Open Source in the Era of Vibe Coding

Add firmware/WebUI version compatibility check with startup warning and serial logging

Maslow-Main ← copilot/fix-197

opened 09:22PM - 20 Sep 25 UTC

+179 -0

This PR implements an automatic version compatibility check that runs during sta…rtup to warn users when firmware and WebUI versions may not be compatible. ## What's Changed - **Automatic Version Check**: After successful firmware connection, the system now compares firmware version with WebUI version - **Smart Version Parsing**: Handles multiple version formats including semantic versioning (`3.6.7`), git hashes (`abc1234`), and mixed formats (`v3.6.7 (devt-abc1234)`) - **User-Friendly Warning**: Shows a clear modal dialog when versions don't match, displaying both version strings and explaining potential issues - **Serial Log Integration**: Version comparison results are now logged to the serial messages textarea for user visibility - **Persistent Modal Display**: Modal warning now appears with proper timing to prevent automatic dismissal during UI initialization - **Non-Intrusive**: Compatible versions pass silently without interrupting the user workflow ## Implementation Details The feature integrates seamlessly into the existing connection flow by adding a single call to `checkVersionCompatibility()` in the `connectsuccess()` function. The implementation includes: - **Version extraction logic** that parses various version string formats - **Compatibility comparison** that handles semantic versioning and git hash matching - **Modal warning dialog** using the existing `alertdlg()` system for UI consistency - **Serial message logging** using the `addMessage()` function to display version info in the visible serial log - **Modal timing control** with 3-second delay and conflict resolution to ensure proper display ## Example Scenarios **Incompatible versions detected:** ``` Firmware: FluidNC v3.5.0 (devt-old123) WebUI: github.com/MitchBradley/ESP3D-WEBUI@new456 → Shows warning popup with version details → Logs "WARNING: Version mismatch detected!" to serial messages ``` **Compatible versions:** ``` Firmware: FluidNC v3.6.7 (devt-abc1234) WebUI: github.com/MitchBradley/ESP3D-WEBUI@abc1234 → Silent pass, no interruption → Logs "Version compatibility check PASSED" to serial messages ``` This helps prevent unexpected behavior or missing features that can occur when firmware and WebUI versions are mismatched, improving the overall user experience by providing early visibility into potential compatibility issues through both popup warnings and serial log messages. Fixes #197. --- 💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click [here](https://survey3.medallia.com/?EAHeSx-AP01bZqG0Ld9QLQ) to start the survey.

popup to warn when the versions of firmware and UI don’t match (beware, this
will be annoying during development as they aren’t going to match)

Add bit change functionality with fixed feedrate F300 while preserving all existing functionality

