = SEGMENT.virtualLength();  // one particle per pixel (if possible)
-#ifdef ESP8266
-  uint32_t particlelimit = ESP8266_MAXPARTICLES_1D; // maximum number of paticles allowed
-#elif ARDUINO_ARCH_ESP32S2
-  uint32_t particlelimit = ESP32S2_MAXPARTICLES_1D; // maximum number of paticles allowed
-#else
-  uint32_t particlelimit = ESP32_MAXPARTICLES_1D; // maximum number of paticles allowed
-#endif
+  uint32_t particlelimit = MAXPARTICLES_1D; // maximum number of paticles allowed
   numberofParticles = min(numberofParticles, particlelimit); // limit to particlelimit
   if (isadvanced) // advanced property array needs ram, reduce number of particles to use the same amount
     numberofParticles = (numberofParticles * sizeof(PSparticle1D)) / (sizeof(PSparticle1D) + sizeof(PSadvancedParticle1D));
   numberofParticles = (numberofParticles * (fraction + 1)) >> 8; // calculate fraction of particles
-  numberofParticles = numberofParticles < 20 ? 20 : numberofParticles; // 20 minimum
+  numberofParticles = numberofParticles < 10 ? 10 : numberofParticles; // 10 minimum
   //make sure it is a multiple of 4 for proper memory alignment (easier than using padding bytes)
   numberofParticles = (numberofParticles+3) & ~0x03; // note: with a separate particle buffer, this is probably unnecessary
-  PSPRINTLN(" calc numparticles:" + String(numberofParticles))
+  PSPRINTLN(" calc numparticles:" + String(numberofParticles));
   return numberofParticles;
 }
 
 uint32_t calculateNumberOfSources1D(const uint32_t requestedsources) {
-#ifdef ESP8266
-   int numberofSources = max(1, min((int)requestedsources,ESP8266_MAXSOURCES_1D)); // limit
-#elif ARDUINO_ARCH_ESP32S2
-  int numberofSources = max(1, min((int)requestedsources, ESP32S2_MAXSOURCES_1D)); // limit
-#else
-  int numberofSources = max(1, min((int)requestedsources, ESP32_MAXSOURCES_1D)); // limit
-#endif
+  int numberofSources = max(1, min((int)requestedsources,MAXSOURCES_1D)); // limit
   // make sure it is a multiple of 4 for proper memory alignment (so minimum is acutally 4)
   numberofSources = (numberofSources+3) & ~0x03;
   return numberofSources;
@@ -1835,10 +1792,11 @@ bool allocateParticleSystemMemory1D(const uint32_t numparticles, const uint32_t
   requiredmemory += sizeof(PSparticleFlags1D) * numparticles;
   requiredmemory += sizeof(PSparticle1D) * numparticles;
   requiredmemory += sizeof(PSsource1D) * numsources;
-  #ifndef ESP8266 // no local buffer on ESP8266
-  requiredmemory += sizeof(CRGB) * SEGMENT.virtualLength();
-  #endif
-  requiredmemory += additionalbytes + 3; // add 3 to ensure room for stuffing bytes to make it 4 byte aligned
+#ifndef WLED_DISABLE_2D
+  if(SEGMENT.is2D())
+    requiredmemory += sizeof(uint32_t) * SEGMENT.maxMappingLength(); // need local buffer for mapped rendering
+#endif
+  requiredmemory += additionalbytes;
   if (isadvanced)
     requiredmemory += sizeof(PSadvancedParticle1D) * numparticles;
   return(SEGMENT.allocateData(requiredmemory));
@@ -1850,9 +1808,19 @@ bool initParticleSystem1D(ParticleSystem1D *&PartSys, const uint32_t requestedso
   if (SEGLEN == 1) return false; // single pixel not supported
   uint32_t numparticles = calculateNumberOfParticles1D(fractionofparticles, advanced);
   uint32_t numsources = calculateNumberOfSources1D(requestedsources);
-  if (!allocateParticleSystemMemory1D(numparticles, numsources, advanced, additionalbytes)) {
-    DEBUG_PRINT(F("PS init failed: memory depleted"));
-    return false;
+  bool allocsuccess = false;
+  while(numparticles >= 10) { // make sure we have at least 10 particles or quit
+    if (allocateParticleSystemMemory1D(numparticles, numsources, advanced, additionalbytes)) {
+      PSPRINT(F("PS 1D alloc succeeded"));
+      allocsuccess = true;
+      break; // allocation succeeded
+    }
+    numparticles = ((numparticles / 2) + 3) & ~0x03; // cut number of particles in half and try again, must be 4 byte aligned
+    PSPRINTLN(F("PS 1D alloc failed, trying with less particles..."));
+  }
+  if (!allocsuccess) {
+    PSPRINTLN(F("PS init failed: memory depleted"));
+    return false; // allocation failed
   }
   PartSys = new (SEGENV.data) ParticleSystem1D(SEGMENT.virtualLength(), numparticles, numsources, advanced); // particle system constructor
   return true;
@@ -1861,18 +1829,17 @@ bool initParticleSystem1D(ParticleSystem1D *&PartSys, const uint32_t requestedso
 // blur a 1D buffer, sub-size blurring can be done using start and size
 // for speed, 32bit variables are used, make sure to limit them to 8bit (0-255) or result is undefined
 // to blur a subset of the buffer, change the size and set start to the desired starting coordinates
-void blur1D(CRGB *colorbuffer, uint32_t size, uint32_t blur, uint32_t start)
+void blur1D(uint32_t *colorbuffer, uint32_t size, uint32_t blur, uint32_t start)
 {
-  CRGB seeppart, carryover;
+  CRGBW seeppart, carryover;
   uint32_t seep = blur >> 1;
   carryover =  BLACK;
   for (uint32_t x = start; x < start + size; x++) {
-    seeppart = colorbuffer[x]; // create copy of current color
-    fast_color_scale(seeppart, seep); // scale it and seep to neighbours
+    seeppart = fast_color_scale(colorbuffer[x], seep); // scale it and seep to neighbours
     if (x > 0) {
-      fast_color_add(colorbuffer[x-1], seeppart);
-      if (carryover) // note: check adds overhead but is faster on average
-        fast_color_add(colorbuffer[x], carryover); // is black on first pass
+      colorbuffer[x-1] = fast_color_add(colorbuffer[x-1], seeppart);
+      if (carryover.color32) // note: check adds overhead but is faster on average
+        colorbuffer[x] = fast_color_add(colorbuffer[x], carryover); // is black on first pass
     }
     carryover = seeppart;
   }
@@ -1921,23 +1888,14 @@ static bool checkBoundsAndWrap(int32_t &position, const int32_t max, const int32
   return true; // particle is in bounds
 }
 
-// fastled color adding is very inaccurate in color preservation (but it is fast)
-// a better color add function is implemented in colors.cpp but it uses 32bit RGBW. to use it colors need to be shifted just to then be shifted back by that function, which is slow
-// this is a fast version for RGB (no white channel, PS does not handle white) and with native CRGB including scaling of second color
-// note: result is stored in c1, not using a return value is faster as the CRGB struct does not need to be copied upon return
-// note2: function is mainly used to add scaled colors, so checking if one color is black is slower
-// note3: scale is 255 when using blur, checking for that makes blur faster
- __attribute__((optimize("O2"))) static void fast_color_add(CRGB &c1, const CRGB &c2, const uint8_t scale) {
+// this is a fast version for CRGBW color adding ignoring white channel (PS does not handle white) including scaling of second color
+// note: function is mainly used to add scaled colors, so checking if one color is black is slower
+// note2: returning CRGBW value is slightly slower as the return value gets written to uint32_t framebuffer
+ __attribute__((optimize("O2"))) static uint32_t fast_color_add(CRGBW c1, const CRGBW c2, const uint8_t scale) {
   uint32_t r, g, b;
-  if (scale < 255) {
-    r = c1.r + ((c2.r * scale) >> 8);
-    g = c1.g + ((c2.g * scale) >> 8);
-    b = c1.b + ((c2.b * scale) >> 8);
-  } else {
-    r = c1.r + c2.r;
-    g = c1.g + c2.g;
-    b = c1.b + c2.b;
-  }
+  r = c1.r + ((c2.r * scale) >> 8);
+  g = c1.g + ((c2.g * scale) >> 8);
+  b = c1.b + ((c2.b * scale) >> 8);
 
   // note: this chained comparison is the fastest method for max of 3 values (faster than std:max() or using xor)
   uint32_t max = (r > g) ? ((r > b) ? r : b) : ((g > b) ? g : b);
@@ -1951,13 +1909,15 @@ static bool checkBoundsAndWrap(int32_t &position, const int32_t max, const int32
     c1.g = (g * newscale) >> 16;
     c1.b = (b * newscale) >> 16;
   }
+  return c1.color32;
 }
 
-// faster than fastled color scaling as it does in place scaling
- __attribute__((optimize("O2"))) static void fast_color_scale(CRGB &c, const uint8_t scale) {
+// fast CRGBW color scaling ignoring white channel (PS does not handle white)
+ __attribute__((optimize("O2"))) static uint32_t fast_color_scale(CRGBW c, const uint8_t scale) {
   c.r = ((c.r * scale) >> 8);
   c.g = ((c.g * scale) >> 8);
   c.b = ((c.b * scale) >> 8);
+  return c.color32;
 }
 
 #endif  // !(defined(WLED_DISABLE_PARTICLESYSTEM2D) && defined(WLED_DISABLE_PARTICLESYSTEM1D))

wled00/FXparticleSystem.h

@@ -37,13 +37,20 @@ static inline int32_t limitSpeed(const int32_t speed) {
 #endif
 
 #ifndef WLED_DISABLE_PARTICLESYSTEM2D
-// memory allocation
-#define ESP8266_MAXPARTICLES 256 // enough up to 16x16 pixels
-#define ESP8266_MAXSOURCES 24
-#define ESP32S2_MAXPARTICLES 1024 // enough up to 32x32 pixels
-#define ESP32S2_MAXSOURCES 64
-#define ESP32_MAXPARTICLES 2048 // enough up to 64x32 pixels
-#define ESP32_MAXSOURCES 128
+// memory allocation (based on reasonable segment size and available FX memory)
+#ifdef ESP8266
+  #define MAXPARTICLES_2D 256
+  #define MAXSOURCES_2D 24
+  #define SOURCEREDUCTIONFACTOR 8
+#elif ARDUINO_ARCH_ESP32S2
+  #define MAXPARTICLES_2D 1024
+  #define MAXSOURCES_2D 64
+  #define SOURCEREDUCTIONFACTOR 6
+#else
+  #define MAXPARTICLES_2D 2048
+  #define MAXSOURCES_2D 128
+  #define SOURCEREDUCTIONFACTOR 4
+#endif
 
 // particle dimensions (subpixel division)
 #define PS_P_RADIUS 64 // subpixel size, each pixel is divided by this for particle movement (must be a power of 2)
@@ -188,7 +195,7 @@ public:
 private:
   //rendering functions
   void render();
-  [[gnu::hot]] void renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGB& color, const bool wrapX, const bool wrapY);
+  [[gnu::hot]] void renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGBW& color, const bool wrapX, const bool wrapY);
   //paricle physics applied by system if flags are set
   void applyGravity(); // applies gravity to all particles
   void handleCollisions();
@@ -200,13 +207,13 @@ private:
   void getParticleXYsize(PSadvancedParticle *advprops, PSsizeControl *advsize, uint32_t &xsize, uint32_t &ysize);
   [[gnu::hot]] void bounce(int8_t &incomingspeed, int8_t &parallelspeed, int32_t &position, const uint32_t maxposition); // bounce on a wall
   // note: variables that are accessed often are 32bit for speed
-  CRGB *framebuffer; // local frame buffer for rendering
+  uint32_t *framebuffer; // frame buffer for rendering. note: using CRGBW as the buffer is slower, ESP compiler seems to optimize this better giving more consistent FPS
   PSsettings2D particlesettings; // settings used when updating particles (can also used by FX to move sources), do not edit properties directly, use functions above
   uint32_t numParticles;  // total number of particles allocated by this system
   uint32_t emitIndex; // index to count through particles to emit so searching for dead pixels is faster
   int32_t collisionHardness;
   uint32_t wallHardness;
-  uint32_t wallRoughness; // randomizes wall collisions  
+  uint32_t wallRoughness; // randomizes wall collisions
   uint32_t particleHardRadius; // hard surface radius of a particle, used for collision detection (32bit for speed)
   uint16_t collisionStartIdx; // particle array start index for collision detection
   uint8_t fireIntesity = 0; // fire intensity, used for fire mode (flash use optimization, better than passing an argument to render function)
@@ -219,7 +226,7 @@ private:
   uint8_t smearBlur; // 2D smeared blurring of full frame
 };
 
-void blur2D(CRGB *colorbuffer, const uint32_t xsize, uint32_t ysize, const uint32_t xblur, const uint32_t yblur, const uint32_t xstart = 0, uint32_t ystart = 0, const bool isparticle = false);
+void blur2D(uint32_t *colorbuffer, const uint32_t xsize, uint32_t ysize, const uint32_t xblur, const uint32_t yblur, const uint32_t xstart = 0, uint32_t ystart = 0, const bool isparticle = false);
 // initialization functions (not part of class)
 bool initParticleSystem2D(ParticleSystem2D *&PartSys, const uint32_t requestedsources, const uint32_t additionalbytes = 0, const bool advanced = false, const bool sizecontrol = false);
 uint32_t calculateNumberOfParticles2D(const uint32_t pixels, const bool advanced, const bool sizecontrol);
@@ -232,12 +239,18 @@ bool allocateParticleSystemMemory2D(const uint32_t numparticles, const uint32_t
 ////////////////////////
 #ifndef WLED_DISABLE_PARTICLESYSTEM1D
 // memory allocation
-#define ESP8266_MAXPARTICLES_1D 320
-#define ESP8266_MAXSOURCES_1D 16
-#define ESP32S2_MAXPARTICLES_1D 1300
-#define ESP32S2_MAXSOURCES_1D 32
-#define ESP32_MAXPARTICLES_1D 2600
-#define ESP32_MAXSOURCES_1D 64
+#ifdef ESP8266
+  #define MAXPARTICLES_1D 320
+  #define MAXSOURCES_1D 16
+#elif ARDUINO_ARCH_ESP32S2
+  #define MAXPARTICLES_1D 1300
+  #define MAXSOURCES_1D 32
+#else
+  #define MAXPARTICLES_1D 2600
+  #define MAXSOURCES_1D 64
+#endif
+
+
 
 // particle dimensions (subpixel division)
 #define PS_P_RADIUS_1D 32 // subpixel size, each pixel is divided by this for particle movement, if this value is changed, also change the shift defines (next two lines)
@@ -351,7 +364,7 @@ public:
 private:
   //rendering functions
   void render(void);
-  [[gnu::hot]] void renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGB &color, const bool wrap);
+  [[gnu::hot]] void renderParticle(const uint32_t particleindex, const uint8_t brightness, const CRGBW &color, const bool wrap);
 
   //paricle physics applied by system if flags are set
   void applyGravity(); // applies gravity to all particles
@@ -363,9 +376,7 @@ private:
   //void updateSize(PSadvancedParticle *advprops, PSsizeControl *advsize); // advanced size control
   [[gnu::hot]] void bounce(int8_t &incomingspeed, int8_t &parallelspeed, int32_t &position, const uint32_t maxposition); // bounce on a wall
   // note: variables that are accessed often are 32bit for speed
-  #ifndef ESP8266
-  CRGB *framebuffer; // local frame buffer for rendering
-  #endif
+  uint32_t *framebuffer; // frame buffer for rendering. note: using CRGBW as the buffer is slower, ESP compiler seems to optimize this better giving more consistent FPS
   PSsettings1D particlesettings; // settings used when updating particles
   uint32_t numParticles;  // total number of particles allocated by this system
   uint32_t emitIndex; // index to count through particles to emit so searching for dead pixels is faster
@@ -386,5 +397,5 @@ bool initParticleSystem1D(ParticleSystem1D *&PartSys, const uint32_t requestedso
 uint32_t calculateNumberOfParticles1D(const uint32_t fraction, const bool isadvanced);
 uint32_t calculateNumberOfSources1D(const uint32_t requestedsources);
 bool allocateParticleSystemMemory1D(const uint32_t numparticles, const uint32_t numsources, const bool isadvanced, const uint32_t additionalbytes);
-void blur1D(CRGB *colorbuffer, uint32_t size, uint32_t blur, uint32_t start);
+void blur1D(uint32_t *colorbuffer, uint32_t size, uint32_t blur, uint32_t start);
 #endif // WLED_DISABLE_PARTICLESYSTEM1D

wled00/bus_manager.cpp

@@ -5,6 +5,8 @@
 #include <Arduino.h>
 #include <IPAddress.h>
 #ifdef ARDUINO_ARCH_ESP32
+#include <ESPmDNS.h>
+#include "src/dependencies/network/Network.h" // for isConnected() (& WiFi)
 #include "driver/ledc.h"
 #include "soc/ledc_struct.h"
   #if !(defined(CONFIG_IDF_TARGET_ESP32C3) || defined(CONFIG_IDF_TARGET_ESP32S2) || defined(CONFIG_IDF_TARGET_ESP32S3))
@@ -20,11 +22,13 @@
 #include "core_esp8266_waveform.h"
 #endif
 #include "const.h"
+#include "colors.h"
 #include "pin_manager.h"
 #include "bus_manager.h"
 #include "bus_wrapper.h"
 #include <bits/unique_ptr.h>
 
+extern char cmDNS[];
 extern bool cctICused;
 extern bool useParallelI2S;
 
@@ -36,25 +40,32 @@ uint8_t realtimeBroadcast(uint8_t type, IPAddress client, uint16_t length, const
 
 //util.cpp
 // PSRAM allocation wrappers
-#ifndef ESP8266
+#if !defined(ESP8266) && !defined(CONFIG_IDF_TARGET_ESP32C3)
 extern "C" {
   void *p_malloc(size_t);           // prefer PSRAM over DRAM
   void *p_calloc(size_t, size_t);   // prefer PSRAM over DRAM
   void *p_realloc(void *, size_t);  // prefer PSRAM over DRAM
+  void *p_realloc_malloc(void *ptr, size_t size); // realloc with malloc fallback, prefer PSRAM over DRAM
   inline void p_free(void *ptr) { heap_caps_free(ptr); }
   void *d_malloc(size_t);           // prefer DRAM over PSRAM
   void *d_calloc(size_t, size_t);   // prefer DRAM over PSRAM
   void *d_realloc(void *, size_t);  // prefer DRAM over PSRAM
+  void *d_realloc_malloc(void *ptr, size_t size); // realloc with malloc fallback, prefer DRAM over PSRAM
   inline void d_free(void *ptr) { heap_caps_free(ptr); }
 }
 #else
+extern "C" {
+  void *realloc_malloc(void *ptr, size_t size);
+}
 #define p_malloc malloc
 #define p_calloc calloc
 #define p_realloc realloc
+#define p_realloc_malloc realloc_malloc
 #define p_free free
 #define d_malloc malloc
 #define d_calloc calloc
 #define d_realloc realloc
+#define d_realloc_malloc realloc_malloc
 #define d_free free
 #endif
 
@@ -95,7 +106,7 @@ void Bus::calculateCCT(uint32_t c, uint8_t &ww, uint8_t &cw) {
   } else {
     cct = (approximateKelvinFromRGB(c) - 1900) >> 5;  // convert K (from RGB value) to relative format
   }
-  
+
   //0 - linear (CCT 127 = 50% warm, 50% cold), 127 - additive CCT blending (CCT 127 = 100% warm, 100% cold)
   if (cct       < _cctBlend) ww = 255;
   else                       ww = ((255-cct) * 255) / (255 - _cctBlend);
@@ -134,6 +145,7 @@ BusDigital::BusDigital(const BusConfig &bc, uint8_t nr)
   if (!isDigital(bc.type) || !bc.count) { DEBUGBUS_PRINTLN(F("Not digial or empty bus!")); return; }
   if (!PinManager::allocatePin(bc.pins[0], true, PinOwner::BusDigital)) { DEBUGBUS_PRINTLN(F("Pin 0 allocated!")); return; }
   _frequencykHz = 0U;
+  _colorSum = 0;
   _pins[0] = bc.pins[0];
   if (is2Pin(bc.type)) {
     if (!PinManager::allocatePin(bc.pins[1], true, PinOwner::BusDigital)) {
@@ -176,80 +188,62 @@ BusDigital::BusDigital(const BusConfig &bc, uint8_t nr)
 //Stay safe with high amperage and have a reasonable safety margin!
 //I am NOT to be held liable for burned down garages or houses!
 
-// To disable brightness limiter we either set output max current to 0 or single LED current to 0
-uint8_t BusDigital::estimateCurrentAndLimitBri() const {
-  bool useWackyWS2815PowerModel = false;
-  byte actualMilliampsPerLed = _milliAmpsPerLed;
-
-  if (_milliAmpsMax < MA_FOR_ESP/BusManager::getNumBusses() || actualMilliampsPerLed == 0) { //0 mA per LED and too low numbers turn off calculation
-    return _bri;
-  }
+// note on ABL implementation:
+// ABL is set up in finalizeInit()
+// scaled color channels are summed in BusDigital::setPixelColor()
+// the used current is estimated and limited in BusManager::show()
+// if limit is set too low, brightness is limited to 1 to at least show some light
+// to disable brightness limiter for a bus, set LED current to 0
 
+void BusDigital::estimateCurrent() {
+  uint32_t actualMilliampsPerLed = _milliAmpsPerLed;
   if (_milliAmpsPerLed == 255) {
-    useWackyWS2815PowerModel = true;
+    // use wacky WS2815 power model, see WLED issue #549
+    _colorSum *= 3; // sum is sum of max value for each color, need to multiply by three to account for clrUnitsPerChannel being 3*255
     actualMilliampsPerLed = 12; // from testing an actual strip
   }
+  // _colorSum has all the values of color channels summed, max would be getLength()*(3*255 + (255 if hasWhite()): convert to milliAmps
+  uint32_t clrUnitsPerChannel = hasWhite() ? 4*255 : 3*255;
+  _milliAmpsTotal = ((uint64_t)_colorSum * actualMilliampsPerLed) / clrUnitsPerChannel + getLength(); // add 1mA standby current per LED to total (WS2812: ~0.7mA, WS2815: ~2mA)
+}
 
-  unsigned powerBudget = (_milliAmpsMax - MA_FOR_ESP/BusManager::getNumBusses()); //80/120mA for ESP power
-  if (powerBudget > getLength()) { //each LED uses about 1mA in standby, exclude that from power budget
-    powerBudget -= getLength();
-  } else {
-    powerBudget = 0;
-  }
+void BusDigital::applyBriLimit(uint8_t newBri) {
+  // a newBri of 0 means calculate per-bus brightness limit
+  if (newBri == 0) {
+    if (_milliAmpsLimit == 0 || _milliAmpsTotal == 0) return; // ABL not used for this bus
+    newBri = 255;
 
-  uint32_t busPowerSum = 0;
-  for (unsigned i = 0; i < getLength(); i++) {  //sum up the usage of each LED
-    uint32_t c = getPixelColor(i); // always returns original or restored color without brightness scaling
-    byte r = R(c), g = G(c), b = B(c), w = W(c);
-
-    if (useWackyWS2815PowerModel) { //ignore white component on WS2815 power calculation
-      busPowerSum += (max(max(r,g),b)) * 3;
+    if (_milliAmpsLimit > getLength()) { // each LED uses about 1mA in standby
+      if (_milliAmpsTotal > _milliAmpsLimit) {
+        // scale brightness down to stay in current limit
+        newBri = ((uint32_t)_milliAmpsLimit * 255) / _milliAmpsTotal + 1; // +1 to avoid 0 brightness
+        _milliAmpsTotal = _milliAmpsLimit;
+      }
     } else {
-      busPowerSum += (r + g + b + w);
+      newBri = 1; // limit too low, set brightness to 1, this will dim down all colors to minimum since we use video scaling
+      _milliAmpsTotal = getLength(); // estimate bus current as minimum
     }
   }
 
-  if (hasWhite()) { //RGBW led total output with white LEDs enabled is still 50mA, so each channel uses less
-    busPowerSum *= 3;
-    busPowerSum >>= 2; //same as /= 4
-  }
-
-  // powerSum has all the values of channels summed (max would be getLength()*765 as white is excluded) so convert to milliAmps
-  BusDigital::_milliAmpsTotal = (busPowerSum * actualMilliampsPerLed * _bri) / (765*255);
-
-  uint8_t newBri = _bri;
-  if (BusDigital::_milliAmpsTotal > powerBudget) {
-    //scale brightness down to stay in current limit
-    unsigned scaleB = powerBudget * 255 / BusDigital::_milliAmpsTotal;
-    newBri = (_bri * scaleB) / 256 + 1;
-    BusDigital::_milliAmpsTotal = powerBudget;
-    //_milliAmpsTotal = (busPowerSum * actualMilliampsPerLed * newBri) / (765*255);
-  }
-  return newBri;
-}
-
-void BusDigital::show() {
-  BusDigital::_milliAmpsTotal = 0;
-  if (!_valid) return;
-
-  uint8_t cctWW = 0, cctCW = 0;
-  unsigned newBri = estimateCurrentAndLimitBri();  // will fill _milliAmpsTotal (TODO: could use PolyBus::CalcTotalMilliAmpere())
-  if (newBri < _bri) {
-    PolyBus::setBrightness(_busPtr, _iType, newBri); // limit brightness to stay within current limits
+  if (newBri < 255) {
+    uint8_t cctWW = 0, cctCW = 0;
     unsigned hwLen = _len;
     if (_type == TYPE_WS2812_1CH_X3) hwLen = NUM_ICS_WS2812_1CH_3X(_len); // only needs a third of "RGB" LEDs for NeoPixelBus
     for (unsigned i = 0; i < hwLen; i++) {
-      // use 0 as color order, actual order does not matter here as we just update the channel values as-is
-      uint32_t c = restoreColorLossy(PolyBus::getPixelColor(_busPtr, _iType, i, 0), _bri);
-      if (hasCCT()) Bus::calculateCCT(c, cctWW, cctCW); // this will unfortunately corrupt (segment) CCT data on every bus
-      PolyBus::setPixelColor(_busPtr, _iType, i, c, 0, (cctCW<<8) | cctWW); // repaint all pixels with new brightness
+      uint8_t co = _colorOrderMap.getPixelColorOrder(i+_start, _colorOrder); // need to revert color order for correct color scaling and CCT calc in case white is swapped
+      uint32_t c = PolyBus::getPixelColor(_busPtr, _iType, i, co);
+      c = color_fade(c, newBri, true); // apply additional dimming  note: using inline version is a bit faster but overhead of getPixelColor() dominates the speed impact by far
+      if (hasCCT()) Bus::calculateCCT(c, cctWW, cctCW);
+      PolyBus::setPixelColor(_busPtr, _iType, i, c, co, (cctCW<<8) | cctWW); // repaint all pixels with new brightness
     }
   }
+
+  _colorSum = 0; // reset for next frame
+}
+
+void BusDigital::show() {
+  if (!_valid) return;
   PolyBus::show(_busPtr, _iType, _skip); // faster if buffer consistency is not important (no skipped LEDs)
-  // restore bus brightness to its original value
-  // this is done right after show, so this is only OK if LED updates are completed before show() returns
-  // or async show has a separate buffer (ESP32 RMT and I2S are ok)
-  if (newBri < _bri) PolyBus::setBrightness(_busPtr, _iType, _bri);
 }
 
 bool BusDigital::canShow() const {
@@ -257,12 +251,6 @@ bool BusDigital::canShow() const {
   return PolyBus::canShow(_busPtr, _iType);
 }
 
-void BusDigital::setBrightness(uint8_t b) {
-  if (_bri == b) return;
-  Bus::setBrightness(b);
-  PolyBus::setBrightness(_busPtr, _iType, b);
-}
-
 //If LEDs are skipped, it is possible to use the first as a status LED.
 //TODO only show if no new show due in the next 50ms
 void BusDigital::setStatusPixel(uint32_t c) {
@@ -276,13 +264,25 @@ void IRAM_ATTR BusDigital::setPixelColor(unsigned pix, uint32_t c) {
   if (!_valid) return;
   if (hasWhite()) c = autoWhiteCalc(c);
   if (Bus::_cct >= 1900) c = colorBalanceFromKelvin(Bus::_cct, c); //color correction from CCT
+  c = color_fade(c, _bri, true); // apply brightness
+
+  if (BusManager::_useABL) {
+    // if using ABL, sum all color channels to estimate current and limit brightness in show()
+    uint8_t r = R(c), g = G(c), b = B(c);
+    if (_milliAmpsPerLed < 255) { // normal ABL
+      _colorSum += r + g + b + W(c);
+    } else { // wacky WS2815 power model, ignore white channel, use max of RGB (issue #549)
+      _colorSum += ((r > g) ? ((r > b) ? r : b) : ((g > b) ? g : b));
+    }
+  }
+
   if (_reversed) pix = _len - pix -1;
   pix += _skip;
-  unsigned co = _colorOrderMap.getPixelColorOrder(pix+_start, _colorOrder);
+  const uint8_t co = _colorOrderMap.getPixelColorOrder(pix+_start, _colorOrder);
   if (_type == TYPE_WS2812_1CH_X3) { // map to correct IC, each controls 3 LEDs
     unsigned pOld = pix;
     pix = IC_INDEX_WS2812_1CH_3X(pix);
-    uint32_t cOld = restoreColorLossy(PolyBus::getPixelColor(_busPtr, _iType, pix, co),_bri);
+    uint32_t cOld = PolyBus::getPixelColor(_busPtr, _iType, pix, co);
     switch (pOld % 3) { // change only the single channel (TODO: this can cause loss because of get/set)
       case 0: c = RGBW32(R(cOld), W(c)   , B(cOld), 0); break;
       case 1: c = RGBW32(W(c)   , G(cOld), B(cOld), 0); break;
@@ -299,17 +299,17 @@ void IRAM_ATTR BusDigital::setPixelColor(unsigned pix, uint32_t c) {
   PolyBus::setPixelColor(_busPtr, _iType, pix, c, co, wwcw);
 }
 
-// returns original color if global buffering is enabled, else returns lossly restored color from bus
+// returns lossly restored color from bus
 uint32_t IRAM_ATTR BusDigital::getPixelColor(unsigned pix) const {
   if (!_valid) return 0;
   if (_reversed) pix = _len - pix -1;
   pix += _skip;
-  const unsigned co = _colorOrderMap.getPixelColorOrder(pix+_start, _colorOrder);
+  const uint8_t co = _colorOrderMap.getPixelColorOrder(pix+_start, _colorOrder);
   uint32_t c = restoreColorLossy(PolyBus::getPixelColor(_busPtr, _iType, (_type==TYPE_WS2812_1CH_X3) ? IC_INDEX_WS2812_1CH_3X(pix) : pix, co),_bri);
   if (_type == TYPE_WS2812_1CH_X3) { // map to correct IC, each controls 3 LEDs
-    unsigned r = R(c);
-    unsigned g = _reversed ? B(c) : G(c); // should G and B be switched if _reversed?
-    unsigned b = _reversed ? G(c) : B(c);
+    uint8_t r = R(c);
+    uint8_t g = _reversed ? B(c) : G(c); // should G and B be switched if _reversed?
+    uint8_t b = _reversed ? G(c) : B(c);
     switch (pix % 3) { // get only the single channel
       case 0: c = RGBW32(g, g, g, g); break;
       case 1: c = RGBW32(r, r, r, r); break;
@@ -461,10 +461,7 @@ void BusPwm::setPixelColor(unsigned pix, uint32_t c) {
   if (Bus::_cct >= 1900 && (_type == TYPE_ANALOG_3CH || _type == TYPE_ANALOG_4CH)) {
     c = colorBalanceFromKelvin(Bus::_cct, c); //color correction from CCT
   }
-  uint8_t r = R(c);
-  uint8_t g = G(c);
-  uint8_t b = B(c);
-  uint8_t w = W(c);
+  uint8_t r = R(c), g = G(c), b = B(c), w = W(c);
 
   switch (_type) {
     case TYPE_ANALOG_1CH: //one channel (white), relies on auto white calculation
@@ -639,10 +636,7 @@ BusOnOff::BusOnOff(const BusConfig &bc)
 void BusOnOff::setPixelColor(unsigned pix, uint32_t c) {
   if (pix != 0 || !_valid) return; //only react to first pixel
   c = autoWhiteCalc(c);
-  uint8_t r = R(c);
-  uint8_t g = G(c);
-  uint8_t b = B(c);
-  uint8_t w = W(c);
+  uint8_t r = R(c), g = G(c), b = B(c), w = W(c);
   _data = bool(r|g|b|w) && bool(_bri) ? 0xFF : 0;
 }
 
@@ -692,6 +686,10 @@ BusNetwork::BusNetwork(const BusConfig &bc)
   _hasCCT = false;
   _UDPchannels = _hasWhite + 3;
   _client = IPAddress(bc.pins[0],bc.pins[1],bc.pins[2],bc.pins[3]);
+  #ifdef ARDUINO_ARCH_ESP32
+  _hostname = bc.text;
+  resolveHostname(); // resolve hostname to IP address if needed
+  #endif
   _data = (uint8_t*)d_calloc(_len, _UDPchannels);
   _valid = (_data != nullptr);
   DEBUGBUS_PRINTF_P(PSTR("%successfully inited virtual strip with type %u and IP %u.%u.%u.%u\n"), _valid?"S":"Uns", bc.type, bc.pins[0], bc.pins[1], bc.pins[2], bc.pins[3]);
@@ -726,6 +724,19 @@ size_t BusNetwork::getPins(uint8_t* pinArray) const {
   return 4;
 }
 
+#ifdef ARDUINO_ARCH_ESP32
+void BusNetwork::resolveHostname() {
+  static unsigned long nextResolve = 0;
+  if (Network.isConnected() && millis() > nextResolve && _hostname.length() > 0) {
+    nextResolve = millis() + 600000; // resolve only every 10 minutes
+    IPAddress clnt;
+    if (strlen(cmDNS) > 0) clnt = MDNS.queryHost(_hostname);
+    else WiFi.hostByName(_hostname.c_str(), clnt);
+    if (clnt != IPAddress()) _client = clnt;
+  }
+}
+#endif
+
 // credit @willmmiles & @netmindz https://github.com/wled/WLED/pull/4056
 std::vector<LEDType> BusNetwork::getLEDTypes() {
   return {
@@ -911,6 +922,13 @@ void BusManager::on() {
       }
     }
   }
+  #else
+  for (auto &bus : busses) if (bus->isVirtual()) {
+    // virtual/network bus should check for IP change if hostname is specified
+    // otherwise there are no endpoints to force DNS resolution
+    BusNetwork &b = static_cast<BusNetwork&>(*bus);
+    b.resolveHostname();
+  }
   #endif
   #ifdef ESP32_DATA_IDLE_HIGH
   esp32RMTInvertIdle();
@@ -930,13 +948,13 @@ void BusManager::off() {
   #ifdef ESP32_DATA_IDLE_HIGH
   esp32RMTInvertIdle();
   #endif
+  _gMilliAmpsUsed = 0; // reset, assume no LED idle current if relay is off
 }
 
 void BusManager::show() {
-  _gMilliAmpsUsed = 0;
+  applyABL(); // apply brightness limit, updates _gMilliAmpsUsed
   for (auto &bus : busses) {
     bus->show();
-    _gMilliAmpsUsed += bus->getUsedCurrent();
   }
 }
 
@@ -969,6 +987,85 @@ bool BusManager::canAllShow() {
   return true;
 }
 
+void BusManager::initializeABL() {
+  _useABL = false; // reset
+  if (_gMilliAmpsMax > 0) {
+    // check global brightness limit
+    for (auto &bus : busses) {
+      if (bus->isDigital() && bus->getLEDCurrent() > 0) {
+        _useABL = true; // at least one bus has valid LED current
+        return;
+      }
+    }
+  } else {
+    // check per bus brightness limit
+    unsigned numABLbuses = 0;
+    for (auto &bus : busses) {
+      if (bus->isDigital() && bus->getLEDCurrent() > 0 && bus->getMaxCurrent() > 0)
+        numABLbuses++; // count ABL enabled buses
+    }
+    if (numABLbuses > 0) {
+      _useABL = true; // at least one bus has ABL set
+      uint32_t ESPshare = MA_FOR_ESP / numABLbuses; // share of ESP current per ABL bus
+      for (auto &bus : busses) {
+        if (bus->isDigital()) {
+          BusDigital &busd = static_cast<BusDigital&>(*bus);
+          uint32_t busLength = busd.getLength();
+          uint32_t busDemand = busLength * busd.getLEDCurrent();
+          uint32_t busMax    = busd.getMaxCurrent();
+          if (busMax > ESPshare)  busMax -= ESPshare;
+          if (busMax < busLength) busMax  = busLength; // give each LED 1mA, ABL will dim down to minimum
+          if (busDemand == 0) busMax = 0; // no LED current set, disable ABL for this bus
+          busd.setCurrentLimit(busMax);
+        }
+      }
+    }
+  }
+}
+
+void BusManager::applyABL() {
+  if (_useABL) {
+    unsigned milliAmpsSum = 0; // use temporary variable to always return a valid _gMilliAmpsUsed to UI
+    unsigned totalLEDs = 0;
+    for (auto &bus : busses) {
+      if (bus->isDigital() && bus->isOk()) {
+        BusDigital &busd = static_cast<BusDigital&>(*bus);
+        busd.estimateCurrent(); // sets _milliAmpsTotal, current is estimated for all buses even if they have the limit set to 0
+        if (_gMilliAmpsMax == 0)
+          busd.applyBriLimit(0); // apply per bus ABL limit, updates _milliAmpsTotal if limit reached
+        milliAmpsSum += busd.getUsedCurrent();
+        totalLEDs += busd.getLength(); // sum total number of LEDs for global Limit
+      }
+    }
+    // check global current limit and apply global ABL limit, total current is summed above
+    if (_gMilliAmpsMax > 0) {
+      uint8_t  newBri = 255;
+      uint32_t globalMax = _gMilliAmpsMax > MA_FOR_ESP ? _gMilliAmpsMax - MA_FOR_ESP : 1; // subtract ESP current consumption, fully limit if too low
+      if (globalMax > totalLEDs) { // check if budget is larger than standby current
+        if (milliAmpsSum > globalMax) {
+          newBri = globalMax * 255 / milliAmpsSum + 1; // scale brightness down to stay in current limit, +1 to avoid 0 brightness
+          milliAmpsSum = globalMax; // update total used current
+        }
+      } else {
+        newBri = 1; // limit too low, set brightness to minimum
+        milliAmpsSum = totalLEDs; // estimate total used current as minimum
+      }
+
+      // apply brightness limit to each bus, if its 255 it will only reset _colorSum
+      for (auto &bus : busses) {
+        if (bus->isDigital() && bus->isOk()) {
+          BusDigital &busd = static_cast<BusDigital&>(*bus);
+          if (busd.getLEDCurrent() > 0)  // skip buses with LED current set to 0
+            busd.applyBriLimit(newBri);
+        }
+      }
+    }
+    _gMilliAmpsUsed = milliAmpsSum;
+  }
+  else
+    _gMilliAmpsUsed = 0; // reset, we have no current estimation without ABL
+}
+
 ColorOrderMap& BusManager::getColorOrderMap() { return _colorOrderMap; }
 
 
@@ -984,3 +1081,4 @@ uint16_t BusDigital::_milliAmpsTotal = 0;
 std::vector<std::unique_ptr<Bus>> BusManager::busses;
 uint16_t BusManager::_gMilliAmpsUsed = 0;
 uint16_t BusManager::_gMilliAmpsMax = ABL_MILLIAMPS_DEFAULT;
+bool BusManager::_useABL = false;

wled00/bus_manager.h

@@ -133,6 +133,7 @@ class Bus {
     virtual uint16_t getUsedCurrent() const                     { return 0; }
     virtual uint16_t getMaxCurrent() const                      { return 0; }
     virtual size_t   getBusSize() const                         { return sizeof(Bus); }
+    virtual const String getCustomText() const                  { return String(); }
 
     inline  bool     hasRGB() const                             { return _hasRgb; }
     inline  bool     hasWhite() const                           { return _hasWhite; }
@@ -215,7 +216,7 @@ class Bus {
     uint8_t  _autoWhiteMode;
     // global Auto White Calculation override
     static uint8_t _gAWM;
-    // _cct has the following menaings (see calculateCCT() & BusManager::setSegmentCCT()):
+    // _cct has the following meanings (see calculateCCT() & BusManager::setSegmentCCT()):
     //    -1 means to extract approximate CCT value in K from RGB (in calcualteCCT())
     //    [0,255] is the exact CCT value where 0 means warm and 255 cold
     //    [1900,10060] only for color correction expressed in K (colorBalanceFromKelvin())
@@ -237,7 +238,6 @@ class BusDigital : public Bus {
 
     void show() override;
     bool canShow() const override;
-    void setBrightness(uint8_t b) override;
     void setStatusPixel(uint32_t c) override;
     [[gnu::hot]] void setPixelColor(unsigned pix, uint32_t c) override;
     void setColorOrder(uint8_t colorOrder) override;
@@ -249,6 +249,9 @@ class BusDigital : public Bus {
     uint16_t getLEDCurrent() const override  { return _milliAmpsPerLed; }
     uint16_t getUsedCurrent() const override { return _milliAmpsTotal; }
     uint16_t getMaxCurrent() const override  { return _milliAmpsMax; }
+    void     setCurrentLimit(uint16_t milliAmps) { _milliAmpsLimit = milliAmps; }
+    void     estimateCurrent(); // estimate used current from summed colors
+    void     applyBriLimit(uint8_t newBri);
     size_t   getBusSize() const override;
     void begin() override;
     void cleanup();
@@ -261,8 +264,10 @@ class BusDigital : public Bus {
     uint8_t  _pins[2];
     uint8_t  _iType;
     uint16_t _frequencykHz;
-    uint8_t  _milliAmpsPerLed;
     uint16_t _milliAmpsMax;
+    uint8_t  _milliAmpsPerLed;
+    uint16_t _milliAmpsLimit;
+    uint32_t _colorSum; // total color value for the bus, updated in setPixelColor(), used to estimate current
     void    *_busPtr;
 
     static uint16_t _milliAmpsTotal; // is overwitten/recalculated on each show()
@@ -277,8 +282,6 @@ class BusDigital : public Bus {
       }
       return c;
     }
-
-    uint8_t  estimateCurrentAndLimitBri() const;
 };
 
 
@@ -342,6 +345,10 @@ class BusNetwork : public Bus {
     size_t getBusSize() const override  { return sizeof(BusNetwork) + (isOk() ? _len * _UDPchannels : 0); }
     void   show() override;
     void   cleanup();
+    #ifdef ARDUINO_ARCH_ESP32
+    void   resolveHostname();
+    const String getCustomText() const override { return _hostname; }
+    #endif
 
     static std::vector<LEDType> getLEDTypes();
 
@@ -351,6 +358,9 @@ class BusNetwork : public Bus {
     uint8_t   _UDPchannels;
     bool      _broadcastLock;
     uint8_t   *_data;
+    #ifdef ARDUINO_ARCH_ESP32
+    String    _hostname;
+    #endif
 };
 
 
@@ -368,8 +378,9 @@ struct BusConfig {
   uint16_t frequency;
   uint8_t milliAmpsPerLed;
   uint16_t milliAmpsMax;
+  String text;
 
-  BusConfig(uint8_t busType, uint8_t* ppins, uint16_t pstart, uint16_t len = 1, uint8_t pcolorOrder = COL_ORDER_GRB, bool rev = false, uint8_t skip = 0, byte aw=RGBW_MODE_MANUAL_ONLY, uint16_t clock_kHz=0U, uint8_t maPerLed=LED_MILLIAMPS_DEFAULT, uint16_t maMax=ABL_MILLIAMPS_DEFAULT)
+  BusConfig(uint8_t busType, uint8_t* ppins, uint16_t pstart, uint16_t len = 1, uint8_t pcolorOrder = COL_ORDER_GRB, bool rev = false, uint8_t skip = 0, byte aw=RGBW_MODE_MANUAL_ONLY, uint16_t clock_kHz=0U, uint8_t maPerLed=LED_MILLIAMPS_DEFAULT, uint16_t maMax=ABL_MILLIAMPS_DEFAULT, String sometext = "")
   : count(std::max(len,(uint16_t)1))
   , start(pstart)
   , colorOrder(pcolorOrder)
@@ -379,6 +390,7 @@ struct BusConfig {
   , frequency(clock_kHz)
   , milliAmpsPerLed(maPerLed)
   , milliAmpsMax(maMax)
+  , text(sometext)
   {
     refreshReq = (bool) GET_BIT(busType,7);
     type = busType & 0x7F;  // bit 7 may be/is hacked to include refresh info (1=refresh in off state, 0=no refresh)
@@ -412,8 +424,8 @@ struct BusConfig {
 };
 
 
-//fine tune power estimation constants for your setup
-//you can set it to 0 if the ESP is powered by USB and the LEDs by external
+// milliamps used by ESP (for power estimation)
+// you can set it to 0 if the ESP is powered by USB and the LEDs by external
 #ifndef MA_FOR_ESP
   #ifdef ESP8266
     #define MA_FOR_ESP         80 //how much mA does the ESP use (Wemos D1 about 80mA)
@@ -428,6 +440,7 @@ namespace BusManager {
   //extern std::vector<Bus*> busses;
   extern uint16_t _gMilliAmpsUsed;
   extern uint16_t _gMilliAmpsMax;
+  extern bool     _useABL;
 
   #ifdef ESP32_DATA_IDLE_HIGH
   void    esp32RMTInvertIdle() ;
@@ -443,6 +456,8 @@ namespace BusManager {
   //inline uint16_t ablMilliampsMax()             { unsigned sum = 0; for (auto &bus : busses) sum += bus->getMaxCurrent(); return sum; }
   inline uint16_t ablMilliampsMax()             { return _gMilliAmpsMax; }  // used for compatibility reasons (and enabling virtual global ABL)
   inline void     setMilliampsMax(uint16_t max) { _gMilliAmpsMax = max;}
+  void            initializeABL();              // setup automatic brightness limiter parameters, call once after buses are initialized
+  void            applyABL();                   // apply automatic brightness limiter, global or per bus
 
   void useParallelOutput(); // workaround for inaccessible PolyBus
   bool hasParallelOutput(); // workaround for inaccessible PolyBus

wled00/bus_wrapper.h

@@ -3,7 +3,7 @@
 #define BusWrapper_h
 
 //#define NPB_CONF_4STEP_CADENCE
-#include "NeoPixelBusLg.h"
+#include "NeoPixelBus.h"
 
 //Hardware SPI Pins
 #define P_8266_HS_MOSI 13
@@ -141,65 +141,65 @@
 /*** ESP8266 Neopixel methods ***/
 #ifdef ESP8266
 //RGB
-#define B_8266_U0_NEO_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266Uart0Ws2813Method, NeoGammaNullMethod> //3 chan, esp8266, gpio1
-#define B_8266_U1_NEO_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266Uart1Ws2813Method, NeoGammaNullMethod> //3 chan, esp8266, gpio2
-#define B_8266_DM_NEO_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266Dma800KbpsMethod, NeoGammaNullMethod>  //3 chan, esp8266, gpio3
-#define B_8266_BB_NEO_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266BitBang800KbpsMethod, NeoGammaNullMethod> //3 chan, esp8266, bb (any pin but 16)
+#define B_8266_U0_NEO_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266Uart0Ws2813Method> //3 chan, esp8266, gpio1
+#define B_8266_U1_NEO_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266Uart1Ws2813Method> //3 chan, esp8266, gpio2
+#define B_8266_DM_NEO_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266Dma800KbpsMethod>  //3 chan, esp8266, gpio3
+#define B_8266_BB_NEO_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266BitBang800KbpsMethod> //3 chan, esp8266, bb (any pin but 16)
 //RGBW
-#define B_8266_U0_NEO_4 NeoPixelBusLg<NeoGrbwFeature, NeoEsp8266Uart0Ws2813Method, NeoGammaNullMethod>   //4 chan, esp8266, gpio1
-#define B_8266_U1_NEO_4 NeoPixelBusLg<NeoGrbwFeature, NeoEsp8266Uart1Ws2813Method, NeoGammaNullMethod>   //4 chan, esp8266, gpio2
-#define B_8266_DM_NEO_4 NeoPixelBusLg<NeoGrbwFeature, NeoEsp8266Dma800KbpsMethod, NeoGammaNullMethod>    //4 chan, esp8266, gpio3
-#define B_8266_BB_NEO_4 NeoPixelBusLg<NeoGrbwFeature, NeoEsp8266BitBang800KbpsMethod, NeoGammaNullMethod> //4 chan, esp8266, bb (any pin)
+#define B_8266_U0_NEO_4 NeoPixelBus<NeoGrbwFeature, NeoEsp8266Uart0Ws2813Method>   //4 chan, esp8266, gpio1
+#define B_8266_U1_NEO_4 NeoPixelBus<NeoGrbwFeature, NeoEsp8266Uart1Ws2813Method>   //4 chan, esp8266, gpio2
+#define B_8266_DM_NEO_4 NeoPixelBus<NeoGrbwFeature, NeoEsp8266Dma800KbpsMethod>    //4 chan, esp8266, gpio3
+#define B_8266_BB_NEO_4 NeoPixelBus<NeoGrbwFeature, NeoEsp8266BitBang800KbpsMethod> //4 chan, esp8266, bb (any pin)
 //400Kbps
-#define B_8266_U0_400_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266Uart0400KbpsMethod, NeoGammaNullMethod>   //3 chan, esp8266, gpio1
-#define B_8266_U1_400_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266Uart1400KbpsMethod, NeoGammaNullMethod>   //3 chan, esp8266, gpio2
-#define B_8266_DM_400_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266Dma400KbpsMethod, NeoGammaNullMethod>     //3 chan, esp8266, gpio3
-#define B_8266_BB_400_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266BitBang400KbpsMethod, NeoGammaNullMethod> //3 chan, esp8266, bb (any pin)
+#define B_8266_U0_400_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266Uart0400KbpsMethod>   //3 chan, esp8266, gpio1
+#define B_8266_U1_400_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266Uart1400KbpsMethod>   //3 chan, esp8266, gpio2
+#define B_8266_DM_400_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266Dma400KbpsMethod>     //3 chan, esp8266, gpio3
+#define B_8266_BB_400_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266BitBang400KbpsMethod> //3 chan, esp8266, bb (any pin)
 //TM1814 (RGBW)
-#define B_8266_U0_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, NeoEsp8266Uart0Tm1814Method, NeoGammaNullMethod>
-#define B_8266_U1_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, NeoEsp8266Uart1Tm1814Method, NeoGammaNullMethod>
-#define B_8266_DM_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, NeoEsp8266DmaTm1814Method, NeoGammaNullMethod>
-#define B_8266_BB_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, NeoEsp8266BitBangTm1814Method, NeoGammaNullMethod>
+#define B_8266_U0_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, NeoEsp8266Uart0Tm1814Method>
+#define B_8266_U1_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, NeoEsp8266Uart1Tm1814Method>
+#define B_8266_DM_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, NeoEsp8266DmaTm1814Method>
+#define B_8266_BB_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, NeoEsp8266BitBangTm1814Method>
 //TM1829 (RGB)
-#define B_8266_U0_TM2_3 NeoPixelBusLg<NeoBrgFeature, NeoEsp8266Uart0Tm1829Method, NeoGammaNullMethod>
-#define B_8266_U1_TM2_3 NeoPixelBusLg<NeoBrgFeature, NeoEsp8266Uart1Tm1829Method, NeoGammaNullMethod>
-#define B_8266_DM_TM2_3 NeoPixelBusLg<NeoBrgFeature, NeoEsp8266DmaTm1829Method, NeoGammaNullMethod>
-#define B_8266_BB_TM2_3 NeoPixelBusLg<NeoBrgFeature, NeoEsp8266BitBangTm1829Method, NeoGammaNullMethod>
+#define B_8266_U0_TM2_3 NeoPixelBus<NeoBrgFeature, NeoEsp8266Uart0Tm1829Method>
+#define B_8266_U1_TM2_3 NeoPixelBus<NeoBrgFeature, NeoEsp8266Uart1Tm1829Method>
+#define B_8266_DM_TM2_3 NeoPixelBus<NeoBrgFeature, NeoEsp8266DmaTm1829Method>
+#define B_8266_BB_TM2_3 NeoPixelBus<NeoBrgFeature, NeoEsp8266BitBangTm1829Method>
 //UCS8903
-#define B_8266_U0_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, NeoEsp8266Uart0Ws2813Method, NeoGammaNullMethod> //3 chan, esp8266, gpio1
-#define B_8266_U1_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, NeoEsp8266Uart1Ws2813Method, NeoGammaNullMethod> //3 chan, esp8266, gpio2
-#define B_8266_DM_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, NeoEsp8266Dma800KbpsMethod, NeoGammaNullMethod>  //3 chan, esp8266, gpio3
-#define B_8266_BB_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, NeoEsp8266BitBang800KbpsMethod, NeoGammaNullMethod> //3 chan, esp8266, bb (any pin but 16)
+#define B_8266_U0_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, NeoEsp8266Uart0Ws2813Method> //3 chan, esp8266, gpio1
+#define B_8266_U1_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, NeoEsp8266Uart1Ws2813Method> //3 chan, esp8266, gpio2
+#define B_8266_DM_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, NeoEsp8266Dma800KbpsMethod>  //3 chan, esp8266, gpio3
+#define B_8266_BB_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, NeoEsp8266BitBang800KbpsMethod> //3 chan, esp8266, bb (any pin but 16)
 //UCS8904 RGBW
-#define B_8266_U0_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, NeoEsp8266Uart0Ws2813Method, NeoGammaNullMethod>   //4 chan, esp8266, gpio1
-#define B_8266_U1_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, NeoEsp8266Uart1Ws2813Method, NeoGammaNullMethod>   //4 chan, esp8266, gpio2
-#define B_8266_DM_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, NeoEsp8266Dma800KbpsMethod, NeoGammaNullMethod>    //4 chan, esp8266, gpio3
-#define B_8266_BB_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, NeoEsp8266BitBang800KbpsMethod, NeoGammaNullMethod> //4 chan, esp8266, bb (any pin)
+#define B_8266_U0_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, NeoEsp8266Uart0Ws2813Method>   //4 chan, esp8266, gpio1
+#define B_8266_U1_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, NeoEsp8266Uart1Ws2813Method>   //4 chan, esp8266, gpio2
+#define B_8266_DM_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, NeoEsp8266Dma800KbpsMethod>    //4 chan, esp8266, gpio3
+#define B_8266_BB_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, NeoEsp8266BitBang800KbpsMethod> //4 chan, esp8266, bb (any pin)
 //APA106
-#define B_8266_U0_APA106_3 NeoPixelBusLg<NeoRbgFeature, NeoEsp8266Uart0Apa106Method, NeoGammaNullMethod> //3 chan, esp8266, gpio1
-#define B_8266_U1_APA106_3 NeoPixelBusLg<NeoRbgFeature, NeoEsp8266Uart1Apa106Method, NeoGammaNullMethod> //3 chan, esp8266, gpio2
-#define B_8266_DM_APA106_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266DmaApa106Method, NeoGammaNullMethod>  //3 chan, esp8266, gpio3
-#define B_8266_BB_APA106_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp8266BitBangApa106Method, NeoGammaNullMethod> //3 chan, esp8266, bb (any pin but 16)
+#define B_8266_U0_APA106_3 NeoPixelBus<NeoRbgFeature, NeoEsp8266Uart0Apa106Method> //3 chan, esp8266, gpio1
+#define B_8266_U1_APA106_3 NeoPixelBus<NeoRbgFeature, NeoEsp8266Uart1Apa106Method> //3 chan, esp8266, gpio2
+#define B_8266_DM_APA106_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266DmaApa106Method>  //3 chan, esp8266, gpio3
+#define B_8266_BB_APA106_3 NeoPixelBus<NeoGrbFeature, NeoEsp8266BitBangApa106Method> //3 chan, esp8266, bb (any pin but 16)
 //FW1906 GRBCW
-#define B_8266_U0_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, NeoEsp8266Uart0Ws2813Method, NeoGammaNullMethod>   //esp8266, gpio1
-#define B_8266_U1_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, NeoEsp8266Uart1Ws2813Method, NeoGammaNullMethod>   //esp8266, gpio2
-#define B_8266_DM_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, NeoEsp8266Dma800KbpsMethod, NeoGammaNullMethod>   //esp8266, gpio3
-#define B_8266_BB_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, NeoEsp8266BitBang800KbpsMethod, NeoGammaNullMethod>   //esp8266, bb
+#define B_8266_U0_FW6_5 NeoPixelBus<NeoGrbcwxFeature, NeoEsp8266Uart0Ws2813Method>   //esp8266, gpio1
+#define B_8266_U1_FW6_5 NeoPixelBus<NeoGrbcwxFeature, NeoEsp8266Uart1Ws2813Method>   //esp8266, gpio2
+#define B_8266_DM_FW6_5 NeoPixelBus<NeoGrbcwxFeature, NeoEsp8266Dma800KbpsMethod>   //esp8266, gpio3
+#define B_8266_BB_FW6_5 NeoPixelBus<NeoGrbcwxFeature, NeoEsp8266BitBang800KbpsMethod>   //esp8266, bb
 //WS2805 GRBCW
-#define B_8266_U0_2805_5 NeoPixelBusLg<NeoGrbwwFeature, NeoEsp8266Uart0Ws2805Method, NeoGammaNullMethod> //esp8266, gpio1
-#define B_8266_U1_2805_5 NeoPixelBusLg<NeoGrbwwFeature, NeoEsp8266Uart1Ws2805Method, NeoGammaNullMethod> //esp8266, gpio2
-#define B_8266_DM_2805_5 NeoPixelBusLg<NeoGrbwwFeature, NeoEsp8266DmaWs2805Method, NeoGammaNullMethod> //esp8266, gpio3
-#define B_8266_BB_2805_5 NeoPixelBusLg<NeoGrbwwFeature, NeoEsp8266BitBangWs2805Method, NeoGammaNullMethod> //esp8266, bb
+#define B_8266_U0_2805_5 NeoPixelBus<NeoGrbwwFeature, NeoEsp8266Uart0Ws2805Method> //esp8266, gpio1
+#define B_8266_U1_2805_5 NeoPixelBus<NeoGrbwwFeature, NeoEsp8266Uart1Ws2805Method> //esp8266, gpio2
+#define B_8266_DM_2805_5 NeoPixelBus<NeoGrbwwFeature, NeoEsp8266DmaWs2805Method> //esp8266, gpio3
+#define B_8266_BB_2805_5 NeoPixelBus<NeoGrbwwFeature, NeoEsp8266BitBangWs2805Method> //esp8266, bb
 //TM1914 (RGB)
-#define B_8266_U0_TM1914_3 NeoPixelBusLg<NeoRgbTm1914Feature, NeoEsp8266Uart0Tm1914Method, NeoGammaNullMethod>
-#define B_8266_U1_TM1914_3 NeoPixelBusLg<NeoRgbTm1914Feature, NeoEsp8266Uart1Tm1914Method, NeoGammaNullMethod>
-#define B_8266_DM_TM1914_3 NeoPixelBusLg<NeoRgbTm1914Feature, NeoEsp8266DmaTm1914Method, NeoGammaNullMethod>
-#define B_8266_BB_TM1914_3 NeoPixelBusLg<NeoRgbTm1914Feature, NeoEsp8266BitBangTm1914Method, NeoGammaNullMethod>
+#define B_8266_U0_TM1914_3 NeoPixelBus<NeoRgbTm1914Feature, NeoEsp8266Uart0Tm1914Method>
+#define B_8266_U1_TM1914_3 NeoPixelBus<NeoRgbTm1914Feature, NeoEsp8266Uart1Tm1914Method>
+#define B_8266_DM_TM1914_3 NeoPixelBus<NeoRgbTm1914Feature, NeoEsp8266DmaTm1914Method>
+#define B_8266_BB_TM1914_3 NeoPixelBus<NeoRgbTm1914Feature, NeoEsp8266BitBangTm1914Method>
 //Sm16825 (RGBWC)
-#define B_8266_U0_SM16825_5 NeoPixelBusLg<NeoRgbwcSm16825eFeature, NeoEsp8266Uart0Ws2813Method, NeoGammaNullMethod>
-#define B_8266_U1_SM16825_5 NeoPixelBusLg<NeoRgbwcSm16825eFeature, NeoEsp8266Uart1Ws2813Method, NeoGammaNullMethod>
-#define B_8266_DM_SM16825_5 NeoPixelBusLg<NeoRgbwcSm16825eFeature, NeoEsp8266Dma800KbpsMethod, NeoGammaNullMethod>
-#define B_8266_BB_SM16825_5 NeoPixelBusLg<NeoRgbwcSm16825eFeature, NeoEsp8266BitBangWs2813Method, NeoGammaNullMethod>
+#define B_8266_U0_SM16825_5 NeoPixelBus<NeoRgbwcSm16825eFeature, NeoEsp8266Uart0Ws2813Method>
+#define B_8266_U1_SM16825_5 NeoPixelBus<NeoRgbwcSm16825eFeature, NeoEsp8266Uart1Ws2813Method>
+#define B_8266_DM_SM16825_5 NeoPixelBus<NeoRgbwcSm16825eFeature, NeoEsp8266Dma800KbpsMethod>
+#define B_8266_BB_SM16825_5 NeoPixelBus<NeoRgbwcSm16825eFeature, NeoEsp8266BitBangWs2813Method>
 #endif
 
