Feature: Calibration tool for measuring actual Rust paint values by VoX · Pull Request #10 · VoX/Bob-Rust-Java

VoX · 2026-04-05T03:02:36Z

Summary

Calibration tool to measure the actual circle sizes, alpha values, and circle shapes from Rust game screenshots and compare against Bob-Rust's hardcoded assumptions.

New files

CalibrationPatternGenerator — generates a 6x6 calibration grid (size x opacity), white on black
ScreenshotAnalyzer — analyzes a screenshot of the painted pattern, auto-detects grid, measures diameter/alpha/shape match per cell
CalibrationRoundTripTest — 5 tests including round-trip verification and 2x scale detection

Usage

Run CalibrationPatternGenerator to create reference pattern
Paint the pattern in Rust (or use debug mode)
Screenshot it
Run ScreenshotAnalyzer path/to/screenshot.png
Get corrected SIZES/ALPHAS arrays + diff images

Output includes

Console table of measured vs expected values
Suggested corrected Java constants
Color-coded diff image (green=correct, red=missing, blue=false positive)

All 41 tests pass.

Provides a CalibrationPatternGenerator that creates a 6x6 reference grid (sizes x alphas) and a ScreenshotAnalyzer that measures painted circles from Rust screenshots to detect mismatches with hardcoded SIZES and ALPHAS constants. Includes grid auto-detection via brightness projections, shape diff image generation, and a copy-paste Java snippet for corrected values. Also makes CircleCache and Scanline public so the calibration package can access the scanline masks. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Copilot

Pull request overview

Adds a calibration workflow to generate a known size/opacity grid pattern and analyze Rust screenshots to measure actual brush diameters, effective alpha, and circle shape fidelity vs the current scanline masks.

Changes:

Introduces CalibrationPatternGenerator to generate a size×opacity grid reference image.
Introduces ScreenshotAnalyzer to auto-detect the grid in a screenshot, measure per-cell diameter/alpha/shape match, and emit a diff image + suggested constants.
Adds CalibrationRoundTripTest to validate round-trip behavior (including 2x scaling) and exposes generator internals needed by the calibration code.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 8 comments.

Show a summary per file

File	Description
src/test/java/com/bobrust/calibration/CalibrationRoundTripTest.java	Adds round-trip + scaling tests for the calibration toolchain
src/main/java/com/bobrust/generator/Scanline.java	Makes `Scanline` public for cross-package calibration usage
src/main/java/com/bobrust/generator/CircleCache.java	Makes `CircleCache` public so calibration code can access scanline masks
src/main/java/com/bobrust/calibration/ScreenshotAnalyzer.java	New screenshot analysis CLI + measurement and diff-image generation
src/main/java/com/bobrust/calibration/CalibrationPatternGenerator.java	New reference-pattern generator CLI (grid of size/opacity circles)

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-05T03:06:27Z

src/main/java/com/bobrust/calibration/ScreenshotAnalyzer.java

+ *   <li>Shape match percentage against Bob-Rust's scanline masks</li>
+ * </ul>
+ *
+ * Usage: java -cp ... com.bobrust.calibration.ScreenshotAnalyzer screenshot.png [reference.png]


The class-level Javadoc says the CLI usage is screenshot.png [reference.png], but main actually treats the optional second argument as the diff output path. Please update the Javadoc usage string to match the implemented arguments (or implement the reference-image parameter if that was intended).

Suggested change

* Usage: java -cp ... com.bobrust.calibration.ScreenshotAnalyzer screenshot.png [reference.png]

* Usage: java -cp ... com.bobrust.calibration.ScreenshotAnalyzer screenshot.png [diff.png]

Copilot · 2026-04-05T03:06:27Z

src/main/java/com/bobrust/calibration/ScreenshotAnalyzer.java

+		// Find the total energy to set a threshold
+		long totalEnergy = 0;
+		for (long v : proj) totalEnergy += v;
+		if (totalEnergy == 0) return null;
+


The comment says totalEnergy is used to set a threshold, but the code only uses it to early-return when the projection is empty. Either add the intended thresholding (to avoid noise peaks) or update the comment to reflect the actual behavior.

Copilot · 2026-04-05T03:06:28Z

src/main/java/com/bobrust/calibration/ScreenshotAnalyzer.java

+		int totalChecked = 0;
+		for (int dy = -halfSize; dy <= halfSize; dy++) {
+			for (int dx = -halfSize; dx <= halfSize; dx++) {
+				int px = cx + dx;
+				int py = cy + dy;
+				if (px < 0 || px >= imgW || py < 0 || py >= imgH) continue;
+				totalChecked++;


totalChecked is incremented but never used. Please remove it (or use it to report/normalize something) to avoid dead code and confusion about what the loop is measuring.

