Require ReadStat 1.2, fix the v1.0.0 release notes - #33
Merged
Merged
Conversation
ReadStat 1.2.0 is now registered, so the bound deferred from #31 can be tightened. ReadStat "0.4.1, 1" admitted versions that cannot work here. ReadStat 0.4.x predates ReadStat_jll and depends on BinaryProvider, so it cannot resolve on the julia 1.10 floor this package now declares. Narrowing to "1.2" rather than "1" is deliberate: ReadStat 1.0 and 1.1 allow ReadStat_jll "1.1.1-1", so only 1.2 guarantees users get the ReadStat_jll 1.1.9 pin, which was the point of the audit that led here. DataValues floor goes to 0.4.13 because that is what ReadStat 1.2.0 requires; advertising 0.4.4 was a claim the resolver would never honour. 0.4.13 is also the newest registered DataValues. Verified against the registry rather than a dev checkout: resolution picks ReadStat v1.2.0 and ReadStat_jll v1.1.9+0, and both test items pass, including the exact-string Show item. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The NEWS entry for v1.0.0 still read "Drop julia 0.7 support / Migrate to Project.toml", both of which describe work from the 0.8.0 era. It would have shipped as the release notes for a change set it does not describe. Replace with what 1.0.0 actually contains. The codecov badge still pointed at branch=master, which no longer exists, so it rendered broken; the CI badge on the line above already uses main. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Gets
maininto a releasable state. Follows queryverse/ReadStat.jl#95 now that ReadStat 1.2.0 is registered.Compat
ReadStat = "0.4.1, 1"admitted versions that cannot work here — ReadStat 0.4.x predatesReadStat_jlland depends onBinaryProvider, so it can't resolve on the julia 1.10 floor this package now declares.Narrowing to
"1.2"rather than the looser"1"is deliberate: ReadStat 1.0 and 1.1 allowReadStat_jll = "1.1.1-1", so only 1.2 guarantees users actually get theReadStat_jll = "1.1.9"pin — which was the point of the audit that led here. Without it we'd quietly give that back.DataValuesfloor goes0.4.4→0.4.13, because that is what ReadStat 1.2.0 requires. The old floor was a claim the resolver would never honour. 0.4.13 is also the newest registered DataValues.Everything else is left alone.
DataValuesandTableShowUtilskeep their0.xranges because they are still pre-1.0 in the registry (0.4.13 and 0.2.7).Release notes
The
NEWS.mdentry for v1.0.0 still read "Drop julia 0.7 support / Migrate to Project.toml" — both describe work from the 0.8.0 era, and would have shipped as the notes for a release whose actual content is the julia 1.10 floor, the TestItems migration, the ReadStat 1.2 requirement and the widened bounds.Also repointed the codecov badge from
branch=master(gone, so the badge rendered broken) tobranch=main.Testing
Resolved against the registry rather than a dev checkout: picks ReadStat v1.2.0 and ReadStat_jll v1.1.9+0, which is the pairing the
"1.2"bound exists to guarantee. Both test items pass, including the exact-stringShowitem — the tripwire for any drift in inferred column types coming from the newer ReadStat.Note on CI history:
mainhas never had a green run, but neither failure was a test failure. The #30 run died ininstall-juliaupwithread ECONNRESETon two macOS-intel jobs (Julia never installed), and the #31 run was cancelled mid-queue during the Actions backup — 16 passed, 8 cancelled, none failed.Version deliberately left at
1.0.0-DEV.🤖 Generated with Claude Code