Add monochrome palette detection and fallback palette resolution by Copilot · Pull Request #145 · jongalloway/DiagramForge

Copilot · 2026-03-24T00:39:50Z

Themes with all-achromatic NodePalette entries (e.g. Prism's all-#FFFFFF palette) had no shared infrastructure to detect the condition or substitute meaningful colors—leading to invisible fills, hue-scoring no-ops, and silent Vibrant() failures.

`ColorUtils` additions

IsAchromatic(hex, saturationThreshold = 0.08) — computes HSL saturation and returns true below the threshold; guards against division-by-zero at extreme lightness values.
IsPaletteMonochrome(palette, backgroundColor?) — returns true when all entries are achromatic, all are the same color, or all match the background (invisible nodes). Returns false for empty palettes.
Vibrant() hardening — detects achromatic input via IsAchromatic and applies an explicit lightness shift (light → −0.35, dark → +0.35, bounded to [0.25, 0.75]) instead of silently running the hue-amplification formula on colorless input.

`ThemePaletteResolver` addition

ResolveEffectivePalette(theme) — pure function returning a usable 8-entry fill palette. Returns NodePalette unchanged when it's chromatic; otherwise falls back to:
1. Linear interpolation of BorderGradientStops to 8 entries (when UseBorderGradients && BorderGradientStops.Count > 1)
2. Hue-rotation of AccentColor/SecondaryColor across 8 evenly-spaced positions

// Prism has all-white NodePalette — ResolveEffectivePalette returns chromatic colors from gradient stops
var palette = ThemePaletteResolver.ResolveEffectivePalette(Theme.Prism);
// → ["#2563EB", …, "#F59E0B"] — 8 interpolated entries, none monochrome

// IsAchromatic / IsPaletteMonochrome as the shared gate before consuming NodePalette directly
bool skip = ColorUtils.IsPaletteMonochrome(theme.NodePalette, theme.BackgroundColor);

These are additive utilities only—no existing rendering paths are changed; per-diagram wiring is handled in follow-on sub-issues.

⚡ Quickly spin up Copilot coding agent tasks from anywhere on your macOS or Windows machine with Raycast.

Co-authored-by: jongalloway <68539+jongalloway@users.noreply.github.com> Agent-Logs-Url: https://github.com/jongalloway/DiagramForge/sessions/27f13f43-5376-482e-8812-a2b0189deafb

Copilot

Pull request overview

Adds shared infrastructure to detect achromatic/monochrome palettes and derive a usable fallback palette, preventing “invisible” node fills and improving robustness of color derivation utilities.

Changes:

Added ColorUtils.IsAchromatic / ColorUtils.IsPaletteMonochrome and hardened ColorUtils.Vibrant() for achromatic inputs.
Added ThemePaletteResolver.ResolveEffectivePalette(theme) to return either the theme palette (when usable) or an 8-color fallback derived from gradient stops or hue rotation.
Added unit tests covering the new utilities and fallback behavior.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 6 comments.

File	Description
tests/DiagramForge.Tests/Models/ThemePaletteResolverTests.cs	Adds tests for `ResolveEffectivePalette` behavior across monochrome and gradient-stop fallback scenarios.
tests/DiagramForge.Tests/Models/ColorUtilsTests.cs	Adds tests for `IsAchromatic`, `IsPaletteMonochrome`, and the new achromatic behavior in `Vibrant()`.
src/DiagramForge/Models/ThemePaletteResolver.cs	Implements `ResolveEffectivePalette` plus internal helpers for gradient sampling and hue-rotation palette derivation.
src/DiagramForge/Models/ColorUtils.cs	Implements achromatic/monochrome detection and modifies `Vibrant()` to handle achromatic inputs explicitly.

Comments suppressed due to low confidence (1)

tests/DiagramForge.Tests/Models/ColorUtilsTests.cs:813

Test name doesn’t match the assertion: IsPaletteMonochrome_MatchesBackground_NoBackground_ReturnsFalse asserts True. Rename the test (e.g., “…ReturnsTrue”) or change the assertion so the name accurately describes the expected behavior.

    public void IsPaletteMonochrome_MatchesBackground_NoBackground_ReturnsFalse()
    {
        // Same chromatic entries, but without a background to compare against —
        // they are the same color, so still "all same color" → true
        var palette = new[] { "#2563EB", "#2563EB", "#2563EB" };
        Assert.True(ColorUtils.IsPaletteMonochrome(palette, backgroundColor: null));
    }

Copilot · 2026-03-24T02:13:46Z

+    /// <param name="hex">Hex color string.</param>
+    /// <param name="saturationThreshold">
+    /// Saturation threshold 0–1; colors with saturation below this value are considered
+    /// achromatic. Defaults to 0.08.
+    /// </param>
+    public static bool IsAchromatic(string hex, double saturationThreshold = 0.08)
+    {
+        var (rRaw, gRaw, bRaw) = ParseHex(hex);
+        double r = rRaw / 255d;
+        double g = gRaw / 255d;
+        double b = bRaw / 255d;
+        double max = Math.Max(r, Math.Max(g, b));
+        double min = Math.Min(r, Math.Min(g, b));
+        double delta = max - min;
+        double lightness = (max + min) / 2;
+        double denominator = 1 - Math.Abs(2 * lightness - 1);
+        // When delta is near zero the color is already achromatic; guard the denominator
+        // (which is also ~0 at lightness=0 or lightness=1) to avoid division by zero.
+        double saturation = delta < 0.0001 || denominator < 0.0001
+            ? 0
+            : delta / denominator;
+        return saturation < saturationThreshold;