Maslow-Main ← copilot/continue-bit-change-work

opened 08:45AM - 18 Oct 25 UTC

+253 -25

- [x] Add bit change functionality from PR #202 - [x] Add bit change button to… tablettab.html (row 3, column 5) - [x] Preserve trace boundary button in dynamic_button_1 (row 2, column 5) - [x] Add bit change height preference field in unified Maslow panel - [x] Add bitChangeHeight to default preferences in prefDefs.js (default: 70mm) - [x] Add bit change functionality to grbl.js with F300 feedrate - [x] Wire up button click handler in tablet.js - [x] Add bit change height support in preferencesdlg.js - [x] Use fixed feedrate F300 for raise bit command - [x] Default bit change height: 70mm - [x] Preserve ALL existing functionality (no removals) - [x] Update probe detection to check $probe/pin setting - [x] Add response parsing and caching for probe pin detection - [x] Update probe state dynamically when response arrives - [x] Combine Frame definition and Maslow settings into unified panel - [x] Fix bit change height preference saving and validation - [x] Initialize bit change height from defaults when not in preferences - [x] Reduce debug logging to essential messages - [x] Build and test the changes ## Summary of Changes This PR adds **only the bit change feature** from PR #202 without removing any existing functionality. All existing features like `jogWithUnitsSafeguard()`, `updateJobBoundsDisplay()`, `fetchYamlFiles()`, and trace boundary button functionality are preserved. ### Debug Logging Reduction **Removed verbose debugging console logs:** Now that the bit change feature is working correctly, removed the extensive debug logging that was added during development: - Removed detailed step-by-step logging from `StartBitChangeProcess()` - Removed verbose probe detection logging - Removed detailed button state update logging - Kept only essential error messages for troubleshooting The console output is now much cleaner while still maintaining critical error logging. ### Bug Fix: Default Value Initialization **Fixed default value initialization for bit change height:** When a system doesn't have `bitChangeHeight` defined in preferences (first-time use), the system now properly uses the default value from `default_preferenceslist[0]`: - Updated `onpreferenceloadedCallback()` to check `default_preferenceslist[0]` for missing probe values - Updated `StartBitChangeProcess()` to check `default_preferenceslist[0]` for missing bit change height - Added fallback to hardcoded default (70mm) if `default_preferenceslist` is unavailable - The default value is now immediately usable without requiring a save operation The bit change feature now works correctly on first use without needing to open and save preferences. ### Bug Fix: Bit Change Height Preference **Fixed saving and reading of bit change height:** - Added validation check for `preferences_bitchangeheight` in `getPreferencesForSave()` - Added `bitChangeHeight` to the saved preferences list - The bit change height value is now properly saved to and loaded from preferences - The value is correctly used when moving to bit change height ### Preferences Organization **Unified Maslow Panel:** The preferences dialog now features a single "Maslow" panel at the top (after language selection) that consolidates all Maslow-specific settings: - **Config filename**: Frame definition configuration (maslow.yaml) - **Bit change height**: Bit change height setting (default: 70mm) This provides better organization by grouping related settings together in one panel rather than splitting them across multiple panels. ### Probe Detection Update Complete probe detection implementation with dynamic state update: - Added `queryProbePin()` function to query `$probe/pin` setting - Added response handler to parse `$/probe/pin=gpio.48:pd` style responses - Stores probe pin value in `grbl.probePin` for caching - Re-evaluates probe availability when response arrives during bit change process - Updates button state dynamically from "Lower bit" to "Probe for bit length" - `isProbeAvailable()` checks cached value and falls back to `grbl.pins` ### Key Implementation Details **Bit Change Command with Fixed F300:** ```javascript const cmd = `$J=G53G90F300Z${probeValues.bitChangeHeight.value}`; ``` **Dynamic Probe State Update:** When the `$/probe/pin` response arrives after the bit change process starts, the system now: 1. Re-evaluates probe availability 2. Updates the `bitChangeState.probeEnabled` flag 3. Updates the button text to "Probe for bit length" if probe is detected **Default Bit Change Height:** 70mm (configurable through preferences in the unified Maslow panel) **No Functionality Removed:** All existing code preserved - only additions made for bit change feature ### Files Changed (6 total) 1. **www/js/grbl.js** - Added bit change logic + probe detection + reduced debug logging (190 lines added/modified) 2. **www/js/prefDefs.js** - Added bitChangeHeight: "70" (2 lines added, 2 modified) 3. **www/js/preferencesdlg.js** - Added bit change height support + fixed saving (4 lines added/modified) 4. **www/js/tablet.js** - Added button event listener (17 lines added) 5. **www/sub/preferencesdlg.html** - Unified Maslow panel with config filename and bit change height (reorganized preferences structure) 6. **www/sub/tablettab.html** - Added bit change button + preserved trace boundary (10 lines added, 14 modified) ### Build Status ✅ Build successful: 130.34 kB (compressed) ✅ No functionality removed ✅ All existing features preserved ✅ Probe detection parses and caches $probe/pin responses ✅ Button updates dynamically when probe response arrives during bit change ✅ Bit change height preference saves and loads correctly ✅ Bit change height uses default value immediately on first use ✅ Default values properly initialized from default_preferenceslist ✅ Debug logging reduced to essential messages only Fixes BarbourSmith/ESP3D-WEBUI#233 <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>[FEATURE REQUEST] continue work on bit change</issue_title> > <issue_description>include all the changes from all the files in the BarbourSmith/ESP3D-WEBUI#202 branch copilot/fix-201 commit cfdd66ff727a3f02f642400d02bc20d8122d5a96 > > change the raise bit to use $J=G53G90F300Z<bit change height> > > check $probe.pin to see if it is NO_PIN (case insensitive), if it is, then change the button to lower bit as you do now > If it is set to anything else, change the button to 'probe length' and when pressed issue G38.6 Z-<bit change height +5mm> F100 > > </issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/ESP3D-WEBUI#233 --- 💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click [here](https://survey3.medallia.com/?EAHeSx-AP01bZqG0Ld9QLQ) to start the survey.

bit change height (including probing if configured)