 /*** ESP32 Neopixel methods ***/
@@ -245,84 +245,84 @@
 #endif
 
 //RGB
-#define B_32_RN_NEO_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp32RmtNWs2812xMethod, NeoGammaNullMethod> // ESP32, S2, S3, C3
-//#define B_32_IN_NEO_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp32I2sNWs2812xMethod, NeoGammaNullMethod> // ESP32 (dynamic I2S selection)
-#define B_32_I2_NEO_3 NeoPixelBusLg<NeoGrbFeature, X1Ws2812xMethod, NeoGammaNullMethod> // ESP32, S2, S3 (automatic I2S selection, see typedef above)
-#define B_32_IP_NEO_3 NeoPixelBusLg<NeoGrbFeature, X8Ws2812xMethod, NeoGammaNullMethod> // parallel I2S (ESP32, S2, S3)
+#define B_32_RN_NEO_3 NeoPixelBus<NeoGrbFeature, NeoEsp32RmtNWs2812xMethod> // ESP32, S2, S3, C3
+//#define B_32_IN_NEO_3 NeoPixelBus<NeoGrbFeature, NeoEsp32I2sNWs2812xMethod> // ESP32 (dynamic I2S selection)
+#define B_32_I2_NEO_3 NeoPixelBus<NeoGrbFeature, X1Ws2812xMethod> // ESP32, S2, S3 (automatic I2S selection, see typedef above)
+#define B_32_IP_NEO_3 NeoPixelBus<NeoGrbFeature, X8Ws2812xMethod> // parallel I2S (ESP32, S2, S3)
 //RGBW
-#define B_32_RN_NEO_4 NeoPixelBusLg<NeoGrbwFeature, NeoEsp32RmtNSk6812Method, NeoGammaNullMethod>
-#define B_32_I2_NEO_4 NeoPixelBusLg<NeoGrbwFeature, X1Sk6812Method, NeoGammaNullMethod>
-#define B_32_IP_NEO_4 NeoPixelBusLg<NeoGrbwFeature, X8Sk6812Method, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_NEO_4 NeoPixelBus<NeoGrbwFeature, NeoEsp32RmtNSk6812Method>
+#define B_32_I2_NEO_4 NeoPixelBus<NeoGrbwFeature, X1Sk6812Method>
+#define B_32_IP_NEO_4 NeoPixelBus<NeoGrbwFeature, X8Sk6812Method> // parallel I2S
 //400Kbps
-#define B_32_RN_400_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp32RmtN400KbpsMethod, NeoGammaNullMethod>
-#define B_32_I2_400_3 NeoPixelBusLg<NeoGrbFeature, X1400KbpsMethod, NeoGammaNullMethod>
-#define B_32_IP_400_3 NeoPixelBusLg<NeoGrbFeature, X8400KbpsMethod, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_400_3 NeoPixelBus<NeoGrbFeature, NeoEsp32RmtN400KbpsMethod>
+#define B_32_I2_400_3 NeoPixelBus<NeoGrbFeature, X1400KbpsMethod>
+#define B_32_IP_400_3 NeoPixelBus<NeoGrbFeature, X8400KbpsMethod> // parallel I2S
 //TM1814 (RGBW)
-#define B_32_RN_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, NeoEsp32RmtNTm1814Method, NeoGammaNullMethod>
-#define B_32_I2_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, X1Tm1814Method, NeoGammaNullMethod>
-#define B_32_IP_TM1_4 NeoPixelBusLg<NeoWrgbTm1814Feature, X8Tm1814Method, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, NeoEsp32RmtNTm1814Method>
+#define B_32_I2_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, X1Tm1814Method>
+#define B_32_IP_TM1_4 NeoPixelBus<NeoWrgbTm1814Feature, X8Tm1814Method> // parallel I2S
 //TM1829 (RGB)
-#define B_32_RN_TM2_3 NeoPixelBusLg<NeoBrgFeature, NeoEsp32RmtNTm1829Method, NeoGammaNullMethod>
-#define B_32_I2_TM2_3 NeoPixelBusLg<NeoBrgFeature, X1Tm1829Method, NeoGammaNullMethod>
-#define B_32_IP_TM2_3 NeoPixelBusLg<NeoBrgFeature, X8Tm1829Method, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_TM2_3 NeoPixelBus<NeoBrgFeature, NeoEsp32RmtNTm1829Method>
+#define B_32_I2_TM2_3 NeoPixelBus<NeoBrgFeature, X1Tm1829Method>
+#define B_32_IP_TM2_3 NeoPixelBus<NeoBrgFeature, X8Tm1829Method> // parallel I2S
 //UCS8903
-#define B_32_RN_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, NeoEsp32RmtNWs2812xMethod, NeoGammaNullMethod>
-#define B_32_I2_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, X1800KbpsMethod, NeoGammaNullMethod>
-#define B_32_IP_UCS_3 NeoPixelBusLg<NeoRgbUcs8903Feature, X8800KbpsMethod, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, NeoEsp32RmtNWs2812xMethod>
+#define B_32_I2_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, X1800KbpsMethod>
+#define B_32_IP_UCS_3 NeoPixelBus<NeoRgbUcs8903Feature, X8800KbpsMethod> // parallel I2S
 //UCS8904
-#define B_32_RN_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, NeoEsp32RmtNWs2812xMethod, NeoGammaNullMethod>
-#define B_32_I2_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, X1800KbpsMethod, NeoGammaNullMethod>
-#define B_32_IP_UCS_4 NeoPixelBusLg<NeoRgbwUcs8904Feature, X8800KbpsMethod, NeoGammaNullMethod>// parallel I2S
+#define B_32_RN_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, NeoEsp32RmtNWs2812xMethod>
+#define B_32_I2_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, X1800KbpsMethod>
+#define B_32_IP_UCS_4 NeoPixelBus<NeoRgbwUcs8904Feature, X8800KbpsMethod>// parallel I2S
 //APA106
-#define B_32_RN_APA106_3 NeoPixelBusLg<NeoGrbFeature, NeoEsp32RmtNApa106Method, NeoGammaNullMethod>
-#define B_32_I2_APA106_3 NeoPixelBusLg<NeoGrbFeature, X1Apa106Method, NeoGammaNullMethod>
-#define B_32_IP_APA106_3 NeoPixelBusLg<NeoGrbFeature, X8Apa106Method, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_APA106_3 NeoPixelBus<NeoGrbFeature, NeoEsp32RmtNApa106Method>
+#define B_32_I2_APA106_3 NeoPixelBus<NeoGrbFeature, X1Apa106Method>
+#define B_32_IP_APA106_3 NeoPixelBus<NeoGrbFeature, X8Apa106Method> // parallel I2S
 //FW1906 GRBCW
-#define B_32_RN_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, NeoEsp32RmtNWs2812xMethod, NeoGammaNullMethod>
-#define B_32_I2_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, X1800KbpsMethod, NeoGammaNullMethod>
-#define B_32_IP_FW6_5 NeoPixelBusLg<NeoGrbcwxFeature, X8800KbpsMethod, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_FW6_5 NeoPixelBus<NeoGrbcwxFeature, NeoEsp32RmtNWs2812xMethod>
+#define B_32_I2_FW6_5 NeoPixelBus<NeoGrbcwxFeature, X1800KbpsMethod>
+#define B_32_IP_FW6_5 NeoPixelBus<NeoGrbcwxFeature, X8800KbpsMethod> // parallel I2S
 //WS2805 RGBWC
-#define B_32_RN_2805_5 NeoPixelBusLg<NeoGrbwwFeature, NeoEsp32RmtNWs2805Method, NeoGammaNullMethod>
-#define B_32_I2_2805_5 NeoPixelBusLg<NeoGrbwwFeature, X1Ws2805Method, NeoGammaNullMethod>
-#define B_32_IP_2805_5 NeoPixelBusLg<NeoGrbwwFeature, X8Ws2805Method, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_2805_5 NeoPixelBus<NeoGrbwwFeature, NeoEsp32RmtNWs2805Method>
+#define B_32_I2_2805_5 NeoPixelBus<NeoGrbwwFeature, X1Ws2805Method>
+#define B_32_IP_2805_5 NeoPixelBus<NeoGrbwwFeature, X8Ws2805Method> // parallel I2S
 //TM1914 (RGB)
-#define B_32_RN_TM1914_3 NeoPixelBusLg<NeoGrbTm1914Feature, NeoEsp32RmtNTm1914Method, NeoGammaNullMethod>
-#define B_32_I2_TM1914_3 NeoPixelBusLg<NeoGrbTm1914Feature, X1Tm1914Method, NeoGammaNullMethod>
-#define B_32_IP_TM1914_3 NeoPixelBusLg<NeoGrbTm1914Feature, X8Tm1914Method, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_TM1914_3 NeoPixelBus<NeoGrbTm1914Feature, NeoEsp32RmtNTm1914Method>
+#define B_32_I2_TM1914_3 NeoPixelBus<NeoGrbTm1914Feature, X1Tm1914Method>
+#define B_32_IP_TM1914_3 NeoPixelBus<NeoGrbTm1914Feature, X8Tm1914Method> // parallel I2S
 //Sm16825 (RGBWC)
-#define B_32_RN_SM16825_5 NeoPixelBusLg<NeoRgbcwSm16825eFeature, NeoEsp32RmtNWs2812xMethod, NeoGammaNullMethod>
-#define B_32_I2_SM16825_5 NeoPixelBusLg<NeoRgbcwSm16825eFeature, X1Ws2812xMethod, NeoGammaNullMethod>
-#define B_32_IP_SM16825_5 NeoPixelBusLg<NeoRgbcwSm16825eFeature, X8Ws2812xMethod, NeoGammaNullMethod> // parallel I2S
+#define B_32_RN_SM16825_5 NeoPixelBus<NeoRgbcwSm16825eFeature, NeoEsp32RmtNWs2812xMethod>
+#define B_32_I2_SM16825_5 NeoPixelBus<NeoRgbcwSm16825eFeature, X1Ws2812xMethod>
+#define B_32_IP_SM16825_5 NeoPixelBus<NeoRgbcwSm16825eFeature, X8Ws2812xMethod> // parallel I2S
 #endif
 
 //APA102
 #ifdef WLED_USE_ETHERNET
 // fix for #2542 (by @BlackBird77)
-#define B_HS_DOT_3 NeoPixelBusLg<DotStarBgrFeature, DotStarEsp32HspiHzMethod, NeoGammaNullMethod> //hardware HSPI (was DotStarEsp32DmaHspi5MhzMethod in NPB @ 2.6.9)
+#define B_HS_DOT_3 NeoPixelBus<DotStarBgrFeature, DotStarEsp32HspiHzMethod> //hardware HSPI (was DotStarEsp32DmaHspi5MhzMethod in NPB @ 2.6.9)
 #else
-#define B_HS_DOT_3 NeoPixelBusLg<DotStarBgrFeature, DotStarSpiHzMethod, NeoGammaNullMethod> //hardware VSPI
+#define B_HS_DOT_3 NeoPixelBus<DotStarBgrFeature, DotStarSpiHzMethod> //hardware VSPI
 #endif
-#define B_SS_DOT_3 NeoPixelBusLg<DotStarBgrFeature, DotStarMethod, NeoGammaNullMethod>    //soft SPI
+#define B_SS_DOT_3 NeoPixelBus<DotStarBgrFeature, DotStarMethod>    //soft SPI
 
 //LPD8806
-#define B_HS_LPD_3 NeoPixelBusLg<Lpd8806GrbFeature, Lpd8806SpiHzMethod, NeoGammaNullMethod>
-#define B_SS_LPD_3 NeoPixelBusLg<Lpd8806GrbFeature, Lpd8806Method, NeoGammaNullMethod>
+#define B_HS_LPD_3 NeoPixelBus<Lpd8806GrbFeature, Lpd8806SpiHzMethod>
+#define B_SS_LPD_3 NeoPixelBus<Lpd8806GrbFeature, Lpd8806Method>
 
 //LPD6803
-#define B_HS_LPO_3 NeoPixelBusLg<Lpd6803GrbFeature, Lpd6803SpiHzMethod, NeoGammaNullMethod>
-#define B_SS_LPO_3 NeoPixelBusLg<Lpd6803GrbFeature, Lpd6803Method, NeoGammaNullMethod>
+#define B_HS_LPO_3 NeoPixelBus<Lpd6803GrbFeature, Lpd6803SpiHzMethod>
+#define B_SS_LPO_3 NeoPixelBus<Lpd6803GrbFeature, Lpd6803Method>
 
 //WS2801
 #ifdef WLED_USE_ETHERNET
-#define B_HS_WS1_3 NeoPixelBusLg<NeoRbgFeature, Ws2801MethodBase<TwoWireHspiImple<SpiSpeedHz>>, NeoGammaNullMethod>
+#define B_HS_WS1_3 NeoPixelBus<NeoRbgFeature, Ws2801MethodBase<TwoWireHspiImple<SpiSpeedHz>>>
 #else
-#define B_HS_WS1_3 NeoPixelBusLg<NeoRbgFeature, Ws2801SpiHzMethod, NeoGammaNullMethod>
+#define B_HS_WS1_3 NeoPixelBus<NeoRbgFeature, Ws2801SpiHzMethod>
 #endif
-#define B_SS_WS1_3 NeoPixelBusLg<NeoRbgFeature, Ws2801Method, NeoGammaNullMethod>
+#define B_SS_WS1_3 NeoPixelBus<NeoRbgFeature, Ws2801Method>
 
 //P9813
-#define B_HS_P98_3 NeoPixelBusLg<P9813BgrFeature, P9813SpiHzMethod, NeoGammaNullMethod>
-#define B_SS_P98_3 NeoPixelBusLg<P9813BgrFeature, P9813Method, NeoGammaNullMethod>
+#define B_HS_P98_3 NeoPixelBus<P9813BgrFeature, P9813SpiHzMethod>
+#define B_SS_P98_3 NeoPixelBus<P9813BgrFeature, P9813Method>
 
 // 48bit & 64bit to 24bit & 32bit RGB(W) conversion
 #define toRGBW32(c) (RGBW32((c>>40)&0xFF, (c>>24)&0xFF, (c>>8)&0xFF, (c>>56)&0xFF))
@@ -896,102 +896,6 @@ class PolyBus {
     }
   }
 
-  static void setBrightness(void* busPtr, uint8_t busType, uint8_t b) {
-    switch (busType) {
-      case I_NONE: break;
-    #ifdef ESP8266
-      case I_8266_U0_NEO_3: (static_cast<B_8266_U0_NEO_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_NEO_3: (static_cast<B_8266_U1_NEO_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_NEO_3: (static_cast<B_8266_DM_NEO_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_NEO_3: (static_cast<B_8266_BB_NEO_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_NEO_4: (static_cast<B_8266_U0_NEO_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_NEO_4: (static_cast<B_8266_U1_NEO_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_NEO_4: (static_cast<B_8266_DM_NEO_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_NEO_4: (static_cast<B_8266_BB_NEO_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_400_3: (static_cast<B_8266_U0_400_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_400_3: (static_cast<B_8266_U1_400_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_400_3: (static_cast<B_8266_DM_400_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_400_3: (static_cast<B_8266_BB_400_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_TM1_4: (static_cast<B_8266_U0_TM1_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_TM1_4: (static_cast<B_8266_U1_TM1_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_TM1_4: (static_cast<B_8266_DM_TM1_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_TM1_4: (static_cast<B_8266_BB_TM1_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_TM2_3: (static_cast<B_8266_U0_TM2_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_TM2_3: (static_cast<B_8266_U1_TM2_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_TM2_3: (static_cast<B_8266_DM_TM2_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_TM2_3: (static_cast<B_8266_BB_TM2_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_UCS_3: (static_cast<B_8266_U0_UCS_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_UCS_3: (static_cast<B_8266_U1_UCS_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_UCS_3: (static_cast<B_8266_DM_UCS_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_UCS_3: (static_cast<B_8266_BB_UCS_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_UCS_4: (static_cast<B_8266_U0_UCS_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_UCS_4: (static_cast<B_8266_U1_UCS_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_UCS_4: (static_cast<B_8266_DM_UCS_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_UCS_4: (static_cast<B_8266_BB_UCS_4*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_APA106_3: (static_cast<B_8266_U0_APA106_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_APA106_3: (static_cast<B_8266_U1_APA106_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_APA106_3: (static_cast<B_8266_DM_APA106_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_APA106_3: (static_cast<B_8266_BB_APA106_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_FW6_5: (static_cast<B_8266_U0_FW6_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_FW6_5: (static_cast<B_8266_U1_FW6_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_FW6_5: (static_cast<B_8266_DM_FW6_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_FW6_5: (static_cast<B_8266_BB_FW6_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_2805_5: (static_cast<B_8266_U0_2805_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_2805_5: (static_cast<B_8266_U1_2805_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_2805_5: (static_cast<B_8266_DM_2805_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_2805_5: (static_cast<B_8266_BB_2805_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_TM1914_3: (static_cast<B_8266_U0_TM1914_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_TM1914_3: (static_cast<B_8266_U1_TM1914_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_TM1914_3: (static_cast<B_8266_DM_TM1914_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_TM1914_3: (static_cast<B_8266_BB_TM1914_3*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U0_SM16825_5: (static_cast<B_8266_U0_SM16825_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_U1_SM16825_5: (static_cast<B_8266_U1_SM16825_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_DM_SM16825_5: (static_cast<B_8266_DM_SM16825_5*>(busPtr))->SetLuminance(b); break;
-      case I_8266_BB_SM16825_5: (static_cast<B_8266_BB_SM16825_5*>(busPtr))->SetLuminance(b); break;
-    #endif
-    #ifdef ARDUINO_ARCH_ESP32
-      // RMT buses
-      case I_32_RN_NEO_3: (static_cast<B_32_RN_NEO_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_NEO_4: (static_cast<B_32_RN_NEO_4*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_400_3: (static_cast<B_32_RN_400_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_TM1_4: (static_cast<B_32_RN_TM1_4*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_TM2_3: (static_cast<B_32_RN_TM2_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_UCS_3: (static_cast<B_32_RN_UCS_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_UCS_4: (static_cast<B_32_RN_UCS_4*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_APA106_3: (static_cast<B_32_RN_APA106_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_FW6_5: (static_cast<B_32_RN_FW6_5*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_2805_5: (static_cast<B_32_RN_2805_5*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_TM1914_3: (static_cast<B_32_RN_TM1914_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_RN_SM16825_5: (static_cast<B_32_RN_SM16825_5*>(busPtr))->SetLuminance(b); break;
-      // I2S1 bus or paralell buses
-      #ifndef CONFIG_IDF_TARGET_ESP32C3
-      case I_32_I2_NEO_3: if (_useParallelI2S) (static_cast<B_32_IP_NEO_3*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_NEO_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_NEO_4: if (_useParallelI2S) (static_cast<B_32_IP_NEO_4*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_NEO_4*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_400_3: if (_useParallelI2S) (static_cast<B_32_IP_400_3*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_400_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_TM1_4: if (_useParallelI2S) (static_cast<B_32_IP_TM1_4*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_TM1_4*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_TM2_3: if (_useParallelI2S) (static_cast<B_32_IP_TM2_3*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_TM2_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_UCS_3: if (_useParallelI2S) (static_cast<B_32_IP_UCS_3*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_UCS_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_UCS_4: if (_useParallelI2S) (static_cast<B_32_IP_UCS_4*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_UCS_4*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_APA106_3: if (_useParallelI2S) (static_cast<B_32_IP_APA106_3*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_APA106_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_FW6_5: if (_useParallelI2S) (static_cast<B_32_IP_FW6_5*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_FW6_5*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_2805_5: if (_useParallelI2S) (static_cast<B_32_IP_2805_5*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_2805_5*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_TM1914_3: if (_useParallelI2S) (static_cast<B_32_IP_TM1914_3*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_TM1914_3*>(busPtr))->SetLuminance(b); break;
-      case I_32_I2_SM16825_5: if (_useParallelI2S) (static_cast<B_32_IP_SM16825_5*>(busPtr))->SetLuminance(b); else (static_cast<B_32_I2_SM16825_5*>(busPtr))->SetLuminance(b); break;
-      #endif
-    #endif
-      case I_HS_DOT_3: (static_cast<B_HS_DOT_3*>(busPtr))->SetLuminance(b); break;
-      case I_SS_DOT_3: (static_cast<B_SS_DOT_3*>(busPtr))->SetLuminance(b); break;
-      case I_HS_LPD_3: (static_cast<B_HS_LPD_3*>(busPtr))->SetLuminance(b); break;
-      case I_SS_LPD_3: (static_cast<B_SS_LPD_3*>(busPtr))->SetLuminance(b); break;
-      case I_HS_LPO_3: (static_cast<B_HS_LPO_3*>(busPtr))->SetLuminance(b); break;
-      case I_SS_LPO_3: (static_cast<B_SS_LPO_3*>(busPtr))->SetLuminance(b); break;
-      case I_HS_WS1_3: (static_cast<B_HS_WS1_3*>(busPtr))->SetLuminance(b); break;
-      case I_SS_WS1_3: (static_cast<B_SS_WS1_3*>(busPtr))->SetLuminance(b); break;
-      case I_HS_P98_3: (static_cast<B_HS_P98_3*>(busPtr))->SetLuminance(b); break;
-      case I_SS_P98_3: (static_cast<B_SS_P98_3*>(busPtr))->SetLuminance(b); break;
-    }
-  }
-
   [[gnu::hot]] static uint32_t getPixelColor(void* busPtr, uint8_t busType, uint16_t pix, uint8_t co) {
     RgbwColor col(0,0,0,0);
     switch (busType) {

wled00/cfg.cpp

@@ -235,7 +235,8 @@ bool deserializeConfig(JsonObject doc, bool fromFS) {
       }
       ledType |= refresh << 7; // hack bit 7 to indicate strip requires off refresh
 
-      busConfigs.emplace_back(ledType, pins, start, length, colorOrder, reversed, skipFirst, AWmode, freqkHz, maPerLed, maMax);
+      String host = elm[F("text")] | String();
+      busConfigs.emplace_back(ledType, pins, start, length, colorOrder, reversed, skipFirst, AWmode, freqkHz, maPerLed, maMax, host);
       doInitBusses = true;  // finalization done in beginStrip()
       if (!Bus::isVirtual(ledType)) s++; // have as many virtual buses as you want
     }
@@ -379,7 +380,7 @@ bool deserializeConfig(JsonObject doc, bool fromFS) {
             DEBUG_PRINTF_P(PSTR("PIN ALLOC error: GPIO%d for touch button #%d is not a touch pin!\n"), btnPin[s], s);
             btnPin[s] = -1;
             PinManager::deallocatePin(pin,PinOwner::Button);
-          }          
+          }
           //if touch pin, enable the touch interrupt on ESP32 S2 & S3
           #ifdef SOC_TOUCH_VERSION_2    // ESP32 S2 and S3 have a function to check touch state but need to attach an interrupt to do so
           else
@@ -518,7 +519,6 @@ bool deserializeConfig(JsonObject doc, bool fromFS) {
   CJSON(briMultiplier, light[F("scale-bri")]);
   CJSON(paletteBlend, light[F("pal-mode")]);
   CJSON(strip.autoSegments, light[F("aseg")]);
-  CJSON(useRainbowWheel, light[F("rw")]);
 
   CJSON(gammaCorrectVal, light["gc"]["val"]); // default 2.2
   float light_gc_bri = light["gc"]["bri"];
@@ -772,9 +772,32 @@ bool deserializeConfig(JsonObject doc, bool fromFS) {
   return (doc["sv"] | true);
 }
 
-
 static const char s_cfg_json[] PROGMEM = "/cfg.json";
 
+bool backupConfig() {
+  return backupFile(s_cfg_json);
+}
+
+bool restoreConfig() {
+  return restoreFile(s_cfg_json);
+}
+
+bool verifyConfig() {
+  return validateJsonFile(s_cfg_json);
+}
+
+// rename config file and reboot
+// if the cfg file doesn't exist, such as after a reset, do nothing
+void resetConfig() {
+  if (WLED_FS.exists(s_cfg_json)) {
+    DEBUG_PRINTLN(F("Reset config"));
+    char backupname[32];
+    snprintf_P(backupname, sizeof(backupname), PSTR("/rst.%s"), &s_cfg_json[1]);
+    WLED_FS.rename(s_cfg_json, backupname);
+    doReboot = true;
+  }
+}
+
 bool deserializeConfigFromFS() {
   [[maybe_unused]] bool success = deserializeConfigSec();
   #ifdef WLED_ADD_EEPROM_SUPPORT
@@ -800,6 +823,7 @@ bool deserializeConfigFromFS() {
 
 void serializeConfigToFS() {
   serializeConfigSec();
+  backupConfig(); // backup before writing new config
 
   DEBUG_PRINTLN(F("Writing settings to /cfg.json..."));
 
@@ -976,6 +1000,7 @@ void serializeConfig(JsonObject root) {
     ins[F("freq")]   = bus->getFrequency();
     ins[F("maxpwr")] = bus->getMaxCurrent();
     ins[F("ledma")]  = bus->getLEDCurrent();
+    ins[F("text")]   = bus->getCustomText();
   }
 
   JsonArray hw_com = hw.createNestedArray(F("com"));
@@ -1040,7 +1065,6 @@ void serializeConfig(JsonObject root) {
   light[F("scale-bri")] = briMultiplier;
   light[F("pal-mode")] = paletteBlend;
   light[F("aseg")] = strip.autoSegments;
-  light[F("rw")] = useRainbowWheel;
 
   JsonObject light_gc = light.createNestedObject("gc");
   light_gc["bri"] = (gammaCorrectBri) ? gammaCorrectVal : 1.0f;  // keep compatibility
@@ -1340,4 +1364,4 @@ void serializeConfigSec() {
   if (f) serializeJson(root, f);
   f.close();
   releaseJSONBufferLock();
-}
+}

wled00/colors.cpp

@@ -8,7 +8,7 @@
  * color blend function, based on FastLED blend function
  * the calculation for each color is: result = (A*(amountOfA) + A + B*(amountOfB) + B) / 256 with amountOfA = 255 - amountOfB
  */
-uint32_t color_blend(uint32_t color1, uint32_t color2, uint8_t blend) {
+uint32_t IRAM_ATTR color_blend(uint32_t color1, uint32_t color2, uint8_t blend) {
   // min / max blend checking is omitted: calls with 0 or 255 are rare, checking lowers overall performance
   const uint32_t TWO_CHANNEL_MASK = 0x00FF00FF;     // mask for R and B channels or W and G if negated (poorman's SIMD; https://github.com/wled/WLED/pull/4568#discussion_r1986587221)
   uint32_t rb1 =  color1       & TWO_CHANNEL_MASK;  // extract R & B channels from color1
@@ -64,26 +64,26 @@ uint32_t color_add(uint32_t c1, uint32_t c2, bool preserveCR)
  * fades color toward black
  * if using "video" method the resulting color will never become black unless it is already black
  */
-
-uint32_t color_fade(uint32_t c1, uint8_t amount, bool video)
-{
+uint32_t IRAM_ATTR color_fade(uint32_t c1, uint8_t amount, bool video) {
+  if (c1 == 0 || amount == 0) return 0; // black or no change
   if (amount == 255) return c1;
-  if (c1 == BLACK || amount == 0) return BLACK;
-  uint32_t scaledcolor; // color order is: W R G B from MSB to LSB
-  uint32_t scale = amount; // 32bit for faster calculation
   uint32_t addRemains = 0;
-  if (!video) scale++; // add one for correct scaling using bitshifts
-  else { // video scaling: make sure colors do not dim to zero if they started non-zero
-    addRemains  = R(c1) ? 0x00010000 : 0;
-    addRemains |= G(c1) ? 0x00000100 : 0;
-    addRemains |= B(c1) ? 0x00000001 : 0;
-    addRemains |= W(c1) ? 0x01000000 : 0;
+
+  if (!video) amount++; // add one for correct scaling using bitshifts
+  else {
+    // video scaling: make sure colors do not dim to zero if they started non-zero unless they distort the hue
+    uint8_t r = byte(c1>>16), g = byte(c1>>8), b = byte(c1), w = byte(c1>>24); // extract r, g, b, w channels
+    uint8_t maxc = (r > g) ? ((r > b) ? r : b) : ((g > b) ? g : b); // determine dominant channel for hue preservation
+    uint8_t quarterMax = maxc >> 2; // note: using half of max results in color artefacts
+    addRemains  = r && r > quarterMax ? 0x00010000 : 0;
+    addRemains |= g && g > quarterMax ? 0x00000100 : 0;
+    addRemains |= b && b > quarterMax ? 0x00000001 : 0;
+    addRemains |= w ? 0x01000000 : 0;
   }
   const uint32_t TWO_CHANNEL_MASK = 0x00FF00FF;
-  uint32_t rb = (((c1 & TWO_CHANNEL_MASK) * scale) >> 8) &  TWO_CHANNEL_MASK; // scale red and blue
-  uint32_t wg = (((c1 >> 8) & TWO_CHANNEL_MASK) * scale) & ~TWO_CHANNEL_MASK; // scale white and green
-  scaledcolor = (rb | wg) + addRemains;
-  return scaledcolor;
+  uint32_t rb = (((c1 & TWO_CHANNEL_MASK) * amount) >> 8) &  TWO_CHANNEL_MASK; // scale red and blue
+  uint32_t wg = (((c1 >> 8) & TWO_CHANNEL_MASK) * amount) & ~TWO_CHANNEL_MASK; // scale white and green
+  return (rb | wg) + addRemains;
 }
 
 /*
@@ -92,7 +92,7 @@ uint32_t color_fade(uint32_t c1, uint8_t amount, bool video)
    note: inputs are 32bit to speed up the function, useful input value ranges are 0-255
  */
 uint32_t adjust_color(uint32_t rgb, uint32_t hueShift, uint32_t lighten, uint32_t brighten) {
-    if(rgb == 0 | hueShift + lighten + brighten == 0) return rgb; // black or no change
+    if (rgb == 0 | hueShift + lighten + brighten == 0) return rgb; // black or no change
     CHSV32 hsv;
     rgb2hsv(rgb, hsv); //convert to HSV
     hsv.h += (hueShift << 8); // shift hue (hue is 16 bits)
@@ -104,8 +104,7 @@ uint32_t adjust_color(uint32_t rgb, uint32_t hueShift, uint32_t lighten, uint32_
 }
 
 // 1:1 replacement of fastled function optimized for ESP, slightly faster, more accurate and uses less flash (~ -200bytes)
-uint32_t ColorFromPaletteWLED(const CRGBPalette16& pal, unsigned index, uint8_t brightness, TBlendType blendType)
-{
+uint32_t ColorFromPaletteWLED(const CRGBPalette16& pal, unsigned index, uint8_t brightness, TBlendType blendType) {
   if (blendType == LINEARBLEND_NOWRAP) {
     index = (index * 0xF0) >> 8; // Blend range is affected by lo4 blend of values, remap to avoid wrapping
   }
@@ -120,16 +119,16 @@ uint32_t ColorFromPaletteWLED(const CRGBPalette16& pal, unsigned index, uint8_t
     else ++entry;
     unsigned f2 = (lo4 << 4);
     unsigned f1 = 256 - f2;
-    red1   = (red1 * f1 + (unsigned)entry->r * f2) >> 8; // note: using color_blend() is 20% slower
+    red1   = (red1   * f1 + (unsigned)entry->r * f2) >> 8; // note: using color_blend() is slower
     green1 = (green1 * f1 + (unsigned)entry->g * f2) >> 8;
-    blue1  = (blue1 * f1 + (unsigned)entry->b * f2) >> 8;
+    blue1  = (blue1  * f1 + (unsigned)entry->b * f2) >> 8;
   }
   if (brightness < 255) { // note: zero checking could be done to return black but that is hardly ever used so it is omitted
-    // actually color_fade(c1, brightness)
+    // actually same as color_fade(), using color_fade() is slower
     uint32_t scale = brightness + 1; // adjust for rounding (bitshift)
-    red1   = (red1 * scale) >> 8; // note: using color_fade() is 30% slower
+    red1   = (red1   * scale) >> 8;
     green1 = (green1 * scale) >> 8;
-    blue1  = (blue1 * scale) >> 8;
+    blue1  = (blue1  * scale) >> 8;
   }
   return RGBW32(red1,green1,blue1,0);
 }
@@ -589,10 +588,13 @@ uint8_t NeoGammaWLEDMethod::gammaT_inv[256];
 void NeoGammaWLEDMethod::calcGammaTable(float gamma)
 {
   float gamma_inv = 1.0f / gamma; // inverse gamma
-  for (size_t i = 0; i < 256; i++) {
+  for (size_t i = 1; i < 256; i++) {
     gammaT[i] = (int)(powf((float)i / 255.0f, gamma) * 255.0f + 0.5f);
-    gammaT_inv[i] = (int)(powf((float)i / 255.0f, gamma_inv) * 255.0f + 0.5f);
+    gammaT_inv[i] = (int)(powf(((float)i - 0.5f) / 255.0f, gamma_inv) * 255.0f + 0.5f);
+    //DEBUG_PRINTF_P(PSTR("gammaT[%d] = %d gammaT_inv[%d] = %d\n"), i, gammaT[i], i, gammaT_inv[i]);
   }
+  gammaT[0] = 0;
+  gammaT_inv[0] = 0;
 }
 
 uint8_t IRAM_ATTR_YN NeoGammaWLEDMethod::Correct(uint8_t value)
@@ -601,21 +603,6 @@ uint8_t IRAM_ATTR_YN NeoGammaWLEDMethod::Correct(uint8_t value)
   return gammaT[value];
 }
 