The XML doc for saturationThreshold says it is a 0–1 value, but the method currently accepts any double without clamping/validation. Either clamp saturationThreshold into [0,1] (similar to Blend/WithOpacity) or update the docs to reflect the actual behavior to avoid surprising results for negative/large thresholds.

Copilot · 2026-03-24T02:13:47Z

+    private static IReadOnlyList<string> BuildPaletteFromHueRotation(
+        string accentColor, string secondaryColor, int count)
+    {
+        bool isLight = ColorUtils.IsLight(accentColor);
+        double hueStep = 360.0 / count;
+        var result = new List<string>(count);
+
+        for (int i = 0; i < count; i++)
+        {
+            string baseColor = i % 2 == 0 ? accentColor : secondaryColor;
+            double rotation = Math.Floor(i / 2.0) * hueStep;
+            result.Add(ColorUtils.RotateHue(baseColor, rotation, isLight));
+        }


BuildPaletteFromHueRotation passes ColorUtils.IsLight(accentColor) as the isLightBackground argument to ColorUtils.RotateHue. That parameter controls clamping based on the canvas/background, not whether the accent color itself is light. This can skew the generated palette (too dark/light) depending on theme background. Consider deriving isLightBackground from theme.BackgroundColor (and passing it down) or adding an explicit parameter so callers can provide the correct background context.

Copilot · 2026-03-24T02:13:47Z

+        // Build fallback from gradient stops when they are available and meaningful.
+        if (theme.UseBorderGradients && theme.BorderGradientStops is { Count: > 1 })
+            return BuildPaletteFromGradientStops(theme.BorderGradientStops, DefaultPaletteSize);
+
+        // Fall back to hue-rotation derivation from the theme's semantic colors.
+        return BuildPaletteFromHueRotation(theme.AccentColor, theme.SecondaryColor, DefaultPaletteSize);


ResolveEffectivePalette can still return a monochrome palette if BorderGradientStops are present but themselves monochrome/achromatic (e.g., two identical stops, or all-gray stops). Since the method’s contract is to return a palette suitable for direct fill use, consider validating the gradient-derived palette with IsPaletteMonochrome (optionally against theme.BackgroundColor) and falling back to the hue-rotation path when the gradient result is still monochrome.

Copilot · 2026-03-24T02:13:47Z

+    ///   <item>
+    ///     If <see cref="Theme.UseBorderGradients"/> is <see langword="true"/> and
+    ///     <see cref="Theme.BorderGradientStops"/> contains more than one stop, the stops
+    ///     are interpolated/cycled to produce <see cref="DefaultPaletteSize"/> entries.


The remarks mention gradient stops are “interpolated/cycled”, but BuildPaletteFromGradientStops only linearly interpolates across the stop range and does not cycle/repeat stops. Please update the comment to match the actual behavior (or implement cycling if that was intended).

Suggested change

/// are interpolated/cycled to produce <see cref="DefaultPaletteSize"/> entries.

/// are linearly interpolated to produce <see cref="DefaultPaletteSize"/> entries.

Copilot · 2026-03-24T02:13:47Z

+    public void IsAchromatic_CustomThreshold_LowSaturationColor_ReturnsTrue()
+    {
+        // A muted blue — below the default threshold but above an even lower custom one
+        // Test that the threshold parameter is respected
+        Assert.False(ColorUtils.IsAchromatic("#4F81BD", saturationThreshold: 0.01));
+    }


Test name doesn’t match the assertion: IsAchromatic_CustomThreshold_LowSaturationColor_ReturnsTrue asserts False. Rename the test (e.g., “…ReturnsFalse”) or adjust the expectation so the name accurately reflects the behavior being validated.

This issue also appears on line 807 of the same file.

Copilot · 2026-03-24T02:13:48Z

+        if (palette.All(c => string.Equals(c, palette[0], StringComparison.OrdinalIgnoreCase)))
+            return true;
+
+        // All entries match the background (nodes would be invisible against the canvas).
+        if (backgroundColor is not null &&
+            palette.All(c => string.Equals(c, backgroundColor, StringComparison.OrdinalIgnoreCase)))
+            return true;


IsPaletteMonochrome compares colors using string.Equals, which treats equivalent hex encodings as different (e.g. #F00 vs #FF0000, or #ABC vs #AABBCC). Because ColorUtils explicitly supports shorthand formats, this can incorrectly return false for palettes that are effectively monochrome or match the background. Consider normalizing colors (e.g., via ParseHex/ParseHexWithAlpha or ExpandShorthand) before doing “all same” / “matches background” comparisons, and compare by RGB(A) values rather than raw strings.

