Repository navigation
Conversation
Went through all six tracked .md files and the comments in backend/app, checking every claim that can be checked mechanically: markdown links, file paths in backticks, environment variables against their use in code, make targets, API paths against the contract snapshot, the twenty feature claims in README, third-party versions against the vendored files and the Dockerfile ARG, the behaviour README attributes to backup.sh and restore.sh, and the file modes, cookie flags, CSP directives and headers SECURITY.md lists. Three things were wrong, all introduced during this release: - CHANGELOG quoted a coverage floor of 81 and a total of 81.99%. The floor is 82 and the total 82.35%; the CSRF tests landed after those lines were written. - CHANGELOG reported the release balance as 77.83% -> 81.99%, true halfway through. It is 77.83% -> 82.35%, 736 -> 844 cases. - README's restore example named a 0.1.0 backup archive. Everything else held: the lockout numbers, the session TTL, the scrypt parameters, the operation count, the cache list, the library versions and the security table all match the code. The remaining edits are register. Several paragraphs added during this release ended on an aphorism, README had the only bold rhetorical question in the repository, and the 1.0.0 heading was an abstract statement where 0.9.0 and 0.8.0 open with what changed. The explanations stay; the punchlines go.
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
One commit. A mechanical pass over all six tracked
.mdfiles and the commentsin
backend/app, run before tagging because the CHANGELOG that ships with 1.0.0is not something a patch release can quietly fix.
What was checked
Markdown links; file paths in backticks; environment variables against their use
in code;
maketargets; API paths againsttests/api_contract.json; the twentyfeature claims in README; third-party versions against the vendored files and the
Dockerfile
ARG; the behaviour README attributes tobackup.shandrestore.sh; and the file modes, cookie flags, CSP directives and headers listedin SECURITY.md's enforcement table.
What was wrong
Three things, all introduced during this release:
and the total 82.35% — the CSRF tests landed after those lines were written.
through. It is 77.83% → 82.35%, 736 → 844 cases.
0.1.0backup archive.Everything else held: lockout thresholds, session TTL, scrypt parameters,
operation count, the cache list, library versions and the security table all
match the code.
The remaining changes are register, not fact. Several paragraphs added during
this release ended on an aphorism, README carried the only bold rhetorical
question in the repository, and the 1.0.0 heading was an abstract statement where
0.9.0 and 0.8.0 open with what changed.
Verification
844 tests, coverage 82.35% against a floor of 82, mypy clean. No code paths
changed; the only file under
backend/appis a docstring edit inmain.py.