-// used for color gamma correction
-uint32_t IRAM_ATTR_YN NeoGammaWLEDMethod::Correct32(uint32_t color)
-{
-  if (!gammaCorrectCol) return color;
-  uint8_t w = W(color);
-  uint8_t r = R(color);
-  uint8_t g = G(color);
-  uint8_t b = B(color);
-  w = gammaT[w];
-  r = gammaT[r];
-  g = gammaT[g];
-  b = gammaT[b];
-  return RGBW32(r, g, b, w);
-}
-
 uint32_t IRAM_ATTR_YN NeoGammaWLEDMethod::inverseGamma32(uint32_t color)
 {
   if (!gammaCorrectCol) return color;

wled00/colors.h

@@ -0,0 +1,144 @@
+#pragma once
+#ifndef WLED_COLORS_H
+#define WLED_COLORS_H
+
+/*
+ * Color structs and color utility functions
+ */
+#include <vector>
+#include "FastLED.h"
+
+#define ColorFromPalette ColorFromPaletteWLED // override fastled version
+
+// CRGBW can be used to manipulate 32bit colors faster. However: if it is passed to functions, it adds overhead compared to a uint32_t color
+// use with caution and pay attention to flash size. Usually converting a uint32_t to CRGBW to extract r, g, b, w values is slower than using bitshifts
+// it can be useful to avoid back and forth conversions between uint32_t and fastled CRGB
+struct CRGBW {
+    union {
+        uint32_t color32; // Access as a 32-bit value (0xWWRRGGBB)
+        struct {
+            uint8_t b;
+            uint8_t g;
+            uint8_t r;
+            uint8_t w;
+        };
+        uint8_t raw[4];   // Access as an array in the order B, G, R, W
+    };
+
+    // Default constructor
+    inline CRGBW() __attribute__((always_inline)) = default;
+
+    // Constructor from a 32-bit color (0xWWRRGGBB)
+    constexpr CRGBW(uint32_t color) __attribute__((always_inline)) : color32(color) {}
+
+    // Constructor with r, g, b, w values
+    constexpr CRGBW(uint8_t red, uint8_t green, uint8_t blue, uint8_t white = 0) __attribute__((always_inline)) : b(blue), g(green), r(red), w(white) {}
+
+    // Constructor from CRGB
+    constexpr CRGBW(CRGB rgb) __attribute__((always_inline)) : b(rgb.b), g(rgb.g), r(rgb.r), w(0) {}
+
+    // Access as an array
+    inline const uint8_t& operator[] (uint8_t x) const __attribute__((always_inline)) { return raw[x]; }
+
+    // Assignment from 32-bit color
+    inline CRGBW& operator=(uint32_t color) __attribute__((always_inline)) { color32 = color; return *this; }
+
+    // Assignment from r, g, b, w
+    inline CRGBW& operator=(const CRGB& rgb) __attribute__((always_inline)) { b = rgb.b; g = rgb.g; r = rgb.r; w = 0; return *this; }
+
+    // Conversion operator to uint32_t
+    inline operator uint32_t() const __attribute__((always_inline)) {
+      return color32;
+    }
+    /*
+    // Conversion operator to CRGB
+    inline operator CRGB() const __attribute__((always_inline)) {
+      return CRGB(r, g, b);
+    }
+
+    CRGBW& scale32 (uint8_t scaledown) // 32bit math
+    {
+      if (color32 == 0) return *this; // 2 extra instructions, worth it if called a lot on black (which probably is true) adding check if scaledown is zero adds much more overhead as its 8bit
+      uint32_t scale = scaledown + 1;
+      uint32_t rb = (((color32 & 0x00FF00FF) * scale) >> 8) & 0x00FF00FF; // scale red and blue
+      uint32_t wg = (((color32 & 0xFF00FF00) >> 8) * scale) & 0xFF00FF00; // scale white and green
+          color32 =  rb | wg;
+      return *this;
+    }*/
+
+};
+
+struct CHSV32 { // 32bit HSV color with 16bit hue for more accurate conversions
+  union {
+    struct {
+        uint16_t h;  // hue
+        uint8_t s;   // saturation
+        uint8_t v;   // value
+    };
+    uint32_t raw;    // 32bit access
+  };
+  inline CHSV32() __attribute__((always_inline)) = default; // default constructor
+
+    /// Allow construction from hue, saturation, and value
+    /// @param ih input hue
+    /// @param is input saturation
+    /// @param iv input value
+  inline CHSV32(uint16_t ih, uint8_t is, uint8_t iv) __attribute__((always_inline)) // constructor from 16bit h, s, v
+        : h(ih), s(is), v(iv) {}
+  inline CHSV32(uint8_t ih, uint8_t is, uint8_t iv) __attribute__((always_inline)) // constructor from 8bit h, s, v
+        : h((uint16_t)ih << 8), s(is), v(iv) {}
+  inline CHSV32(const CHSV& chsv) __attribute__((always_inline))  // constructor from CHSV
+    : h((uint16_t)chsv.h << 8), s(chsv.s), v(chsv.v) {}
+  inline operator CHSV() const { return CHSV((uint8_t)(h >> 8), s, v); } // typecast to CHSV
+};
+extern bool gammaCorrectCol;
+// similar to NeoPixelBus NeoGammaTableMethod but allows dynamic changes (superseded by NPB::NeoGammaDynamicTableMethod)
+class NeoGammaWLEDMethod {
+  public:
+    [[gnu::hot]] static uint8_t Correct(uint8_t value);             // apply Gamma to single channel
+    [[gnu::hot]] static uint32_t inverseGamma32(uint32_t color);    // apply inverse Gamma to RGBW32 color
+    static void calcGammaTable(float gamma);                        // re-calculates & fills gamma tables
+    static inline uint8_t rawGamma8(uint8_t val) { return gammaT[val]; }  // get value from Gamma table (WLED specific, not used by NPB)
+    static inline uint8_t rawInverseGamma8(uint8_t val) { return gammaT_inv[val]; }  // get value from inverse Gamma table (WLED specific, not used by NPB)
+    static inline uint32_t Correct32(uint32_t color) { // apply Gamma to RGBW32 color (WLED specific, not used by NPB)
+      if (!gammaCorrectCol) return color; // no gamma correction
+      uint8_t  w = byte(color>>24), r = byte(color>>16), g = byte(color>>8), b = byte(color); // extract r, g, b, w channels
+      w = gammaT[w]; r = gammaT[r]; g = gammaT[g]; b = gammaT[b];
+      return (uint32_t(w) << 24) | (uint32_t(r) << 16) | (uint32_t(g) << 8) | uint32_t(b);
+    }
+  private:
+    static uint8_t gammaT[];
+    static uint8_t gammaT_inv[];
+};
+#define gamma32(c) NeoGammaWLEDMethod::Correct32(c)
+#define gamma8(c)  NeoGammaWLEDMethod::rawGamma8(c)
+#define gamma32inv(c) NeoGammaWLEDMethod::inverseGamma32(c)
+#define gamma8inv(c)  NeoGammaWLEDMethod::rawInverseGamma8(c)
+[[gnu::hot, gnu::pure]] uint32_t color_blend(uint32_t c1, uint32_t c2 , uint8_t blend);
+inline uint32_t color_blend16(uint32_t c1, uint32_t c2, uint16_t b) { return color_blend(c1, c2, b >> 8); };
+[[gnu::hot, gnu::pure]] uint32_t color_add(uint32_t, uint32_t, bool preserveCR = false);
+[[gnu::hot, gnu::pure]] uint32_t adjust_color(uint32_t rgb, uint32_t hueShift, uint32_t lighten, uint32_t brighten);
+[[gnu::hot, gnu::pure]] uint32_t ColorFromPaletteWLED(const CRGBPalette16 &pal, unsigned index, uint8_t brightness = (uint8_t)255U, TBlendType blendType = LINEARBLEND);
+CRGBPalette16 generateHarmonicRandomPalette(const CRGBPalette16 &basepalette);
+CRGBPalette16 generateRandomPalette();
+void loadCustomPalettes();
+extern std::vector<CRGBPalette16> customPalettes;
+inline size_t getPaletteCount() { return 13 + GRADIENT_PALETTE_COUNT + customPalettes.size(); }
+inline uint32_t colorFromRgbw(byte* rgbw) { return uint32_t((byte(rgbw[3]) << 24) | (byte(rgbw[0]) << 16) | (byte(rgbw[1]) << 8) | (byte(rgbw[2]))); }
+void hsv2rgb(const CHSV32& hsv, uint32_t& rgb);
+void colorHStoRGB(uint16_t hue, byte sat, byte* rgb);
+void rgb2hsv(const uint32_t rgb, CHSV32& hsv);
+inline CHSV rgb2hsv(const CRGB c) { CHSV32 hsv; rgb2hsv((uint32_t((byte(c.r) << 16) | (byte(c.g) << 8) | (byte(c.b)))), hsv); return CHSV(hsv); } // CRGB to hsv
+void colorKtoRGB(uint16_t kelvin, byte* rgb);
+void colorCTtoRGB(uint16_t mired, byte* rgb); //white spectrum to rgb
+void colorXYtoRGB(float x, float y, byte* rgb); // only defined if huesync disabled TODO
+void colorRGBtoXY(const byte* rgb, float* xy); // only defined if huesync disabled TODO
+void colorFromDecOrHexString(byte* rgb, const char* in);
+bool colorFromHexString(byte* rgb, const char* in);
+uint32_t colorBalanceFromKelvin(uint16_t kelvin, uint32_t rgb);
+uint16_t approximateKelvinFromRGB(uint32_t rgb);
+void setRandomColor(byte* rgb);
+
+[[gnu::hot, gnu::pure]] uint32_t color_fade(uint32_t c1, uint8_t amount, bool video = false);
+
+#endif

wled00/const.h

@@ -546,8 +546,8 @@ static_assert(WLED_MAX_BUSSES <= 32, "WLED_MAX_BUSSES exceeds hard limit");
   #endif
 #endif
 
-//#define MIN_HEAP_SIZE
-#define MIN_HEAP_SIZE 2048
+// minimum heap size required to process web requests
+#define MIN_HEAP_SIZE 8192
 
 // Web server limits
 #ifdef ESP8266

wled00/data/common.js

@@ -13,7 +13,7 @@ function isN(n)     { return !isNaN(parseFloat(n)) && isFinite(n); } // isNumber
 // https://stackoverflow.com/questions/3885817/how-do-i-check-that-a-number-is-float-or-integer
 function isF(n)     { return n === +n && n !== (n|0); } // isFloat
 function isI(n)     { return n === +n && n === (n|0); } // isInteger
-function toggle(el) { gId(el).classList.toggle("hide"); gId('No'+el).classList.toggle("hide"); }
+function toggle(el) { gId(el).classList.toggle("hide"); let n = gId('No'+el); if (n) n.classList.toggle("hide"); }
 function tooltip(cont=null) {
 	d.querySelectorAll((cont?cont+" ":"")+"[title]").forEach((element)=>{
 		element.addEventListener("pointerover", ()=>{

wled00/data/index.htm

@@ -329,7 +329,7 @@
 	<div>
 		<button class="btn ibtn" onclick="requestJson()">Refresh</button>
 		<button class="btn ibtn" onclick="toggleNodes()">Instance List</button>
-		<button class="btn ibtn" onclick="window.open(getURL('/update'),'_self');">Update WLED</button>
+		<button class="btn ibtn" onclick="window.open(getURL('/update'),'_self');" id="updBt">Update WLED</button>
 		<button class="btn ibtn" id="resetbtn" onclick="cnfReset()">Reboot WLED</button>
 	</div>
 	<br>

wled00/data/index.js

@@ -685,6 +685,7 @@ function parseInfo(i) {
 		gId("filter2D").classList.remove('hide');
 		gId('bs').querySelectorAll('option[data-type="2D"]').forEach((o,i)=>{o.style.display='';});
 	}
+	gId("updBt").style.display = (i.opt & 1) ? '':'none';
 //	if (i.noaudio) {
 //		gId("filterVol").classList.add("hide");
 //		gId("filterFreq").classList.add("hide");
@@ -1527,6 +1528,9 @@ function readState(s,command=false)
 			case  3:
 				errstr = "Buffer locked!";
 				break;
+			case  7:
+				errstr = "No RAM for buffer!";
+				break;
 			case  8:
 				errstr = "Effect RAM depleted!";
 				break;

wled00/data/settings_2D.htm

@@ -107,6 +107,7 @@ Y: <input name="P${i}Y" type="number" min="0" max="255" value="0" oninput="UI()"
 				Sf[`P${p}H`].value = ph;
 			}
 		}
+		UI(); // Update the preview after generating panels
 	}
 
 	function expand(o,i)

wled00/data/settings_leds.htm

@@ -7,7 +7,6 @@
 	<script src="common.js" async type="text/javascript"></script>
 	<script>
 		var maxB=1,maxD=1,maxA=1,maxV=0,maxM=4000,maxPB=2048,maxL=1664,maxCO=5; //maximum bytes for LED allocation: 4kB for 8266, 32kB for 32
-		var oMaxB=1;
 		var customStarts=false,startsDirty=[];
 		function off(n)    { gN(n).value = -1;}
 		// these functions correspond to C macros found in const.h
@@ -43,8 +42,8 @@
 			if (loc) d.Sf.action = getURL('/settings/leds');
 		}
 		function bLimits(b,v,p,m,l,o=5,d=2,a=6) {
-			oMaxB = maxB = b; // maxB - max buses (can be changed if using ESP32 parallel I2S): 20 - ESP32, 14 - S3/S2, 6 - C3, 4 - 8266
-			maxD  = d; // maxD - max digital channels (can be changed if using ESP32 parallel I2S): 17 - ESP32, 12 - S3/S2, 2 - C3, 3 - 8266
+			maxB  = b; // maxB - max physical (analog + digital) buses: 32 - ESP32, 14 - S3/S2, 6 - C3, 4 - 8266
+			maxD  = d; // maxD - max digital channels (can be changed if using ESP32 parallel I2S): 16 - ESP32, 12 - S3/S2, 2 - C3, 3 - 8266
 			maxA  = a; // maxA - max analog channels: 16 - ESP32, 8 - S3/S2, 6 - C3, 5 - 8266
 			maxV  = v; // maxV - min virtual buses: 6 - ESP32/S3, 4 - S2/C3, 3 - ESP8266 (only used to distinguish S2/S3)
 			maxPB = p; // maxPB - max LEDs per bus
@@ -52,6 +51,11 @@
 			maxL  = l; // maxL - max LEDs (will serve to determine ESP >1664 == ESP32)
 			maxCO = o; // maxCO - max Color Order mappings
 		}
+		function is8266() { return maxA ==  5 && maxD ==  3; } // NOTE: see const.h
+		function is32()   { return maxA == 16 && maxD == 16; } // NOTE: see const.h
+		function isC3()   { return maxA ==  6 && maxD ==  2; } // NOTE: see const.h
+		function isS2()   { return maxA ==  8 && maxD == 12 && maxV == 4; } // NOTE: see const.h
+		function isS3()   { return maxA ==  8 && maxD == 12 && maxV == 6; } // NOTE: see const.h
 		function pinsOK() {
 			var ok = true;
 			var nList = d.Sf.querySelectorAll("#mLC input[name^=L]");
@@ -212,7 +216,6 @@
 			let busMA = 0;
 			let sLC = 0, sPC = 0, sDI = 0, maxLC = 0;
 			const abl = d.Sf.ABL.checked;
-			maxB = oMaxB; // TODO make sure we start with all possible buses
 			let setPinConfig = (n,t) => {
 				let p0d = "GPIO:";
 				let p1d = "";
@@ -271,7 +274,7 @@
 				gRGBW |= hasW(t);                                                           // RGBW checkbox
 				gId("co"+n).style.display = (isVir(t) || isAna(t)) ? "none":"inline";       // hide color order for PWM
 				gId("dig"+n+"w").style.display = (isDig(t) && hasW(t)) ? "inline":"none";   // show swap channels dropdown
-				gId("dig"+n+"w").querySelector("[data-opt=CCT]").disabled = !hasCCT(t);     // disable WW/CW swapping	
+				gId("dig"+n+"w").querySelector("[data-opt=CCT]").disabled = !hasCCT(t);     // disable WW/CW swapping
 				if (!(isDig(t) && hasW(t))) d.Sf["WO"+n].value = 0;                         // reset swapping
 				gId("dig"+n+"c").style.display = (isAna(t)) ? "none":"inline";              // hide count for analog
 				gId("dig"+n+"r").style.display = (isVir(t)) ? "none":"inline";              // hide reversed for virtual
@@ -281,6 +284,8 @@
 				gId("dig"+n+"l").style.display = (isD2P(t) || isPWM(t)) ? "inline":"none";  // bus clock speed / PWM speed (relative) (not On/Off)
 				gId("rev"+n).innerHTML = isAna(t) ? "Inverted output":"Reversed";           // change reverse text for analog else (rotated 180°)
 				//gId("psd"+n).innerHTML = isAna(t) ? "Index:":"Start:";                      // change analog start description
+				gId("net"+n+"h").style.display = isNet(t) && !is8266() ? "block" : "none";  // show host field for network types except on ESP8266
+				if (!isNet(t) || is8266()) d.Sf["HS"+n].value = "";                         // cleart host field if not network type or ESP8266
 			});
 			// display global white channel overrides
 			gId("wc").style.display = (gRGBW) ? 'inline':'none';
@@ -289,11 +294,16 @@
 				d.Sf.CR.checked = false;
 			}
 			// update start indexes, max values, calculate current, etc
+			let sameType = 0;
 			var nList = d.Sf.querySelectorAll("#mLC input[name^=L]");
 			nList.forEach((LC,i)=>{
 				let nm = LC.name.substring(0,2);  // field name
 				let n  = LC.name.substring(2);    // bus number
 				let t  = parseInt(d.Sf["LT"+n].value); // LED type SELECT
+				if (isDig(t)) {
+					if (sameType == 0) sameType = t; // first bus type
+					else if (sameType != t) sameType = -1; // different bus type
+				}
 				// do we have a led count field
 				if (nm=="LC") {
 					let c = parseInt(LC.value,10); //get LED count
@@ -351,18 +361,13 @@
 						else LC.style.color = d.ro_gpio.some((e)=>e==parseInt(LC.value)) ? "orange" : "#fff";
 					}
 			});
-			const S2 = (oMaxB == 14) && (maxV == 4);
-			const S3 = (oMaxB == 14) && (maxV == 6);
-			if (oMaxB == 32 || S2 || S3) { // TODO: crude ESP32 & S2/S3 detection
-				if (maxLC > 300 || dC <= 2) {
+			if (is32() || isS2() || isS3()) {
+				if (maxLC > 600 || dC < 2 || sameType <= 0) {
 					d.Sf["PR"].checked = false;
 					gId("prl").classList.add("hide");
 				} else
 					gId("prl").classList.remove("hide");
-				// S2 supports mono I2S as well as parallel so we need to take that into account; S3 only supports parallel
-				maxD = (S2 || S3 ? 4 : 8) + (d.Sf["PR"].checked ? 8 : S2); // TODO: use bLimits() : 4/8RMT + (x1/x8 parallel) I2S1
-				maxB = oMaxB - (d.Sf["PR"].checked ? 0 : 7 + S3); // S2 (maxV==4) does support mono I2S
-			}
+			} else d.Sf["PR"].checked = false;
 			// distribute ABL current if not using PPL
 			enPPL(sDI);
 
@@ -463,6 +468,7 @@ mA/LED: <select name="LAsel${s}" onchange="enLA(this,'${s}');UI();">
 <span id="p2d${s}"></span><input type="number" name="L2${s}" class="s" onchange="UI();pinUpd(this);"/>
 <span id="p3d${s}"></span><input type="number" name="L3${s}" class="s" onchange="UI();pinUpd(this);"/>
 <span id="p4d${s}"></span><input type="number" name="L4${s}" class="s" onchange="UI();pinUpd(this);"/>
+<div id="net${s}h" class="hide">Host: <input type="text" name="HS${s}" maxlength="32" pattern="[a-zA-Z0-9_\\-]*" onchange="UI()"/>.local</div>
 <div id="dig${s}r" style="display:inline"><br><span id="rev${s}">Reversed</span>: <input type="checkbox" name="CV${s}"></div>
 <div id="dig${s}s" style="display:inline"><br>Skip first LEDs: <input type="number" name="SL${s}" min="0" max="255" value="0" oninput="UI()"></div>
 <div id="dig${s}f" style="display:inline"><br><span id="off${s}">Off Refresh</span>: <input id="rf${s}" type="checkbox" name="RF${s}"></div>
@@ -479,15 +485,17 @@ mA/LED: <select name="LAsel${s}" onchange="enLA(this,'${s}');UI();">
 							if (type.t != undefined && type.t != "") {
 								opt.setAttribute('data-type', type.t);
 							}
-							sel.appendChild(opt);			
+							sel.appendChild(opt);
 						}
 					}
 				});
 				enLA(d.Sf["LAsel"+s],s); // update LED mA
 				// disable inappropriate LED types
-				let sel = d.getElementsByName("LT"+s)[0]
-				if (i >= maxB || digitalB >= maxD) disable(sel,'option[data-type="D"]'); // NOTE: see isDig()
-				if (i >= maxB || twopinB >= 1)     disable(sel,'option[data-type="2P"]'); // NOTE: see isD2P()
+				let sel = d.getElementsByName("LT"+s)[0];
+				// 32 & S2 supports mono I2S as well as parallel so we need to take that into account; S3 only supports parallel
+				let maxDB = maxD - (is32() || isS2() || isS3() ? (!d.Sf["PR"].checked)*8 - (!isS3()) : 0); // adjust max digital buses if parallel I2S is not used
+				if (digitalB >= maxDB) disable(sel,'option[data-type="D"]'); // NOTE: see isDig()
+				if (twopinB  >= 2)     disable(sel,'option[data-type="2P"]'); // NOTE: see isD2P() (we will only allow 2 2pin buses)
 				disable(sel,`option[data-type^="${'A'.repeat(maxA-analogB+1)}"]`); // NOTE: see isPWM()
 				sel.selectedIndex = sel.querySelector('option:not(:disabled)').index;
 			}
@@ -586,7 +594,7 @@ Swap: <select id="xw${s}" name="XW${s}">
 			var cs = false;
 			for (var i=1; i < gEBCN("iST").length; i++) {
 				var s = chrID(i);
-				var p = chrID(i-1); // cover edge case 'A' previous char being '9' 
+				var p = chrID(i-1); // cover edge case 'A' previous char being '9'
 				var v = parseInt(gId("ls"+p).value) + parseInt(gN("LC"+p).value);
 				if (v != parseInt(gId("ls"+s).value)) {cs = true; startsDirty[i] = true;}
 			}
@@ -617,7 +625,7 @@ Swap: <select id="xw${s}" name="XW${s}">
 
 			function receivedText(e) {
 				let lines = e.target.result;
-				let c = JSON.parse(lines); 
+				let c = JSON.parse(lines);
 				if (c.hw) {
 					if (c.hw.led) {
 						// remove all existing outputs
@@ -900,7 +908,6 @@ Swap: <select id="xw${s}" name="XW${s}">
 			<option value="3">None (not recommended)</option>
 		</select><br>
 		Use harmonic <i>Random Cycle</i> palette: <input type="checkbox" name="TH"><br>
-		Use &quot;rainbow&quot; color wheel: <input type="checkbox" name="RW"><br>
 		Target refresh rate: <input type="number" class="s" min="0" max="250" name="FR" oninput="UI()" required> FPS
 		<div id="fpsNone" class="warn" style="display: none;">&#9888; Unlimited FPS Mode is experimental &#9888;<br></div>
 		<div id="fpsHigh" class="warn" style="display: none;">&#9888; High FPS Mode is experimental.<br></div>

wled00/data/settings_sec.htm

@@ -53,13 +53,13 @@
 		Factory reset: <input type="checkbox" name="RS"><br>
 		All settings and presets will be erased.<br><br>
 		<div class="warn">&#9888; Unencrypted transmission. An attacker on the same network can intercept form data!</div>
-		<hr>
+		<span id="OTA"><hr>
 		<h3>Software Update</h3>
 		<button type="button" onclick="U()">Manual OTA Update</button><br>
 		<div id="aOTA">Enable ArduinoOTA: <input type="checkbox" name="AO"></div>
 		Only allow update from same network/WiFi: <input type="checkbox" name="SU"><br>
 		<i class="warn">&#9888; If you are using multiple VLANs (i.e. IoT or guest network) either set PIN or disable this option.<br>
-			Disabling this option will make your device less secure.</i><br>
+			Disabling this option will make your device less secure.</i><br></span>
 		<hr id="backup">
 		<h3>Backup & Restore</h3>
 		<div class="warn">&#9888; Restoring presets/configuration will OVERWRITE your current presets/configuration.<br>
@@ -76,7 +76,7 @@
 		A huge thank you to everyone who helped me create WLED!<br><br>
 		(c) 2016-2024 Christian Schwinne <br>
 		<i>Licensed under the <a href="https://github.com/wled-dev/WLED/blob/main/LICENSE" target="_blank">EUPL v1.2 license</a></i><br><br>
-		Server message: <span class="sip"> Response error! </span><hr>
+		Installed version: <span class="sip">WLED ##VERSION##</span><hr>
 		<div id="toast"></div>
 		<button type="button" onclick="B()">Back</button><button type="submit">Save</button>
 	</form>

wled00/data/settings_wifi.htm

@@ -258,10 +258,10 @@ Static subnet mask:<br>
 			<h3>Ethernet Type</h3>
 			<select name="ETH">
 				<option value="0">None</option>
+				<option value="6">IoTorero/ESP32Deux/RGB2Go</option>
 				<option value="9">ABC! WLED V43 & compatible</option>
 				<option value="2">ESP32-POE</option>
 				<option value="11">ESP32-POE-WROVER</option>
-				<option value="6">ESP32Deux/RGB2Go Tetra</option>
 				<option value="7">KIT-VE</option>
 				<option value="12">LILYGO T-POE Pro</option>
 				<option value="8">QuinLED-Dig-Octa & T-ETH-POE</option>

wled00/data/style.css

@@ -79,6 +79,9 @@ input {
 input:disabled {
   color: #888;
 }
+input:invalid {
+  color: #f00;
+}
 input[type="text"],
 input[type="number"],
 input[type="password"],
@@ -202,4 +205,4 @@ td {
   #btns select {
     width: 144px;
   }
-}
+}

wled00/data/update.htm

@@ -27,7 +27,7 @@
 <body onload="GetV()">
 	<h2>WLED Software Update</h2>
 	<form method='POST' action='./update' id='upd' enctype='multipart/form-data' onsubmit="toggle('upd')">
-		Installed version: <span class="sip">##VERSION##</span><br>
+		Installed version: <span class="sip">WLED ##VERSION##</span><br>
 		Download the latest binary: <a href="https://github.com/wled-dev/WLED/releases" target="_blank" 
 		style="vertical-align: text-bottom; display: inline-flex;">
 		<img src="https://img.shields.io/github/release/wled-dev/WLED.svg?style=flat-square"></a><br>

wled00/e131.cpp

@@ -191,7 +191,7 @@ void handleDMXData(uint16_t uni, uint16_t dmxChannels, uint8_t* e131_data, uint8
         // only change brightness if value changed
         if (bri != e131_data[dataOffset]) {                                        
           bri = e131_data[dataOffset];
-          strip.setBrightness(scaledBri(bri), false);
+          strip.setBrightness(bri, false);
           stateUpdated(CALL_MODE_WS_SEND);
         }
         return;

wled00/fcn_declare.h

@@ -24,6 +24,10 @@ void handleIO();
 void IRAM_ATTR touchButtonISR();
 
 //cfg.cpp
+bool backupConfig();
+bool restoreConfig();
+bool verifyConfig();
+void resetConfig();
 bool deserializeConfig(JsonObject doc, bool fromFS = false);
 bool deserializeConfigFromFS();
 bool deserializeConfigSec();
@@ -69,133 +73,6 @@ typedef struct WiFiConfig {
   }
 } wifi_config;
 
-//colors.cpp
-#define ColorFromPalette ColorFromPaletteWLED // override fastled version
-
-// CRGBW can be used to manipulate 32bit colors faster. However: if it is passed to functions, it adds overhead compared to a uint32_t color
-// use with caution and pay attention to flash size. Usually converting a uint32_t to CRGBW to extract r, g, b, w values is slower than using bitshifts
-// it can be useful to avoid back and forth conversions between uint32_t and fastled CRGB
-struct CRGBW {
-    union {
-        uint32_t color32; // Access as a 32-bit value (0xWWRRGGBB)
-        struct {
-            uint8_t b;
-            uint8_t g;
-            uint8_t r;
-            uint8_t w;
-        };
-        uint8_t raw[4];   // Access as an array in the order B, G, R, W
-    };
-
-    // Default constructor
-    inline CRGBW() __attribute__((always_inline)) = default;
-
-    // Constructor from a 32-bit color (0xWWRRGGBB)
-    constexpr CRGBW(uint32_t color) __attribute__((always_inline)) : color32(color) {}
-
-    // Constructor with r, g, b, w values
-    constexpr CRGBW(uint8_t red, uint8_t green, uint8_t blue, uint8_t white = 0) __attribute__((always_inline)) : b(blue), g(green), r(red), w(white) {}
-
-    // Constructor from CRGB
-    constexpr CRGBW(CRGB rgb) __attribute__((always_inline)) : b(rgb.b), g(rgb.g), r(rgb.r), w(0) {}
-
-    // Access as an array
-    inline const uint8_t& operator[] (uint8_t x) const __attribute__((always_inline)) { return raw[x]; }
-
-    // Assignment from 32-bit color
-    inline CRGBW& operator=(uint32_t color) __attribute__((always_inline)) { color32 = color; return *this; }
-
-    // Assignment from r, g, b, w
-    inline CRGBW& operator=(const CRGB& rgb) __attribute__((always_inline)) { b = rgb.b; g = rgb.g; r = rgb.r; w = 0; return *this; }
-
-    // Conversion operator to uint32_t
-    inline operator uint32_t() const __attribute__((always_inline)) {
-      return color32;
-    }
-    /*
-    // Conversion operator to CRGB
-    inline operator CRGB() const __attribute__((always_inline)) {
-      return CRGB(r, g, b);
-    }
-
-    CRGBW& scale32 (uint8_t scaledown) // 32bit math
-    {
-      if (color32 == 0) return *this; // 2 extra instructions, worth it if called a lot on black (which probably is true) adding check if scaledown is zero adds much more overhead as its 8bit
-      uint32_t scale = scaledown + 1;
-      uint32_t rb = (((color32 & 0x00FF00FF) * scale) >> 8) & 0x00FF00FF; // scale red and blue
-      uint32_t wg = (((color32 & 0xFF00FF00) >> 8) * scale) & 0xFF00FF00; // scale white and green
-          color32 =  rb | wg;
-      return *this;
-    }*/
-
-};
-
-struct CHSV32 { // 32bit HSV color with 16bit hue for more accurate conversions
-  union {
-    struct {
-        uint16_t h;  // hue
-        uint8_t s;   // saturation
-        uint8_t v;   // value
-    };
-    uint32_t raw;    // 32bit access
-  };
-  inline CHSV32() __attribute__((always_inline)) = default; // default constructor
-
-    /// Allow construction from hue, saturation, and value
-    /// @param ih input hue
-    /// @param is input saturation
-    /// @param iv input value
-  inline CHSV32(uint16_t ih, uint8_t is, uint8_t iv) __attribute__((always_inline)) // constructor from 16bit h, s, v
-        : h(ih), s(is), v(iv) {}
-  inline CHSV32(uint8_t ih, uint8_t is, uint8_t iv) __attribute__((always_inline)) // constructor from 8bit h, s, v
-        : h((uint16_t)ih << 8), s(is), v(iv) {}
-  inline CHSV32(const CHSV& chsv) __attribute__((always_inline))  // constructor from CHSV
-    : h((uint16_t)chsv.h << 8), s(chsv.s), v(chsv.v) {}
-  inline operator CHSV() const { return CHSV((uint8_t)(h >> 8), s, v); } // typecast to CHSV
-};
-// similar to NeoPixelBus NeoGammaTableMethod but allows dynamic changes (superseded by NPB::NeoGammaDynamicTableMethod)
-class NeoGammaWLEDMethod {
-  public:
-    [[gnu::hot]] static uint8_t Correct(uint8_t value);         // apply Gamma to single channel
-    [[gnu::hot]] static uint32_t Correct32(uint32_t color);     // apply Gamma to RGBW32 color (WLED specific, not used by NPB)
-    [[gnu::hot]] static uint32_t inverseGamma32(uint32_t color); // apply inverse Gamma to RGBW32 color
-    static void calcGammaTable(float gamma);                    // re-calculates & fills gamma tables
-    static inline uint8_t rawGamma8(uint8_t val) { return gammaT[val]; }  // get value from Gamma table (WLED specific, not used by NPB)
-    static inline uint8_t rawInverseGamma8(uint8_t val) { return gammaT_inv[val]; }  // get value from inverse Gamma table (WLED specific, not used by NPB)
-  private:
-    static uint8_t gammaT[];
-    static uint8_t gammaT_inv[];
-};
-#define gamma32(c) NeoGammaWLEDMethod::Correct32(c)
-#define gamma8(c)  NeoGammaWLEDMethod::rawGamma8(c)
-#define gamma32inv(c) NeoGammaWLEDMethod::inverseGamma32(c)
-#define gamma8inv(c)  NeoGammaWLEDMethod::rawInverseGamma8(c)
-[[gnu::hot, gnu::pure]] uint32_t color_blend(uint32_t c1, uint32_t c2 , uint8_t blend);
-inline uint32_t color_blend16(uint32_t c1, uint32_t c2, uint16_t b) { return color_blend(c1, c2, b >> 8); };
-[[gnu::hot, gnu::pure]] uint32_t color_add(uint32_t, uint32_t, bool preserveCR = false);
-[[gnu::hot, gnu::pure]] uint32_t color_fade(uint32_t c1, uint8_t amount, bool video=false);
-[[gnu::hot, gnu::pure]] uint32_t adjust_color(uint32_t rgb, uint32_t hueShift, uint32_t lighten, uint32_t brighten);
-[[gnu::hot, gnu::pure]] uint32_t ColorFromPaletteWLED(const CRGBPalette16 &pal, unsigned index, uint8_t brightness = (uint8_t)255U, TBlendType blendType = LINEARBLEND);
-CRGBPalette16 generateHarmonicRandomPalette(const CRGBPalette16 &basepalette);
-CRGBPalette16 generateRandomPalette();
-void loadCustomPalettes();
-extern std::vector<CRGBPalette16> customPalettes;
-inline size_t getPaletteCount() { return 13 + GRADIENT_PALETTE_COUNT + customPalettes.size(); }
-inline uint32_t colorFromRgbw(byte* rgbw) { return uint32_t((byte(rgbw[3]) << 24) | (byte(rgbw[0]) << 16) | (byte(rgbw[1]) << 8) | (byte(rgbw[2]))); }
-void hsv2rgb(const CHSV32& hsv, uint32_t& rgb);
-void colorHStoRGB(uint16_t hue, byte sat, byte* rgb);
-void rgb2hsv(const uint32_t rgb, CHSV32& hsv);
-inline CHSV rgb2hsv(const CRGB c) { CHSV32 hsv; rgb2hsv((uint32_t((byte(c.r) << 16) | (byte(c.g) << 8) | (byte(c.b)))), hsv); return CHSV(hsv); } // CRGB to hsv
-void colorKtoRGB(uint16_t kelvin, byte* rgb);
-void colorCTtoRGB(uint16_t mired, byte* rgb); //white spectrum to rgb
-void colorXYtoRGB(float x, float y, byte* rgb); // only defined if huesync disabled TODO
-void colorRGBtoXY(const byte* rgb, float* xy); // only defined if huesync disabled TODO
-void colorFromDecOrHexString(byte* rgb, const char* in);
-bool colorFromHexString(byte* rgb, const char* in);
-uint32_t colorBalanceFromKelvin(uint16_t kelvin, uint32_t rgb);
-uint16_t approximateKelvinFromRGB(uint32_t rgb);
-void setRandomColor(byte* rgb);
-
 //dmx_output.cpp
 void initDMXOutput();
 void handleDMXOutput();
@@ -223,6 +100,11 @@ inline bool writeObjectToFileUsingId(const String &file, uint16_t id, const Json
 inline bool writeObjectToFile(const String &file, const char* key, const JsonDocument* content) { return writeObjectToFile(file.c_str(), key, content); };
 inline bool readObjectFromFileUsingId(const String &file, uint16_t id, JsonDocument* dest, const JsonDocument* filter = nullptr) { return readObjectFromFileUsingId(file.c_str(), id, dest); };
 inline bool readObjectFromFile(const String &file, const char* key, JsonDocument* dest, const JsonDocument* filter = nullptr) { return readObjectFromFile(file.c_str(), key, dest); };
+bool copyFile(const char* src_path, const char* dst_path);
+bool backupFile(const char* filename);
+bool restoreFile(const char* filename);
+bool validateJsonFile(const char* filename);
+void dumpFilesToSerial();
 