Add real-time connection monitoring to calibration popup

Maslow-Main ← copilot/add-connection-status-display

opened 07:05AM - 18 Oct 25 UTC

+192 -12

## Task: Add connection monitoring to calibration popup ### Implementation Comp…lete ✅ Successfully implemented real-time connection monitoring in the calibration popup that displays the number of active WebSocket and Telnet connections. ### Final Implementation The feature now correctly handles $CI command responses that come via WebSocket. Based on user testing with actual firmware, the $CI response format is: ``` $CI usbcdc macros websocket telnet ok ``` ### Key Discovery: WebSocket Response Handling Through iterative debugging with console logs, we discovered that **FluidNC responds to `$CI` commands via WebSocket, not HTTP**. This explained why: - HTTP responses were always empty - Channel names appeared in console logs before HTTP response was processed - They were logged from WebSocket message handlers ### Changes Made 1. **HTML (www/sub/tablettab.html)**: - Split state label row into left (state) and right (connection info) columns - Uses Bootstrap grid for responsive layout 2. **JavaScript (www/js/tablet.js)**: - `startConnectionInfoPolling()` - Queries firmware with `$CI` every second - `stopConnectionInfoPolling()` - Cleanup when popup closes - Accumulates WebSocket responses over 500ms before processing - Counts "websocket" and "telnet" occurrences - Updates display with "Web:X Tel:Y" format - Color-coded backgrounds: - 🔴 Red (#ffcccc) - >1 websocket connection - 🟢 Green (#ccffcc) - 1 websocket connection - ⚪ Gray (#eeeeee) - 0 websocket connections - **Clean code**: Debug logging removed - **FIXED**: Added filtering in `tabletShowMessage()` to prevent $CI channel names from appearing in Serial Messages text area 3. **JavaScript (www/js/grbl.js)**: - Added WebSocket message handler for $CI responses - Captures channel names: "usbcdc", "macros", "websocket", "telnet" - Routes to `accumulateConnectionInfo()` function - Precise matching prevents capturing "ok" or other messages - **Clean code**: Debug logging removed 4. **JavaScript (www/js/socket.js)**: - Filters $CI channel names from both `Monitor_output_Update()` AND `console.log()` - **Uses exact matching** (^word$ in regex terms) - filters "websocket" but NOT "websocket connection established" - Prevents clutter in both console AND WebSocket message handling 5. **JavaScript (www/js/printercmd.js)**: - Direct `SendPrinterCommand("$CI", false, null, null)` call - Avoids callback conflicts with position updates - **Clean code**: Debug logging removed 6. **JavaScript (www/js/http.js)**: - **Clean code**: Debug logging removed 7. **Test Server (fluidnc-web-sim.py)**: - Mock $CI handler for local testing - Returns proper plain-text format ### How It Works 1. Calibration popup opens → Start polling $CI every 1000ms 2. $CI command sent via HTTP (triggers firmware to respond) 3. Firmware responds asynchronously via WebSocket with channel names 4. WebSocket handler captures: usbcdc, macros, websocket (×N), telnet (×N) 5. **Channel names filtered at MULTIPLE levels:** - socket.js: Prevents calling `Monitor_output_Update()` and `console.log()` - tablet.js: `tabletShowMessage()` filters before calling `addMessage()` (Serial Messages display) - This ensures complete filtering from all output locations 6. After 500ms, accumulated channel names are processed 7. Count occurrences: websocketCount, telnetCount 8. Display updates: "Web:1 Tel:1" with green background 9. Popup closes → Stop polling ### Console Filtering Details The filter uses **exact line matching** (equivalent to `^telnet$` in regex) at multiple levels: - ✅ Filtered from Serial Messages: "websocket", "telnet", "usbcdc", "macros" (when they appear alone) - ❌ NOT filtered: "websocket connection established", "telnet: connected to 192.168.1.100" This ensures that $CI polling doesn't clutter either the browser console OR the Serial Messages panel while preserving meaningful status messages that happen to contain these words. ### Testing Tested with actual FluidNC firmware via telnet: ``` $CI usbcdc macros websocket telnet ok ``` Expected behavior: - **Serial Messages panel is completely clean** (bare channel names fully filtered) - Browser console is clean (bare channel names filtered, no debug output) - Messages containing these words in context are preserved in both places - Connection info display updates correctly showing "Web:X Tel:Y" - Color-coded background reflects connection status ### Benefits - **Prevents calibration conflicts**: Visual warning for multiple browsers - **Color-coded feedback**: Instant recognition of connection status - **Clean Serial Messages output**: $CI bare channel names completely filtered from text area at source - **Clean console output**: No debug logging cluttering the console - **Minimal overhead**: Only polls when calibration popup is open - **Non-intrusive**: Seamlessly integrated into existing UI - **Real-time updates**: Refreshes every second - **Production ready**: All debug code removed, clean output everywhere ### Build Info - Final build size: 129.54 kB (compressed) - All filtering working correctly at multiple levels Fixes BarbourSmith/ESP3D-WEBUI#231 <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>[FEATURE REQUEST] avoid duplicate browsers when calibrating</issue_title> > <issue_description>In the popup that you get when you click the setup gear, add a display (updated once a second) that shows webconnect and telnet connections as reported by the firmware with the '$CI' command > If there is more than one webconnect listed, make the background red, if only one is listed, make the background green. > > Currently the top of this shows the state, split that line in half with the left showing the state and have this new display in the right half.</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/ESP3D-WEBUI#231 --- 💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more [Copilot coding agent tips](https://gh.io/copilot-coding-agent-tips) in the docs.

shows number of connections on setup popup (turns red if there are multiple
browser connections, since that messes up the current calibration)

Display git version alongside firmware size during builds

Maslow-Main ← copilot/add-version-display-on-builds

opened 01:45AM - 24 Oct 25 UTC

+11 -0

## Summary This PR implements the feature request from issue #242 to display the… git version alongside the firmware size when compiling with gulp and in GitHub Actions workflow outputs. ## Changes Modified `gulpfile.js` to display the git version information during the build process: - Added a global variable `buildVersion` to store the version computed by `git describe --tags --always --dirty` - Modified the `replaceVersion()` function to save the computed version for later use - Modified the `compress()` function to display the version before showing the file size Modified `.github/workflows/compile-webui.yml` to display version in PR comments: - Added a "Get version" step that captures the git version using `git describe --tags --always --dirty` - Modified the PR comment template to include the version in the "Build Details" section ## Build Output The gulp build now displays both the version and the compressed file size: ``` Version: v1.13-2-g89351d3 [01:50:07] Size index.html.gz : 129.74 kB ``` The GitHub Actions workflow now includes version in the PR comment: ``` **Build Details:** - Version: `v1.13-2-g89351d3` - Commit: `5b090efbca628c3e8ea3317bb66c676944bbdd47` - Language: `en` (English) - WebUI size: 129.74 KB (compressed) - Build time: ~3 seconds ``` This provides better visibility into which version is being built, making it easier to track builds and identify potential issues. The version information is especially useful when troubleshooting or comparing different builds. ## Testing Tested with multiple language builds: - English (`gulp package --lang en`) - Spanish (`gulp package --lang es`) - French (`gulp package --lang fr`) - Multi-language (`gulp package`) All builds correctly display the version information alongside the file size. GitHub Actions workflow YAML syntax validated successfully. ## Security CodeQL analysis completed with 0 alerts - no security vulnerabilities introduced. Fixes MaslowCNC/ESP3D-WEBUI#242 <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>[FEATURE REQUEST] show version on builds</issue_title> > <issue_description>I gulpfile.js you compute: > var version = execSync('git describe --tags --always --dirty') > > This should be displayed along with the firmware size when compiling</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes MaslowCNC/ESP3D-WEBUI#242 --- 💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more [Copilot coding agent tips](https://gh.io/copilot-coding-agent-tips) in the docs.