Suggested change

int totalChecked = 0;

for (int dy = -halfSize; dy <= halfSize; dy++) {

for (int dx = -halfSize; dx <= halfSize; dx++) {

int px = cx + dx;

int py = cy + dy;

if (px < 0 || px >= imgW || py < 0 || py >= imgH) continue;

totalChecked++;

for (int dy = -halfSize; dy <= halfSize; dy++) {

for (int dx = -halfSize; dx <= halfSize; dx++) {

int px = cx + dx;

int py = cy + dy;

if (px < 0 || px >= imgW || py < 0 || py >= imgH) continue;

Copilot · 2026-04-05T03:06:28Z

src/main/java/com/bobrust/calibration/ScreenshotAnalyzer.java

+		if (totalExpected == 0) return 0.0;
+		// F1-like score: penalize both misses and false positives
+		int errors = (totalExpected - matched) + falsePositives;
+		int possible = totalExpected + falsePositives;
+		return possible > 0 ? 100.0 * (possible - errors) / possible : 100.0;


The score labeled as "F1-like" is effectively matched / (totalExpected + falsePositives) (IoU/Jaccard-style), not an F1 score. Either rename the metric/comment to what it is, or change the formula to a true F1 (2TP/(2TP+FP+FN)) so the reported percentage matches the description.

Copilot · 2026-04-05T03:06:28Z

src/main/java/com/bobrust/calibration/CalibrationPatternGenerator.java

+		// White with given alpha, composited onto black background:
+		// result = alpha/255 * 255 = alpha


This comment describes drawing "white with given alpha" composited onto black, but the code actually writes an opaque pixel with RGB set to alpha (gray). Please update the comment to match what is written, since this affects how consumers interpret the generated pattern (alpha channel vs brightness).

Suggested change

// White with given alpha, composited onto black background:

// result = alpha/255 * 255 = alpha

// Store the composited-on-black result directly as an opaque grayscale pixel:

// RGB = alpha, with a fully opaque alpha channel.

Copilot · 2026-04-05T03:06:28Z

src/main/java/com/bobrust/generator/CircleCache.java

 import org.apache.logging.log4j.Logger;

-class CircleCache {
+public class CircleCache {
 	private static final Logger LOGGER = LogManager.getLogger(BobRustPainter.class);



Making this class public exposes CIRCLE_CACHE/CIRCLE_CACHE_LENGTH as part of the public API, and those arrays remain mutable from outside the package. If this is only needed for calibration tooling, consider keeping the class/package internals non-public and exposing a narrow accessor (or returning defensive copies / unmodifiable views) to avoid accidental mutation by external callers.

Copilot · 2026-04-05T03:06:29Z

src/main/java/com/bobrust/generator/Scanline.java

 	public int y;
 	public int x1;
 	public int x2; // inclusive


Changing Scanline to public exposes a mutable data type (public fields) as API surface. Consider making the fields final (or using an immutable type like a record) so callers can’t mutate cached scanlines and corrupt shape masks at runtime.

Suggested change

public int y;

public int x1;

public int x2; // inclusive

public final int y;

public final int x1;

public final int x2; // inclusive

Copilot · 2026-04-05T03:06:29Z

src/test/java/com/bobrust/calibration/CalibrationRoundTripTest.java

+		// All bottom-row detected cells should have correct alpha
+		int bottomRowDetected = 0;
+		for (int col = 0; col < numSizes; col++) {
+			if (detected[numAlphas - 1][col]) bottomRowDetected++;
+		}
+		assertTrue(alphaMatches >= bottomRowDetected * 0.8,
+			"Too few alpha matches in bottom row: " + alphaMatches + "/" + bottomRowDetected);


The comment says "All bottom-row detected cells should have correct alpha", but the assertion only requires 80% (alphaMatches >= bottomRowDetected * 0.8). Please align the comment/message with the actual tolerance, or tighten the assertion if the intent is truly 100% for the round-trip case.

Suggested change

// All bottom-row detected cells should have correct alpha

int bottomRowDetected = 0;

for (int col = 0; col < numSizes; col++) {

if (detected[numAlphas - 1][col]) bottomRowDetected++;

}

assertTrue(alphaMatches >= bottomRowDetected * 0.8,

"Too few alpha matches in bottom row: " + alphaMatches + "/" + bottomRowDetected);

// At least 80% of bottom-row detected cells should have correct alpha

int bottomRowDetected = 0;

for (int col = 0; col < numSizes; col++) {

if (detected[numAlphas - 1][col]) bottomRowDetected++;

}

assertTrue(alphaMatches >= bottomRowDetected * 0.8,

"Too few alpha matches in bottom row (expected at least 80%): "

+ alphaMatches + "/" + bottomRowDetected);

Copilot AI review requested due to automatic review settings April 5, 2026 03:02

Copilot started reviewing on behalf of VoX April 5, 2026 03:03 View session

Copilot AI reviewed Apr 5, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Feature: Calibration tool for measuring actual Rust paint values#10

Feature: Calibration tool for measuring actual Rust paint values#10
VoX wants to merge 1 commit intomainfrom
feature/calibration-tool

VoX commented Apr 5, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Copilot AI Apr 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	* Usage: java -cp ... com.bobrust.calibration.ScreenshotAnalyzer screenshot.png [reference.png]
	* Usage: java -cp ... com.bobrust.calibration.ScreenshotAnalyzer screenshot.png [diff.png]

		// White with given alpha, composited onto black background:
		// result = alpha/255 * 255 = alpha

Conversation

VoX commented Apr 5, 2026

Summary

New files

Usage

Output includes

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 5, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants