test: trigger GitHub Actions#3
Merged
Merged
Conversation
|
Warning Rate limit exceeded
To keep reviews running without waiting, you can enable usage-based add-on for your organization. This allows additional reviews beyond the hourly cap. Account admins can enable it under billing. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (12)
📝 WalkthroughWalkthroughAdds DocSync GitHub Action, related workflows, and runtime code to detect documentation drift for JS/TS repos, compare PR changes to a snapshot, optionally generate AI-backed docs, create companion PRs/comments, and update snapshots on default-branch pushes. Changes
Sequence Diagram(s)sequenceDiagram
participant GH as GitHub Actions
participant Action as DocSync Action
participant API as GitHub API
participant Parser as File Parser
participant AI as AI Service
participant FS as File System
GH->>Action: Trigger on pull_request / pull_request_target
Action->>API: List PR files (paginated)
API-->>Action: Changed file list
Action->>FS: Download selected JS/TS files
FS-->>Action: Files available
Action->>FS: Load baseline snapshot
FS-->>Action: Snapshot data
Action->>Parser: Parse files -> constructs
Parser-->>Action: Parsed structures
Action->>Action: analyzeDrift(baseline, current)
alt Drift exceeds threshold
Action->>AI: Generate docs (if API key present)
AI-->>Action: Generated markdown
Action->>API: Create companion PR with docs
API-->>Action: PR info
end
Action->>API: Post PR comment (optional)
Action->>GH: Write job summary & set outputs
Action->>FS: Cleanup temp dir
sequenceDiagram
participant GH as GitHub Actions
participant Action as DocSync Action
participant API as GitHub API
participant Parser as File Parser
participant FS as File System
GH->>Action: Trigger on push to default branch
Action->>API: Get repository default branch
API-->>Action: Repo metadata
Action->>FS: Scan & parse trackable files
FS-->>Action: Parsed structures
Action->>FS: Load existing snapshot (if any)
FS-->>Action: Snapshot data
alt Snapshot exists
Action->>Action: analyzeDrift(snapshot, current)
Action->>GH: Write analysis summary
end
Action->>FS: Write new snapshot
FS-->>Action: Snapshot saved
Action->>GH: Set outputs (files-analyzed, drift-score, drift-detected)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 inconclusive)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Review rate limit: 0/1 reviews remaining, refill in 20 minutes and 4 seconds.Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 6
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/workflows/docsync.yml:
- Around line 79-83: Replace the local action reference "uses: ./" with the
published action identifier so consumers don't resolve to their repo; update the
workflow step that currently has uses: ./ to use the released action tag (for
example, ishwar-prog/docsync@v1) so the workflow references the published action
version instead of a local path.
In `@src/action/index.js`:
- Around line 128-137: The validateInputs function currently calls
process.exit(1) after core.setFailed which bypasses the outer try/catch; remove
those process.exit calls and instead throw an Error (or rethrow) with the same
message so the error propagates to the top-level catch handler. Specifically, in
validateInputs replace the two process.exit(1) usages (the githubToken check and
the driftThreshold check) with thrown Error objects carrying the failure
message, keeping the core.setFailed call if you want the immediate log but
ensuring the thrown Error lets the surrounding async error handler run.
In `@src/action/pr-handler.js`:
- Around line 179-190: The current condition conflates true "no drift" with
"drift below threshold" by using if (!driftReport.hasDrift ||
driftReport.driftScore < inputs.driftThreshold); change the logic in the PR
handler so you first check driftReport.hasDrift (true/false) and only treat the
"no drift" path as when hasDrift is false; add a separate branch for hasDrift
=== true && driftReport.driftScore < inputs.driftThreshold that records/labels
"drift_below_threshold" (or similar) and does not call postNoDriftComment, and
ensure reporter.writeSummary receives distinct status values (e.g., 'in_sync' vs
'drift_below_threshold'); update references to postNoDriftComment and
reporter.writeSummary accordingly so messaging and labels reflect the actual
state.
- Around line 81-86: The handler exits early in multiple places without
guaranteeing the action outputs, so initialize the full output contract near the
start of the PR handler (e.g., after you read pullNumber/pr) by calling
core.setOutput('drift-detected','false') and core.setOutput('drift-score','0')
as defaults, then update/override those outputs later where real results are
computed; also audit every early-return branch (the draft check that inspects
pr.draft and other early returns that reference pullNumber/pr) and ensure each
branch either uses the initialized defaults or explicitly sets both
'drift-detected' and 'drift-score' before returning.
