Add unit tests for spiral shape and arrow shape tools

[tarone-saloni]

Summary

Adds #[cfg(test)] unit test coverage to two currently untested shape tool files, as tracked in #3965.

• spiral_shape.rs — 5 tests
• arrow_shape.rs — 5 tests

Tests follow the same pattern as ellipse_shape.rs and line_shape.rs. Node inputs are read directly from document.network_interface (no graph evaluation required), following the line_shape.rs approach.

Note on spiral tests: The spiral tests use raw editor.editor.handle_message(...) calls instead of the EditorTestUtils helpers. This is because handle_message internally calls eval_graph (the instrumented graph evaluator), which does not support SpiralType as a node input and would panic. Reading spiral parameters via extract_spiral_parameters (which queries document.network_interface directly) works without evaluation.

---

Tests added

spiral_shape.rs

Test	What it checks
spiral_draw_simple	outer_radius matches drag distance (~40 px)
spiral_archimedean_inner_radius_default	inner_radius == 0.0 for Archimedean type
spiral_logarithmic_inner_radius_default	inner_radius == 0.1 for Logarithmic type
spiral_cancel_rmb	No layer created when drag cancelled with RMB
spiral_default_turns	turns >= 1.0 after a draw

arrow_shape.rs

Test	What it checks
arrow_draw_simple	arrow_to matches a horizontal 100 px drag
arrow_draw_diagonal	arrow_to length matches a 3-4-5 (60×80→100) drag
arrow_cancel_rmb	No layer created when drag cancelled with RMB
arrow_snap_angle_shift	Angle snaps to nearest 15° multiple with SHIFT held
arrow_zero_length_no_layer	Zero-length drag produces no meaningful arrow_to (guarded by the < 1e-6 check)

Test plan

• cargo test -p graphite-editor spiral — 5 passed, 0 failed
• cargo test -p graphite-editor arrow — 5 passed, 0 failed

[cubic-dev-ai]

No issues found across 1 file

Confidence score: 5/5

• Automated review surfaced no issues in the provided summaries.
• No files require special attention.

[gemini-code-assist]

Code Review

This pull request introduces a comprehensive suite of unit tests for shape utility functions, covering angle wrapping, coordinate calculations for arcs, stars, and polygons, and hit-testing logic. The feedback identifies opportunities to improve test maintainability and readability by implementing a custom macro for floating-point approximations and utilizing idiomatic vector comparison methods from the glam library.

[gemini-code-assist]

There are many floating-point comparisons in these tests using the pattern (a - b).abs() < 1e-10. To improve readability and provide better failure messages, consider defining a local assert_approx_eq! macro at the top of the tests module.

macro_rules! assert_approx_eq {
    ($a:expr, $b:expr, $eps:expr) => {
        assert!(
            ($a - $b).abs() < $eps,
            "assertion failed: `(left - right).abs() < epsilon`\n  left: `{:?}`\n right: `{:?}`\nepsilon: `{:?}`",
            $a, $b, $eps
        );
    };
    ($a:expr, $b:expr) => {
        assert_approx_eq!($a, $b, 1e-10);
    };
}

This would allow you to write this assertion as assert_approx_eq!(wrap_to_tau(PI), PI); and can be used for all scalar float comparisons in this test module, making them more concise and readable.

[gemini-code-assist]

These per-component assertions can be simplified by using glam::DVec2::abs_diff_eq for a more concise and idiomatic vector comparison.

For example, these four lines can be replaced with:

let expected = DVec2::new(1., 0.);
assert!(start.abs_diff_eq(expected, 1e-10), "start point mismatch");
assert!(end.abs_diff_eq(expected, 1e-10), "end point mismatch");

This approach can be applied to other vector comparisons in this file (e.g., in arc_endpoints_with_identity_viewport_matches_no_viewport, star_vertex_*, and polygon_vertex_* tests) to make them cleaner and more readable.

[cubic-dev-ai]

1 issue found across 2 files (changes from recent commits).

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="editor/src/messages/tool/common_functionality/shapes/arrow_shape.rs">

<violation number="1" location="editor/src/messages/tool/common_functionality/shapes/arrow_shape.rs:212">
P2: `arrow_zero_length_no_layer` can pass without asserting any invariant because assertions are conditional on `get_arrow_to()` returning `Some`.</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}

[cubic-dev-ai]

P2: arrow_zero_length_no_layer can pass without asserting any invariant because assertions are conditional on get_arrow_to() returning Some.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At editor/src/messages/tool/common_functionality/shapes/arrow_shape.rs, line 212:

<comment>`arrow_zero_length_no_layer` can pass without asserting any invariant because assertions are conditional on `get_arrow_to()` returning `Some`.</comment>

<file context>
@@ -114,3 +114,103 @@ impl Arrow {
+		editor.drag_tool(ToolType::Shape, 50., 50., 50., 50., ModifierKeys::empty()).await;
+
+		// Either no layer created, or arrow_to is zero/near-zero
+		if let Some(arrow_to) = get_arrow_to(&editor) {
+			assert!(arrow_to.length() < 1e-4, "Zero-length drag should produce no meaningful arrow_to, got {arrow_to:?}");
+		}
</file context>

Suggested change

	if let Some(arrow_to) = get_arrow_to(&editor) {
	assert!(arrow_to.length() < 1e-4, "Zero-length drag should produce no meaningful arrow_to, got {arrow_to:?}");
	}
	let layer_count = editor.active_document().metadata().all_layers().count();
	if layer_count == 0 {
	assert!(get_arrow_to(&editor).is_none(), "No arrow data should exist when no layer is created");
	} else {
	assert_eq!(layer_count, 1, "Zero-length drag should not create extra layers");
	let arrow_to = get_arrow_to(&editor).expect("Expected arrow_to for created arrow layer");
	assert!(arrow_to.length() < 1e-4, "Zero-length drag should produce no meaningful arrow_to, got {arrow_to:?}");
	}