change the notice that is sent out when you have copilot compile code that
includes the version number as reported, not just the commit id

UI PRs to close

test for release

UI PRs that may need to be abandoned/restarted due to age

It's time to replace `Gulp`

Maslow-Main ← aersida:gulp_down_the_bun

opened 11:54PM - 14 Nov 24 UTC

aersida

+9738 -8620

We're stuck on `gulp` v4, and the upgrade path is painful. More importantly is t…hat `gulp` is effectively abandonware. In reality we need to ditch `gulp` and its way of doing things with something better. Right now, the thing that is 'better' is also much more comprehensive, and built on the knowledge from the last several years of how to do JS/TS based projects. And that is [bun](bun.sh) This PR shifts over to bun, replacing all of the gulpisms along the way. As much as possible, plain js/ts techniques are used, the idea is that things should just work, even if bun was not the tech chosen.

replace gulp (with copilot support, this may be much easier to do, or it may
become a horrid mess)

Implement right-click to move functionality on gcode canvas with left-click only view cycling

Maslow-Main ← copilot/fix-155

opened 05:59PM - 10 Aug 25 UTC

+174 -2

- [x] Implement right-click context menu on gcode canvas - [x] Add "Move here" o…ption with coordinates - [x] Fix left-click only view cycling - [x] Add coordinate conversion from canvas to machine coordinates - [x] Add smart view restriction to top-down views --- 💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click [here](https://survey.alchemer.com/s3/8343779/Copilot-Coding-agent) to start the survey.

change right click behavior on the canvas (I just kicked off instructions to
maslowbot to see if this can be recovered)

UI PR to leave as-is

Add alternate calibration algorithm with Gauss-Newton Least Squares

Maslow-Main ← copilot/add-alternate-calibration-routine

opened 07:25AM - 13 Oct 25 UTC

+430 -2

- [x] Include all changes from PR #220 commit 81b71aea (6 files) - [x] Fix findF…itnessGaussNewtonLeastSquares to add missing setTimeout for calibration restart - [x] Fix sendCommand calls to use `kinematics/MaslowKinematics/` prefix - [x] Add missing calibration parameters (blX=0, blY=0, trY=0) - now sends all 8 parameters - [x] Rename algorithm option to descriptive name - [x] Fix parameter mapping bug: trY should use y2, brY should be 0 - [x] Add sendCalibrationEvent call for consistency with classic algorithm - [x] Build and test - Build successful (130.07 kB) **Summary of Changes:** 1. Added Calibration Algorithm dropdown to preferences UI (Maslow Classic / Gauss-Newton Least Squares with outlier detection) 2. Integrated Gauss-Newton Least Squares algorithm into calculatesCalibrationStuff.js 3. Fixed sendCommand to use correct `$/kinematics/MaslowKinematics/` prefix for all parameters 4. Added missing setTimeout to restart calibration process for next stage 5. Added commands to set blX=0.0, blY=0.0, and brY=0.0 (these are always 0 by definition) 6. Now sends all 8 corner parameters: tlX, tlY, trX, trY, blX, blY, brX, brY 7. Renamed algorithm option to "Gauss-Newton Least Squares with outlier detection" for clarity 8. Fixed parameter mapping bug: trY now correctly uses y2 parameter, brY is always 0 9. **Added sendCalibrationEvent call** to match classic algorithm behavior and ensure proper event flow <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>[FEATURE REQUEST] allow alternate calibration calculation</issue_title> > <issue_description>comment (not for copilot) > adds support (in UI and in code) to have multiple calibration routines. all must accept the same input and output the same output > > add the Gauss-Newton Least Squares algorithm (as requested by David Lang from Grok and shown in the test page at http://lang.hm/maslow/calibrator_grok.html > > Instructions for copilot > > include all changes to all 6 files in PR BarbourSmith/ESP3D-WEBUI#220 commit 81b71aea0b6a0a3e70a627356c8c274c8494711c on branch copilot/add-algorithm-selection-option. > > Then fix the following problems > > the old findMaxFitness function also had the lines > ``` > // This restarts calibration process for the next stage > setTimeout(() => { onCalibrationButtonsClick('$CAL', 'Calibrate'); }, 2000); > ``` > These need to be in the new function as well > > the lines > > sendCommand($/Maslow_tlX=${x3.toFixed(1)}); > sendCommand($/Maslow_tlY=${y3.toFixed(1)}); > sendCommand($/Maslow_trX=${x2.toFixed(1)}); > sendCommand($/Maslow_brY=${y2.toFixed(1)}); > > are incorrect, instead of Maslow_ they need to have kinematics/MaslowKinematics/ > > fix this</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/ESP3D-WEBUI#221 --- 💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more [Copilot coding agent tips](https://gh.io/copilot-coding-agent-tips) in the docs.