In `@src/action/push-handler.js`:
- Around line 60-63: In the early-return branch that checks
scanResult.totalFound === 0 in src/action/push-handler.js, set the Action
outputs (using core.setOutput) to sensible defaults before returning so
downstream steps don't receive blank values; specifically, inside that if block
assign the outputs your workflow expects (e.g., an empty list/empty string for
file/artifact outputs, zero for counts, and a status like "no_changes") and then
return. Ensure you update the same if block that contains the
scanResult.totalFound === 0 check so all expected outputs are always emitted.
In `@src/action/reporter.js`:
- Around line 36-40: The summary for the in_sync branch uses pullNumber directly
and can render "PR `#null`" on non-PR events; update the in_sync handling (the
block using status === 'in_sync' and summary.addRaw) to guard against a
null/undefined pullNumber by formatting the string conditionally (e.g., use
`pullNumber ? \`PR #${pullNumber}\` : 'push event'` or similar) before passing
it to summary.addRaw so the message never contains "PR `#null`".
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 4cab1c70-f036-4b90-82a1-774c19c00c88
⛔ Files ignored due to path filters (1)
package-lock.jsonis excluded by!**/package-lock.json
📒 Files selected for processing (9)
.github/workflows/ci.yml.github/workflows/docsync.ymlaction.ymlpackage.jsonsrc/action/index.jssrc/action/pr-handler.jssrc/action/push-handler.jssrc/action/reporter.jssrc/utils/logger.js
There was a problem hiding this comment.
Actionable comments posted: 6
♻️ Duplicate comments (1)
src/action/reporter.js (1)
27-31:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winHandle
drift_below_thresholdexplicitly in the summary.
src/action/pr-handler.jsnow passes a dedicateddrift_below_thresholdstatus, but this method still routes every non-in_synccase through the full “Documentation Drift Detected” path. That makes the summary read like an actionable failure instead of an intentional below-threshold no-op.Also applies to: 36-46
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@src/action/reporter.js` around lines 27 - 31, The summary generator in src/action/reporter.js currently treats every non-'in_sync' status as a full "Documentation Drift Detected" case; add an explicit branch for the new 'drift_below_threshold' status in the function that reads params.status (the method that builds the human-readable summary using params.driftReport, params.generationSummary, params.pullNumber, and params.companionPR) so that when status === 'drift_below_threshold' it returns a short, non-actionable summary (e.g., "Drift detected but below threshold — no action required") and avoids the full failure/error path used for 'drift_detected'; also update the analogous logic block referenced around lines 36-46 to handle 'drift_below_threshold' similarly.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/workflows/docsync.yml:
- Around line 66-74: The workflow contains a CI step named "Install
dependencies" that runs "npm ci", which should be removed because the template
action doesn't need to install consumer repo dependencies and this breaks
pnpm/yarn or lockfile-less repos; delete or disable the entire step block that
runs npm ci (the step with name "Install dependencies" and run: npm ci) or
replace it with a no-op/commented placeholder, ensuring no other steps reference
it.
- Around line 21-22: The workflow currently triggers on pull_request_target
types [opened, synchronize, reopened] which misses the draft -> ready_for_review
transition; update the pull_request_target types array in
.github/workflows/docsync.yml to include "ready_for_review" so that the handler
in src/action/pr-handler.js (which skips draft PRs) will run when a draft is
marked ready and the PR will be analyzed.
- Line 51: Replace floating action refs with immutable commit SHAs: locate the
uses entries for actions/checkout@v4, actions/setup-node@v4, and
ishwar-prog/docsync@v1 in the workflow and replace each tag (e.g., `@v4`, `@v1`)
with the corresponding full commit SHA (format @<40-char-sha>) from the official
repository commit for that release; ensure you fetch and paste the exact commit
SHA for each action, update the three refs (actions/checkout,
actions/setup-node, ishwar-prog/docsync), and scan the workflow for any other
floating refs to pin similarly.
