feat(comment-commands): auto suggest reviewers on pr open/update via git blame by juliethecao · Pull Request #5651 · apache/texera

juliethecao · 2026-06-12T22:54:04Z

What changes were proposed in this PR?

This PR adds an automatic reviewer suggestion CI job to .github/workflows/comment-commands.yml. When a PR is opened or updated (pull_request: opened, synchronize, reopened), the CI automatically runs git blame -p at the base commit on each changed file to identify who most recently touched that code. Candidates are split into two groups:

Committers — collaborators who can be formally review-requested via GitHub's API
Non-committer contributors — have context but cannot be review-requested; the author can @-mention them to notify

The CI posts a comment in this format:
Suggested reviewers (based on git blame of changed files):
Committers — can be formally requested: @ alice, @ bob
Non-committer contributors — cc to notify: @ carol
Use /request-review @ alice to request a review, or cc @ carol to notify them.

On every subsequent push, the job finds the existing suggestion comment (via a hidden HTML marker ) and edits it in place, keeping the PR timeline clean. The CI never sends a review request on its own as the author must explicitly use /request-review @ user.

Files with status added are skipped before git blame is attempted since they did not exist at the base commit.

Any related issues, documentation, discussions?

Closes #5611

How was this PR tested?

Unit tests: 83 tests across 9 suites were written locally (https://github.com/juliethecao/texera/tree/cc-test) to cover the core JavaScript logic extracted from the workflow: git blame -p output parsing, candidate ranking, comment body generation, find-or-update marker logic, author/bot exclusion, @ mention parsing, file status filtering, candidate accumulation, and MARKER integrity. Tests were not checked in as the logic lives inside a GitHub Actions script rather than a standalone module.

Manual CI test: A test PR was opened on a personal fork (juliethecao#9) against the feature branch as the base. The suggest-reviewers job triggered on open, ran git blame on the changed files, and posted the suggestion comment. Closing and reopening the PR confirmed the comment was updated in place rather than duplicated.

Was this PR authored or co-authored using generative AI tooling?

Co-authored with Claude Sonnet 4.6 in compliance with ASF guidelines

…t-review-no-args feat: assign reviewer on /request-review with no arguments

…t-review-no-args test(comment-commands): add blame-parser tests

Signed-off-by: oplaws <juucceey@gmail.com>

Co-authored-by: Matthew B. <mgball@uci.edu> Signed-off-by: Julie Cao <116243642+juliethecao@users.noreply.github.com>

…er detection

…git blame Add a suggest-reviewers CI job that runs on pull_request (opened, synchronize, reopened). It blame-scans changed files, splits candidates into committers (requestable) and non-committer contributors (cc-only), then posts or updates a single suggestion comment per PR. Explicit @mentions are still required; the suggestion comment provides the candidates to choose from.

codecov-commenter · 2026-06-12T22:55:35Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 52.91%. Comparing base (fa5fcbb) to head (50cd0f2).
⚠️ Report is 22 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #5651      +/-   ##
============================================
- Coverage     52.96%   52.91%   -0.06%     
- Complexity     2523     2628     +105     
============================================
  Files          1075     1090      +15     
  Lines         41743    42208     +465     
  Branches       4513     4531      +18     
============================================
+ Hits          22109    22333     +224     
- Misses        18330    18569     +239     
- Partials       1304     1306       +2

Flag	Coverage Δ		*Carryforward flag
access-control-service	`70.91% <ø> (-0.52%)`	⬇️
agent-service	`34.36% <ø> (ø)`
amber	`53.03% <ø> (-0.73%)`	⬇️
computing-unit-managing-service	`1.65% <ø> (ø)`
config-service	`56.71% <ø> (ø)`
file-service	`57.06% <ø> (ø)`
frontend	`47.86% <ø> (+0.48%)`	⬆️
pyamber	`90.68% <ø> (-0.04%)`	⬇️
python	`90.74% <ø> (ø)`		Carriedforward from 8f9566f
workflow-compiling-service	`58.69% <ø> (ø)`

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

github-actions · 2026-06-12T22:58:47Z

✅ No material benchmark regressions detected

🟢 7 better · 🔴 0 worse · ⚪ 8 noise (<±5%) · 0 without baseline

Compared against main 73c76f5 benchmarked on this same runner, so the delta is largely free of cross-runner hardware noise. The "7d avg" column still reflects the gh-pages dashboard. Treat <±5% as noise unless repeated.

