Stop repeating the __original_version claim - #22
Merged
Merged
Conversation
All three were checked against the GitHub API rather than against memory. "Two maintainers approved it." One did: 777arc, 2026-08-17T21:52:30Z, nine minutes after the PR opened. The reviews endpoint returns exactly one review. Teque5 drove the discussion and wrote the PR, which is participation and not a second approval; the line now says so, and says what it used to say. "The accessor shipped as declared_version, not as __original_version. Reading the diff would have given the wrong name." Wrong on both counts. The string __original_version appears nowhere upstream in that spelling -- not in the issue, not in the PR body -- and the PR has named _declared_version since it was opened on 2026-08-17, so reading the diff would have given the right name. The claim is removed and replaced with what actually happened. What actually happened is better than the version being corrected, which is why it is now recorded instead. The maintainer's 2026-08-17 comment ended with three candidate names -- .declared_version, .file_version, .original_version -- and asked which. The reply from here on 2026-08-19 picked declared_version over original_version. It shipped as public declared_version. The shortlist was upstream's, the pick was endorsed from here, and neither side invented it alone. Third: the paragraph introducing the get_global_info() note said the detail "differs from what was predicted", while the note itself says it was predicted correctly from a stand-in and then confirmed. The lead-in contradicted its own content and is rewritten. Everything else in the file was checked and holds: issue filed 2026-08-10, PR opened 2026-08-17 (seven days), the PR body carrying deepcopy, the self.version removal, "Closes #159" and "increment to v1.13.0", and the monthly call being named by the maintainer in-thread. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The claim corrected in the previous commit had been copied into four other files. Each said the pull request named the attribute __original_version and that it shipped under a different name. Neither half is true: the PR body has read `add self._declared_version` since it was opened on 2026-08-17, and the string __original_version appears nowhere upstream in that spelling. What actually happened is now written where the wrong version was. The maintainer's issue comment ended with three candidates -- .declared_version, .file_version, .original_version -- and asked which; the reply from here picked the first, and that is what shipped. ROADMAP carried the second error too, "two maintainers approved", and now says one approved nine minutes after the PR opened. tests/test_io.py keeps its conclusion -- membership in FIXED_SIGMF_VERSIONS is earned by measuring an installed release, not by reading a diff -- but it needed a reason that is true. The old one was that the diff would have given the wrong name. The real one is that the half of the fix which matters downstream is invisible in the diff: get_global_info() still returns the library's spec version, and no changelog line said so. Comment only; no behaviour change. The remaining matches for "original_version" in the tree are deliberate: the maintainer's three-name shortlist, quoted, and the self-correction in docs/sigmf-python-issue-draft.md that explains what the old claim said. 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.
Two commits. The first is the correction that missed the last merge; the second removes the same wrong claim from four more files.
Why the first commit is here
f237985corrected three claims indocs/sigmf-python-issue-draft.md, but I pushed it after #21 had already been merged. A merged PR does not move its head, so it never reachedmain.I reported that at the time as a GitHub indexing lag. It was not — the timestamps are unambiguous:
The commit is cherry-picked here unchanged.
What was wrong, and what is true
Four files said the pull request named the attribute
__original_versionand that it shipped under a different name. Both halves are false, checked against the GitHub API:add self._declared_version to preserved declared versionsince it was opened on 2026-08-17.__original_versionappears nowhere upstream in that spelling — not in issue #159, not in PR #160.What actually happened, from the issue thread:
The shortlist was upstream's, the pick was endorsed from here, and it shipped as public
declared_version.ROADMAP.mdcarried the second error as well — "two maintainers approved". The reviews endpoint returns exactly one review:777arc,2026-08-17T21:52:30Z, nine minutes after the PR opened.Files
docs/sigmf-python-issue-draft.mdCHANGELOG.md[0.5.0]sigmf entrydocs/release-notes/v0.5.0.mdROADMAP.mdtests/test_io.pyThe test comment needed a new reason, not just a new name
FIXED_SIGMF_VERSIONSmembership is earned by measuring an installed release rather than reading a diff. The stated reason was that the diff would have given the wrong name — which was the false claim. The conclusion survives; the reason is replaced with one that holds: the half of the fix that matters downstream is invisible in the diff.get_global_info()still returns the library's spec version, and no changelog line said so.Deliberate remaining matches
grep original_versionstill hits five lines: the maintainer's three-name shortlist, quoted in two files, and the self-correction indocs/sigmf-python-issue-draft.mdthat records what the old claim said and why it was wrong. Removing those would delete the correction rather than make it.Verification
🤖 Generated with Claude Code