In `@src/action/index.js`:
- Around line 110-119: readInputs currently exposes configPath but nothing
consumes it, so workflows using with: config-path get ignored; update the
pipeline to pass that configPath into the config loader and handlers: change
loadConfig(cwd) in push-handler.js (and any PR handler entry points) to accept
the configPath from readInputs (e.g., call loadConfig(cwd, inputs.configPath) or
similar), and modify src/utils/config.js's loader to accept and resolve the
provided configPath instead of always resolving 'docsync.yaml'; ensure
readInputs' configPath value is propagated from the action entry point into
whatever function calls loadConfig.
- Around line 29-34: The ESLint "no-undef" failures come from CommonJS files
using Node globals (require, process); fix by enabling the Node environment:
either add a file-level pragma ("/* eslint-env node */" and optionally "/*
globals require, process */") at the top of each CommonJS entry (e.g., the file
that imports core and github and references process.env and RUNNER_DEBUG, and
similarly in pr-handler.js and push-handler.js), or update the central
eslint.config.cjs to include env: { node: true } so require(), process, and
other Node globals are recognized; pick one approach and apply consistently
across files that export/require CommonJS modules (references: the require()
calls and process.env/RUNNER_DEBUG usages and the handlePullRequest/handlePush
imports).
In `@src/action/pr-handler.js`:
- Around line 3-16: This file has unused bindings causing ESLint failures:
remove the unused requires and parameters or make them used; specifically delete
the unused imports github and readSnapshot (or use them where needed), and if
you don't intend to call parseFiles here either remove its import or integrate
parseFiles() into the flow; change any catch((_)=>...) to catch((err)=>...) or
omit the unused `_` variable, and remove the unused driftReport parameter from
the postNoDriftComment function signature (or use it inside the function).
Update the top-level requires and the postNoDriftComment declaration accordingly
so no unused symbols remain.
---
Duplicate comments:
In `@src/action/reporter.js`:
- Around line 27-31: The summary generator in src/action/reporter.js currently
treats every non-'in_sync' status as a full "Documentation Drift Detected" case;
add an explicit branch for the new 'drift_below_threshold' status in the
function that reads params.status (the method that builds the human-readable
summary using params.driftReport, params.generationSummary, params.pullNumber,
and params.companionPR) so that when status === 'drift_below_threshold' it
returns a short, non-actionable summary (e.g., "Drift detected but below
threshold — no action required") and avoids the full failure/error path used for
'drift_detected'; also update the analogous logic block referenced around lines
36-46 to handle 'drift_below_threshold' similarly.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 4b85d906-3cfa-4a78-9a0a-c41520d70b10
📒 Files selected for processing (7)
.github/workflows/ci.yml.github/workflows/docsync.ymleslint.config.cjssrc/action/index.jssrc/action/pr-handler.jssrc/action/push-handler.jssrc/action/reporter.js
🚧 Files skipped from review as they are similar to previous changes (1)
- .github/workflows/ci.yml
There was a problem hiding this comment.
Actionable comments posted: 6
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@src/action/pr-handler.js`:
- Around line 414-474: The current flow always creates a timestamped
companionBranch and a new PR (createRef + pulls.create), causing duplicate
companion PRs; make companion PR creation idempotent by using a deterministic
branch name (e.g., `docsync/pr-${pullNumber}` instead of including Date.now()),
then before createRef/check createOrUpdateFileContents call try to fetch the ref
with octokit.rest.git.getRef for that branch and update it (update the ref) if
it exists or create it if not; next, search for an existing companion PR for
that head and source PR (use octokit.rest.pulls.list or search for open PRs with
head `${owner}:${companionBranch}`) and if found call octokit.rest.pulls.update
to update title/body (and ensure the branch is pushed), otherwise call
octokit.rest.pulls.create as currently done; reference companionBranch,
octokit.rest.git.createRef/getRef, octokit.rest.git.updateRef,
octokit.rest.repos.createOrUpdateFileContents, and
octokit.rest.pulls.create/pulls.update to implement this behavior.
- Around line 58-60: handlePullRequest is not loading the DocSync config
(inputs.configPath) so it uses extension-only filtering and the hardcoded
companion-doc path, ignoring track/ignore/output.dir/output.format; load the
same DocSync config used by the push path at the start of async function
handlePullRequest (using the project's existing config loader) and merge/apply
its track, ignore and output settings when computing companion doc paths and
filters instead of using hardcoded defaults; make the identical change in the
other PR-related blocks that mirror this logic so they all derive behavior from
the loaded config rather than hardcoded values.
