Skip to content

Run fork integration's GitHub API jobs on allowlisted runners #1101

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
david-siqi-liu merged 2 commits into from
Oct 12, 2026
Merged

Run fork integration's GitHub API jobs on allowlisted runners #1101

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
297f250
Run fork integration's GitHub API jobs on allowlisted runners
david-siqi-liu Oct 12, 2026
277b3dd
Tighten the fork integration runner and PR-read contract tests
david-siqi-liu Oct 12, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 12 additions & 6 deletions .github/workflows/fork-integration.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,9 @@ jobs:
startsWith(github.event.comment.body, '/integration-test ')) &&
contains(fromJSON('["MEMBER", "OWNER"]'), github.event.comment.author_association)
}}
runs-on: ubuntu-latest
runs-on:
group: databricks-protected-runner-group
labels: linux-ubuntu-latest
timeout-minutes: 5
permissions:
pull-requests: write
Expand All @@ -47,9 +49,9 @@ jobs:
[[ "$COMMENT_BODY" =~ ^/integration-test([[:space:]]+([0-9a-f]{7,40}))?[[:space:]]*$ ]] \
|| refuse "use \`/integration-test\`, or \`/integration-test <sha>\` with the head commit you reviewed."
requested="${BASH_REMATCH[2]}"
pr=$(gh api "repos/$GITHUB_REPOSITORY/pulls/$PR_NUMBER")
sha=$(jq -r '.head.sha' <<<"$pr")
head_repo=$(jq -r '.head.repo.full_name // empty' <<<"$pr")
# gh's built-in --jq; the protected runner image does not guarantee a jq binary.
read -r sha head_repo < <(gh api "repos/$GITHUB_REPOSITORY/pulls/$PR_NUMBER" \
--jq '[.head.sha, (.head.repo.full_name // "")] | @tsv') || true
[[ "$sha" =~ ^[0-9a-f]{40}$ ]] || refuse "could not read the PR head commit."
[ -n "$head_repo" ] || refuse "the fork repository no longer exists."
# Naming the commit binds the run to what was reviewed, not to a later push.
Expand All @@ -71,7 +73,9 @@ jobs:
startsWith(github.event.comment.body, '/integration-test ')) &&
!contains(fromJSON('["MEMBER", "OWNER"]'), github.event.comment.author_association)
}}
runs-on: ubuntu-latest
runs-on:
group: databricks-protected-runner-group
labels: linux-ubuntu-latest
timeout-minutes: 5
permissions:
pull-requests: write
Expand Down Expand Up @@ -104,7 +108,9 @@ jobs:
name: Report fork integration
needs: [authorize, integration]
if: ${{ always() && needs.authorize.result == 'success' }}
runs-on: ubuntu-latest
runs-on:
group: databricks-protected-runner-group
labels: linux-ubuntu-latest
timeout-minutes: 5
permissions:
statuses: write
Expand Down
19 changes: 16 additions & 3 deletions tests/test_integration_contract.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -224,6 +224,8 @@ def test_fork_integration_runs_only_member_requested_pr_heads():


HEAD = "0123456789abcdef0123456789abcdef01234567"
# Verified against the live API on the protected runner; a text change here must be re-verified.
PR_HEAD_JQ = '[.head.sha, (.head.repo.full_name // "")] | @tsv'


@pytest.mark.skipif(sys.platform == "win32", reason="the authorize step runs bash on Linux")
Expand All @@ -247,13 +249,14 @@ def test_fork_integration_authorize_pins_the_pr_head_or_the_reviewed_commit(
script = textwrap.dedent(authorize.split(" run: |\n", 1)[1])
bin_dir = tmp_path / "bin"
bin_dir.mkdir()
pr = {"head": {"sha": HEAD, "repo": {"full_name": head_repo} if head_repo else None}}
(tmp_path / "pr.json").write_text(json.dumps(pr))
# What real gh prints for that query; any other --jq expression gets no output and is refused.
(tmp_path / "pr.tsv").write_text(f"{HEAD}\t{head_repo or ''}\n")
fake_gh = bin_dir / "gh"
fake_gh.write_text(
"#!/bin/sh\n"
f'echo "$*" >> "{tmp_path}/calls"\n'
f'case "$*" in "api repos/databricks/unity-gateway/pulls/7") cat "{tmp_path}/pr.json";; esac\n'
f"case \"$*\" in 'api repos/databricks/unity-gateway/pulls/7 --jq {PR_HEAD_JQ}')"
f' cat "{tmp_path}/pr.tsv";; esac\n'
)
fake_gh.chmod(0o755)
output = tmp_path / "github-output"
Expand Down Expand Up @@ -284,6 +287,16 @@ def test_fork_integration_authorize_pins_the_pr_head_or_the_reviewed_commit(
assert reason in calls and "statuses/" not in calls


def test_fork_integration_github_api_jobs_run_on_allowlisted_runners():
# The org IP allow list rejects GitHub API calls from GitHub-hosted runners.
workflow = _workflow("fork-integration.yml")
protected = " runs-on:\n group: databricks-protected-runner-group\n labels: linux-ubuntu-latest\n"

assert workflow.count(protected) == 3
assert "runs-on: ubuntu-latest" not in workflow
assert not re.search(r"(?m)^\s*[^#\n]*\bjq\s+-", workflow)


def test_no_workflow_uses_pull_request_target():
# GitHub disables pull_request_target by default in public repos from 2026-11-02.
workflows = sorted((Path(__file__).parent.parent / ".github/workflows").glob("*.yml"))
Expand Down
Loading