Suggested change

if (palette.All(c => string.Equals(c, palette[0], StringComparison.OrdinalIgnoreCase)))

return true;

// All entries match the background (nodes would be invisible against the canvas).

if (backgroundColor is not null &&

palette.All(c => string.Equals(c, backgroundColor, StringComparison.OrdinalIgnoreCase)))

return true;

{

var referenceColor = ParseHexWithAlpha(palette[0]);

if (palette.All(c => ParseHexWithAlpha(c) == referenceColor))

return true;

}

// All entries match the background (nodes would be invisible against the canvas).

if (backgroundColor is not null)

{

var background = ParseHexWithAlpha(backgroundColor);

if (palette.All(c => ParseHexWithAlpha(c) == background))

return true;

}

jongalloway · 2026-03-25T05:29:42Z

@copilot apply changes based on the comments in this thread

…ackground, validate gradient palette, fix test names Co-authored-by: jongalloway <68539+jongalloway@users.noreply.github.com> Agent-Logs-Url: https://github.com/jongalloway/DiagramForge/sessions/4b46141f-444c-46cc-9873-d40f91adb1a3

Copilot · 2026-03-25T05:36:03Z

@copilot apply changes based on the comments in this thread

Applied all six review comments in commit 0e5b06f:

saturationThreshold clamping — IsAchromatic now clamps the parameter to [0,1] via Math.Clamp, and the doc updated to mention the clamp.
isLightBackground source — BuildPaletteFromHueRotation now takes an explicit isLightBackground parameter; ResolveEffectivePalette derives it from theme.BackgroundColor instead of the accent color.
Monochrome gradient validation — ResolveEffectivePalette now checks the gradient-derived palette with IsPaletteMonochrome before returning it; if the gradient stops are themselves monochrome, it falls through to hue-rotation.
Doc comment — "interpolated/cycled" → "linearly interpolated".
Test name mismatches — IsAchromatic_CustomThreshold_LowSaturationColor_ReturnsTrue → ReturnsFalse; IsPaletteMonochrome_MatchesBackground_NoBackground_ReturnsFalse → ReturnsTrue.
Hex normalization — IsPaletteMonochrome now compares via ParseHexWithAlpha tuples so #FFF and #FFFFFF are treated as equal. New tests cover #FFF/#FFFFFF equivalence and threshold clamping edge cases.

Initial plan

026bf48

Copilot AI assigned Copilot and jongalloway Mar 24, 2026

Copilot started work on behalf of jongalloway March 24, 2026 00:39 View session

Copilot AI linked an issue Mar 24, 2026 that may be closed by this pull request

Add monochrome palette detection and fallback palette resolution #139

Closed

5 tasks

Add monochrome palette detection and fallback palette resolution

b764eac

Co-authored-by: jongalloway <68539+jongalloway@users.noreply.github.com> Agent-Logs-Url: https://github.com/jongalloway/DiagramForge/sessions/27f13f43-5376-482e-8812-a2b0189deafb

Copilot AI changed the title ~~[WIP] Add monochrome palette detection and fallback palette resolution~~ Add monochrome palette detection and fallback palette resolution Mar 24, 2026

Copilot AI requested a review from jongalloway March 24, 2026 00:52

Copilot finished work on behalf of jongalloway March 24, 2026 00:52

jongalloway marked this pull request as ready for review March 24, 2026 02:10

jongalloway requested a review from Copilot March 24, 2026 02:10

Copilot started reviewing on behalf of jongalloway March 24, 2026 02:10 View session

Copilot AI reviewed Mar 24, 2026

View reviewed changes

Copilot started work on behalf of jongalloway March 25, 2026 05:29 View session

Copilot finished work on behalf of jongalloway March 25, 2026 05:36

jongalloway merged commit f73e1c8 into main Mar 25, 2026
5 checks passed

jongalloway deleted the copilot/add-monochrome-palette-detection branch March 25, 2026 05:40

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add monochrome palette detection and fallback palette resolution#145

Add monochrome palette detection and fallback palette resolution#145
jongalloway merged 3 commits into
mainfrom
copilot/add-monochrome-palette-detection

Copilot AI commented Mar 24, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 24, 2026

Uh oh!

Copilot AI Mar 24, 2026

Uh oh!

Copilot AI Mar 24, 2026

Uh oh!

Copilot AI Mar 24, 2026

Uh oh!

Copilot AI Mar 24, 2026

Uh oh!

Copilot AI Mar 24, 2026

Uh oh!

jongalloway commented Mar 25, 2026

Uh oh!

Copilot AI commented Mar 25, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

	/// are interpolated/cycled to produce <see cref="DefaultPaletteSize"/> entries.
	/// are linearly interpolated to produce <see cref="DefaultPaletteSize"/> entries.

Conversation

Copilot AI commented Mar 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

ColorUtils additions

ThemePaletteResolver addition

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

jongalloway commented Mar 25, 2026

Uh oh!

Copilot AI commented Mar 25, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Copilot AI commented Mar 24, 2026 •

edited

Loading

`ColorUtils` additions

`ThemePaletteResolver` addition