In `@src/action/push-handler.js`:
- Around line 41-45: The call to octokit.rest.repos.get(...) that sets
defaultBranch is currently outside the existing try/catch, so network or
permission failures escape core.setFailed; move the repo lookup (the
octokit.rest.repos.get call and assignment to defaultBranch) into the same try
block that wraps the rest of the handler logic (and likewise move the other
external repo/default-branch lookup at the later site referenced) so any thrown
errors are caught and handled by the existing catch which calls core.setFailed;
ensure you keep the same variable name defaultBranch and do not create duplicate
declarations when relocating these calls.
- Line 3: Remove the unused "path" import to satisfy the linter: delete the
`const path = require('path');` statement in src/action/push-handler.js (the
unused symbol is `path`) so no-unused-vars CI error remains; ensure no other
code depends on `path` before removing and run tests/linter after change.
In `@src/utils/config.js`:
- Around line 56-62: The loadConfig function updates the missing-file warning to
use the configurable configFilename, but later log messages still hardcode
"docsync.yaml"; update all diagnostic messages in loadConfig (including success
and parse-error logs) to reference the configFilename or configPath variable
instead of the literal 'docsync.yaml' so messages reflect custom filenames (look
for usages in loadConfig of logger.warn/logger.info/logger.error and the
variables configFilename and configPath).
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: ba32cec0-b89f-4e5f-bba2-ed8c37ca8c1a
📒 Files selected for processing (5)
.github/workflows/docsync.ymleslint.config.cjssrc/action/pr-handler.jssrc/action/push-handler.jssrc/utils/config.js
🚧 Files skipped from review as they are similar to previous changes (1)
- eslint.config.cjs
Comment on lines
+58
to
+60
| async function handlePullRequest({ octokit, context, inputs }) { | ||
| const { repo } = context; | ||
| const pr = context.payload.pull_request; |
There was a problem hiding this comment.
Load DocSync config for PR runs too.
inputs.configPath is available here, but this handler never loads config. The extension-only filter and hardcoded companion-doc path ignore track, ignore, output.dir, and output.format, so PR drift can disagree with the snapshot built by src/action/push-handler.js and generate docs into the wrong location.
Also applies to: 107-111, 423-430
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/action/pr-handler.js` around lines 58 - 60, handlePullRequest is not
loading the DocSync config (inputs.configPath) so it uses extension-only
filtering and the hardcoded companion-doc path, ignoring
track/ignore/output.dir/output.format; load the same DocSync config used by the
push path at the start of async function handlePullRequest (using the project's
existing config loader) and merge/apply its track, ignore and output settings
when computing companion doc paths and filters instead of using hardcoded
defaults; make the identical change in the other PR-related blocks that mirror
this logic so they all derive behavior from the loaded config rather than
hardcoded values.
Comment on lines
+414
to
+474
| // Create companion branch | ||
| const companionBranch = `docsync/pr-${pullNumber}-${Date.now()}`; | ||
| await octokit.rest.git.createRef({ | ||
| owner, repo, | ||
| ref: `refs/heads/${companionBranch}`, | ||
| sha: baseSha, | ||
| }); | ||
| core.info(`Created branch: ${companionBranch}`); | ||
|
|
||
| // Commit documentation file | ||
| await octokit.rest.repos.createOrUpdateFileContents({ | ||
| owner, repo, | ||
| path: 'docs/api-reference.md', | ||
| message: `docs: auto-generate documentation for PR #${pullNumber} [DocSync]`, | ||
| content: Buffer.from(docContent, 'utf8').toString('base64'), | ||
| branch: companionBranch, | ||
| }); | ||
| core.info('Committed documentation file'); | ||
|
|
||
| // Build PR body | ||
| const aiInfo = generationSummary | ||
| ? `\n| AI Model | ${generationSummary.results[0]?.model || 'Llama 3.3 70B'} |` | ||
| : ''; | ||
|
|
||
| const prBody = `## 📄 DocSync — Automated Documentation Update | ||
|
|
||
| This PR was automatically created by **DocSync** because PR #${pullNumber} introduced code changes that caused documentation drift. | ||
|
|
||
| ### Drift Summary | ||
|
|
||
| | Metric | Value | | ||
| |--------|-------| | ||
| | Drift Score | **${driftReport.driftScore}/100** | | ||
| | Files Affected | ${driftReport.summary.filesAffected} | | ||
| | Total Changes | ${driftReport.summary.totalChanges} |${aiInfo} | ||
|
|
||
| ### Changes Detected | ||
|
|
||
| ${Object.entries(driftReport.summary.byType) | ||
| .map(([type, count]) => `- **${type}**: ${count} occurrence(s)`) | ||
| .join('\n')} | ||
|
|
||
| ### Review Instructions | ||
|
|
||
| 1. Open \`docs/api-reference.md\` to see the generated documentation | ||
| 2. Review for accuracy — AI documentation is thorough but not omniscient | ||
| 3. Make any necessary corrections | ||
| 4. Merge this PR alongside or after PR #${pullNumber} | ||
|
|
||
| --- | ||
| *🤖 Generated by [DocSync](https://github.com/ishwar-prog/docsync) — Docs that stay true to your code*`; | ||
|
|
||
| // Open the PR | ||
| const { data: companionPR } = await octokit.rest.pulls.create({ | ||
| owner, repo, | ||
| title: `📄 DocSync: Update docs for PR #${pullNumber} — "${prTitle}"`, | ||
| body: prBody, | ||
| head: companionBranch, | ||
| base: baseBranch, | ||
| draft: false, | ||
| }); |
There was a problem hiding this comment.
