Repository navigation
docs: record ecosystem status, and withdraw the Swift fixture - #17
Merged
Merged
Conversation
RodolpheKouyoumdjian
force-pushed
the
feat/demo-swift
branch
from
August 29, 2026 14:53
bbaa5ac to
135e7d1
Compare
The Swift fixture is withdrawn rather than shipped. It claimed swift-argument-parser 0.4.4 → 1.0.3 removed the top-level `ArrayParsingStrategy` enum; diffing `public enum` declarations across the two tags appeared to confirm it. The oracle disproved it — the upgraded tree still built. The type was not removed, it changed from an `enum` to a `public struct` of the same name with the same `.upToNextOption` member, so consumer code is unaffected. The diff method was wrong, not the library. This is the second time the oracle has caught an incorrect claim of mine before it reached anyone, which is what it is for. docs/ecosystem-status.md records, per ecosystem, what ships and what does not and why: the Swift finding above, CocoaPods needing Xcode that Codespaces cannot provide, and conan/vcpkg/arduino/opam as not yet built with the specific obstacle each carries. Refs #6. Co-Authored-By: Claude Sonnet 5 <[email protected]>
RodolpheKouyoumdjian
force-pushed
the
feat/demo-swift
branch
from
August 29, 2026 14:57
135e7d1 to
2346da5
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR no longer adds a Swift demo. It removes one I had built, because the
oracle proved it was not a real break.
What happened
I built
swift-argument-parser0.4.4 → 1.0.3 on the claim that version 1.0removed the top-level
ArrayParsingStrategyenum. I verified that by diffingpublic enumdeclarations across both tags' sources, and it looked clean.The oracle disproved it in CI: the upgraded tree still built.
The type was never removed. It changed from an
enumto apublic structwiththe same name and the same
.upToNextOptionmember, so consumer code isunaffected. My diff only compared
public enumlines, so a reclassification readas a removal. The method was wrong, not the library.
This is the second time the oracle has caught a bad claim of mine before it
reached anyone — the first was the Rust fixture's error code. That is exactly
the check the earlier review asked for, doing its job.
I also probed
swift-argument-parser1.0.3 → 1.5.0: no public symbol removalsat all, so it cannot host a fixture under §10 either.
Separately, Drift returns nothing for either Swift pair — no findings, no
changelog sections,
safe-to-upgrade. SwiftPM identifies packages by git URLwith no central registry, so Drift declares
surface: noneand partial evidencefor Swift. Both halves would need solving before a Swift demo is worth shipping.
What this PR adds instead
docs/ecosystem-status.md— per ecosystem: what ships, what its Drift resultactually is, and for the rest the specific obstacle. Covers the Swift finding
above, CocoaPods needing Xcode that Codespaces (Linux) cannot provide per §40,
and
conan/vcpkg/arduino/opamas not yet built with the reason eachcarries.
Refs #6.