add alternate calibration calculation, will need to be rebased to continue to be
useful

firmware PRs to merge

Improve memory usage through string and container optimizations with shared log buffer

Maslow-Main ← copilot/improve-memory-usage

opened 12:22AM - 18 Oct 25 UTC

+231 -125

## Problem The FluidNC firmware was running low on available dynamic memory dur…ing operation, requiring optimization to reduce memory footprint and improve runtime stability. ## Solution This PR implements targeted memory optimizations that eliminate temporary heap allocations and reduce container overhead, resulting in measurable memory savings: - **Flash memory**: Reduced from 65.3% to 64.5% (saved 25,976 bytes / ~0.84%) - **RAM**: Static allocation increased by 1,180 bytes (for shared buffer), but eliminates dynamic heap allocations during runtime ## Changes ### 1. String Allocation Optimizations Replaced heap-allocating string operations with stack-based `snprintf()`: **`log_telem_pt_csv()` in Maslow.cpp** - **Before**: Used ~40 `std::to_string()` calls with string concatenation, creating numerous temporary heap allocations per telemetry data point - **After**: Single `snprintf()` call with a 512-byte stack buffer - **Impact**: Eliminated ~40 temporary allocations per telemetry point, reduced code size, improved execution speed **`formatBytes()` in NutsBolts.cpp** - **Before**: Used `std::ostringstream` with significant heap overhead - **After**: `snprintf()` with 32-byte stack buffer - **Impact**: Eliminated heap allocations and removed dependency on `<sstream>` and `<iomanip>` headers **`IP_string()` in NutsBolts.cpp** - **Before**: 4 separate `std::to_string()` calls plus concatenation - **After**: Single `snprintf()` call with 16-byte stack buffer - **Impact**: Eliminated 4+ temporary allocations per IP address formatting ### 2. Container Optimizations Replaced `std::map` containers with const arrays where enums are sequential: **AlarmNames (Protocol.cpp)** - Converted `std::map<ExecAlarm, const char*>` to `const char*[]` array - ExecAlarm enum is sequential (0-12), making direct array indexing efficient - Added `alarmString()` accessor function for type safety - Eliminates std::map tree structure overhead (node allocations, pointers) **StateName (System.cpp)** - Converted `std::map<State, const char*>` to `const char*[]` array - State enum is sequential (0-9) - Added `stateName()` accessor function - O(1) array access vs O(log n) map lookup ### 3. Shared Log Buffer Implementation Added a single 1400-byte static buffer with mutex protection for thread-safe log formatting: **Implementation (Logging.cpp/h)** - Created `getLogBuffer()` and `releaseLogBuffer()` functions for controlled access - Uses FreeRTOS semaphore for thread safety - Eliminates dynamic `std::string` allocations in log macros **Converted 20+ High-Concatenation Log Calls** - Changed all log calls with >3 concatenations to use `snprintf()` with the shared buffer - Pattern: `getLogBuffer()` → `snprintf()` → `log_*()` → `releaseLogBuffer()` - Files modified: Maslow.cpp, Motors/* (6 files), Kinematics/MaslowKinematics.cpp, Spindles/* (3 files), Stepping.cpp, Uart.cpp, Machine/I2CBus.cpp, NutsBolts.h **Benefits:** - Eliminates heap fragmentation from temporary string allocations - Predictable memory usage with static buffer - Thread-safe with mutex protection - Reduced code size and improved performance ### 4. Code Quality Improvements - Used `constexpr` for array size constants - Improved variable naming (e.g., `stateIndex`, `alarmIndex`) - Removed unused headers ## Technical Details These optimizations work by: 1. **Stack vs Heap**: Stack-allocated buffers are faster and don't contribute to heap fragmentation 2. **Reducing Temporaries**: Each `std::to_string()` allocates memory for the string object and internal buffer; concatenating 40+ creates enormous overhead 3. **Map Overhead**: Each std::map node requires 24-32 bytes (key, value, left/right/parent pointers, color bit) 4. **Array Efficiency**: Direct indexed access with better cache locality and no allocation overhead 5. **Shared Buffer**: Single static buffer eliminates dynamic allocations at runtime, trading predictable static RAM for better heap management ## Testing - ✅ Clean build succeeds without warnings - ✅ Memory usage verified (both Flash and RAM) - ✅ Code review completed and feedback addressed - ✅ Security scan passed (no vulnerabilities) - ✅ Thread safety verified with FreeRTOS semaphore - ✅ No functional changes - maintains identical logging behavior ## Files Changed Modified 24 files total: - `FluidNC/src/Maslow/Maslow.cpp` - Telemetry string optimization + log buffer usage - `FluidNC/src/NutsBolts.cpp` - Utility function optimizations - `FluidNC/src/NutsBolts.h` - constrain_with_message template conversion - `FluidNC/src/ProcessSettings.cpp` - Updated to use new accessor functions - `FluidNC/src/Protocol.cpp` - AlarmNames optimization - `FluidNC/src/Protocol.h` - Updated declarations - `FluidNC/src/System.cpp` - StateName optimization - `FluidNC/src/System.h` - Updated declarations - `FluidNC/src/Logging.cpp` - Shared buffer implementation - `FluidNC/src/Logging.h` - Shared buffer function declarations - `FluidNC/src/Motors/*` - 6 motor driver files converted to use log buffer - `FluidNC/src/Spindles/*` - 3 spindle files converted to use log buffer - `FluidNC/src/Kinematics/MaslowKinematics.cpp` - Log buffer usage - `FluidNC/src/Stepping.cpp` - Log buffer usage - `FluidNC/src/Uart.cpp` - Log buffer usage - `FluidNC/src/Machine/I2CBus.cpp` - Log buffer usage Net change: +231 insertions, -125 deletions Fixes BarbourSmith/FluidNC#439 <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>Improve memory usage</issue_title> > <issue_description>As our program has grown we're running out of available dynamic memory while running. Please improve memory usage to reduce the amount of memory used while running.</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/FluidNC#439 --- ✨ Let Copilot coding agent [set things up for you](https://github.com/BarbourSmith/FluidNC/issues/new?title=✨+Set+up+Copilot+instructions&body=Configure%20instructions%20for%20this%20repository%20as%20documented%20in%20%5BBest%20practices%20for%20Copilot%20coding%20agent%20in%20your%20repository%5D%28https://gh.io/copilot-coding-agent-tips%29%2E%0A%0A%3COnboard%20this%20repo%3E&assignees=copilot) — coding agent works faster and does higher quality work when set up for your repo.