Make companion PR creation idempotent.
This always creates a timestamped branch and then opens a new PR. Because .github/workflows/docsync.yml runs on synchronize with open-companion-pr: 'true', pushing multiple updates to the same source PR will accumulate multiple open DocSync PRs instead of refreshing one companion branch/PR.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/action/pr-handler.js` around lines 414 - 474, The current flow always
creates a timestamped companionBranch and a new PR (createRef + pulls.create),
causing duplicate companion PRs; make companion PR creation idempotent by using
a deterministic branch name (e.g., `docsync/pr-${pullNumber}` instead of
including Date.now()), then before createRef/check createOrUpdateFileContents
call try to fetch the ref with octokit.rest.git.getRef for that branch and
update it (update the ref) if it exists or create it if not; next, search for an
existing companion PR for that head and source PR (use octokit.rest.pulls.list
or search for open PRs with head `${owner}:${companionBranch}`) and if found
call octokit.rest.pulls.update to update title/body (and ensure the branch is
pushed), otherwise call octokit.rest.pulls.create as currently done; reference
companionBranch, octokit.rest.git.createRef/getRef, octokit.rest.git.updateRef,
octokit.rest.repos.createOrUpdateFileContents, and
octokit.rest.pulls.create/pulls.update to implement this behavior.
Comment on lines
+41
to
+45
| const { data: repoData } = await octokit.rest.repos.get({ | ||
| owner: repo.owner, | ||
| repo: repo.repo, | ||
| }); | ||
| const defaultBranch = repoData.default_branch; |
There was a problem hiding this comment.
Move the default-branch lookup inside the existing try.
octokit.rest.repos.get(...) is the first external call in this handler, but it sits outside the catch that calls core.setFailed(...). A permission or network failure here escapes as an unhandled action error.
🔧 Suggested fix
- const pushedBranch = context.ref.replace('refs/heads/', '');
- const { data: repoData } = await octokit.rest.repos.get({
- owner: repo.owner,
- repo: repo.repo,
- });
- const defaultBranch = repoData.default_branch;
-
- if (pushedBranch !== defaultBranch) {
- core.info(`Push to ${pushedBranch} (not default branch ${defaultBranch}) — skipping snapshot update`);
- return;
- }
-
- core.info(`Push to default branch ${defaultBranch} — updating snapshot`);
-
- const cwd = process.cwd();
-
try {
+ const pushedBranch = context.ref.replace('refs/heads/', '');
+ const { data: repoData } = await octokit.rest.repos.get({
+ owner: repo.owner,
+ repo: repo.repo,
+ });
+ const defaultBranch = repoData.default_branch;
+
+ if (pushedBranch !== defaultBranch) {
+ core.info(`Push to ${pushedBranch} (not default branch ${defaultBranch}) — skipping snapshot update`);
+ return;
+ }
+
+ core.info(`Push to default branch ${defaultBranch} — updating snapshot`);
+
+ const cwd = process.cwd();Also applies to: 56-57
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/action/push-handler.js` around lines 41 - 45, The call to
octokit.rest.repos.get(...) that sets defaultBranch is currently outside the
existing try/catch, so network or permission failures escape core.setFailed;
move the repo lookup (the octokit.rest.repos.get call and assignment to
defaultBranch) into the same try block that wraps the rest of the handler logic
(and likewise move the other external repo/default-branch lookup at the later
site referenced) so any thrown errors are caught and handled by the existing
catch which calls core.setFailed; ensure you keep the same variable name
defaultBranch and do not create duplicate declarations when relocating these
calls.
