Release 0.4.0 - #8
Merged
Merged
Conversation
The description of the mobile verification was written when the linker's own evidence was the ceiling: the executable was expected to be stripped, and the symbol half of the check was documented as attempted and normally unable to run. That stopped being true two commits later. The release simulator build keeps its symbol table - 65267 defined symbols - and the check now reads Box3D's own native ones out of it, 415 of them in the run on main. The CHANGELOG matters more than the README here, because [Unreleased] is what becomes the release notes, and shipping it unchanged would have published a claim weaker than what CI actually does. Both now also record why only _b3 names count. An AOT method carries its parameter types in its mangled name, so a plain search for b3 matches managed symbols that are in the binary whether or not the archive survived the link. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Minor rather than patch: the packages reach Android and iOS, and a platform is a capability. Nothing in the library changed - package validation against 0.3.0 reports no difference in any target framework - so there is no fix to record and no break to warn about. The baseline moves to 0.3.0 in the same commit. It was left at 0.2.0 while 0.3.0 shipped, which is the gap Directory.Build.targets already describes having made once with 0.1.0, so the comment now says when to move it rather than only that it should be moved. The CHANGELOG's link references were two releases behind: [Unreleased] still compared against v0.2.0 and 0.3.0 had no reference at all, so its heading rendered as plain text. Co-Authored-By: Claude Opus 5 (1M context) <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.
Two commits, in the order they have to happen.
Say what the iOS check proves now. The
[Unreleased]block described theiOS verification as checking "for evidence that the package handed Box3D's
archive to the linker", which was the honest ceiling when it was written and
stopped being so two commits later.
[Unreleased]is what becomes the releasenotes, so publishing it unchanged would have shipped a claim weaker than what
CI does. The run on main reports 65267 defined symbols in the simulator
executable, 415 of them Box3D's own
_b3names — the executable is notstripped and the symbol check runs. The README's "attempted and normally cannot
run" paragraph is replaced for the same reason, and both now record why only
_b3names count.Release 0.4.0. Minor rather than patch: the packages reach Android and iOS
and nothing in the library changed.
PackageValidationBaselineVersionmoves to0.3.0 in the same commit — it was left at 0.2.0 while 0.3.0 shipped, which is
the gap
Directory.Build.targetsalready describes having made once with0.1.0. The CHANGELOG's link references were two releases behind:
[Unreleased]still compared against
v0.2.0and[0.3.0]had no reference at all.Packing locally against the new baseline reports no difference, which is the
claim the 0.4.0 entry makes.
🤖 Generated with Claude Code