memory usage improvements (unless you want to split out only the maslow-related
work)

Fix Work Coordinate Offset (WCO) initialization to prevent NAN values

Maslow-Main ← copilot/fix-wco-nan-issue

opened 10:19PM - 16 Oct 25 UTC

+18 -10

## Problem Work Coordinate Offsets (WCO) were being set to NAN when the stored …values were unknown or invalid, causing machines to be unable to report their position. This occurred during initialization when reading coordinate system data from NVS (non-volatile storage). ## Root Cause The `Coordinates` class had two critical initialization issues: 1. **Uninitialized array**: The `_currentValue[MAX_N_AXIS]` array was not initialized in the constructor, leaving it with garbage values (potentially NAN) 2. **Unsafe error handling**: When `nvs_get_blob()` returned `ESP_ERR_NVS_INVALID_LENGTH`, the `load()` function returned `true`, allowing the uninitialized or partially-populated array to be used This meant that on first boot, or when NVS data was corrupted or had mismatched length, the coordinate values could contain NAN, breaking position reporting. ## Solution Made minimal, focused changes to ensure coordinates always default to safe zero values: ### FluidNC/src/Settings.h - Initialize `_currentValue` array to 0.0f for all axes in the `Coordinates` constructor - Guarantees valid values even before NVS data is loaded ### FluidNC/src/Settings.cpp - Verify the length of data read from NVS matches the expected buffer size - Return `false` (instead of `true`) when `ESP_ERR_NVS_INVALID_LENGTH` occurs or length mismatches are detected - Returning `false` triggers `setDefault()` in calling code, which properly initializes all coordinates to 0.0 - Added clarifying comment about `nvs_get_blob()` `len` parameter behavior ## Impact Work Coordinate Offsets now safely default to 0,0 instead of NAN in error scenarios: - First boot when NVS has never been written - Invalid or mismatched data length in storage - Corrupted or incomplete coordinate data This follows the "safe defaults" principle: it's acceptable for the home position to default to 0,0, but NAN values break position reporting entirely. ## Testing - ✅ Firmware builds successfully - ✅ Filesystem builds successfully - ✅ Clean build verified - ✅ Code formatted with clang-format - ✅ Code review completed with no issues Fixes the issue where machines could be unable to report position due to NAN coordinate offsets. <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>Don't set WCO to NAN</issue_title> > <issue_description>The work coordinate offsets (basically the home position) should not be set to NAN if they are unknown, they should be set to 0. It is OK for the home position to default to 0,0, but defaulting to NAN can cause the machine to be unable to report it's position.</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/FluidNC#434 --- 💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more [Copilot coding agent tips](https://gh.io/copilot-coding-agent-tips) in the docs.