Comment on lines
+87
to
+89
| // Update the snapshot to reflect the current state | ||
| writeSnapshot(cwd, parsedFiles, { repoRoot: cwd }); | ||
| core.info(`Snapshot updated: ${parsedFiles.length} file(s), ${parsedFiles.reduce((s, f) => s + f.constructCount, 0)} construct(s)`); |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== writeSnapshot implementation ==="
sed -n '1,260p' src/core/snapshot.js | nl -ba | sed -n '/writeSnapshot/,/module.exports/p'
echo
echo "=== push handler snapshot write ==="
sed -n '1,140p' src/action/push-handler.js | nl -ba | sed -n '80,100p'
echo
echo "=== workflow steps that could persist the updated snapshot ==="
rg -nC2 'git (add|commit|push)|createOrUpdateFileContents|peter-evans/create-pull-request' .github/workflows src/actionRepository: ishwar-prog/docsync
Length of output: 161
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Checking writeSnapshot implementation ==="
head -300 src/core/snapshot.js | grep -A 50 "function writeSnapshot\|const writeSnapshot\|writeSnapshot ="
echo
echo "=== Checking push handler lines 80-100 ==="
sed -n '80,100p' src/action/push-handler.js
echo
echo "=== Full push handler to understand context ==="
wc -l src/action/push-handler.js
echo
echo "=== Checking for git operations in workflow ==="
rg "git commit|git push|createOrUpdateFileContents|create-pull-request" .github/workflows/
echo
echo "=== Checking full docsync.yml workflow ==="
cat .github/workflows/docsync.ymlRepository: ishwar-prog/docsync
Length of output: 2971
🏁 Script executed:
cat .github/workflows/docsync.ymlRepository: ishwar-prog/docsync
Length of output: 3625
🏁 Script executed:
grep -r "git commit\|git push\|createOrUpdateFileContents\|gitAction\|updateFileContents" src/action/ src/core/Repository: ishwar-prog/docsync
Length of output: 146
Commit the updated snapshot to the repository so baseline changes persist.
writeSnapshot() writes the snapshot only to the ephemeral runner filesystem. The push handler never commits or pushes this change, and the workflow has no follow-up persistence step. The updated snapshot disappears when the job ends, so the baseline never advances on push.
Comment on lines
+56
to
+62
| function loadConfig(cwd = process.cwd(), configFilename = 'docsync.yaml') { | ||
| const configPath = path.resolve(cwd, configFilename); | ||
|
|
||
| // If no config file exists, that's fine — use all defaults. | ||
| // This is important: the tool should work without any configuration. | ||
| if (!fs.pathExistsSync(configPath)) { | ||
| logger.warn('No docsync.yaml found. Using default configuration.'); | ||
| logger.warn(`No ${configFilename} found. Using default configuration.`); |
There was a problem hiding this comment.
Use the configurable filename in the remaining loadConfig() diagnostics.
This change updates the missing-file warning, but the success and parse-error messages later in the same function still hardcode docsync.yaml. Custom config paths will be misleading to debug.
🔧 Suggested fix
- logger.info(`Config loaded from ${chalk.cyan('docsync.yaml')}`);
+ logger.info(`Config loaded from ${chalk.cyan(configFilename)}`);
...
- logger.error(`Failed to parse docsync.yaml: ${error.message}`);
+ logger.error(`Failed to parse ${configFilename}: ${error.message}`);🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@src/utils/config.js` around lines 56 - 62, The loadConfig function updates
the missing-file warning to use the configurable configFilename, but later log
messages still hardcode "docsync.yaml"; update all diagnostic messages in
loadConfig (including success and parse-error logs) to reference the
configFilename or configPath variable instead of the literal 'docsync.yaml' so
messages reflect custom filenames (look for usages in loadConfig of
logger.warn/logger.info/logger.error and the variables configFilename and
configPath).
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.
Summary by CodeRabbit
New Features
Chores
Quality