 //hue.cpp
 void handleHue();
@@ -433,6 +315,7 @@ class Usermod {
     virtual void onMqttConnect(bool sessionPresent) {}                       // fired when MQTT connection is established (so usermod can subscribe)
     virtual bool onMqttMessage(char* topic, char* payload) { return false; } // fired upon MQTT message received (wled topic)
     virtual bool onEspNowMessage(uint8_t* sender, uint8_t* payload, uint8_t len) { return false; } // fired upon ESP-NOW message received
+    virtual bool onUdpPacket(uint8_t* payload, size_t len) { return false; } //fired upon UDP packet received
     virtual void onUpdateBegin(bool) {}                                      // fired prior to and after unsuccessful firmware update
     virtual void onStateChange(uint8_t mode) {}                              // fired upon WLED state change
     virtual uint16_t getId() {return USERMOD_ID_UNSPECIFIED;}
@@ -472,6 +355,7 @@ namespace UsermodManager {
 #ifndef WLED_DISABLE_ESPNOW
   bool onEspNowMessage(uint8_t* sender, uint8_t* payload, uint8_t len);
 #endif
+  bool onUdpPacket(uint8_t* payload, size_t len);
   void onUpdateBegin(bool);
   void onStateChange(uint8_t);
   Usermod* lookup(uint16_t mod_id);
@@ -551,28 +435,39 @@ inline uint8_t hw_random8(uint32_t upperlimit) { return (hw_random8() * upperlim
 inline uint8_t hw_random8(uint32_t lowerlimit, uint32_t upperlimit) { uint32_t range = upperlimit - lowerlimit; return lowerlimit + hw_random8(range); }; // input range 0-255
 
 // PSRAM allocation wrappers
-#ifndef ESP8266
+#if !defined(ESP8266) && !defined(CONFIG_IDF_TARGET_ESP32C3)
 extern "C" {
   void *p_malloc(size_t);           // prefer PSRAM over DRAM
   void *p_calloc(size_t, size_t);   // prefer PSRAM over DRAM
   void *p_realloc(void *, size_t);  // prefer PSRAM over DRAM
+  void *p_realloc_malloc(void *ptr, size_t size); // realloc with malloc fallback, prefer PSRAM over DRAM
   inline void p_free(void *ptr) { heap_caps_free(ptr); }
   void *d_malloc(size_t);           // prefer DRAM over PSRAM
   void *d_calloc(size_t, size_t);   // prefer DRAM over PSRAM
   void *d_realloc(void *, size_t);  // prefer DRAM over PSRAM
+  void *d_realloc_malloc(void *ptr, size_t size); // realloc with malloc fallback, prefer DRAM over PSRAM
   inline void d_free(void *ptr) { heap_caps_free(ptr); }
 }
 #else
+extern "C" {
+  void *realloc_malloc(void *ptr, size_t size);
+}
 #define p_malloc malloc
 #define p_calloc calloc
 #define p_realloc realloc
+#define p_realloc_malloc realloc_malloc
 #define p_free free
 #define d_malloc malloc
 #define d_calloc calloc
 #define d_realloc realloc
+#define d_realloc_malloc realloc_malloc
 #define d_free free
 #endif
 
+void handleBootLoop();   // detect and handle bootloops
+#ifndef ESP8266
+void bootloopCheckOTA(); // swap boot image if bootloop is detected instead of restoring config
+#endif
 // RAII guard class for the JSON Buffer lock
 // Modeled after std::lock_guard
 class JSONBufferGuard {

wled00/file.cpp

@@ -439,3 +439,156 @@ bool handleFileRead(AsyncWebServerRequest* request, String path){
   }
   return false;
 }
+
+// copy a file, delete destination file if incomplete to prevent corrupted files
+bool copyFile(const char* src_path, const char* dst_path) {
+  DEBUG_PRINTF("copyFile from %s to %s\n", src_path, dst_path);
+  if(!WLED_FS.exists(src_path)) {
+   DEBUG_PRINTLN(F("file not found"));
+   return false;
+  }
+
+  bool success = true; // is set to false on error
+  File src = WLED_FS.open(src_path, "r");
+  File dst = WLED_FS.open(dst_path, "w");
+
+  if (src && dst) {
+    uint8_t buf[128]; // copy file in 128-byte blocks
+    while (src.available() > 0) {
+      size_t bytesRead = src.read(buf, sizeof(buf));
+      if (bytesRead == 0) {
+        success = false;
+        break; // error, no data read
+      }
+      size_t bytesWritten = dst.write(buf, bytesRead);
+      if (bytesWritten != bytesRead) {
+        success = false;
+        break; // error, not all data written
+      }
+    }
+  } else {
+    success = false; // error, could not open files
+  }
+  if(src) src.close();
+  if(dst) dst.close();
+  if (!success) {
+    DEBUG_PRINTLN(F("copy failed"));
+    WLED_FS.remove(dst_path); // delete incomplete file
+  }
+  return success;
+}
+
+// compare two files, return true if identical
+bool compareFiles(const char* path1, const char* path2) {
+  DEBUG_PRINTF("compareFile %s and %s\n", path1, path2);
+  if (!WLED_FS.exists(path1) || !WLED_FS.exists(path2)) {
+    DEBUG_PRINTLN(F("file not found"));
+    return false;
+  }
+
+  bool identical = true; // set to false on mismatch
+  File f1 = WLED_FS.open(path1, "r");
+  File f2 = WLED_FS.open(path2, "r");
+
+  if (f1 && f2) {
+    uint8_t buf1[128], buf2[128];
+    while (f1.available() > 0 || f2.available() > 0) {
+      size_t len1 = f1.read(buf1, sizeof(buf1));
+      size_t len2 = f2.read(buf2, sizeof(buf2));
+
+      if (len1 != len2) {
+        identical = false;
+        break; // files differ in size or read failed
+      }
+
+      if (memcmp(buf1, buf2, len1) != 0) {
+        identical = false;
+        break; // files differ in content
+      }
+    }
+  } else {
+    identical = false; // error opening files
+  }
+
+  if (f1) f1.close();
+  if (f2) f2.close();
+  return identical;
+}
+
+static const char s_backup_fmt[] PROGMEM = "/bkp.%s";
+
+bool backupFile(const char* filename) {
+  DEBUG_PRINTF("backup %s \n", filename);
+  if (!validateJsonFile(filename)) {
+    DEBUG_PRINTLN(F("broken file"));
+    return false;
+  }
+  char backupname[32];
+  snprintf_P(backupname, sizeof(backupname), s_backup_fmt, filename + 1); // skip leading '/' in filename
+
+  if (copyFile(filename, backupname)) {
+    DEBUG_PRINTLN(F("backup ok"));
+    return true;
+  }
+  DEBUG_PRINTLN(F("backup failed"));
+  return false;
+}
+
+bool restoreFile(const char* filename) {
+  DEBUG_PRINTF("restore %s \n", filename);
+  char backupname[32];
+  snprintf_P(backupname, sizeof(backupname), s_backup_fmt, filename + 1); // skip leading '/' in filename
+
+  if (!WLED_FS.exists(backupname)) {
+    DEBUG_PRINTLN(F("no backup found"));
+    return false;
+  }
+
+  if (!validateJsonFile(backupname)) {
+    DEBUG_PRINTLN(F("broken backup"));
+    return false;
+  }
+
+  if (copyFile(backupname, filename)) {
+    DEBUG_PRINTLN(F("restore ok"));
+    return true;
+  }
+  DEBUG_PRINTLN(F("restore failed"));
+  return false;
+}
+
+bool validateJsonFile(const char* filename) {
+  if (!WLED_FS.exists(filename)) return false;
+  File file = WLED_FS.open(filename, "r");
+  if (!file) return false;
+  StaticJsonDocument<0> doc, filter; // https://arduinojson.org/v6/how-to/validate-json/
+  bool result = deserializeJson(doc, file, DeserializationOption::Filter(filter)) == DeserializationError::Ok;
+  file.close();
+  if (!result) {
+    DEBUG_PRINTF_P(PSTR("Invalid JSON file %s\n"), filename);
+  } else {
+    DEBUG_PRINTF_P(PSTR("Valid JSON file %s\n"), filename);
+  }
+  return result;
+}
+
+// print contents of all files in root dir to Serial except wsec files
+void dumpFilesToSerial() {
+  File rootdir = WLED_FS.open("/", "r");
+  File rootfile = rootdir.openNextFile();
+  while (rootfile) {
+    size_t len = strlen(rootfile.name());
+    // skip files starting with "wsec" and dont end in .json
+    if (strncmp(rootfile.name(), "wsec", 4) != 0 && len >= 6 && strcmp(rootfile.name() + len - 5, ".json") == 0) {
+      Serial.println(rootfile.name());
+      while (rootfile.available()) {
+        Serial.write(rootfile.read());
+      }
+      Serial.println();
+      Serial.println();
+    }
+    rootfile.close();
+    rootfile = rootdir.openNextFile();
+  }
+}
+

wled00/image_loader.cpp

@@ -58,7 +58,7 @@ void drawPixelCallback(int16_t x, int16_t y, uint8_t red, uint8_t green, uint8_t
   // set multiple pixels if upscaling
   for (int16_t i = 0; i < (activeSeg->width()+(gifWidth-1)) / gifWidth; i++) {
     for (int16_t j = 0; j < (activeSeg->height()+(gifHeight-1)) / gifHeight; j++) {
-      activeSeg->setPixelColorXY(outX + i, outY + j, gamma8(red), gamma8(green), gamma8(blue));
+      activeSeg->setPixelColorXY(outX + i, outY + j, red, green, blue);
     }
   }
 }

wled00/json.cpp

@@ -312,7 +312,7 @@ static bool deserializeSegment(JsonObject elem, byte it, byte presetId = 0)
     jsonTransitionOnce = true;
     if (seg.isInTransition()) seg.startTransition(0); // setting transition time to 0 will stop transition in next frame
     strip.setTransition(0);
-    strip.setBrightness(scaledBri(bri), true);
+    strip.setBrightness(bri, true);
 
     // freeze and init to black
     if (!seg.freeze) {

wled00/led.cpp

@@ -57,7 +57,7 @@ void toggleOnOff()
 //scales the brightness with the briMultiplier factor
 byte scaledBri(byte in)
 {
-  unsigned val = ((uint16_t)in*briMultiplier)/100;
+  unsigned val = ((unsigned)in*briMultiplier)/100;
   if (val > 255) val = 255;
   return (byte)val;
 }
@@ -68,7 +68,7 @@ void applyBri() {
   if (realtimeOverride || !(realtimeMode && arlsForceMaxBri))
   {
     //DEBUG_PRINTF_P(PSTR("Applying strip brightness: %d (%d,%d)\n"), (int)briT, (int)bri, (int)briOld);
-    strip.setBrightness(scaledBri(briT));
+    strip.setBrightness(briT);
   }
 }
 

wled00/mqtt.cpp

@@ -11,6 +11,11 @@
 #warning "MQTT topics length > 32 is not recommended for compatibility with usermods!"
 #endif
 
+static const char* sTopicFormat PROGMEM = "%.*s/%s";
+
+// parse payload for brightness, ON/OFF or toggle
+// briLast is used to remember last brightness value in case of ON/OFF or toggle
+// bri is set to 0 if payload is "0" or "OFF" or "false"
 static void parseMQTTBriPayload(char* payload)
 {
   if      (strstr(payload, "ON") || strstr(payload, "on") || strstr(payload, "true")) {bri = briLast; stateUpdated(CALL_MODE_DIRECT_CHANGE);}
@@ -30,22 +35,18 @@ static void onMqttConnect(bool sessionPresent)
   char subuf[MQTT_MAX_TOPIC_LEN + 9];
 
   if (mqttDeviceTopic[0] != 0) {
-    strlcpy(subuf, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
+    mqtt->subscribe(mqttDeviceTopic, 0);
+    snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "col");
     mqtt->subscribe(subuf, 0);
-    strcat_P(subuf, PSTR("/col"));
-    mqtt->subscribe(subuf, 0);
-    strlcpy(subuf, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
-    strcat_P(subuf, PSTR("/api"));
+    snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "api");
     mqtt->subscribe(subuf, 0);
   }
 
   if (mqttGroupTopic[0] != 0) {
-    strlcpy(subuf, mqttGroupTopic, MQTT_MAX_TOPIC_LEN + 1);
+    mqtt->subscribe(mqttGroupTopic, 0);
+    snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttGroupTopic, "col");
     mqtt->subscribe(subuf, 0);
-    strcat_P(subuf, PSTR("/col"));
-    mqtt->subscribe(subuf, 0);
-    strlcpy(subuf, mqttGroupTopic, MQTT_MAX_TOPIC_LEN + 1);
-    strcat_P(subuf, PSTR("/api"));
+    snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttGroupTopic, "api");
     mqtt->subscribe(subuf, 0);
   }
 
@@ -54,8 +55,7 @@ static void onMqttConnect(bool sessionPresent)
   DEBUG_PRINTLN(F("MQTT ready"));
 
 #ifndef USERMOD_SMARTNEST
-  strlcpy(subuf, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
-  strcat_P(subuf, PSTR("/status"));
+  snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "status");
   mqtt->publish(subuf, 0, true, "online"); // retain message for a LWT
 #endif
 
@@ -136,7 +136,7 @@ static void onMqttMessage(char* topic, char* payload, AsyncMqttClientMessageProp
 }
 
 // Print adapter for flat buffers
-namespace { 
+namespace {
 class bufferPrint : public Print {
   char* _buf;
   size_t _size, _offset;
@@ -172,21 +172,21 @@ void publishMqtt()
   char subuf[MQTT_MAX_TOPIC_LEN + 16];
 
   sprintf_P(s, PSTR("%u"), bri);
-  strlcpy(subuf, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
-  strcat_P(subuf, PSTR("/g"));
+  snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "g");
   mqtt->publish(subuf, 0, retainMqttMsg, s);         // optionally retain message (#2263)
 
   sprintf_P(s, PSTR("#%06X"), (colPri[3] << 24) | (colPri[0] << 16) | (colPri[1] << 8) | (colPri[2]));
-  strlcpy(subuf, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
-  strcat_P(subuf, PSTR("/c"));
+  snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "c");
   mqtt->publish(subuf, 0, retainMqttMsg, s);         // optionally retain message (#2263)
 
+  snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "status");
+  mqtt->publish(subuf, 0, true, "online");  // retain message for a LWT
+
   // TODO: use a DynamicBufferList.  Requires a list-read-capable MQTT client API.
   DynamicBuffer buf(1024);
   bufferPrint pbuf(buf.data(), buf.size());
   XML_response(pbuf);
-  strlcpy(subuf, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
-  strcat_P(subuf, PSTR("/v"));
+  snprintf_P(subuf, sizeof(subuf)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "v");
   mqtt->publish(subuf, 0, retainMqttMsg, buf.data(), pbuf.size());   // optionally retain message (#2263)
   #endif
 }
@@ -213,14 +213,26 @@ bool initMqtt()
   {
     mqtt->setServer(mqttIP, mqttPort);
   } else {
-    mqtt->setServer(mqttServer, mqttPort);
+    #ifdef ARDUINO_ARCH_ESP32
+    String mqttMDNS = mqttServer;
+    mqttMDNS.toLowerCase(); // make sure we have a lowercase hostname
+    int pos = mqttMDNS.indexOf(F(".local"));
+    if (pos > 0) mqttMDNS.remove(pos); // remove .local domain if present (and anything following it)
+    if (strlen(cmDNS) > 0 && mqttMDNS.length() > 0 && mqttMDNS.indexOf('.') < 0) { // if mDNS is enabled and server does not have domain
+      mqttIP = MDNS.queryHost(mqttMDNS.c_str());
+      if (mqttIP != IPAddress()) // if MDNS resolved the hostname
+        mqtt->setServer(mqttIP, mqttPort);
+      else
+        mqtt->setServer(mqttServer, mqttPort);
+    } else
+    #endif
+      mqtt->setServer(mqttServer, mqttPort);
   }
   mqtt->setClientId(mqttClientID);
   if (mqttUser[0] && mqttPass[0]) mqtt->setCredentials(mqttUser, mqttPass);
 
   #ifndef USERMOD_SMARTNEST
-  strlcpy(mqttStatusTopic, mqttDeviceTopic, MQTT_MAX_TOPIC_LEN + 1);
-  strcat_P(mqttStatusTopic, PSTR("/status"));
+  snprintf_P(mqttStatusTopic, sizeof(mqttStatusTopic)-1, sTopicFormat, MQTT_MAX_TOPIC_LEN, mqttDeviceTopic, "status");
   mqtt->setWill(mqttStatusTopic, 0, true, "offline"); // LWT message
   #endif
   mqtt->setKeepAlive(MQTT_KEEP_ALIVE_TIME);

wled00/set.cpp

@@ -141,6 +141,7 @@ void handleSettingsSet(AsyncWebServerRequest *request, byte subPage)
     unsigned colorOrder, type, skip, awmode, channelSwap, maPerLed;
     unsigned length, start, maMax;
     uint8_t pins[5] = {255, 255, 255, 255, 255};
+    String text;
 
     // this will set global ABL max current used when per-port ABL is not used
     unsigned ablMilliampsMax = request->arg(F("MA")).toInt();
@@ -174,6 +175,7 @@ void handleSettingsSet(AsyncWebServerRequest *request, byte subPage)
       char sp[4] = "SP"; sp[2] = offset+s; sp[3] = 0; //bus clock speed (DotStar & PWM)
       char la[4] = "LA"; la[2] = offset+s; la[3] = 0; //LED mA
       char ma[4] = "MA"; ma[2] = offset+s; ma[3] = 0; //max mA
+      char hs[4] = "HS"; hs[2] = offset+s; hs[3] = 0; //hostname (for network types, custom text for others)
       if (!request->hasArg(lp)) {
         DEBUG_PRINTF_P(PSTR("# of buses: %d\n"), s+1);
         break;
@@ -224,9 +226,10 @@ void handleSettingsSet(AsyncWebServerRequest *request, byte subPage)
         maMax = request->arg(ma).toInt() * request->hasArg(F("PPL")); // if PP-ABL is disabled maMax (per bus) must be 0
       }
       type |= request->hasArg(rf) << 7; // off refresh override
+      text = request->arg(hs).substring(0,31);
       // actual finalization is done in WLED::loop() (removing old busses and adding new)
       // this may happen even before this loop is finished so we do "doInitBusses" after the loop
-      busConfigs.emplace_back(type, pins, start, length, colorOrder | (channelSwap<<4), request->hasArg(cv), skip, awmode, freq, maPerLed, maMax);
+      busConfigs.emplace_back(type, pins, start, length, colorOrder | (channelSwap<<4), request->hasArg(cv), skip, awmode, freq, maPerLed, maMax, text);
       busesChanged = true;
     }
     //doInitBusses = busesChanged; // we will do that below to ensure all input data is processed
@@ -348,7 +351,6 @@ void handleSettingsSet(AsyncWebServerRequest *request, byte subPage)
     t = request->arg(F("TP")).toInt();
     randomPaletteChangeTime = MIN(255,MAX(1,t));
     useHarmonicRandomPalette = request->hasArg(F("TH"));
-    useRainbowWheel = request->hasArg(F("RW"));
 
     nightlightTargetBri = request->arg(F("TB")).toInt();
     t = request->arg(F("TL")).toInt();

