Repository navigation
fix(triage): classify maintainer revision requests as rejected, not inconclusive - #61
Merged
Merged
Conversation
…nconclusive
FixVerifier's prompt only described reporter confirm/deny semantics
('does the fix resolve the issue for you'). A maintainer reviewing the
diff itself -- confirming the fix technically works but asking for a
simpler implementation and an explicit retriage -- didn't match either
the positive or negative examples, so it fell into 'inconclusive',
which the workflow handles by silently doing nothing: no label change,
no comment, no retriage (see withastro/astro#18590).
Broaden the negative-confirmation definition to cover a maintainer (or
anyone with technical authority) asking for the candidate fix to be
revised or redone, or explicitly requesting a retriage, even when they
say the approach otherwise works. That verdict now chains into
retriage() (per the earlier fix for #18600), so feedback like this
reaches a fresh triage attempt instead of disappearing.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Broaden
FixVerifier's negative-confirmation definition to cover a maintainer (or anyone with technical authority) reviewing the candidate fix's diff and asking for it to be revised/simplified, or explicitly requesting a retriage — even when they confirm the approach technically works.Why
FixVerifier's prompt only described reporter confirm/deny semantics ("does the preview release fix your issue"). On withastro/astro#18590, a maintainer reviewed the diff, confirmed the diagnosis and that the fix passes tests, but asked for a simpler implementation and explicitly said "please retriage with the simpler fix." That comment matched neither the positive nor negative examples, so it fell intoinconclusive— which the workflow handles by doing nothing: no label change, no comment, no retriage. From the outside it looked like the bot ignored the comment entirely.This is the same class of bug fixed in #60 (silent no-op on a verdict that actually carries actionable feedback), just hitting the
inconclusivebucket of the classifier instead ofrejected.How
FixVerifierprompt: a maintainer-authority revision request or an explicit retriage request is a negative confirmation, not inconclusive.rejectedalready chains straight intoretriage()with the full conversation (from fix(triage): chain rejected fixes straight into retriage #60), so once classified correctly the maintainer's simpler-fix suggestion reaches a fresh triage attempt.Testing
tsc --noEmitclean.vitest run: 459/459 pass.