Dashboard · Run

	config	throughput	MB/s	latency	max Δ latest / 7d
🟢	bs=10 sw=10 sl=64	443	0.271	21,143/32,328/32,328 us	🟢 -15.6% / 🟢 -11.1%
⚪	bs=100 sw=10 sl=64	954	0.582	102,805/135,526/135,526 us	⚪ within ±5% / 🟢 -8.4%
🟢	bs=1000 sw=10 sl=64	1,122	0.685	891,170/913,914/913,914 us	🟢 -13.2% / 🟢 -10.7%

Baseline details

Latest main 73c76f5 from same runner

config	metric	PR	latest main	7d avg	Δ latest	Δ 7d
bs=10 sw=10 sl=64	throughput	443 tuples/sec	401 tuples/sec	410.82 tuples/sec	+10.5%	+7.8%
bs=10 sw=10 sl=64	MB/s	0.271 MB/s	0.245 MB/s	0.251 MB/s	+10.6%	+8.1%
bs=10 sw=10 sl=64	p50	21,143 us	25,059 us	23,785 us	-15.6%	-11.1%
bs=10 sw=10 sl=64	p95	32,328 us	34,416 us	34,980 us	-6.1%	-7.6%
bs=10 sw=10 sl=64	p99	32,328 us	34,416 us	34,980 us	-6.1%	-7.6%
bs=100 sw=10 sl=64	throughput	954 tuples/sec	949 tuples/sec	891.94 tuples/sec	+0.5%	+7.0%
bs=100 sw=10 sl=64	MB/s	0.582 MB/s	0.579 MB/s	0.544 MB/s	+0.5%	+6.9%
bs=100 sw=10 sl=64	p50	102,805 us	104,273 us	112,277 us	-1.4%	-8.4%
bs=100 sw=10 sl=64	p95	135,526 us	129,827 us	139,802 us	+4.4%	-3.1%
bs=100 sw=10 sl=64	p99	135,526 us	129,827 us	139,802 us	+4.4%	-3.1%
bs=1000 sw=10 sl=64	throughput	1,122 tuples/sec	1,104 tuples/sec	1,041 tuples/sec	+1.6%	+7.8%
bs=1000 sw=10 sl=64	MB/s	0.685 MB/s	0.674 MB/s	0.635 MB/s	+1.6%	+7.8%
bs=1000 sw=10 sl=64	p50	891,170 us	894,019 us	972,714 us	-0.3%	-8.4%
bs=1000 sw=10 sl=64	p95	913,914 us	1,052,542 us	1,023,057 us	-13.2%	-10.7%
bs=1000 sw=10 sl=64	p99	913,914 us	1,052,542 us	1,023,057 us	-13.2%	-10.7%

Raw CSV

config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,451.04,200,128000,443,0.271,21142.67,32328.44,32328.44
1,100,10,64,20,2097.51,2000,1280000,954,0.582,102805.09,135526.12,135526.12
2,1000,10,64,20,17825.67,20000,12800000,1122,0.685,891169.73,913913.99,913913.99

…lback

Yicong-Huang · 2026-06-13T01:28:00Z

I see on your fork run

Suggested reviewers (based on git blame of changed files):

Committers — none identified

No candidates found from blame history.

This is not ideal right? for a new file, there is no blame history. will it go to blame the parent folder or related file?

Could you test on three cases:

adding a new file
modifying a file
deleting a file

To see the recommendations of each?

juliethecao · 2026-06-13T02:39:28Z

Case	Current behavior	Why
Adding a new file	`git blame` fails (caught), skips file → "No candidates found"	File didn't exist at base commit as there's nothing to blame
Modifying a file	`git blame` succeeds → suggests committers/non-committers	This is the primary intended use case
Deleting a file	Skipped entirely (`status === 'removed'`) → "No candidates found"	File existed at base so blame is technically possible, but see note below

For all three test PRs, I used .github/welcome-first-time-contributor.txt as the test file to add it, modify it, and delete it respectively.

Add case: test: add new file to test suggest-reviewers CI (add case) juliethecao/texera#10
Modify case: test: modify existing file to test suggest-reviewers CI (modify case) juliethecao/texera#11
Delete case: test: delete existing file to test suggest-reviewers CI (delete case) juliethecao/texera#12

Let me know if the modify case should use a more representative file (e.g. something with richer blame history across more contributors).

On the delete case: I currently skip deleted files, but since the file existed at the base commit, git blame would technically work. That said, I'm not sure it's the right call as whoever is deleting a file likely already knows what they're doing and is the most appropriate reviewer themselves. Happy to change the logic if you think surfacing blame candidates for deletions would be useful.