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient.cpp

@@ -1,877 +0,0 @@
-#include "AsyncMqttClient.hpp"
-
-AsyncMqttClient::AsyncMqttClient()
-: _connected(false)
-, _connectPacketNotEnoughSpace(false)
-, _disconnectFlagged(false)
-, _tlsBadFingerprint(false)
-, _lastClientActivity(0)
-, _lastServerActivity(0)
-, _lastPingRequestTime(0)
-, _host(nullptr)
-, _useIp(false)
-#if ASYNC_TCP_SSL_ENABLED
-, _secure(false)
-#endif
-, _port(0)
-, _keepAlive(15)
-, _cleanSession(true)
-, _clientId(nullptr)
-, _username(nullptr)
-, _password(nullptr)
-, _willTopic(nullptr)
-, _willPayload(nullptr)
-, _willPayloadLength(0)
-, _willQos(0)
-, _willRetain(false)
-, _parsingInformation { .bufferState = AsyncMqttClientInternals::BufferState::NONE }
-, _currentParsedPacket(nullptr)
-, _remainingLengthBufferPosition(0)
-, _nextPacketId(1) {
-  _client.onConnect([](void* obj, AsyncClient* c) { (static_cast<AsyncMqttClient*>(obj))->_onConnect(c); }, this);
-  _client.onDisconnect([](void* obj, AsyncClient* c) { (static_cast<AsyncMqttClient*>(obj))->_onDisconnect(c); }, this);
-  _client.onError([](void* obj, AsyncClient* c, int8_t error) { (static_cast<AsyncMqttClient*>(obj))->_onError(c, error); }, this);
-  _client.onTimeout([](void* obj, AsyncClient* c, uint32_t time) { (static_cast<AsyncMqttClient*>(obj))->_onTimeout(c, time); }, this);
-  _client.onAck([](void* obj, AsyncClient* c, size_t len, uint32_t time) { (static_cast<AsyncMqttClient*>(obj))->_onAck(c, len, time); }, this);
-  _client.onData([](void* obj, AsyncClient* c, void* data, size_t len) { (static_cast<AsyncMqttClient*>(obj))->_onData(c, static_cast<char*>(data), len); }, this);
-  _client.onPoll([](void* obj, AsyncClient* c) { (static_cast<AsyncMqttClient*>(obj))->_onPoll(c); }, this);
-
-#ifdef ESP32
-  sprintf(_generatedClientId, "esp32%06x", (uint32_t)ESP.getEfuseMac());
-  _xSemaphore = xSemaphoreCreateMutex();
-#elif defined(ESP8266)
-  sprintf(_generatedClientId, "esp8266%06x", (uint32_t)ESP.getChipId());
-#endif
-  _clientId = _generatedClientId;
-
-  setMaxTopicLength(128);
-}
-
-AsyncMqttClient::~AsyncMqttClient() {
-  delete _currentParsedPacket;
-  delete[] _parsingInformation.topicBuffer;
-#ifdef ESP32
-  vSemaphoreDelete(_xSemaphore);
-#endif
-}
-
-AsyncMqttClient& AsyncMqttClient::setKeepAlive(uint16_t keepAlive) {
-  _keepAlive = keepAlive;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setClientId(const char* clientId) {
-  _clientId = clientId;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setCleanSession(bool cleanSession) {
-  _cleanSession = cleanSession;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setMaxTopicLength(uint16_t maxTopicLength) {
-  _parsingInformation.maxTopicLength = maxTopicLength;
-  delete[] _parsingInformation.topicBuffer;
-  _parsingInformation.topicBuffer = new char[maxTopicLength + 1];
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setCredentials(const char* username, const char* password) {
-  _username = username;
-  _password = password;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setWill(const char* topic, uint8_t qos, bool retain, const char* payload, size_t length) {
-  _willTopic = topic;
-  _willQos = qos;
-  _willRetain = retain;
-  _willPayload = payload;
-  _willPayloadLength = length;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setServer(IPAddress ip, uint16_t port) {
-  _useIp = true;
-  _ip = ip;
-  _port = port;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::setServer(const char* host, uint16_t port) {
-  _useIp = false;
-  _host = host;
-  _port = port;
-  return *this;
-}
-
-#if ASYNC_TCP_SSL_ENABLED
-AsyncMqttClient& AsyncMqttClient::setSecure(bool secure) {
-  _secure = secure;
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::addServerFingerprint(const uint8_t* fingerprint) {
-  std::array<uint8_t, SHA1_SIZE> newFingerprint;
-  memcpy(newFingerprint.data(), fingerprint, SHA1_SIZE);
-  _secureServerFingerprints.push_back(newFingerprint);
-  return *this;
-}
-#endif
-
-AsyncMqttClient& AsyncMqttClient::onConnect(AsyncMqttClientInternals::OnConnectUserCallback callback) {
-  _onConnectUserCallbacks.push_back(callback);
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::onDisconnect(AsyncMqttClientInternals::OnDisconnectUserCallback callback) {
-  _onDisconnectUserCallbacks.push_back(callback);
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::onSubscribe(AsyncMqttClientInternals::OnSubscribeUserCallback callback) {
-  _onSubscribeUserCallbacks.push_back(callback);
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::onUnsubscribe(AsyncMqttClientInternals::OnUnsubscribeUserCallback callback) {
-  _onUnsubscribeUserCallbacks.push_back(callback);
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::onMessage(AsyncMqttClientInternals::OnMessageUserCallback callback) {
-  _onMessageUserCallbacks.push_back(callback);
-  return *this;
-}
-
-AsyncMqttClient& AsyncMqttClient::onPublish(AsyncMqttClientInternals::OnPublishUserCallback callback) {
-  _onPublishUserCallbacks.push_back(callback);
-  return *this;
-}
-
-void AsyncMqttClient::_freeCurrentParsedPacket() {
-  delete _currentParsedPacket;
-  _currentParsedPacket = nullptr;
-}
-
-void AsyncMqttClient::_clear() {
-  _lastPingRequestTime = 0;
-  _connected = false;
-  _disconnectFlagged = false;
-  _connectPacketNotEnoughSpace = false;
-  _tlsBadFingerprint = false;
-  _freeCurrentParsedPacket();
-
-  _pendingPubRels.clear();
-  _pendingPubRels.shrink_to_fit();
-
-  _toSendAcks.clear();
-  _toSendAcks.shrink_to_fit();
-
-  _nextPacketId = 1;
-  _parsingInformation.bufferState = AsyncMqttClientInternals::BufferState::NONE;
-}
-
-/* TCP */
-void AsyncMqttClient::_onConnect(AsyncClient* client) {
-  (void)client;
-
-#if ASYNC_TCP_SSL_ENABLED
-  if (_secure && _secureServerFingerprints.size() > 0) {
-    SSL* clientSsl = _client.getSSL();
-
-    bool sslFoundFingerprint = false;
-    for (std::array<uint8_t, SHA1_SIZE> fingerprint : _secureServerFingerprints) {
-      if (ssl_match_fingerprint(clientSsl, fingerprint.data()) == SSL_OK) {
-        sslFoundFingerprint = true;
-        break;
-      }
-    }
-
-    if (!sslFoundFingerprint) {
-      _tlsBadFingerprint = true;
-      _client.close(true);
-      return;
-    }
-  }
-#endif
-
-  char fixedHeader[5];
-  fixedHeader[0] = AsyncMqttClientInternals::PacketType.CONNECT;
-  fixedHeader[0] = fixedHeader[0] << 4;
-  fixedHeader[0] = fixedHeader[0] | AsyncMqttClientInternals::HeaderFlag.CONNECT_RESERVED;
-
-  uint16_t protocolNameLength = 4;
-  char protocolNameLengthBytes[2];
-  protocolNameLengthBytes[0] = protocolNameLength >> 8;
-  protocolNameLengthBytes[1] = protocolNameLength & 0xFF;
-
-  char protocolLevel[1];
-  protocolLevel[0] = 0x04;
-
-  char connectFlags[1];
-  connectFlags[0] = 0;
-  if (_cleanSession) connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.CLEAN_SESSION;
-  if (_username != nullptr) connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.USERNAME;
-  if (_password != nullptr) connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.PASSWORD;
-  if (_willTopic != nullptr) {
-    connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.WILL;
-    if (_willRetain) connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.WILL_RETAIN;
-    switch (_willQos) {
-      case 0:
-        connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.WILL_QOS0;
-        break;
-      case 1:
-        connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.WILL_QOS1;
-        break;
-      case 2:
-        connectFlags[0] |= AsyncMqttClientInternals::ConnectFlag.WILL_QOS2;
-        break;
-    }
-  }
-
-  char keepAliveBytes[2];
-  keepAliveBytes[0] = _keepAlive >> 8;
-  keepAliveBytes[1] = _keepAlive & 0xFF;
-
-  uint16_t clientIdLength = strlen(_clientId);
-  char clientIdLengthBytes[2];
-  clientIdLengthBytes[0] = clientIdLength >> 8;
-  clientIdLengthBytes[1] = clientIdLength & 0xFF;
-
-  // Optional fields
-  uint16_t willTopicLength = 0;
-  char willTopicLengthBytes[2];
-  uint16_t willPayloadLength = _willPayloadLength;
-  char willPayloadLengthBytes[2];
-  if (_willTopic != nullptr) {
-    willTopicLength = strlen(_willTopic);
-    willTopicLengthBytes[0] = willTopicLength >> 8;
-    willTopicLengthBytes[1] = willTopicLength & 0xFF;
-
-    if (_willPayload != nullptr && willPayloadLength == 0) willPayloadLength = strlen(_willPayload);
-
-    willPayloadLengthBytes[0] = willPayloadLength >> 8;
-    willPayloadLengthBytes[1] = willPayloadLength & 0xFF;
-  }
-
-  uint16_t usernameLength = 0;
-  char usernameLengthBytes[2];
-  if (_username != nullptr) {
-    usernameLength = strlen(_username);
-    usernameLengthBytes[0] = usernameLength >> 8;
-    usernameLengthBytes[1] = usernameLength & 0xFF;
-  }
-
-  uint16_t passwordLength = 0;
-  char passwordLengthBytes[2];
-  if (_password != nullptr) {
-    passwordLength = strlen(_password);
-    passwordLengthBytes[0] = passwordLength >> 8;
-    passwordLengthBytes[1] = passwordLength & 0xFF;
-  }
-
-  uint32_t remainingLength = 2 + protocolNameLength + 1 + 1 + 2 + 2 + clientIdLength;  // always present
-  if (_willTopic != nullptr) remainingLength += 2 + willTopicLength + 2 + willPayloadLength;
-  if (_username != nullptr) remainingLength += 2 + usernameLength;
-  if (_password != nullptr) remainingLength += 2 + passwordLength;
-  uint8_t remainingLengthLength = AsyncMqttClientInternals::Helpers::encodeRemainingLength(remainingLength, fixedHeader + 1);
-
-  uint32_t neededSpace = 1 + remainingLengthLength;
-  neededSpace += 2;
-  neededSpace += protocolNameLength;
-  neededSpace += 1;
-  neededSpace += 1;
-  neededSpace += 2;
-  neededSpace += 2;
-  neededSpace += clientIdLength;
-  if (_willTopic != nullptr) {
-    neededSpace += 2;
-    neededSpace += willTopicLength;
-
-    neededSpace += 2;
-    if (_willPayload != nullptr) neededSpace += willPayloadLength;
-  }
-  if (_username != nullptr) {
-    neededSpace += 2;
-    neededSpace += usernameLength;
-  }
-  if (_password != nullptr) {
-    neededSpace += 2;
-    neededSpace += passwordLength;
-  }
-
-  SEMAPHORE_TAKE();
-  if (_client.space() < neededSpace) {
-    _connectPacketNotEnoughSpace = true;
-    _client.close(true);
-    SEMAPHORE_GIVE();
-    return;
-  }
-
-  _client.add(fixedHeader, 1 + remainingLengthLength);
-  _client.add(protocolNameLengthBytes, 2);
-  _client.add("MQTT", protocolNameLength);
-  _client.add(protocolLevel, 1);
-  _client.add(connectFlags, 1);
-  _client.add(keepAliveBytes, 2);
-  _client.add(clientIdLengthBytes, 2);
-  _client.add(_clientId, clientIdLength);
-  if (_willTopic != nullptr) {
-    _client.add(willTopicLengthBytes, 2);
-    _client.add(_willTopic, willTopicLength);
-
-    _client.add(willPayloadLengthBytes, 2);
-    if (_willPayload != nullptr) _client.add(_willPayload, willPayloadLength);
-  }
-  if (_username != nullptr) {
-    _client.add(usernameLengthBytes, 2);
-    _client.add(_username, usernameLength);
-  }
-  if (_password != nullptr) {
-    _client.add(passwordLengthBytes, 2);
-    _client.add(_password, passwordLength);
-  }
-  _client.send();
-  _lastClientActivity = millis();
-  SEMAPHORE_GIVE();
-}
-
-void AsyncMqttClient::_onDisconnect(AsyncClient* client) {
-  (void)client;
-  if (!_disconnectFlagged) {
-    AsyncMqttClientDisconnectReason reason;
-
-    if (_connectPacketNotEnoughSpace) {
-      reason = AsyncMqttClientDisconnectReason::ESP8266_NOT_ENOUGH_SPACE;
-    } else if (_tlsBadFingerprint) {
-      reason = AsyncMqttClientDisconnectReason::TLS_BAD_FINGERPRINT;
-    } else {
-      reason = AsyncMqttClientDisconnectReason::TCP_DISCONNECTED;
-    }
-    for (auto callback : _onDisconnectUserCallbacks) callback(reason);
-  }
-  _clear();
-}
-
-void AsyncMqttClient::_onError(AsyncClient* client, int8_t error) {
-  (void)client;
-  (void)error;
-  // _onDisconnect called anyway
-}
-
-void AsyncMqttClient::_onTimeout(AsyncClient* client, uint32_t time) {
-  (void)client;
-  (void)time;
-  // disconnection will be handled by ping/pong management
-}
-
-void AsyncMqttClient::_onAck(AsyncClient* client, size_t len, uint32_t time) {
-  (void)client;
-  (void)len;
-  (void)time;
-}
-
-void AsyncMqttClient::_onData(AsyncClient* client, char* data, size_t len) {
-  (void)client;
-  size_t currentBytePosition = 0;
-  char currentByte;
-  do {
-    switch (_parsingInformation.bufferState) {
-      case AsyncMqttClientInternals::BufferState::NONE:
-        currentByte = data[currentBytePosition++];
-        _parsingInformation.packetType = currentByte >> 4;
-        _parsingInformation.packetFlags = (currentByte << 4) >> 4;
-        _parsingInformation.bufferState = AsyncMqttClientInternals::BufferState::REMAINING_LENGTH;
-        _lastServerActivity = millis();
-        switch (_parsingInformation.packetType) {
-          case AsyncMqttClientInternals::PacketType.CONNACK:
-            _currentParsedPacket = new AsyncMqttClientInternals::ConnAckPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onConnAck, this, std::placeholders::_1, std::placeholders::_2));
-            break;
-          case AsyncMqttClientInternals::PacketType.PINGRESP:
-            _currentParsedPacket = new AsyncMqttClientInternals::PingRespPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onPingResp, this));
-            break;
-          case AsyncMqttClientInternals::PacketType.SUBACK:
-            _currentParsedPacket = new AsyncMqttClientInternals::SubAckPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onSubAck, this, std::placeholders::_1, std::placeholders::_2));
-            break;
-          case AsyncMqttClientInternals::PacketType.UNSUBACK:
-            _currentParsedPacket = new AsyncMqttClientInternals::UnsubAckPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onUnsubAck, this, std::placeholders::_1));
-            break;
-          case AsyncMqttClientInternals::PacketType.PUBLISH:
-            _currentParsedPacket = new AsyncMqttClientInternals::PublishPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onMessage, this, std::placeholders::_1, std::placeholders::_2, std::placeholders::_3, std::placeholders::_4, std::placeholders::_5, std::placeholders::_6, std::placeholders::_7, std::placeholders::_8, std::placeholders::_9), std::bind(&AsyncMqttClient::_onPublish, this, std::placeholders::_1, std::placeholders::_2));
-            break;
-          case AsyncMqttClientInternals::PacketType.PUBREL:
-            _currentParsedPacket = new AsyncMqttClientInternals::PubRelPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onPubRel, this, std::placeholders::_1));
-            break;
-          case AsyncMqttClientInternals::PacketType.PUBACK:
-            _currentParsedPacket = new AsyncMqttClientInternals::PubAckPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onPubAck, this, std::placeholders::_1));
-            break;
-          case AsyncMqttClientInternals::PacketType.PUBREC:
-            _currentParsedPacket = new AsyncMqttClientInternals::PubRecPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onPubRec, this, std::placeholders::_1));
-            break;
-          case AsyncMqttClientInternals::PacketType.PUBCOMP:
-            _currentParsedPacket = new AsyncMqttClientInternals::PubCompPacket(&_parsingInformation, std::bind(&AsyncMqttClient::_onPubComp, this, std::placeholders::_1));
-            break;
-          default:
-            break;
-        }
-        break;
-      case AsyncMqttClientInternals::BufferState::REMAINING_LENGTH:
-        currentByte = data[currentBytePosition++];
-        _remainingLengthBuffer[_remainingLengthBufferPosition++] = currentByte;
-        if (currentByte >> 7 == 0) {
-          _parsingInformation.remainingLength = AsyncMqttClientInternals::Helpers::decodeRemainingLength(_remainingLengthBuffer);
-          _remainingLengthBufferPosition = 0;
-          if (_parsingInformation.remainingLength > 0) {
-            _parsingInformation.bufferState = AsyncMqttClientInternals::BufferState::VARIABLE_HEADER;
-          } else {
-            // PINGRESP is a special case where it has no variable header, so the packet ends right here
-            _parsingInformation.bufferState = AsyncMqttClientInternals::BufferState::NONE;
-            _onPingResp();
-          }
-        }
-        break;
-      case AsyncMqttClientInternals::BufferState::VARIABLE_HEADER:
-        _currentParsedPacket->parseVariableHeader(data, len, &currentBytePosition);
-        break;
-      case AsyncMqttClientInternals::BufferState::PAYLOAD:
-        _currentParsedPacket->parsePayload(data, len, &currentBytePosition);
-        break;
-      default:
-        currentBytePosition = len;
-    }
-  } while (currentBytePosition != len);
-}
-
-void AsyncMqttClient::_onPoll(AsyncClient* client) {
-  if (!_connected) return;
-
-  // if there is too much time the client has sent a ping request without a response, disconnect client to avoid half open connections
-  if (_lastPingRequestTime != 0 && (millis() - _lastPingRequestTime) >= (_keepAlive * 1000 * 2)) {
-    disconnect();
-    return;
-  // send ping to ensure the server will receive at least one message inside keepalive window
-  } else if (_lastPingRequestTime == 0 && (millis() - _lastClientActivity) >= (_keepAlive * 1000 * 0.7)) {
-    _sendPing();
-
-  // send ping to verify if the server is still there (ensure this is not a half connection)
-  } else if (_connected && _lastPingRequestTime == 0 && (millis() - _lastServerActivity) >= (_keepAlive * 1000 * 0.7)) {
-    _sendPing();
-  }
-
-  // handle to send ack packets
-
-  _sendAcks();
-
-  // handle disconnect
-
-  if (_disconnectFlagged) {
-    _sendDisconnect();
-  }
-}
-
-/* MQTT */
-void AsyncMqttClient::_onPingResp() {
-  _freeCurrentParsedPacket();
-  _lastPingRequestTime = 0;
-}
-
-void AsyncMqttClient::_onConnAck(bool sessionPresent, uint8_t connectReturnCode) {
-  (void)sessionPresent;
-  _freeCurrentParsedPacket();
-
-  if (connectReturnCode == 0) {
-    _connected = true;
-    for (auto callback : _onConnectUserCallbacks) callback(sessionPresent);
-  } else {
-    for (auto callback : _onDisconnectUserCallbacks) callback(static_cast<AsyncMqttClientDisconnectReason>(connectReturnCode));
-    _disconnectFlagged = true;
-  }
-}
-
-void AsyncMqttClient::_onSubAck(uint16_t packetId, char status) {
-  _freeCurrentParsedPacket();
-
-  for (auto callback : _onSubscribeUserCallbacks) callback(packetId, status);
-}
-
-void AsyncMqttClient::_onUnsubAck(uint16_t packetId) {
-  _freeCurrentParsedPacket();
-
-  for (auto callback : _onUnsubscribeUserCallbacks) callback(packetId);
-}
-
-void AsyncMqttClient::_onMessage(char* topic, char* payload, uint8_t qos, bool dup, bool retain, size_t len, size_t index, size_t total, uint16_t packetId) {
-  bool notifyPublish = true;
-
-  if (qos == 2) {
-    for (AsyncMqttClientInternals::PendingPubRel pendingPubRel : _pendingPubRels) {
-      if (pendingPubRel.packetId == packetId) {
-        notifyPublish = false;
-        break;
-      }
-    }
-  }
-
-  if (notifyPublish) {
-    AsyncMqttClientMessageProperties properties;
-    properties.qos = qos;
-    properties.dup = dup;
-    properties.retain = retain;
-
-    for (auto callback : _onMessageUserCallbacks) callback(topic, payload, properties, len, index, total);
-  }
-}
-
-void AsyncMqttClient::_onPublish(uint16_t packetId, uint8_t qos) {
-  AsyncMqttClientInternals::PendingAck pendingAck;
-
-  if (qos == 1) {
-    pendingAck.packetType = AsyncMqttClientInternals::PacketType.PUBACK;
-    pendingAck.headerFlag = AsyncMqttClientInternals::HeaderFlag.PUBACK_RESERVED;
-    pendingAck.packetId = packetId;
-    _toSendAcks.push_back(pendingAck);
-  } else if (qos == 2) {
-    pendingAck.packetType = AsyncMqttClientInternals::PacketType.PUBREC;
-    pendingAck.headerFlag = AsyncMqttClientInternals::HeaderFlag.PUBREC_RESERVED;
-    pendingAck.packetId = packetId;
-    _toSendAcks.push_back(pendingAck);
-
-    bool pubRelAwaiting = false;
-    for (AsyncMqttClientInternals::PendingPubRel pendingPubRel : _pendingPubRels) {
-      if (pendingPubRel.packetId == packetId) {
-        pubRelAwaiting = true;
-        break;
-      }
-    }
-
-    if (!pubRelAwaiting) {
-      AsyncMqttClientInternals::PendingPubRel pendingPubRel;
-      pendingPubRel.packetId = packetId;
-      _pendingPubRels.push_back(pendingPubRel);
-    }
-
-    _sendAcks();
-  }
-
-  _freeCurrentParsedPacket();
-}
-
-void AsyncMqttClient::_onPubRel(uint16_t packetId) {
-  _freeCurrentParsedPacket();
-
-  AsyncMqttClientInternals::PendingAck pendingAck;
-  pendingAck.packetType = AsyncMqttClientInternals::PacketType.PUBCOMP;
-  pendingAck.headerFlag = AsyncMqttClientInternals::HeaderFlag.PUBCOMP_RESERVED;
-  pendingAck.packetId = packetId;
-  _toSendAcks.push_back(pendingAck);
-
-  for (size_t i = 0; i < _pendingPubRels.size(); i++) {
-    if (_pendingPubRels[i].packetId == packetId) {
-      _pendingPubRels.erase(_pendingPubRels.begin() + i);
-      _pendingPubRels.shrink_to_fit();
-    }
-  }
-
-  _sendAcks();
-}
-
-void AsyncMqttClient::_onPubAck(uint16_t packetId) {
-  _freeCurrentParsedPacket();
-
-  for (auto callback : _onPublishUserCallbacks) callback(packetId);
-}
-
-void AsyncMqttClient::_onPubRec(uint16_t packetId) {
-  _freeCurrentParsedPacket();
-
-  AsyncMqttClientInternals::PendingAck pendingAck;
-  pendingAck.packetType = AsyncMqttClientInternals::PacketType.PUBREL;
-  pendingAck.headerFlag = AsyncMqttClientInternals::HeaderFlag.PUBREL_RESERVED;
-  pendingAck.packetId = packetId;
-  _toSendAcks.push_back(pendingAck);
-
-  _sendAcks();
-}
-
-void AsyncMqttClient::_onPubComp(uint16_t packetId) {
-  _freeCurrentParsedPacket();
-
-  for (auto callback : _onPublishUserCallbacks) callback(packetId);
-}
-
-bool AsyncMqttClient::_sendPing() {
-  char fixedHeader[2];
-  fixedHeader[0] = AsyncMqttClientInternals::PacketType.PINGREQ;
-  fixedHeader[0] = fixedHeader[0] << 4;
-  fixedHeader[0] = fixedHeader[0] | AsyncMqttClientInternals::HeaderFlag.PINGREQ_RESERVED;
-  fixedHeader[1] = 0;
-
-  size_t neededSpace = 2;
-
-  SEMAPHORE_TAKE(false);
-  if (_client.space() < neededSpace) { SEMAPHORE_GIVE(); return false; }
-
-  _client.add(fixedHeader, 2);
-  _client.send();
-  _lastClientActivity = millis();
-  _lastPingRequestTime = millis();
-
-  SEMAPHORE_GIVE();
-  return true;
-}
-
-void AsyncMqttClient::_sendAcks() {
-  uint8_t neededAckSpace = 2 + 2;
-
-  SEMAPHORE_TAKE();
-  for (size_t i = 0; i < _toSendAcks.size(); i++) {
-    if (_client.space() < neededAckSpace) break;
-
-    AsyncMqttClientInternals::PendingAck pendingAck = _toSendAcks[i];
-
-    char fixedHeader[2];
-    fixedHeader[0] = pendingAck.packetType;
-    fixedHeader[0] = fixedHeader[0] << 4;
-    fixedHeader[0] = fixedHeader[0] | pendingAck.headerFlag;
-    fixedHeader[1] = 2;
-
-    char packetIdBytes[2];
-    packetIdBytes[0] = pendingAck.packetId >> 8;
-    packetIdBytes[1] = pendingAck.packetId & 0xFF;
-
-    _client.add(fixedHeader, 2);
-    _client.add(packetIdBytes, 2);
-    _client.send();
-
-    _toSendAcks.erase(_toSendAcks.begin() + i);
-    _toSendAcks.shrink_to_fit();
-
-    _lastClientActivity = millis();
-  }
-  SEMAPHORE_GIVE();
-}
-
-bool AsyncMqttClient::_sendDisconnect() {
-  if (!_connected) return true;
-
-  const uint8_t neededSpace = 2;
-
-  SEMAPHORE_TAKE(false);
-
-  if (_client.space() < neededSpace) { SEMAPHORE_GIVE(); return false; }
-
-  char fixedHeader[2];
-  fixedHeader[0] = AsyncMqttClientInternals::PacketType.DISCONNECT;
-  fixedHeader[0] = fixedHeader[0] << 4;
-  fixedHeader[0] = fixedHeader[0] | AsyncMqttClientInternals::HeaderFlag.DISCONNECT_RESERVED;
-  fixedHeader[1] = 0;
-
-  _client.add(fixedHeader, 2);
-  _client.send();
-  _client.close(true);
-
-  _disconnectFlagged = false;
-
-  SEMAPHORE_GIVE();
-  return true;
-}
-
-uint16_t AsyncMqttClient::_getNextPacketId() {
-  uint16_t nextPacketId = _nextPacketId;
-
-  if (_nextPacketId == 65535) _nextPacketId = 0;  // 0 is forbidden
-  _nextPacketId++;
-
-  return nextPacketId;
-}
-
-bool AsyncMqttClient::connected() const {
-  return _connected;
-}
-
-void AsyncMqttClient::connect() {
-  if (_connected) return;
-
-#if ASYNC_TCP_SSL_ENABLED
-  if (_useIp) {
-    _client.connect(_ip, _port, _secure);
-  } else {
-    _client.connect(_host, _port, _secure);
-  }
-#else
-  if (_useIp) {
-    _client.connect(_ip, _port);
-  } else {
-    _client.connect(_host, _port);
-  }
-#endif
-}
-
-void AsyncMqttClient::disconnect(bool force) {
-  if (!_connected) return;
-
-  if (force) {
-    _client.close(true);
-  } else {
-    _disconnectFlagged = true;
-    _sendDisconnect();
-  }
-}
-
-uint16_t AsyncMqttClient::subscribe(const char* topic, uint8_t qos) {
-  if (!_connected) return 0;
-
-  char fixedHeader[5];
-  fixedHeader[0] = AsyncMqttClientInternals::PacketType.SUBSCRIBE;
-  fixedHeader[0] = fixedHeader[0] << 4;
-  fixedHeader[0] = fixedHeader[0] | AsyncMqttClientInternals::HeaderFlag.SUBSCRIBE_RESERVED;
-
-  uint16_t topicLength = strlen(topic);
-  char topicLengthBytes[2];
-  topicLengthBytes[0] = topicLength >> 8;
-  topicLengthBytes[1] = topicLength & 0xFF;
-
-  char qosByte[1];
-  qosByte[0] = qos;
-
-  uint8_t remainingLengthLength = AsyncMqttClientInternals::Helpers::encodeRemainingLength(2 + 2 + topicLength + 1, fixedHeader + 1);
-
-  size_t neededSpace = 0;
-  neededSpace += 1 + remainingLengthLength;
-  neededSpace += 2;
-  neededSpace += 2;
-  neededSpace += topicLength;
-  neededSpace += 1;
-
-  SEMAPHORE_TAKE(0);
-  if (_client.space() < neededSpace) { SEMAPHORE_GIVE(); return 0; }
-
-  uint16_t packetId = _getNextPacketId();
-  char packetIdBytes[2];
-  packetIdBytes[0] = packetId >> 8;
-  packetIdBytes[1] = packetId & 0xFF;
-
-  _client.add(fixedHeader, 1 + remainingLengthLength);
-  _client.add(packetIdBytes, 2);
-  _client.add(topicLengthBytes, 2);
-  _client.add(topic, topicLength);
-  _client.add(qosByte, 1);
-  _client.send();
-  _lastClientActivity = millis();
-
-  SEMAPHORE_GIVE();
-  return packetId;
-}
-
-uint16_t AsyncMqttClient::unsubscribe(const char* topic) {
-  if (!_connected) return 0;
-
-  char fixedHeader[5];
-  fixedHeader[0] = AsyncMqttClientInternals::PacketType.UNSUBSCRIBE;
-  fixedHeader[0] = fixedHeader[0] << 4;
-  fixedHeader[0] = fixedHeader[0] | AsyncMqttClientInternals::HeaderFlag.UNSUBSCRIBE_RESERVED;
-
-  uint16_t topicLength = strlen(topic);
-  char topicLengthBytes[2];
-  topicLengthBytes[0] = topicLength >> 8;
-  topicLengthBytes[1] = topicLength & 0xFF;
-
-  uint8_t remainingLengthLength = AsyncMqttClientInternals::Helpers::encodeRemainingLength(2 + 2 + topicLength, fixedHeader + 1);
-
-  size_t neededSpace = 0;
-  neededSpace += 1 + remainingLengthLength;
-  neededSpace += 2;
-  neededSpace += 2;
-  neededSpace += topicLength;
-
-  SEMAPHORE_TAKE(0);
-  if (_client.space() < neededSpace) { SEMAPHORE_GIVE(); return 0; }
-
-  uint16_t packetId = _getNextPacketId();
-  char packetIdBytes[2];
-  packetIdBytes[0] = packetId >> 8;
-  packetIdBytes[1] = packetId & 0xFF;
-
-  _client.add(fixedHeader, 1 + remainingLengthLength);
-  _client.add(packetIdBytes, 2);
-  _client.add(topicLengthBytes, 2);
-  _client.add(topic, topicLength);
-  _client.send();
-  _lastClientActivity = millis();
-
-  SEMAPHORE_GIVE();
-  return packetId;
-}
-
-uint16_t AsyncMqttClient::publish(const char* topic, uint8_t qos, bool retain, const char* payload, size_t length, bool dup, uint16_t message_id) {
-  if (!_connected) return 0;
-
-  char fixedHeader[5];
-  fixedHeader[0] = AsyncMqttClientInternals::PacketType.PUBLISH;
-  fixedHeader[0] = fixedHeader[0] << 4;
-  if (dup) fixedHeader[0] |= AsyncMqttClientInternals::HeaderFlag.PUBLISH_DUP;
-  if (retain) fixedHeader[0] |= AsyncMqttClientInternals::HeaderFlag.PUBLISH_RETAIN;
-  switch (qos) {
-    case 0:
-      fixedHeader[0] |= AsyncMqttClientInternals::HeaderFlag.PUBLISH_QOS0;
-      break;
-    case 1:
-      fixedHeader[0] |= AsyncMqttClientInternals::HeaderFlag.PUBLISH_QOS1;
-      break;
-    case 2:
-      fixedHeader[0] |= AsyncMqttClientInternals::HeaderFlag.PUBLISH_QOS2;
-      break;
-  }
-
-  uint16_t topicLength = strlen(topic);
-  char topicLengthBytes[2];
-  topicLengthBytes[0] = topicLength >> 8;
-  topicLengthBytes[1] = topicLength & 0xFF;
-
-  uint32_t payloadLength = length;
-  if (payload != nullptr && payloadLength == 0) payloadLength = strlen(payload);
-
-  uint32_t remainingLength = 2 + topicLength + payloadLength;
-  if (qos != 0) remainingLength += 2;
-  uint8_t remainingLengthLength = AsyncMqttClientInternals::Helpers::encodeRemainingLength(remainingLength, fixedHeader + 1);
-
-  size_t neededSpace = 0;
-  neededSpace += 1 + remainingLengthLength;
-  neededSpace += 2;
-  neededSpace += topicLength;
-  if (qos != 0) neededSpace += 2;
-  if (payload != nullptr) neededSpace += payloadLength;
-
-  SEMAPHORE_TAKE(0);
-  if (_client.space() < neededSpace) { SEMAPHORE_GIVE(); return 0; }
-
-  uint16_t packetId = 0;
-  char packetIdBytes[2];
-  if (qos != 0) {
-    if (dup && message_id > 0) {
-      packetId = message_id;
-    } else {
-      packetId = _getNextPacketId();
-    }
-
-    packetIdBytes[0] = packetId >> 8;
-    packetIdBytes[1] = packetId & 0xFF;
-  }
-
-  _client.add(fixedHeader, 1 + remainingLengthLength);
-  _client.add(topicLengthBytes, 2);
-  _client.add(topic, topicLength);
-  if (qos != 0) _client.add(packetIdBytes, 2);
-  if (payload != nullptr) _client.add(payload, payloadLength);
-  _client.send();
-  _lastClientActivity = millis();
-
-  SEMAPHORE_GIVE();
-  if (qos != 0) {
-    return packetId;
-  } else {
-    return 1;
-  }
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient.h

@@ -1,6 +0,0 @@
-#ifndef SRC_ASYNCMQTTCLIENT_H_
-#define SRC_ASYNCMQTTCLIENT_H_
-
-#include "AsyncMqttClient.hpp"
-
-#endif  // SRC_ASYNCMQTTCLIENT_H_

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient.hpp

@@ -1,166 +0,0 @@
-#pragma once
-
-#include <functional>
-#include <vector>
-
-#include "Arduino.h"
-
-#ifdef ESP32
-#include <AsyncTCP.h>
-#include <freertos/semphr.h>
-#elif defined(ESP8266)
-#include <ESPAsyncTCP.h>
-#else
-#error Platform not supported
-#endif
-
-#if ASYNC_TCP_SSL_ENABLED
-#include <tcp_axtls.h>
-#define SHA1_SIZE 20
-#endif
-
-#include "AsyncMqttClient/Flags.hpp"
-#include "AsyncMqttClient/ParsingInformation.hpp"
-#include "AsyncMqttClient/MessageProperties.hpp"
-#include "AsyncMqttClient/Helpers.hpp"
-#include "AsyncMqttClient/Callbacks.hpp"
-#include "AsyncMqttClient/DisconnectReasons.hpp"
-#include "AsyncMqttClient/Storage.hpp"
-
-#include "AsyncMqttClient/Packets/Packet.hpp"
-#include "AsyncMqttClient/Packets/ConnAckPacket.hpp"
-#include "AsyncMqttClient/Packets/PingRespPacket.hpp"
-#include "AsyncMqttClient/Packets/SubAckPacket.hpp"
-#include "AsyncMqttClient/Packets/UnsubAckPacket.hpp"
-#include "AsyncMqttClient/Packets/PublishPacket.hpp"
-#include "AsyncMqttClient/Packets/PubRelPacket.hpp"
-#include "AsyncMqttClient/Packets/PubAckPacket.hpp"
-#include "AsyncMqttClient/Packets/PubRecPacket.hpp"
-#include "AsyncMqttClient/Packets/PubCompPacket.hpp"
-
-#if ESP32
-#define SEMAPHORE_TAKE(X) if (xSemaphoreTake(_xSemaphore, 1000 / portTICK_PERIOD_MS) != pdTRUE) { return X; }  // Waits max 1000ms
-#define SEMAPHORE_GIVE() xSemaphoreGive(_xSemaphore);
-#elif defined(ESP8266)
-#define SEMAPHORE_TAKE(X) void()
-#define SEMAPHORE_GIVE() void()
-#endif
-
-class AsyncMqttClient {
- public:
-  AsyncMqttClient();
-  ~AsyncMqttClient();
-
-  AsyncMqttClient& setKeepAlive(uint16_t keepAlive);
-  AsyncMqttClient& setClientId(const char* clientId);
-  AsyncMqttClient& setCleanSession(bool cleanSession);
-  AsyncMqttClient& setMaxTopicLength(uint16_t maxTopicLength);
-  AsyncMqttClient& setCredentials(const char* username, const char* password = nullptr);
-  AsyncMqttClient& setWill(const char* topic, uint8_t qos, bool retain, const char* payload = nullptr, size_t length = 0);
-  AsyncMqttClient& setServer(IPAddress ip, uint16_t port);
-  AsyncMqttClient& setServer(const char* host, uint16_t port);
-#if ASYNC_TCP_SSL_ENABLED
-  AsyncMqttClient& setSecure(bool secure);
-  AsyncMqttClient& addServerFingerprint(const uint8_t* fingerprint);
-#endif
-
-  AsyncMqttClient& onConnect(AsyncMqttClientInternals::OnConnectUserCallback callback);
-  AsyncMqttClient& onDisconnect(AsyncMqttClientInternals::OnDisconnectUserCallback callback);
-  AsyncMqttClient& onSubscribe(AsyncMqttClientInternals::OnSubscribeUserCallback callback);
-  AsyncMqttClient& onUnsubscribe(AsyncMqttClientInternals::OnUnsubscribeUserCallback callback);
-  AsyncMqttClient& onMessage(AsyncMqttClientInternals::OnMessageUserCallback callback);
-  AsyncMqttClient& onPublish(AsyncMqttClientInternals::OnPublishUserCallback callback);
-
-  bool connected() const;
-  void connect();
-  void disconnect(bool force = false);
-  uint16_t subscribe(const char* topic, uint8_t qos);
-  uint16_t unsubscribe(const char* topic);
-  uint16_t publish(const char* topic, uint8_t qos, bool retain, const char* payload = nullptr, size_t length = 0, bool dup = false, uint16_t message_id = 0);
-
- private:
-  AsyncClient _client;
-
-  bool _connected;
-  bool _connectPacketNotEnoughSpace;
-  bool _disconnectFlagged;
-  bool _tlsBadFingerprint;
-  uint32_t _lastClientActivity;
-  uint32_t _lastServerActivity;
-  uint32_t _lastPingRequestTime;
-
-  char _generatedClientId[13 + 1];  // esp8266abc123
-  IPAddress _ip;
-  const char* _host;
-  bool _useIp;
-#if ASYNC_TCP_SSL_ENABLED
-  bool _secure;
-#endif
-  uint16_t _port;
-  uint16_t _keepAlive;
-  bool _cleanSession;
-  const char* _clientId;
-  const char* _username;
-  const char* _password;
-  const char* _willTopic;
-  const char* _willPayload;
-  uint16_t _willPayloadLength;
-  uint8_t _willQos;
-  bool _willRetain;
-
-#if ASYNC_TCP_SSL_ENABLED
-  std::vector<std::array<uint8_t, SHA1_SIZE>> _secureServerFingerprints;
-#endif
-
-  std::vector<AsyncMqttClientInternals::OnConnectUserCallback> _onConnectUserCallbacks;
-  std::vector<AsyncMqttClientInternals::OnDisconnectUserCallback> _onDisconnectUserCallbacks;
-  std::vector<AsyncMqttClientInternals::OnSubscribeUserCallback> _onSubscribeUserCallbacks;
-  std::vector<AsyncMqttClientInternals::OnUnsubscribeUserCallback> _onUnsubscribeUserCallbacks;
-  std::vector<AsyncMqttClientInternals::OnMessageUserCallback> _onMessageUserCallbacks;
-  std::vector<AsyncMqttClientInternals::OnPublishUserCallback> _onPublishUserCallbacks;
-
-  AsyncMqttClientInternals::ParsingInformation _parsingInformation;
-  AsyncMqttClientInternals::Packet* _currentParsedPacket;
-  uint8_t _remainingLengthBufferPosition;
-  char _remainingLengthBuffer[4];
-
-  uint16_t _nextPacketId;
-
-  std::vector<AsyncMqttClientInternals::PendingPubRel> _pendingPubRels;
-
-  std::vector<AsyncMqttClientInternals::PendingAck> _toSendAcks;
-
-#ifdef ESP32
-  SemaphoreHandle_t _xSemaphore = nullptr;
-#endif
-
-  void _clear();
-  void _freeCurrentParsedPacket();
-
-  // TCP
-  void _onConnect(AsyncClient* client);
-  void _onDisconnect(AsyncClient* client);
-  static void _onError(AsyncClient* client, int8_t error);
-  void _onTimeout(AsyncClient* client, uint32_t time);
-  static void _onAck(AsyncClient* client, size_t len, uint32_t time);
-  void _onData(AsyncClient* client, char* data, size_t len);
-  void _onPoll(AsyncClient* client);
-
-  // MQTT
-  void _onPingResp();
-  void _onConnAck(bool sessionPresent, uint8_t connectReturnCode);
-  void _onSubAck(uint16_t packetId, char status);
-  void _onUnsubAck(uint16_t packetId);
-  void _onMessage(char* topic, char* payload, uint8_t qos, bool dup, bool retain, size_t len, size_t index, size_t total, uint16_t packetId);
-  void _onPublish(uint16_t packetId, uint8_t qos);
-  void _onPubRel(uint16_t packetId);
-  void _onPubAck(uint16_t packetId);
-  void _onPubRec(uint16_t packetId);
-  void _onPubComp(uint16_t packetId);
-
-  bool _sendPing();
-  void _sendAcks();
-  bool _sendDisconnect();
-
-  uint16_t _getNextPacketId();
-};