fix work coordinate checks to avaoid NaN output

Cleanup stale state variables by converting to state machine references

Maslow-Main ← copilot/cleanup-stale-state-variables

opened 09:07PM - 13 Oct 25 UTC

+9 -31

## Problem As identified in the analysis from PR #419, there were redundant sta…te tracking variables duplicated between the `Maslow_` and `Calibration` classes. These flags (`retractingTL/TR/BL/BR`, `extendingALL`, `complyALL`) were being used alongside the state machine, creating multiple sources of truth and increasing the risk of state variables getting out of sync. The code comments themselves acknowledged this issue: ```cpp extendingALL = true; //This should be replaced by state variables ``` ```cpp bool extendingALL = false; //This is replaced by the state machine. Delete ``` ## Solution This PR removes 8 redundant boolean flags and converts all their usages to reference the state engine variable through `calibration.getCurrentState()`. The state machine is now the single source of truth for calibration state. ### Removed Variables **From `Maslow_` class:** - `retractingTL`, `retractingTR`, `retractingBL`, `retractingBR` - replaced with `getCurrentState() == RETRACTING` - `extendingALL` - replaced with `getCurrentState() == EXTENDING` - `complyALL` - replaced with `getCurrentState() == RELEASE_TENSION` **From `Calibration` class:** - `extendingALL` - redundant with state machine - `complyALL` - redundant with state machine ### Changes **Cooling fan control** (Maslow.cpp:242): ```cpp // Before else if (calibration.calibrationInProgress || extendingALL || retractingTL || retractingTR || retractingBL || retractingBR) // After else if (calibration.calibrationInProgress || calibration.getCurrentState() == EXTENDING || calibration.getCurrentState() == RETRACTING || calibration.getCurrentState() == RELEASE_TENSION) ``` **Telemetry** (Maslow.cpp:1199-1200): ```cpp // Before: stored as member variables data.extendingALL = extendingALL; data.complyALL = complyALL; // After: computed on-demand from state machine data.extendingALL = (calibration.getCurrentState() == EXTENDING); data.complyALL = (calibration.getCurrentState() == RELEASE_TENSION); ``` **State transition cleanup** (Calibration.cpp:403-405): ```cpp // Before: checking individual flags if (!retractingTL && !retractingBL && !retractingBR && !retractingTR && !extendingALL && !complyALL && !calibrationInProgress && !takeSlack && !checkOverides()) // After: checking state machine states if (currentState != RETRACTING && currentState != EXTENDING && currentState != RELEASE_TENSION && !calibrationInProgress && !takeSlack && !checkOverides()) ``` ## Benefits - ✅ **Single source of truth** - State machine is now the authoritative source for calibration state - ✅ **Reduced complexity** - 22 fewer lines of code to maintain - ✅ **Improved reliability** - Eliminates risk of flags getting out of sync with state machine - ✅ **Memory savings** - 60 bytes saved in flash memory - ✅ **Backward compatible** - Telemetry data structure unchanged for API compatibility ## Testing - ✅ Clean build successful on wifi_s3 target - ✅ No compilation errors or warnings - ✅ Flash usage: 2,011,661 bytes (60 bytes reduction) - ✅ RAM usage: 138,056 bytes (unchanged) - ✅ All state machine logic preserved and functioning correctly ## Note The `retractingTL/TR/BL/BR` and `extendedTL/TR/BL/BR` flags within the `Calibration` class are intentionally kept. These serve a different purpose - tracking which individual belts are actively retracting/extending within the RETRACTING/EXTENDING states. They represent sub-state for fine-grained belt control and are not redundant with the global state machine. Fixes #420 <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>cleanup stale state variables</issue_title> > <issue_description>per the analysis in https://github.com/BarbourSmith/FluidNC/pull/419/commits/b125a7dbe2d142a5c53734716816ad7369393b1d there are state variables that can be cleaned up. > > please do so by converting everything to reference the state engine variable</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/FluidNC#420 --- ✨ Let Copilot coding agent [set things up for you](https://github.com/BarbourSmith/FluidNC/issues/new?title=✨+Set+up+Copilot+instructions&body=Configure%20instructions%20for%20this%20repository%20as%20documented%20in%20%5BBest%20practices%20for%20Copilot%20coding%20agent%20in%20your%20repository%5D%28https://gh.io/copilot-coding-agent-tips%29%2E%0A%0A%3COnboard%20this%20repo%3E&assignees=copilot) — coding agent works faster and does higher quality work when set up for your repo.

clean up old state variables (small size reduction, fewer landmines for future
development)

process changes

Add Copilot instructions for trailing whitespace removal and dead code detection

Maslow-Main ← copilot/fix-23011aa8-1e2a-4a26-835c-279fe6b77fbb

opened 12:28AM - 05 Oct 25 UTC

+21 -1

## Overview This PR addresses issue #343 by adding comprehensive instructions t…o `.github/copilot-instructions.md` that require Copilot to check for and remove trailing whitespace and detect dead/unreachable code in every commit. ## Changes Made Updated the Copilot instructions file with new guidelines integrated into three key sections: ### 1. Pull Request Best Practices - Minimizing Changes Added two new requirements to the checklist: - **Remove trailing whitespace**: Always remove trailing whitespace from modified lines before committing - **Clean up dead code**: Remove any unused code, variables, or functions that changes make obsolete ### 2. New Section - Code Quality Checks (REQUIRED for every commit) Created a comprehensive new section with detailed instructions for: **Trailing Whitespace Removal:** - Use `git diff --check` to identify trailing whitespace issues - Remove trailing spaces from the end of all modified lines - Ensure no trailing whitespace is introduced in new code **Dead Code Detection:** - Check for unused variables, functions, or classes introduced or left behind - Verify that all code paths are reachable - Remove unnecessary commented-out code blocks - Remove unused imports and includes that become unnecessary - Use compiler warnings like `-Wunused-variable` and `-Wunused-function` for detection **Pre-commit Validation:** - Run `git diff --check` to detect trailing whitespace before using `report_progress` - Review the diff for any unreachable or dead code - Fix any issues before committing ### 3. Validation Scenarios - After Making Changes Added two new validation steps to the standard workflow: - **Step 5**: Trailing whitespace check using `git diff --check` - **Step 6**: Dead code review for unreachable code, unused variables/functions ## Impact These instructions ensure that: - All future Copilot commits will be free of trailing whitespace - Dead code is identified and removed during development rather than accumulating - Code quality remains consistent across all automated contributions - PRs are cleaner and easier to review The changes follow the existing structure and formatting of the instructions file and integrate seamlessly into the established development workflow. <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>copilot adds trailing whitespace and does not always cleanup dead code</issue_title> > <issue_description>add instructions to copilot that as part of every new commit, it should check for new trailing whitespace and remove it, and check if the changes made in that commit result in dead/unreachable code in the file.</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/FluidNC#360 --- 💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more [Copilot coding agent tips](https://gh.io/copilot-coding-agent-tips) in the docs.

