fix(autofix): verify reviewdog actually posted before claiming "click Apply"
The sticky summary comment was stating "Posted formatting suggestions inline. Click Apply suggestion on each" even when reviewdog landed zero inline review comments — typical case: the formatter touched lines outside the PR's added range, so `-filter-mode=added` (correctly) filtered everything out. The script unconditionally set `posted=true` after running reviewdog regardless of whether any comments were actually created, leaving the user staring at a sticky that promised buttons that didn't exist. The publish job now snapshots the count of `github-actions[bot]` review comments before and after reviewdog. If the delta is zero, surface a new `diff-no-overlap` UI state that tells the user plainly: "Formatter found fixable issues, but they're on lines outside this PR's added range — there's nothing to click here. Run locally: npm run lint:fix && npm run format." Plus a matching `gitnexus/autofix` Check Run conclusion (still neutral, distinct title) so agents reading `gh pr checks` see the same signal. Three states are now machine-distinguishable in the sticky's gitnexus-autofix JSON block: suggestions-posted (delta > 0), diff-no-overlap (delta == 0), skipped-too-large (>3k lines).
This commit is contained in:
Gergo Magyar
@@ -131,10 +131,13 @@ jobs:
|
||||
if: steps.meta.outputs.changed_lines != '0'
|
||||
env:
|
||||
REVIEWDOG_GITHUB_API_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
GH_REPO: ${{ github.repository }}
|
||||
CI_REPO_OWNER: ${{ github.repository_owner }}
|
||||
CI_REPO_NAME: ${{ github.event.repository.name }}
|
||||
CI_PULL_REQUEST: ${{ steps.meta.outputs.pr_number }}
|
||||
CI_COMMIT: ${{ steps.meta.outputs.head_sha }}
|
||||
PR: ${{ steps.meta.outputs.pr_number }}
|
||||
# Pull `changed_lines` through env so bash gets a real
|
||||
# variable (and shellcheck SC2170 doesn't fire on `-gt` against
|
||||
# a `${{ }}`-interpolated literal).
|
||||
@@ -158,10 +161,19 @@ jobs:
|
||||
exit 0
|
||||
fi
|
||||
|
||||
# Snapshot the count of bot-authored review comments BEFORE
|
||||
# reviewdog. Reviewdog with `-filter-mode=added` silently posts
|
||||
# nothing when the patch's lines don't overlap the PR's added
|
||||
# range (typical case: formatter touched code that wasn't part
|
||||
# of this PR). It exits 0 either way, so the only reliable
|
||||
# signal is "did new review comments actually show up?".
|
||||
before=$(gh api "repos/${GH_REPO}/pulls/${PR}/comments" \
|
||||
--paginate --jq '[.[] | select(.user.login == "github-actions[bot]")] | length')
|
||||
|
||||
# `-f.diff.strip=1` matches `git diff` output (a/foo b/foo).
|
||||
# `-filter-mode=added` only suggests on lines the PR added,
|
||||
# which avoids re-suggesting on already-resolved threads when
|
||||
# the contributor re-adds the autoformat label.
|
||||
# which avoids re-suggesting on already-resolved threads on
|
||||
# subsequent pushes.
|
||||
reviewdog \
|
||||
-f=diff -f.diff.strip=1 \
|
||||
-name="prettier+eslint" \
|
||||
@@ -170,7 +182,24 @@ jobs:
|
||||
-level=warning \
|
||||
-fail-on-error=false < "$patch"
|
||||
|
||||
echo "posted=true" >> "$GITHUB_OUTPUT"
|
||||
# Verify reviewdog actually posted suggestions. Without this
|
||||
# check, a "0 review comments" run still emits a sticky claiming
|
||||
# "Click Apply suggestion" — confusing because there's nothing
|
||||
# to click.
|
||||
after=$(gh api "repos/${GH_REPO}/pulls/${PR}/comments" \
|
||||
--paginate --jq '[.[] | select(.user.login == "github-actions[bot]")] | length')
|
||||
delta=$((after - before))
|
||||
echo "Bot review comments: before=$before after=$after delta=$delta"
|
||||
|
||||
if [ "$delta" -gt 0 ]; then
|
||||
echo "posted=true" >> "$GITHUB_OUTPUT"
|
||||
else
|
||||
# Patch had fixes but reviewdog couldn't surface them as
|
||||
# suggestions — almost always because the formatter touched
|
||||
# lines outside the PR's added range, which `-filter-mode=added`
|
||||
# correctly filters out. Sticky tells the user to apply locally.
|
||||
echo "posted=no-overlap" >> "$GITHUB_OUTPUT"
|
||||
fi
|
||||
|
||||
- name: Upsert sticky summary comment
|
||||
# Only post when ci-quality found something fixable (= the
|
||||
@@ -203,6 +232,14 @@ jobs:
|
||||
if [ "${POSTED}" = "skipped-too-large" ]; then
|
||||
ui_state="skipped-too-large"
|
||||
prose="Diff is **${CHANGED}** lines — too large for inline suggestions (GitHub caps the review-comment API at ~3000). Run locally: \`npm run lint:fix && npm run format\`."
|
||||
elif [ "${POSTED}" = "no-overlap" ]; then
|
||||
# Reviewdog ran but couldn't post any inline suggestions —
|
||||
# the formatter touched lines outside this PR's added range,
|
||||
# which \`-filter-mode=added\` correctly filters out. There's
|
||||
# nothing for **Apply suggestion** to click; the user has to
|
||||
# apply locally.
|
||||
ui_state="diff-no-overlap"
|
||||
prose="Formatter found fixable issues, but they're on lines outside this PR's added range — there's nothing to click here. Run locally: \`npm run lint:fix && npm run format\`."
|
||||
else
|
||||
ui_state="suggestions-posted"
|
||||
prose="Posted formatting / unused-import suggestions inline. Click **Apply suggestion** on each, or run locally: \`npm run lint:fix && npm run format\`."
|
||||
@@ -297,6 +334,10 @@ jobs:
|
||||
conclusion="neutral"
|
||||
title="Diff too large for inline suggestions (${CHANGED} lines)"
|
||||
summary="GitHub caps the review-comment API at ~3000 lines. Run \`npm run lint:fix && npm run format\` locally."
|
||||
elif [ "${POSTED}" = "no-overlap" ]; then
|
||||
conclusion="neutral"
|
||||
title="Formatter changes don't overlap PR's added range"
|
||||
summary="Reviewdog couldn't post inline suggestions because the formatter touched lines outside this PR's added range. Run \`npm run lint:fix && npm run format\` locally."
|
||||
else
|
||||
conclusion="neutral"
|
||||
title="Suggestions posted"
|
||||
|
||||
Reference in New Issue
Block a user