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Callbacks.hpp

@@ -1,28 +0,0 @@
-#pragma once
-
-#include <functional>
-
-#include "DisconnectReasons.hpp"
-#include "MessageProperties.hpp"
-
-namespace AsyncMqttClientInternals {
-// user callbacks
-typedef std::function<void(bool sessionPresent)> OnConnectUserCallback;
-typedef std::function<void(AsyncMqttClientDisconnectReason reason)> OnDisconnectUserCallback;
-typedef std::function<void(uint16_t packetId, uint8_t qos)> OnSubscribeUserCallback;
-typedef std::function<void(uint16_t packetId)> OnUnsubscribeUserCallback;
-typedef std::function<void(char* topic, char* payload, AsyncMqttClientMessageProperties properties, size_t len, size_t index, size_t total)> OnMessageUserCallback;
-typedef std::function<void(uint16_t packetId)> OnPublishUserCallback;
-
-// internal callbacks
-typedef std::function<void(bool sessionPresent, uint8_t connectReturnCode)> OnConnAckInternalCallback;
-typedef std::function<void()> OnPingRespInternalCallback;
-typedef std::function<void(uint16_t packetId, char status)> OnSubAckInternalCallback;
-typedef std::function<void(uint16_t packetId)> OnUnsubAckInternalCallback;
-typedef std::function<void(char* topic, char* payload, uint8_t qos, bool dup, bool retain, size_t len, size_t index, size_t total, uint16_t packetId)> OnMessageInternalCallback;
-typedef std::function<void(uint16_t packetId, uint8_t qos)> OnPublishInternalCallback;
-typedef std::function<void(uint16_t packetId)> OnPubRelInternalCallback;
-typedef std::function<void(uint16_t packetId)> OnPubAckInternalCallback;
-typedef std::function<void(uint16_t packetId)> OnPubRecInternalCallback;
-typedef std::function<void(uint16_t packetId)> OnPubCompInternalCallback;
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/DisconnectReasons.hpp

@@ -1,15 +0,0 @@
-#pragma once
-
-enum class AsyncMqttClientDisconnectReason : int8_t {
-  TCP_DISCONNECTED = 0,
-
-  MQTT_UNACCEPTABLE_PROTOCOL_VERSION = 1,
-  MQTT_IDENTIFIER_REJECTED = 2,
-  MQTT_SERVER_UNAVAILABLE = 3,
-  MQTT_MALFORMED_CREDENTIALS = 4,
-  MQTT_NOT_AUTHORIZED = 5,
-
-  ESP8266_NOT_ENOUGH_SPACE = 6,
-
-  TLS_BAD_FINGERPRINT = 7
-};

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Flags.hpp

@@ -1,57 +0,0 @@
-#pragma once
-
-namespace AsyncMqttClientInternals {
-constexpr struct {
-  const uint8_t RESERVED    = 0;
-  const uint8_t CONNECT     = 1;
-  const uint8_t CONNACK     = 2;
-  const uint8_t PUBLISH     = 3;
-  const uint8_t PUBACK      = 4;
-  const uint8_t PUBREC      = 5;
-  const uint8_t PUBREL      = 6;
-  const uint8_t PUBCOMP     = 7;
-  const uint8_t SUBSCRIBE   = 8;
-  const uint8_t SUBACK      = 9;
-  const uint8_t UNSUBSCRIBE = 10;
-  const uint8_t UNSUBACK    = 11;
-  const uint8_t PINGREQ     = 12;
-  const uint8_t PINGRESP    = 13;
-  const uint8_t DISCONNECT  = 14;
-  const uint8_t RESERVED2   = 1;
-} PacketType;
-
-constexpr struct {
-  const uint8_t CONNECT_RESERVED     = 0x00;
-  const uint8_t CONNACK_RESERVED     = 0x00;
-  const uint8_t PUBLISH_DUP          = 0x08;
-  const uint8_t PUBLISH_QOS0         = 0x00;
-  const uint8_t PUBLISH_QOS1         = 0x02;
-  const uint8_t PUBLISH_QOS2         = 0x04;
-  const uint8_t PUBLISH_QOSRESERVED  = 0x06;
-  const uint8_t PUBLISH_RETAIN       = 0x01;
-  const uint8_t PUBACK_RESERVED      = 0x00;
-  const uint8_t PUBREC_RESERVED      = 0x00;
-  const uint8_t PUBREL_RESERVED      = 0x02;
-  const uint8_t PUBCOMP_RESERVED     = 0x00;
-  const uint8_t SUBSCRIBE_RESERVED   = 0x02;
-  const uint8_t SUBACK_RESERVED      = 0x00;
-  const uint8_t UNSUBSCRIBE_RESERVED = 0x02;
-  const uint8_t UNSUBACK_RESERVED    = 0x00;
-  const uint8_t PINGREQ_RESERVED     = 0x00;
-  const uint8_t PINGRESP_RESERVED    = 0x00;
-  const uint8_t DISCONNECT_RESERVED  = 0x00;
-  const uint8_t RESERVED2_RESERVED   = 0x00;
-} HeaderFlag;
-
-constexpr struct {
-  const uint8_t USERNAME      = 0x80;
-  const uint8_t PASSWORD      = 0x40;
-  const uint8_t WILL_RETAIN   = 0x20;
-  const uint8_t WILL_QOS0     = 0x00;
-  const uint8_t WILL_QOS1     = 0x08;
-  const uint8_t WILL_QOS2     = 0x10;
-  const uint8_t WILL          = 0x04;
-  const uint8_t CLEAN_SESSION = 0x02;
-  const uint8_t RESERVED      = 0x00;
-} ConnectFlag;
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Helpers.hpp

@@ -1,38 +0,0 @@
-#pragma once
-
-namespace AsyncMqttClientInternals {
-class Helpers {
- public:
-  static uint32_t decodeRemainingLength(char* bytes) {
-    uint32_t multiplier = 1;
-    uint32_t value = 0;
-    uint8_t currentByte = 0;
-    uint8_t encodedByte;
-    do {
-      encodedByte = bytes[currentByte++];
-      value += (encodedByte & 127) * multiplier;
-      multiplier *= 128;
-    } while ((encodedByte & 128) != 0);
-
-    return value;
-  }
-
-  static uint8_t encodeRemainingLength(uint32_t remainingLength, char* destination) {
-    uint8_t currentByte = 0;
-    uint8_t bytesNeeded = 0;
-
-    do {
-      uint8_t encodedByte = remainingLength % 128;
-      remainingLength /= 128;
-      if (remainingLength > 0) {
-        encodedByte = encodedByte | 128;
-      }
-
-      destination[currentByte++] = encodedByte;
-      bytesNeeded++;
-    } while (remainingLength > 0);
-
-    return bytesNeeded;
-  }
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/MessageProperties.hpp

@@ -1,7 +0,0 @@
-#pragma once
-
-struct AsyncMqttClientMessageProperties {
-  uint8_t qos;
-  bool dup;
-  bool retain;
-};

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/ConnAckPacket.cpp

@@ -1,30 +0,0 @@
-#include "ConnAckPacket.hpp"
-
-using AsyncMqttClientInternals::ConnAckPacket;
-
-ConnAckPacket::ConnAckPacket(ParsingInformation* parsingInformation, OnConnAckInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _sessionPresent(false)
-, _connectReturnCode(0) {
-}
-
-ConnAckPacket::~ConnAckPacket() {
-}
-
-void ConnAckPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _sessionPresent = (currentByte << 7) >> 7;
-  } else {
-    _connectReturnCode = currentByte;
-    _parsingInformation->bufferState = BufferState::NONE;
-    _callback(_sessionPresent, _connectReturnCode);
-  }
-}
-
-void ConnAckPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/ConnAckPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class ConnAckPacket : public Packet {
- public:
-  explicit ConnAckPacket(ParsingInformation* parsingInformation, OnConnAckInternalCallback callback);
-  ~ConnAckPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnConnAckInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  bool _sessionPresent;
-  uint8_t _connectReturnCode;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/Packet.hpp

@@ -1,11 +0,0 @@
-#pragma once
-
-namespace AsyncMqttClientInternals {
-class Packet {
- public:
-  virtual ~Packet() {}
-
-  virtual void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) = 0;
-  virtual void parsePayload(char* data, size_t len, size_t* currentBytePosition) = 0;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PingRespPacket.cpp

@@ -1,21 +0,0 @@
-#include "PingRespPacket.hpp"
-
-using AsyncMqttClientInternals::PingRespPacket;
-
-PingRespPacket::PingRespPacket(ParsingInformation* parsingInformation, OnPingRespInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback) {
-}
-
-PingRespPacket::~PingRespPacket() {
-}
-
-void PingRespPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}
-
-void PingRespPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PingRespPacket.hpp

@@ -1,21 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class PingRespPacket : public Packet {
- public:
-  explicit PingRespPacket(ParsingInformation* parsingInformation, OnPingRespInternalCallback callback);
-  ~PingRespPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnPingRespInternalCallback _callback;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubAckPacket.cpp

@@ -1,30 +0,0 @@
-#include "PubAckPacket.hpp"
-
-using AsyncMqttClientInternals::PubAckPacket;
-
-PubAckPacket::PubAckPacket(ParsingInformation* parsingInformation, OnPubAckInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _packetIdMsb(0)
-, _packetId(0) {
-}
-
-PubAckPacket::~PubAckPacket() {
-}
-
-void PubAckPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _parsingInformation->bufferState = BufferState::NONE;
-    _callback(_packetId);
-  }
-}
-
-void PubAckPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubAckPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class PubAckPacket : public Packet {
- public:
-  explicit PubAckPacket(ParsingInformation* parsingInformation, OnPubAckInternalCallback callback);
-  ~PubAckPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnPubAckInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  char _packetIdMsb;
-  uint16_t _packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubCompPacket.cpp

@@ -1,30 +0,0 @@
-#include "PubCompPacket.hpp"
-
-using AsyncMqttClientInternals::PubCompPacket;
-
-PubCompPacket::PubCompPacket(ParsingInformation* parsingInformation, OnPubCompInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _packetIdMsb(0)
-, _packetId(0) {
-}
-
-PubCompPacket::~PubCompPacket() {
-}
-
-void PubCompPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _parsingInformation->bufferState = BufferState::NONE;
-    _callback(_packetId);
-  }
-}
-
-void PubCompPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubCompPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class PubCompPacket : public Packet {
- public:
-  explicit PubCompPacket(ParsingInformation* parsingInformation, OnPubCompInternalCallback callback);
-  ~PubCompPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnPubCompInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  char _packetIdMsb;
-  uint16_t _packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubRecPacket.cpp

@@ -1,30 +0,0 @@
-#include "PubRecPacket.hpp"
-
-using AsyncMqttClientInternals::PubRecPacket;
-
-PubRecPacket::PubRecPacket(ParsingInformation* parsingInformation, OnPubRecInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _packetIdMsb(0)
-, _packetId(0) {
-}
-
-PubRecPacket::~PubRecPacket() {
-}
-
-void PubRecPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _parsingInformation->bufferState = BufferState::NONE;
-    _callback(_packetId);
-  }
-}
-
-void PubRecPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubRecPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class PubRecPacket : public Packet {
- public:
-  explicit PubRecPacket(ParsingInformation* parsingInformation, OnPubRecInternalCallback callback);
-  ~PubRecPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnPubRecInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  char _packetIdMsb;
-  uint16_t _packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubRelPacket.cpp

@@ -1,30 +0,0 @@
-#include "PubRelPacket.hpp"
-
-using AsyncMqttClientInternals::PubRelPacket;
-
-PubRelPacket::PubRelPacket(ParsingInformation* parsingInformation, OnPubRelInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _packetIdMsb(0)
-, _packetId(0) {
-}
-
-PubRelPacket::~PubRelPacket() {
-}
-
-void PubRelPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _parsingInformation->bufferState = BufferState::NONE;
-    _callback(_packetId);
-  }
-}
-
-void PubRelPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PubRelPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class PubRelPacket : public Packet {
- public:
-  explicit PubRelPacket(ParsingInformation* parsingInformation, OnPubRelInternalCallback callback);
-  ~PubRelPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnPubRelInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  char _packetIdMsb;
-  uint16_t _packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PublishPacket.cpp

@@ -1,91 +0,0 @@
-#include "PublishPacket.hpp"
-
-using AsyncMqttClientInternals::PublishPacket;
-
-PublishPacket::PublishPacket(ParsingInformation* parsingInformation, OnMessageInternalCallback dataCallback, OnPublishInternalCallback completeCallback)
-: _parsingInformation(parsingInformation)
-, _dataCallback(dataCallback)
-, _completeCallback(completeCallback)
-, _dup(false)
-, _qos(0)
-, _retain(0)
-, _bytePosition(0)
-, _topicLengthMsb(0)
-, _topicLength(0)
-, _ignore(false)
-, _packetIdMsb(0)
-, _packetId(0)
-, _payloadLength(0)
-, _payloadBytesRead(0) {
-    _dup = _parsingInformation->packetFlags & HeaderFlag.PUBLISH_DUP;
-    _retain = _parsingInformation->packetFlags & HeaderFlag.PUBLISH_RETAIN;
-    char qosMasked = _parsingInformation->packetFlags & 0x06;
-    switch (qosMasked) {
-      case HeaderFlag.PUBLISH_QOS0:
-        _qos = 0;
-        break;
-      case HeaderFlag.PUBLISH_QOS1:
-        _qos = 1;
-        break;
-      case HeaderFlag.PUBLISH_QOS2:
-        _qos = 2;
-        break;
-    }
-}
-
-PublishPacket::~PublishPacket() {
-}
-
-void PublishPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition == 0) {
-    _topicLengthMsb = currentByte;
-  } else if (_bytePosition == 1) {
-    _topicLength = currentByte | _topicLengthMsb << 8;
-    if (_topicLength > _parsingInformation->maxTopicLength) {
-      _ignore = true;
-    } else {
-      _parsingInformation->topicBuffer[_topicLength] = '\0';
-    }
-  } else if (_bytePosition >= 2 && _bytePosition < 2 + _topicLength) {
-    // Starting from here, _ignore might be true
-    if (!_ignore) _parsingInformation->topicBuffer[_bytePosition - 2] = currentByte;
-    if (_bytePosition == 2 + _topicLength - 1 && _qos == 0) {
-      _preparePayloadHandling(_parsingInformation->remainingLength - (_bytePosition + 1));
-      return;
-    }
-  } else if (_bytePosition == 2 + _topicLength) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _preparePayloadHandling(_parsingInformation->remainingLength - (_bytePosition + 1));
-  }
-  _bytePosition++;
-}
-
-void PublishPacket::_preparePayloadHandling(uint32_t payloadLength) {
-  _payloadLength = payloadLength;
-  if (payloadLength == 0) {
-    _parsingInformation->bufferState = BufferState::NONE;
-    if (!_ignore) {
-      _dataCallback(_parsingInformation->topicBuffer, nullptr, _qos, _dup, _retain, 0, 0, 0, _packetId);
-      _completeCallback(_packetId, _qos);
-    }
-  } else {
-    _parsingInformation->bufferState = BufferState::PAYLOAD;
-  }
-}
-
-void PublishPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  size_t remainToRead = len - (*currentBytePosition);
-  if (_payloadBytesRead + remainToRead > _payloadLength) remainToRead = _payloadLength - _payloadBytesRead;
-
-  if (!_ignore) _dataCallback(_parsingInformation->topicBuffer, data + (*currentBytePosition), _qos, _dup, _retain, remainToRead, _payloadBytesRead, _payloadLength, _packetId);
-  _payloadBytesRead += remainToRead;
-  (*currentBytePosition) += remainToRead;
-
-  if (_payloadBytesRead == _payloadLength) {
-    _parsingInformation->bufferState = BufferState::NONE;
-    if (!_ignore) _completeCallback(_packetId, _qos);
-  }
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/PublishPacket.hpp

@@ -1,38 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../Flags.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class PublishPacket : public Packet {
- public:
-  explicit PublishPacket(ParsingInformation* parsingInformation, OnMessageInternalCallback dataCallback, OnPublishInternalCallback completeCallback);
-  ~PublishPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnMessageInternalCallback _dataCallback;
-  OnPublishInternalCallback _completeCallback;
-
-  void _preparePayloadHandling(uint32_t payloadLength);
-
-  bool _dup;
-  uint8_t _qos;
-  bool _retain;
-
-  uint8_t _bytePosition;
-  char _topicLengthMsb;
-  uint16_t _topicLength;
-  bool _ignore;
-  char _packetIdMsb;
-  uint16_t _packetId;
-  uint32_t _payloadLength;
-  uint32_t _payloadBytesRead;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/SubAckPacket.cpp

@@ -1,46 +0,0 @@
-#include "SubAckPacket.hpp"
-
-using AsyncMqttClientInternals::SubAckPacket;
-
-SubAckPacket::SubAckPacket(ParsingInformation* parsingInformation, OnSubAckInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _packetIdMsb(0)
-, _packetId(0) {
-}
-
-SubAckPacket::~SubAckPacket() {
-}
-
-void SubAckPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _parsingInformation->bufferState = BufferState::PAYLOAD;
-  }
-}
-
-void SubAckPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  char status = data[(*currentBytePosition)++];
-
-  /* switch (status) {
-    case 0:
-      Serial.println("Success QoS 0");
-      break;
-    case 1:
-      Serial.println("Success QoS 1");
-      break;
-    case 2:
-      Serial.println("Success QoS 2");
-      break;
-    case 0x80:
-      Serial.println("Failure");
-      break;
-  } */
-
-  _parsingInformation->bufferState = BufferState::NONE;
-  _callback(_packetId, status);
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/SubAckPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class SubAckPacket : public Packet {
- public:
-  explicit SubAckPacket(ParsingInformation* parsingInformation, OnSubAckInternalCallback callback);
-  ~SubAckPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnSubAckInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  char _packetIdMsb;
-  uint16_t _packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/UnsubAckPacket.cpp

@@ -1,30 +0,0 @@
-#include "UnsubAckPacket.hpp"
-
-using AsyncMqttClientInternals::UnsubAckPacket;
-
-UnsubAckPacket::UnsubAckPacket(ParsingInformation* parsingInformation, OnUnsubAckInternalCallback callback)
-: _parsingInformation(parsingInformation)
-, _callback(callback)
-, _bytePosition(0)
-, _packetIdMsb(0)
-, _packetId(0) {
-}
-
-UnsubAckPacket::~UnsubAckPacket() {
-}
-
-void UnsubAckPacket::parseVariableHeader(char* data, size_t len, size_t* currentBytePosition) {
-  char currentByte = data[(*currentBytePosition)++];
-  if (_bytePosition++ == 0) {
-    _packetIdMsb = currentByte;
-  } else {
-    _packetId = currentByte | _packetIdMsb << 8;
-    _parsingInformation->bufferState = BufferState::NONE;
-    _callback(_packetId);
-  }
-}
-
-void UnsubAckPacket::parsePayload(char* data, size_t len, size_t* currentBytePosition) {
-  (void)data;
-  (void)currentBytePosition;
-}

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Packets/UnsubAckPacket.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-#include "Arduino.h"
-#include "Packet.hpp"
-#include "../ParsingInformation.hpp"
-#include "../Callbacks.hpp"
-
-namespace AsyncMqttClientInternals {
-class UnsubAckPacket : public Packet {
- public:
-  explicit UnsubAckPacket(ParsingInformation* parsingInformation, OnUnsubAckInternalCallback callback);
-  ~UnsubAckPacket();
-
-  void parseVariableHeader(char* data, size_t len, size_t* currentBytePosition);
-  void parsePayload(char* data, size_t len, size_t* currentBytePosition);
-
- private:
-  ParsingInformation* _parsingInformation;
-  OnUnsubAckInternalCallback _callback;
-
-  uint8_t _bytePosition;
-  char _packetIdMsb;
-  uint16_t _packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/ParsingInformation.hpp

@@ -1,21 +0,0 @@
-#pragma once
-
-namespace AsyncMqttClientInternals {
-enum class BufferState : uint8_t {
-  NONE = 0,
-  REMAINING_LENGTH = 2,
-  VARIABLE_HEADER = 3,
-  PAYLOAD = 4
-};
-
-struct ParsingInformation {
-  BufferState bufferState;
-
-  uint16_t maxTopicLength;
-  char* topicBuffer;
-
-  uint8_t packetType;
-  uint16_t packetFlags;
-  uint32_t remainingLength;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/AsyncMqttClient/Storage.hpp

@@ -1,13 +0,0 @@
-#pragma once
-
-namespace AsyncMqttClientInternals {
-struct PendingPubRel {
-  uint16_t packetId;
-};
-
-struct PendingAck {
-  uint8_t packetType;
-  uint8_t headerFlag;
-  uint16_t packetId;
-};
-}  // namespace AsyncMqttClientInternals

wled00/src/dependencies/async-mqtt-client/LICENSE

@@ -1,21 +0,0 @@
-The MIT License (MIT)
-
-Copyright (c) 2015 Marvin Roger
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.

wled00/src/dependencies/async-mqtt-client/README.md

@@ -1,18 +0,0 @@
-Async MQTT client for ESP8266 and ESP32 (Github: https://github.com/marvinroger/async-mqtt-client)
-=============================
-
-[![Build Status](https://img.shields.io/travis/marvinroger/async-mqtt-client/master.svg?style=flat-square)](https://travis-ci.org/marvinroger/async-mqtt-client)
-
-An Arduino for ESP8266 and ESP32 asynchronous [MQTT](http://mqtt.org/) client implementation, built on [me-no-dev/ESPAsyncTCP (ESP8266)](https://github.com/me-no-dev/ESPAsyncTCP) | [me-no-dev/AsyncTCP (ESP32)](https://github.com/me-no-dev/AsyncTCP) .
-## Features
-
-* Compliant with the 3.1.1 version of the protocol
-* Fully asynchronous
-* Subscribe at QoS 0, 1 and 2
-* Publish at QoS 0, 1 and 2
-* SSL/TLS support
-* Available in the [PlatformIO registry](http://platformio.org/lib/show/346/AsyncMqttClient)
-
-## Requirements, installation and usage
-
-The project is documented in the [/docs folder](docs).

wled00/src/dependencies/network/Network.cpp

@@ -73,17 +73,13 @@ void NetworkClass::localMAC(uint8_t* MAC)
 
 bool NetworkClass::isConnected()
 {
-#if defined(ARDUINO_ARCH_ESP32) && defined(WLED_USE_ETHERNET)
-  return (WiFi.localIP()[0] != 0 && WiFi.status() == WL_CONNECTED) || ETH.localIP()[0] != 0;
-#else
-  return (WiFi.localIP()[0] != 0 && WiFi.status() == WL_CONNECTED);
-#endif
+  return (WiFi.localIP()[0] != 0 && WiFi.status() == WL_CONNECTED) || isEthernet();
 }
 
 bool NetworkClass::isEthernet()
 {
 #if defined(ARDUINO_ARCH_ESP32) && defined(WLED_USE_ETHERNET)
-  return (ETH.localIP()[0] != 0);
+  return (ETH.localIP()[0] != 0) && ETH.linkUp();
 #endif
   return false;
 }

wled00/udp.cpp

@@ -424,7 +424,7 @@ void realtimeLock(uint32_t timeoutMs, byte md)
     }
     // if strip is off (bri==0) and not already in RTM
     if (briT == 0) {
-      strip.setBrightness(scaledBri(briLast), true);
+      strip.setBrightness(briLast, true);
     }
   }
 
@@ -434,14 +434,14 @@ void realtimeLock(uint32_t timeoutMs, byte md)
   realtimeMode = md;
 
   if (realtimeOverride) return;
-  if (arlsForceMaxBri) strip.setBrightness(scaledBri(255), true);
+  if (arlsForceMaxBri) strip.setBrightness(255, true);
   if (briT > 0 && md == REALTIME_MODE_GENERIC) strip.show();
 }
 
 void exitRealtime() {
   if (!realtimeMode) return;
   if (realtimeOverride == REALTIME_OVERRIDE_ONCE) realtimeOverride = REALTIME_OVERRIDE_NONE;
-  strip.setBrightness(scaledBri(bri), true);
+  strip.setBrightness(bri, true);
   realtimeTimeout = 0; // cancel realtime mode immediately
   realtimeMode = REALTIME_MODE_INACTIVE; // inform UI immediately
   realtimeIP[0] = 0;
@@ -565,93 +565,82 @@ void handleNotifications()
     return;
   }
 
-  if (!receiveDirect) return;
+  if (receiveDirect) {
+    //TPM2.NET
+    if (udpIn[0] == 0x9c) {
+      //WARNING: this code assumes that the final TMP2.NET payload is evenly distributed if using multiple packets (ie. frame size is constant)
+      //if the number of LEDs in your installation doesn't allow that, please include padding bytes at the end of the last packet
+      byte tpmType = udpIn[1];
+      if (tpmType == 0xaa) { //TPM2.NET polling, expect answer
+        sendTPM2Ack(); return;
+      }
+      if (tpmType != 0xda) return; //return if notTPM2.NET data
 
-  //TPM2.NET
-  if (udpIn[0] == 0x9c)
-  {
-    //WARNING: this code assumes that the final TMP2.NET payload is evenly distributed if using multiple packets (ie. frame size is constant)
-    //if the number of LEDs in your installation doesn't allow that, please include padding bytes at the end of the last packet
-    byte tpmType = udpIn[1];
-    if (tpmType == 0xaa) { //TPM2.NET polling, expect answer
-      sendTPM2Ack(); return;
+      realtimeIP = (isSupp) ? notifier2Udp.remoteIP() : notifierUdp.remoteIP();
+      realtimeLock(realtimeTimeoutMs, REALTIME_MODE_TPM2NET);
+      if (realtimeOverride) return;
+
+      tpmPacketCount++; //increment the packet count
+      if (tpmPacketCount == 1) tpmPayloadFrameSize = (udpIn[2] << 8) + udpIn[3]; //save frame size for the whole payload if this is the first packet
+      byte packetNum = udpIn[4]; //starts with 1!
+      byte numPackets = udpIn[5];
+
+      unsigned id = (tpmPayloadFrameSize/3)*(packetNum-1); //start LED
+      unsigned totalLen = strip.getLengthTotal();
+      for (size_t i = 6; i < tpmPayloadFrameSize + 4U && id < totalLen; i += 3, id++) {
+        setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], 0);
+      }
+      if (tpmPacketCount == numPackets) { //reset packet count and show if all packets were received
+        tpmPacketCount = 0;
+        if (useMainSegmentOnly) strip.trigger();
+        else                    strip.show();
+      }
+      return;
     }
-    if (tpmType != 0xda) return; //return if notTPM2.NET data
 
-    realtimeIP = (isSupp) ? notifier2Udp.remoteIP() : notifierUdp.remoteIP();
-    realtimeLock(realtimeTimeoutMs, REALTIME_MODE_TPM2NET);
-    if (realtimeOverride) return;
+    //UDP realtime: 1 warls 2 drgb 3 drgbw 4 dnrgb 5 dnrgbw
+    if (udpIn[0] > 0 && udpIn[0] < 6) {
+      realtimeIP = (isSupp) ? notifier2Udp.remoteIP() : notifierUdp.remoteIP();
+      DEBUG_PRINTLN(realtimeIP);
+      if (packetSize < 2) return;
 
-    tpmPacketCount++; //increment the packet count
-    if (tpmPacketCount == 1) tpmPayloadFrameSize = (udpIn[2] << 8) + udpIn[3]; //save frame size for the whole payload if this is the first packet
-    byte packetNum = udpIn[4]; //starts with 1!
-    byte numPackets = udpIn[5];
+      if (udpIn[1] == 0) {
+        realtimeTimeout = 0; // cancel realtime mode immediately
+        return;
+      } else {
+        realtimeLock(udpIn[1]*1000 +1, REALTIME_MODE_UDP);
+      }
+      if (realtimeOverride) return;
 
-    unsigned id = (tpmPayloadFrameSize/3)*(packetNum-1); //start LED
-    unsigned totalLen = strip.getLengthTotal();
-    for (size_t i = 6; i < tpmPayloadFrameSize + 4U && id < totalLen; i += 3, id++) {
-      setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], 0);
-    }
-    if (tpmPacketCount == numPackets) { //reset packet count and show if all packets were received
-      tpmPacketCount = 0;
+      unsigned totalLen = strip.getLengthTotal();
+      if (udpIn[0] == 1 && packetSize > 5) { //warls
+        for (size_t i = 2; i < packetSize -3; i += 4) {
+          setRealtimePixel(udpIn[i], udpIn[i+1], udpIn[i+2], udpIn[i+3], 0);
+        }
+      } else if (udpIn[0] == 2 && packetSize > 4) { //drgb
+        for (size_t i = 2, id = 0; i < packetSize -2 && id < totalLen; i += 3, id++)
+          {
+            setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], 0);
+          }
+      } else if (udpIn[0] == 3 && packetSize > 6) { //drgbw
+          for (size_t i = 2, id = 0; i < packetSize -3 && id < totalLen; i += 4, id++) {
+            setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], udpIn[i+3]);
+          }
+      } else if (udpIn[0] == 4 && packetSize > 7) { //dnrgb
+        unsigned id = ((udpIn[3] << 0) & 0xFF) + ((udpIn[2] << 8) & 0xFF00);
+        for (size_t i = 4; i < packetSize -2 && id < totalLen; i += 3, id++) {
+          setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], 0);
+        }
+      } else if (udpIn[0] == 5 && packetSize > 8) { //dnrgbw
+        unsigned id = ((udpIn[3] << 0) & 0xFF) + ((udpIn[2] << 8) & 0xFF00);
+        for (size_t i = 4; i < packetSize -2 && id < totalLen; i += 4, id++) {
+          setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], udpIn[i+3]);
+        }
+      }
       if (useMainSegmentOnly) strip.trigger();
       else                    strip.show();
-    }
-    return;
-  }
-
-  //UDP realtime: 1 warls 2 drgb 3 drgbw 4 dnrgb 5 dnrgbw
-  if (udpIn[0] > 0 && udpIn[0] < 6)
-  {
-    realtimeIP = (isSupp) ? notifier2Udp.remoteIP() : notifierUdp.remoteIP();
-    DEBUG_PRINTLN(realtimeIP);
-    if (packetSize < 2) return;
-
-    if (udpIn[1] == 0) {
-      realtimeTimeout = 0; // cancel realtime mode immediately
       return;
-    } else {
-      realtimeLock(udpIn[1]*1000 +1, REALTIME_MODE_UDP);
     }
-    if (realtimeOverride) return;
-
-    unsigned totalLen = strip.getLengthTotal();
-    if (udpIn[0] == 1 && packetSize > 5) //warls
-    {
-      for (size_t i = 2; i < packetSize -3; i += 4)
-      {
-        setRealtimePixel(udpIn[i], udpIn[i+1], udpIn[i+2], udpIn[i+3], 0);
-      }
-    } else if (udpIn[0] == 2 && packetSize > 4) //drgb
-    {
-      for (size_t i = 2, id = 0; i < packetSize -2 && id < totalLen; i += 3, id++)
-      {
-        setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], 0);
-      }
-    } else if (udpIn[0] == 3 && packetSize > 6) //drgbw
-    {
-      for (size_t i = 2, id = 0; i < packetSize -3 && id < totalLen; i += 4, id++)
-      {
-        setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], udpIn[i+3]);
-      }
-    } else if (udpIn[0] == 4 && packetSize > 7) //dnrgb
-    {
-      unsigned id = ((udpIn[3] << 0) & 0xFF) + ((udpIn[2] << 8) & 0xFF00);
-      for (size_t i = 4; i < packetSize -2 && id < totalLen; i += 3, id++)
-      {
-        setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], 0);
-      }
-    } else if (udpIn[0] == 5 && packetSize > 8) //dnrgbw
-    {
-      unsigned id = ((udpIn[3] << 0) & 0xFF) + ((udpIn[2] << 8) & 0xFF00);
-      for (size_t i = 4; i < packetSize -2 && id < totalLen; i += 4, id++)
-      {
-        setRealtimePixel(id, udpIn[i], udpIn[i+1], udpIn[i+2], udpIn[i+3]);
-      }
-    }
-    if (useMainSegmentOnly) strip.trigger();
-    else                    strip.show();
-    return;
   }
 
