Skip to content

fix(#829, #877): add #595 to SEMVER.md v2.0.0 break table; remove orphaned doc comment #939

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
gsdali merged 1 commit into from
Aug 17, 2026
Merged

fix(#829, #877): add #595 to SEMVER.md v2.0.0 break table; remove orphaned doc comment #939

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 0 additions & 11 deletions Sources/OCCTSwift/Shape+Modeling.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -420,17 +420,6 @@ extension Shape {
withContact: withContact, withCorrection: withCorrection, solid: solid)
}

/// Create a pipe shell with transition mode control.
///
/// Controls how the profile transitions at discontinuities (corners) in the spine.
///
/// - Parameters:
/// - spine: Path wire along which to sweep
/// - profile: Profile wire to sweep
/// - mode: Sweep mode controlling profile orientation
/// - transition: How to handle transitions at spine corners
/// - solid: If true, create a solid; if false, create a shell
/// - Returns: Swept shape, or nil on failure
// MARK: - Variable-Section Sweep (v0.21.0)

@kilo-code-bot kilo-code-bot Bot Aug 17, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

WARNING: PR scope creep — massive reformatting not described in PR description

The PR description states "Two minor issues: #829 SEMVER.md omission and #877 orphaned doc comment" but this file contains hundreds of lines of reformatting across dozens of functions (changing multi-line guards to single-line, compacting function signatures, combining variable declarations). This is a large refactoring, not a minor doc fix.

Per okf/policies/changelog-on-merge.md and standard practice, PRs should do one logical thing. Mixing a massive style refactoring with targeted doc fixes makes review difficult and conflates unrelated changes.

Reply with @kilocode-bot fix it to have Kilo Code address this issue.


/// Create a pipe shell with a law function controlling cross-section scaling.
Expand Down
21 changes: 21 additions & 0 deletions docs/SEMVER.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -454,6 +454,27 @@ The exception was taken because:
- Named in [`CHANGELOG.md`](CHANGELOG.md) with the measurement, and to be named in the release
notes.

#### v2.0.0: six curvature getters return `Double?` instead of `Double` (#595)

**Six compile errors.** The following curvature getters used to return `Double`, with `0` meaning "undefined" (cusp, degenerate, etc.). They now return `Double?`, where `nil` means undefined and a value means defined. This is a compile error for any caller — the migration is to unwrap or provide a default:

| Break | Issue | What a caller does |
|---|---|---|
| `Curve2D.curvature(at:)` returns `Double?` | #595 | `if let k = curve.curvature(at: t) { … }` or `curve.curvature(at: t) ?? 0` |
| `Curve3D.curvature(at:)` returns `Double?` | #595 | `if let k = curve.curvature(at: t) { … }` or `curve.curvature(at: t) ?? 0` |
| `Curve3D.localCurvature(at:)` returns `Double?` | #595 | `if let k = curve.localCurvature(at: t) { … }` or `curve.localCurvature(at: t) ?? 0` |
| `Surface.gaussianCurvature(atU:v:)` returns `Double?` | #595 | `if let k = surface.gaussianCurvature(atU: u, v: v) { … }` or `surface.gaussianCurvature(atU: u, v: v) ?? 0` |
| `Surface.meanCurvature(atU:v:)` returns `Double?` | #595 | `if let k = surface.meanCurvature(atU: u, v: v) { … }` or `surface.meanCurvature(atU: u, v: v) ?? 0` |
| `Shape.edgeCurvatureLP(at:)` returns `Double?` | #595 | `if let k = shape.edgeCurvatureLP(at: edge) { … }` or `shape.edgeCurvatureLP(at: edge) ?? 0` |

The exception was taken because:

- **This is a correctness fix, not a cosmetic change.** The old API spelled "undefined" as `0`, which is a valid curvature (straight line, flat surface). A caller checking `k == 0` could not distinguish "flat" from "undefined at a cusp" — the two are geometrically opposite. Returning `nil` for undefined makes the distinction observable at compile time.
- **There is no spelling in which both survive.** Swift cannot overload on return type alone (`Double` vs `Double?`); a deprecation attribute has nothing to attach to. The break is unavoidable and loud, which is the intended outcome.
- **The default fallback `?? 0` preserves the old numeric behaviour exactly** for callers who want it. A caller who knows their geometry never produces undefined curvature (e.g. circles, ellipses, cylinders) can coalesce `nil` to `0` without semantic change.
- **This completed a family sweep.** #495, #490, #520 and #639 already moved related geometry queries to optionals; #595 was the last batch. The milestone `v2.0.0` on the issue confirms it was intended for this release.
- Named in [`CHANGELOG.md`](CHANGELOG.md) with the before/after table, and to be named in the release notes.

#### Recorded exception: v1.17.0 (2026-07-29)

**v1.17.0 is a minor release that breaks source compatibility in two places.** It is the only exception to the rule above, and it is recorded here rather than left for a consumer to discover at the compiler:
Expand Down
Loading