always cleanup whitespace and dead code

when someone creates and issue, add them to the PR (checks fail, but I cannot
debug)

firmware PR to close

Add safety check to detect belt feeding with no encoder movement

Maslow-Main ← copilot/stop-belt-on-encoder-error

opened 09:08PM - 11 Oct 25 UTC

+115 -37

- [x] Explore repository structure and understand Maslow CNC system - [x] Unders…tand current safety_control implementation - [x] Identify how motor power, belt speed, and encoder positions are tracked - [x] Implement belt feeding detection logic with static tracking arrays - [x] Add encoder movement tracking per belt (start position and time) - [x] Add 1-second timer for detection with 0.5mm movement threshold - [x] Add logging when condition is detected (belt label, motor power, position delta) - [x] Trigger eStop alarm state when condition is met - [x] Successfully build firmware with new safety check - [x] Verify no compilation errors - [x] Apply clang-format to maintain code style consistency - [x] Verify build still succeeds after formatting - [x] Add debug logging to diagnose reported issue with $TLO command - [x] Change logging to INFO level and increase frequency to 50ms - [x] Add comprehensive diagnostic logging for 5-second feeding issue - [x] Fix critical axis label mapping bug (TL was detected as BR!) - [x] Disable verbose diagnostic logging after confirming system works <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>if any belt is feeding out for more than 1 second with no encoder movement, stop and go into alarm state</issue_title> > <issue_description>if any belt is feeding out for more than one second with no movement of the encoder, stop all movement, log which belt was not reporting movement and trigger the alarm/locked mode</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/FluidNC#406 --- 💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click [here](https://survey3.medallia.com/?EAHeSx-AP01bZqG0Ld9QLQ) to start the survey.

Improve Maslow motor definitions for A-D axis configurability

Maslow-Main ← copilot/fix-75c8e386-2497-4e96-b38a-5f20b3ac1e15

opened 09:08AM - 06 Oct 25 UTC

+268 -10

- [x] Add mm_per_revolution as configurable parameter in MotorUnit - [x] Add dir…ection_inverted flag in MotorUnit for motor reversal - [x] Add encoder_type configuration in MotorUnit - [x] Add pin configurations (forward_pin, backward_pin, readback_pin, encoder_address, pwm_channel_1, pwm_channel_2) to MotorUnit - [x] Add motor configuration parameters to maslow.yaml with hardcoded defaults - [x] Update MachineConfig to parse motor configuration parameters - [x] Apply motor configuration to MotorUnit instances during initialization - [x] Implement fallback mechanism with USE_HARDCODED_MOTOR_PINS define for easy removal later - [x] Use axis naming convention (a, b, c, d) matching maslow.yaml structure - [x] Resolve merge conflicts with Maslow-Main branch ## Latest Changes Resolved merge conflicts with Maslow-Main branch. The conflict in `FluidNC/src/Maslow/MotorUnit.cpp` was resolved by: - Keeping all new getter/setter methods added in this PR - Integrating new `setPosition()` and `getRawEncoderAngle()` methods from Maslow-Main - Removing trailing whitespace Build verified successful. All motor configuration parameters remain fully functional with axis naming convention (a, b, c, d). <details> <summary>Original prompt</summary> > > ---- > > *This section details on the original issue you should resolve* > > <issue_title>improve maslow motor definitions for a-d axis</issue_title> > <issue_description>_mmPerRevolution should be configured in the motor definitions of maslow.yaml not be a private constant > > the motor definition should have a direction flag that lets you configure the motors to run the opposite direction for that axis > > the fact that we use a AS5600 encoder should be configured in the motor definition (although right now, it's the only option > > the pins needed to talk to each encoder and to control each motor should be in the motor section > > please implement each of these as a separate commit</issue_description> > > ## Comments on the Issue (you are @copilot in this section) > > <comments> > </comments> > </details> Fixes BarbourSmith/FluidNC#369 --- ✨ Let Copilot coding agent [set things up for you](https://github.com/BarbourSmith/FluidNC/issues/new?title=✨+Set+up+Copilot+instructions&body=Configure%20instructions%20for%20this%20repository%20as%20documented%20in%20%5BBest%20practices%20for%20Copilot%20coding%20agent%20in%20your%20repository%5D%28https://gh.io/copilot-coding-agent-tips%29%2E%0A%0A%3COnboard%20this%20repo%3E&assignees=copilot) — coding agent works faster and does higher quality work when set up for your repo.

Save belt positions to NVS with state-based validity tracking

Maslow-Main ← copilot/fix-327

opened 11:20PM - 18 Sep 25 UTC