   // API over UDP
@@ -669,6 +658,8 @@ void handleNotifications()
     }
     releaseJSONBufferLock();
   }
+
+  UsermodManager::onUdpPacket(udpIn, packetSize);
 }
 
 

wled00/um_manager.cpp

@@ -68,6 +68,10 @@ bool UsermodManager::onEspNowMessage(uint8_t* sender, uint8_t* payload, uint8_t
   return false;
 }
 #endif
+bool UsermodManager::onUdpPacket(uint8_t* payload, size_t len) {
+  for (auto mod = _usermod_table_begin; mod < _usermod_table_end; ++mod) if ((*mod)->onUdpPacket(payload, len)) return true;
+  return false;
+}
 void UsermodManager::onUpdateBegin(bool init) { for (auto mod = _usermod_table_begin; mod < _usermod_table_end; ++mod) (*mod)->onUpdateBegin(init); } // notify usermods that update is to begin
 void UsermodManager::onStateChange(uint8_t mode) { for (auto mod = _usermod_table_begin; mod < _usermod_table_end; ++mod) (*mod)->onStateChange(mode); } // notify usermods that WLED state changed
 

wled00/util.cpp

@@ -1,6 +1,16 @@
 #include "wled.h"
 #include "fcn_declare.h"
 #include "const.h"
+#ifdef ESP8266
+#include "user_interface.h" // for bootloop detection
+#else
+#include <Update.h>
+#if ESP_IDF_VERSION >= ESP_IDF_VERSION_VAL(4, 4, 0)
+  #include "esp32/rtc.h"    // for bootloop detection
+#elif ESP_IDF_VERSION >= ESP_IDF_VERSION_VAL(3, 3, 0)
+  #include "soc/rtc.h"
+#endif
+#endif
 
 
 //helper to get int value at a position in string
@@ -619,7 +629,8 @@ int32_t hw_random(int32_t lowerlimit, int32_t upperlimit) {
   return hw_random(diff) + lowerlimit;
 }
 
-#ifndef ESP8266
+#if !defined(ESP8266) && !defined(CONFIG_IDF_TARGET_ESP32C3) // ESP8266 does not support PSRAM, ESP32-C3 does not have PSRAM
+// p_x prefer PSRAM, d_x prefer DRAM
 void *p_malloc(size_t size) {
   int caps1 = MALLOC_CAP_SPIRAM  | MALLOC_CAP_8BIT;
   int caps2 = MALLOC_CAP_DEFAULT | MALLOC_CAP_8BIT;
@@ -640,6 +651,14 @@ void *p_realloc(void *ptr, size_t size) {
   return heap_caps_realloc(ptr, size, caps2);
 }
 
+// realloc with malloc fallback, original buffer is freed if realloc fails but not copied!
+void *p_realloc_malloc(void *ptr, size_t size) {
+  void *newbuf = p_realloc(ptr, size); // try realloc first
+  if (newbuf) return newbuf; // realloc successful
+  p_free(ptr); // free old buffer if realloc failed
+  return p_malloc(size); // fallback to malloc
+}
+
 void *p_calloc(size_t count, size_t size) {
   int caps1 = MALLOC_CAP_SPIRAM  | MALLOC_CAP_8BIT;
   int caps2 = MALLOC_CAP_DEFAULT | MALLOC_CAP_8BIT;
@@ -654,7 +673,7 @@ void *d_malloc(size_t size) {
   int caps1 = MALLOC_CAP_DEFAULT | MALLOC_CAP_8BIT;
   int caps2 = MALLOC_CAP_SPIRAM  | MALLOC_CAP_8BIT;
   if (psramSafe) {
-    if (size > MIN_HEAP_SIZE) std::swap(caps1, caps2);  // prefer PSRAM for large alloactions
+    if (heap_caps_get_largest_free_block(caps1) < 3*MIN_HEAP_SIZE && size > MIN_HEAP_SIZE) std::swap(caps1, caps2);  // prefer PSRAM for large alloactions & when DRAM is low
     return heap_caps_malloc_prefer(size, 2, caps1, caps2); // otherwise prefer DRAM
   }
   return heap_caps_malloc(size, caps1);
@@ -664,12 +683,20 @@ void *d_realloc(void *ptr, size_t size) {
   int caps1 = MALLOC_CAP_DEFAULT | MALLOC_CAP_8BIT;
   int caps2 = MALLOC_CAP_SPIRAM  | MALLOC_CAP_8BIT;
   if (psramSafe) {
-    if (size > MIN_HEAP_SIZE) std::swap(caps1, caps2);  // prefer PSRAM for large alloactions
+    if (heap_caps_get_largest_free_block(caps1) < 3*MIN_HEAP_SIZE && size > MIN_HEAP_SIZE) std::swap(caps1, caps2);  // prefer PSRAM for large alloactions & when DRAM is low
     return heap_caps_realloc_prefer(ptr, size, 2, caps1, caps2); // otherwise prefer DRAM
   }
   return heap_caps_realloc(ptr, size, caps1);
 }
 
+// realloc with malloc fallback, original buffer is freed if realloc fails but not copied!
+void *d_realloc_malloc(void *ptr, size_t size) {
+  void *newbuf = d_realloc(ptr, size); // try realloc first
+  if (newbuf) return newbuf; // realloc successful
+  d_free(ptr); // free old buffer if realloc failed
+  return d_malloc(size); // fallback to malloc
+}
+
 void *d_calloc(size_t count, size_t size) {
   int caps1 = MALLOC_CAP_DEFAULT | MALLOC_CAP_8BIT;
   int caps2 = MALLOC_CAP_SPIRAM  | MALLOC_CAP_8BIT;
@@ -679,8 +706,157 @@ void *d_calloc(size_t count, size_t size) {
   }
   return heap_caps_calloc(count, size, caps1);
 }
+#else // ESP8266 & ESP32-C3
+// realloc with malloc fallback, original buffer is freed if realloc fails but not copied!
+void *realloc_malloc(void *ptr, size_t size) {
+  void *newbuf = realloc(ptr, size); // try realloc first
+  if (newbuf) return newbuf; // realloc successful
+  free(ptr); // free old buffer if realloc failed
+  return malloc(size); // fallback to malloc
+}
 #endif
 
+// bootloop detection and handling
+// checks if the ESP reboots multiple times due to a crash or watchdog timeout
+// if a bootloop is detected: restore settings from backup, then reset settings, then switch boot image (and repeat)
+
+#define BOOTLOOP_INTERVAL_MILLIS 120000  // time limit between crashes: 120 seconds (2 minutes)
+#define BOOTLOOP_THRESHOLD       5     // number of consecutive crashes to trigger bootloop detection
+#define BOOTLOOP_ACTION_RESTORE  0     // default action: restore config from /bkp.cfg.json
+#define BOOTLOOP_ACTION_RESET    1     // if restore does not work, reset config (rename /cfg.json to /rst.cfg.json)
+#define BOOTLOOP_ACTION_OTA      2     // swap the boot partition
+#define BOOTLOOP_ACTION_DUMP     3     // nothing seems to help, dump files to serial and reboot (until hardware reset)
+
+// Platform-agnostic abstraction
+enum class ResetReason {
+  Power,
+  Software,
+  Crash,
+  Brownout
+};
+
+#ifdef ESP8266
+// Place variables in RTC memory via references, since RTC memory is not exposed via the linker in the Non-OS SDK
+// Use an offset of 32 as there's some hints that the first 128 bytes of "user" memory are used by the OTA system
+// Ref: https://github.com/esp8266/Arduino/blob/78d0d0aceacc1553f45ad8154592b0af22d1eede/cores/esp8266/Esp.cpp#L168
+static volatile uint32_t& bl_last_boottime = *(RTC_USER_MEM + 32);
+static volatile uint32_t& bl_crashcounter = *(RTC_USER_MEM + 33);
+static volatile uint32_t& bl_actiontracker = *(RTC_USER_MEM + 34);
+
+static inline ResetReason rebootReason() {
+  uint32_t resetReason = system_get_rst_info()->reason;
+  if (resetReason == REASON_EXCEPTION_RST
+      || resetReason == REASON_WDT_RST
+      || resetReason == REASON_SOFT_WDT_RST)
+      return ResetReason::Crash;
+  if (resetReason == REASON_SOFT_RESTART)
+    return ResetReason::Software;
+  return ResetReason::Power;
+}
+
+static inline uint32_t getRtcMillis() { return system_get_rtc_time() / 160; };  // rtc ticks ~160000Hz
+
+#else
+// variables in RTC_NOINIT memory persist between reboots (but not on hardware reset)
+RTC_NOINIT_ATTR static uint32_t bl_last_boottime;
+RTC_NOINIT_ATTR static uint32_t bl_crashcounter;
+RTC_NOINIT_ATTR static uint32_t bl_actiontracker;
+
+static inline ResetReason rebootReason() {
+  esp_reset_reason_t reason = esp_reset_reason();
+  if (reason == ESP_RST_BROWNOUT) return ResetReason::Brownout;
+  if (reason == ESP_RST_SW) return ResetReason::Software;
+  if (reason == ESP_RST_PANIC || reason == ESP_RST_WDT || reason == ESP_RST_INT_WDT || reason == ESP_RST_TASK_WDT) return ResetReason::Crash;
+  return ResetReason::Power;
+}
+
+#if ESP_IDF_VERSION >= ESP_IDF_VERSION_VAL(4, 4, 0)
+static inline uint32_t getRtcMillis() { return esp_rtc_get_time_us() / 1000; }
+#elif ESP_IDF_VERSION >= ESP_IDF_VERSION_VAL(3, 3, 0)
+static inline uint32_t getRtcMillis() { return rtc_time_slowclk_to_us(rtc_time_get(), rtc_clk_slow_freq_get_hz()) / 1000; }
+#endif
+
+void bootloopCheckOTA() { bl_actiontracker = BOOTLOOP_ACTION_OTA; } // swap boot image if bootloop is detected instead of restoring config
+
+#endif
+
+// detect bootloop by checking the reset reason and the time since last boot
+static bool detectBootLoop() {
+  uint32_t rtctime = getRtcMillis();
+  bool result = false;
+
+  switch(rebootReason()) {
+    case ResetReason::Power:
+      bl_actiontracker = BOOTLOOP_ACTION_RESTORE; // init action tracker if not an intentional reboot (e.g. from OTA or bootloop handler)
+      // fall through
+    case ResetReason::Software:
+      // no crash detected, reset counter
+      bl_crashcounter = 0;
+      break;
+
+    case ResetReason::Crash:
+    {
+      DEBUG_PRINTLN(F("crash detected!"));
+      uint32_t rebootinterval = rtctime - bl_last_boottime;
+      if (rebootinterval < BOOTLOOP_INTERVAL_MILLIS) {
+        bl_crashcounter++;
+        if (bl_crashcounter >= BOOTLOOP_THRESHOLD) {
+          DEBUG_PRINTLN(F("!BOOTLOOP DETECTED!"));
+          bl_crashcounter = 0;
+          result = true;
+        }
+      } else {
+        // Reset counter on long intervals to track only consecutive short-interval crashes
+        bl_crashcounter = 0;
+        // TODO: crash reporting goes here
+      }
+      break;
+    }
+
+    case ResetReason::Brownout:
+      // crash due to brownout can't be detected unless using flash memory to store bootloop variables
+      DEBUG_PRINTLN(F("brownout detected"));
+      //restoreConfig(); // TODO: blindly restoring config if brownout detected is a bad idea, need a better way (if at all)
+      break;
+  }
+
+  bl_last_boottime = rtctime; // store current runtime for next reboot
+
+  return result;
+}
+
+void handleBootLoop() {
+  DEBUG_PRINTF_P(PSTR("checking for bootloop: time %d, counter %d, action %d\n"), bl_last_boottime, bl_crashcounter, bl_actiontracker);
+  if (!detectBootLoop()) return; // no bootloop detected
+
+  switch(bl_actiontracker) {
+    case BOOTLOOP_ACTION_RESTORE:
+      restoreConfig();
+      ++bl_actiontracker;
+      break;
+    case BOOTLOOP_ACTION_RESET:
+      resetConfig();
+      ++bl_actiontracker;
+      break;
+    case BOOTLOOP_ACTION_OTA:
+#ifndef ESP8266
+      if(Update.canRollBack()) {
+        DEBUG_PRINTLN(F("Swapping boot partition..."));
+        Update.rollBack(); // swap boot partition
+      }
+      ++bl_actiontracker;
+      break;
+#else
+      // fall through
+#endif
+    case BOOTLOOP_ACTION_DUMP:
+      dumpFilesToSerial();
+      break;
+  }
+
+  ESP.restart(); // restart cleanly and don't wait for another crash
+}
+
 /*
  * Fixed point integer based Perlin noise functions by @dedehai
  * Note: optimized for speed and to mimic fastled inoise functions, not for accuracy or best randomness

wled00/wled.cpp

@@ -1,7 +1,10 @@
 #define WLED_DEFINE_GLOBAL_VARS //only in one source file, wled.cpp!
 #include "wled.h"
 #include "wled_ethernet.h"
-#include <Arduino.h>
+#ifdef WLED_ENABLE_AOTA
+  #define NO_OTA_PORT
+  #include <ArduinoOTA.h>
+#endif
 
 #if defined(ARDUINO_ARCH_ESP32) && defined(WLED_DISABLE_BROWNOUT_DET)
 #include "soc/soc.h"
@@ -105,8 +108,8 @@ void WLED::loop()
   if (!realtimeMode || realtimeOverride || (realtimeMode && useMainSegmentOnly))  // block stuff if WARLS/Adalight is enabled
   {
     if (apActive) dnsServer.processNextRequest();
-    #ifndef WLED_DISABLE_OTA
-    if (WLED_CONNECTED && aOtaEnabled && !otaLock && correctPIN) ArduinoOTA.handle();
+    #ifdef WLED_ENABLE_AOTA
+    if (Network.isConnected() && aOtaEnabled && !otaLock && correctPIN) ArduinoOTA.handle();
     #endif
     handleNightlight();
     yield();
@@ -187,12 +190,10 @@ void WLED::loop()
     doInitBusses = false;
     DEBUG_PRINTLN(F("Re-init busses."));
     bool aligned = strip.checkSegmentAlignment(); //see if old segments match old bus(ses)
-    BusManager::removeAll();
     strip.finalizeInit(); // will create buses and also load default ledmap if present
-    BusManager::setBrightness(bri); // fix re-initialised bus' brightness #4005
     if (aligned) strip.makeAutoSegments();
     else strip.fixInvalidSegments();
-    BusManager::setBrightness(bri); // fix re-initialised bus' brightness
+    BusManager::setBrightness(scaledBri(bri)); // fix re-initialised bus' brightness #4005 and #4824
     configNeedsWrite = true;
   }
   if (loadLedmap >= 0) {
@@ -407,6 +408,9 @@ void WLED::setup()
     DEBUGFS_PRINTLN(F("FS failed!"));
     errorFlag = ERR_FS_BEGIN;
   }
+
+  handleBootLoop(); // check for bootloop and take action (requires WLED_FS)
+
 #ifdef WLED_ADD_EEPROM_SUPPORT
   else deEEP();
 #else
@@ -422,6 +426,11 @@ void WLED::setup()
   WLED_SET_AP_SSID(); // otherwise it is empty on first boot until config is saved
   multiWiFi.push_back(WiFiConfig(CLIENT_SSID,CLIENT_PASS)); // initialise vector with default WiFi
 
+  if(!verifyConfig()) {
+    if(!restoreConfig()) {
+      resetConfig();
+    }
+  }
   DEBUG_PRINTLN(F("Reading config"));
   bool needsCfgSave = deserializeConfigFromFS();
   DEBUG_PRINTF_P(PSTR("heap %u\n"), ESP.getFreeHeap());
@@ -471,7 +480,7 @@ void WLED::setup()
   if (mqttClientID[0] == 0)    sprintf_P(mqttClientID, PSTR("WLED-%*s"), 6, escapedMac.c_str() + 6);
 #endif
 
-#ifndef WLED_DISABLE_OTA
+#ifdef WLED_ENABLE_AOTA
   if (aOtaEnabled) {
     ArduinoOTA.onStart([]() {
       #ifdef ESP8266
@@ -711,9 +720,8 @@ void WLED::initInterfaces()
     alexaInit();
 #endif
 
-#ifndef WLED_DISABLE_OTA
-  if (aOtaEnabled)
-    ArduinoOTA.begin();
+#ifdef WLED_ENABLE_AOTA
+  if (aOtaEnabled) ArduinoOTA.begin();
 #endif
 
   // Set up mDNS responder:
@@ -784,7 +792,7 @@ void WLED::handleConnection()
     if (stac != stacO) {
       stacO = stac;
       DEBUG_PRINTF_P(PSTR("Connected AP clients: %d\n"), (int)stac);
-      if (!WLED_CONNECTED && wifiConfigured) {        // trying to connect, but not connected
+      if (!Network.isConnected() && wifiConfigured) {        // trying to connect, but not connected
         if (stac)
           WiFi.disconnect();        // disable search so that AP can work
         else
@@ -859,7 +867,7 @@ void WLED::handleConnection()
 }
 
 // If status LED pin is allocated for other uses, does nothing
-// else blink at 1Hz when WLED_CONNECTED is false (no WiFi, ?? no Ethernet ??)
+// else blink at 1Hz when Network.isConnected() is false (no WiFi, ?? no Ethernet ??)
 // else blink at 2Hz when MQTT is enabled but not connected
 // else turn the status LED off
 #if defined(STATUSLED)
@@ -873,7 +881,7 @@ void WLED::handleStatusLED()
   }
   #endif
 
-  if (WLED_CONNECTED) {
+  if (Network.isConnected()) {
     c = RGBW32(0,255,0,0);
     ledStatusType = 2;
   } else if (WLED_MQTT_CONNECTED) {

wled00/wled.h

@@ -21,6 +21,12 @@
 
 // You are required to disable over-the-air updates:
 //#define WLED_DISABLE_OTA         // saves 14kb
+#ifdef WLED_ENABLE_AOTA
+  #if defined(WLED_DISABLE_OTA)
+    #warning WLED_DISABLE_OTA was defined but it will be ignored due to WLED_ENABLE_AOTA.
+  #endif
+  #undef WLED_DISABLE_OTA
+#endif
 
 // You can choose some of these features to disable:
 //#define WLED_DISABLE_ALEXA       // saves 11kb
@@ -121,10 +127,6 @@
 #endif
 #include <WiFiUdp.h>
 #include <DNSServer.h>
-#ifndef WLED_DISABLE_OTA
-  #define NO_OTA_PORT
-  #include <ArduinoOTA.h>
-#endif
 #include <SPIFFSEditor.h>
 #include "src/dependencies/time/TimeLib.h"
 #include "src/dependencies/timezone/Timezone.h"
@@ -153,7 +155,7 @@
 
 #include "src/dependencies/e131/ESPAsyncE131.h"
 #ifndef WLED_DISABLE_MQTT
-#include "src/dependencies/async-mqtt-client/AsyncMqttClient.h"
+#include <AsyncMqttClient.h>
 #endif
 
 #define ARDUINOJSON_DECODE_UNICODE 0
@@ -192,6 +194,7 @@ using PSRAMDynamicJsonDocument = BasicJsonDocument<PSRAM_Allocator>;
 #include "fcn_declare.h"
 #include "NodeStruct.h"
 #include "pin_manager.h"
+#include "colors.h"
 #include "bus_manager.h"
 #include "FX.h"
 
@@ -588,7 +591,7 @@ WLED_GLOBAL bool otaLock        _INIT(true);     // prevents OTA firmware update
 WLED_GLOBAL bool otaLock        _INIT(false);     // prevents OTA firmware updates without password. ALWAYS enable if system exposed to any public networks
 #endif
 WLED_GLOBAL bool wifiLock       _INIT(false);     // prevents access to WiFi settings when OTA lock is enabled
-#ifndef WLED_DISABLE_OTA
+#ifdef WLED_ENABLE_AOTA
 WLED_GLOBAL bool aOtaEnabled    _INIT(true);      // ArduinoOTA allows easy updates directly from the IDE. Careful, it does not auto-disable when OTA lock is on
 #else
 WLED_GLOBAL bool aOtaEnabled    _INIT(false);     // ArduinoOTA allows easy updates directly from the IDE. Careful, it does not auto-disable when OTA lock is on
@@ -623,7 +626,6 @@ WLED_GLOBAL unsigned long transitionStartTime;
 WLED_GLOBAL bool          jsonTransitionOnce       _INIT(false);  // flag to override transitionDelay (playlist, JSON API: "live" & "seg":{"i"} & "tt")
 WLED_GLOBAL uint8_t       randomPaletteChangeTime  _INIT(5);      // amount of time [s] between random palette changes (min: 1s, max: 255s)
 WLED_GLOBAL bool          useHarmonicRandomPalette _INIT(true);   // use *harmonic* random palette generation (nicer looking) or truly random
-WLED_GLOBAL bool          useRainbowWheel          _INIT(false);  // use "rainbow" color wheel instead of "spectrum" color wheel
 
 // nightlight
 WLED_GLOBAL bool nightlightActive _INIT(false);
@@ -732,10 +734,10 @@ WLED_GLOBAL bool receiveNotificationPalette    _INIT(true);       // apply palet
 WLED_GLOBAL bool receiveSegmentOptions         _INIT(false);      // apply segment options
 WLED_GLOBAL bool receiveSegmentBounds          _INIT(false);      // apply segment bounds (start, stop, offset)
 WLED_GLOBAL bool receiveDirect _INIT(true);                       // receive UDP/Hyperion realtime
-WLED_GLOBAL bool notifyDirect _INIT(false);                       // send notification if change via UI or HTTP API
-WLED_GLOBAL bool notifyButton _INIT(false);                       // send if updated by button or infrared remote
+WLED_GLOBAL bool notifyDirect _INIT(true);                        // send notification if change via UI or HTTP API
+WLED_GLOBAL bool notifyButton _INIT(true);                        // send if updated by button or infrared remote
 WLED_GLOBAL bool notifyAlexa  _INIT(false);                       // send notification if updated via Alexa
-WLED_GLOBAL bool notifyHue    _INIT(true);                        // send notification if Hue light changes
+WLED_GLOBAL bool notifyHue    _INIT(false);                       // send notification if Hue light changes
 #endif
 
 // effects
@@ -1024,11 +1026,7 @@ WLED_GLOBAL volatile uint8_t jsonBufferLock _INIT(0);
   WLED_GLOBAL unsigned loops _INIT(0);
 #endif
 
-#ifdef ARDUINO_ARCH_ESP32
-  #define WLED_CONNECTED (WiFi.status() == WL_CONNECTED || ETH.localIP()[0] != 0)
-#else
-  #define WLED_CONNECTED (WiFi.status() == WL_CONNECTED)
-#endif
+#define WLED_CONNECTED (Network.isConnected())
 
 #ifndef WLED_AP_SSID_UNIQUE
   #define WLED_SET_AP_SSID() do { \

wled00/wled_server.cpp

@@ -1,9 +1,11 @@
 #include "wled.h"
 
-#ifdef ESP8266
-  #include <Updater.h>
-#else
-  #include <Update.h>
+#ifndef WLED_DISABLE_OTA
+  #ifdef ESP8266
+    #include <Updater.h>
+  #else
+    #include <Update.h>
+  #endif
 #endif
 #include "html_ui.h"
 #include "html_settings.h"
@@ -387,6 +389,7 @@ void initServer()
   createEditHandler(correctPIN);
 
   static const char _update[] PROGMEM = "/update";
+#ifndef WLED_DISABLE_OTA
   //init ota page
   server.on(_update, HTTP_GET, [](AsyncWebServerRequest *request){
     if (otaLock) {
@@ -408,6 +411,9 @@ void initServer()
       serveMessage(request, 500, F("Update failed!"), F("Please check your file and retry!"), 254);
     } else {
       serveMessage(request, 200, F("Update successful!"), FPSTR(s_rebooting), 131);
+      #ifndef ESP8266
+      bootloopCheckOTA(); // let the bootloop-checker know there was an OTA update
+      #endif
       doReboot = true;
     }
   },[](AsyncWebServerRequest *request, String filename, size_t index, uint8_t *data, size_t len, bool isFinal){
@@ -426,8 +432,9 @@ void initServer()
       UsermodManager::onUpdateBegin(true); // notify usermods that update is about to begin (some may require task de-init)
       lastEditTime = millis(); // make sure PIN does not lock during update
       strip.suspend();
-      #ifdef ESP8266
+      backupConfig(); // backup current config in case the update ends badly
       strip.resetSegments();  // free as much memory as you can
+      #ifdef ESP8266
       Update.runAsync(true);
       #endif
       Update.begin((ESP.getFreeSketchSpace() - 0x1000) & 0xFFFFF000);
@@ -446,14 +453,17 @@ void initServer()
       }
     }
   });
+#else
+  const auto notSupported = [](AsyncWebServerRequest *request){
+    serveMessage(request, 501, FPSTR(s_notimplemented), F("This build does not support OTA update."), 254);
+  };
+  server.on(_update, HTTP_GET, notSupported);
+  server.on(_update, HTTP_POST, notSupported, [](AsyncWebServerRequest *request, String filename, size_t index, uint8_t *data, size_t len, bool isFinal){});
+#endif
 
 #ifdef WLED_ENABLE_DMX
   server.on(F("/dmxmap"), HTTP_GET, [](AsyncWebServerRequest *request){
-    request->send_P(200, FPSTR(CONTENT_TYPE_HTML), PAGE_dmxmap     , dmxProcessor);
-  });
-#else
-  server.on(F("/dmxmap"), HTTP_GET, [](AsyncWebServerRequest *request){
-    serveMessage(request, 501, FPSTR(s_notimplemented), F("DMX support is not enabled in this build."), 254);
+    request->send_P(200, FPSTR(CONTENT_TYPE_HTML), PAGE_dmxmap, dmxProcessor);
   });
 #endif
 
@@ -657,6 +667,7 @@ void serveSettings(AsyncWebServerRequest* request, bool post) {
     case SUBPAGE_DMX     :  content = PAGE_settings_dmx;  len = PAGE_settings_dmx_length;  break;
 #endif
     case SUBPAGE_UM      :  content = PAGE_settings_um;   len = PAGE_settings_um_length;   break;
+#ifndef WLED_DISABLE_OTA
     case SUBPAGE_UPDATE  :  content = PAGE_update;        len = PAGE_update_length;
       #ifdef ARDUINO_ARCH_ESP32
       if (request->hasArg(F("revert")) && inLocalSubnet(request->client()->remoteIP()) && Update.canRollBack()) {
@@ -670,6 +681,7 @@ void serveSettings(AsyncWebServerRequest* request, bool post) {
       }
       #endif
       break;
+#endif
 #ifndef WLED_DISABLE_2D
     case SUBPAGE_2D      :  content = PAGE_settings_2D;   len = PAGE_settings_2D_length;   break;
 #endif

wled00/xml.cpp

@@ -26,7 +26,8 @@ void XML_response(Print& dest)
   );
 }
 
-static void extractPin(Print& settingsScript, const JsonObject &obj, const char *key) {
+static void extractPin(Print& settingsScript, const JsonObject &obj, const char *key)
+{
   if (obj[key].is<JsonArray>()) {
     JsonArray pins = obj[key].as<JsonArray>();
     for (JsonVariant pv : pins) {
@@ -37,6 +38,22 @@ static void extractPin(Print& settingsScript, const JsonObject &obj, const char
   }
 }
 
+void fillWLEDVersion(char *buf, size_t len)
+{
+  if (!buf || len == 0) return;
+
+  snprintf_P(buf,len,PSTR("WLED %s (%d)<br>\\\"%s\\\"<br>(Processor: %s)"),
+    versionString,
+    VERSION,
+    releaseString,
+  #if defined(ARDUINO_ARCH_ESP32)
+    ESP.getChipModel()
+  #else
+    "ESP8266"
+  #endif
+  );
+}
+
 // print used pins by scanning JsonObject (1 level deep)
 static void fillUMPins(Print& settingsScript, const JsonObject &mods)
 {
@@ -72,7 +89,8 @@ static void fillUMPins(Print& settingsScript, const JsonObject &mods)
   }
 }
 
-void appendGPIOinfo(Print& settingsScript) {
+void appendGPIOinfo(Print& settingsScript)
+{
   settingsScript.print(F("d.um_p=[-1")); // has to have 1 element
   if (i2c_sda > -1 && i2c_scl > -1) {
     settingsScript.printf_P(PSTR(",%d,%d"), i2c_sda, i2c_scl);
@@ -311,6 +329,7 @@ void getSettingsJS(byte subPage, Print& settingsScript)
       char sp[4] = "SP"; sp[2] = offset+s; sp[3] = 0; //bus clock speed
       char la[4] = "LA"; la[2] = offset+s; la[3] = 0; //LED current
       char ma[4] = "MA"; ma[2] = offset+s; ma[3] = 0; //max per-port PSU current
+      char hs[4] = "HS"; hs[2] = offset+s; hs[3] = 0; //hostname (for network types, custom text for others)
       settingsScript.print(F("addLEDs(1);"));
       uint8_t pins[5];
       int nPins = bus->getPins(pins);
@@ -350,6 +369,7 @@ void getSettingsJS(byte subPage, Print& settingsScript)
       printSetFormValue(settingsScript,sp,speed);
       printSetFormValue(settingsScript,la,bus->getLEDCurrent());
       printSetFormValue(settingsScript,ma,bus->getMaxCurrent());
+      printSetFormValue(settingsScript,hs,bus->getCustomText().c_str());
       sumMa += bus->getMaxCurrent();
     }
     printSetFormValue(settingsScript,PSTR("MA"),BusManager::ablMilliampsMax() ? BusManager::ablMilliampsMax() : sumMa);
@@ -380,7 +400,6 @@ void getSettingsJS(byte subPage, Print& settingsScript)
     printSetFormValue(settingsScript,PSTR("TL"),nightlightDelayMinsDefault);
     printSetFormValue(settingsScript,PSTR("TW"),nightlightMode);
     printSetFormIndex(settingsScript,PSTR("PB"),paletteBlend);
-    printSetFormCheckbox(settingsScript,PSTR("RW"),useRainbowWheel);
     printSetFormValue(settingsScript,PSTR("RL"),rlyPin);
     printSetFormCheckbox(settingsScript,PSTR("RM"),rlyMde);
     printSetFormCheckbox(settingsScript,PSTR("RO"),rlyOpenDrain);
@@ -593,11 +612,14 @@ void getSettingsJS(byte subPage, Print& settingsScript)
     printSetFormCheckbox(settingsScript,PSTR("AO"),aOtaEnabled);
     printSetFormCheckbox(settingsScript,PSTR("SU"),otaSameSubnet);
     char tmp_buf[128];
-    snprintf_P(tmp_buf,sizeof(tmp_buf),PSTR("WLED %s (build %d)"),versionString,VERSION);
+    fillWLEDVersion(tmp_buf,sizeof(tmp_buf));
     printSetClassElementHTML(settingsScript,PSTR("sip"),0,tmp_buf);
     settingsScript.printf_P(PSTR("sd=\"%s\";"), serverDescription);
-    #ifdef WLED_DISABLE_OTA
     //hide settings if not compiled
+    #ifdef WLED_DISABLE_OTA
+    settingsScript.print(F("toggle('OTA');"));  // hide update section
+    #endif
+    #ifndef WLED_ENABLE_AOTA
     settingsScript.print(F("toggle('aOTA');"));  // hide ArduinoOTA checkbox
     #endif
   }
@@ -652,16 +674,7 @@ void getSettingsJS(byte subPage, Print& settingsScript)
   if (subPage == SUBPAGE_UPDATE) // update
   {
     char tmp_buf[128];
-    snprintf_P(tmp_buf,sizeof(tmp_buf),PSTR("WLED %s<br>%s<br>(%s build %d)"),
-      versionString,
-      releaseString,
-    #if defined(ARDUINO_ARCH_ESP32)
-      ESP.getChipModel(),
-    #else
-      "esp8266",
-    #endif
-      VERSION);
-
+    fillWLEDVersion(tmp_buf,sizeof(tmp_buf));
     printSetClassElementHTML(settingsScript,PSTR("sip"),0,tmp_buf);
     #ifndef ARDUINO_ARCH_ESP32
     settingsScript.print(F("toggle('rev');"));  // hide revert button on ESP8266