Ma77Ball

I left some quick changes below. After those are addressed, the design LGTM!

…t renaming Co-authored-by: Matthew B. <mgball@uci.edu> Signed-off-by: Julie Cao <116243642+juliethecao@users.noreply.github.com>

…naming logic Co-authored-by: Matthew B. <mgball@uci.edu> Signed-off-by: Julie Cao <116243642+juliethecao@users.noreply.github.com>

Yicong-Huang · 2026-06-15T03:28:57Z

@juliethecao let's make sure the suggestion won't actually tag a person. for example in juliethecao#11

if you @Ma77Ball he will receive a github notification. if you @Ma77Ball (wrapped in code) he will not receive a notification. let's wrap it in code in suggestion. It is up to the author to actually make the request.

Sorry Matthew for using your account as an example

juliethecao and others added 14 commits May 28, 2026 14:59

feat: assign reviewer on /request-review with no arguments

166a843

Merge pull request #1 from juliethecao/feat-assign-reviewer-on-reques…

b74122c

…t-review-no-args feat: assign reviewer on /request-review with no arguments

test(comment-commands): add blame-parser tests

e837c23

Merge pull request #2 from juliethecao/feat-assign-reviewer-on-reques…

5e5c441

…t-review-no-args test(comment-commands): add blame-parser tests

Update request-review-test.js

ce6bde1

Signed-off-by: oplaws <juucceey@gmail.com>

Merge branch 'apache:main' into main

1c8542c

Merge branch 'apache:main' into main

c38e0aa

test(comment-commands): remove blame-parser tests

fe33960

Update .github/workflows/comment-commands.yml

21a5dd9

Co-authored-by: Matthew B. <mgball@uci.edu> Signed-off-by: Julie Cao <116243642+juliethecao@users.noreply.github.com>

fix(comment-commands): guard against edge cases in blame-based review…

058e91b

…er detection

Merge branch 'apache:main' into main

dbc77d3

Merge branch 'apache:main' into main

74ce5ab

fix(suggest-reviewers): skip added files in git blame loop

894fdab

github-actions Bot assigned juliethecao Jun 12, 2026

github-actions Bot added feature ci changes related to CI labels Jun 12, 2026

docs(comment-commands): remove stale comment about scrapped blame fal…

0c808f3

…lback

Ma77Ball suggested changes Jun 14, 2026

View reviewed changes

Comment thread .github/workflows/comment-commands.yml Outdated

Comment thread .github/workflows/comment-commands.yml Outdated

juliethecao and others added 2 commits June 14, 2026 17:37

Update .github/workflows/comment-commands.yml with pull_request_targe…

8f9566f

…t renaming Co-authored-by: Matthew B. <mgball@uci.edu> Signed-off-by: Julie Cao <116243642+juliethecao@users.noreply.github.com>

Update .github/workflows/comment-commands.yml with previous_filename …

50cd0f2

…naming logic Co-authored-by: Matthew B. <mgball@uci.edu> Signed-off-by: Julie Cao <116243642+juliethecao@users.noreply.github.com>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(comment-commands): auto suggest reviewers on pr open/update via git blame#5651

feat(comment-commands): auto suggest reviewers on pr open/update via git blame#5651
juliethecao wants to merge 17 commits into
apache:mainfrom
juliethecao:feat-auto-suggest-reviewers-on-PR-open/update-via-git-blame

juliethecao commented Jun 12, 2026

Uh oh!

codecov-commenter commented Jun 12, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Jun 12, 2026 •

edited

Loading

Uh oh!

Yicong-Huang commented Jun 13, 2026

Uh oh!

juliethecao commented Jun 13, 2026 •

edited

Loading

Uh oh!

Ma77Ball left a comment

Uh oh!

Uh oh!

Uh oh!

Yicong-Huang commented Jun 15, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

juliethecao commented Jun 12, 2026

What changes were proposed in this PR?

Any related issues, documentation, discussions?

How was this PR tested?

Was this PR authored or co-authored using generative AI tooling?

Uh oh!

codecov-commenter commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

github-actions Bot commented Jun 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ No material benchmark regressions detected

Uh oh!

Yicong-Huang commented Jun 13, 2026

Uh oh!

juliethecao commented Jun 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Ma77Ball left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Yicong-Huang commented Jun 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

codecov-commenter commented Jun 12, 2026 •

edited

Loading

github-actions Bot commented Jun 12, 2026 •

edited

Loading

juliethecao commented Jun 13, 2026 •

edited

Loading

Yicong-Huang commented Jun 15, 2026 •

edited

Loading