Files
sglang/.github/workflows/cancel-unfinished-pr-tests.yml

235 lines
9.5 KiB
YAML

name: Cancel Unfinished PR Runs
on:
workflow_dispatch:
inputs:
workflows:
description: 'Space-separated list of workflow filenames to cancel'
required: true
type: string
default: 'pr-test.yml pr-test-extra.yml'
include_highest_priority:
description: 'Also cancel runs from PRs labelled highest-priority'
required: false
type: boolean
default: false
include_rerun_test:
description: 'Also cancel /rerun-test dispatched runs (rerun-test.yml)'
required: false
type: boolean
default: false
permissions:
actions: write # Needed to cancel runs
contents: read # Needed to read repo info
pull-requests: read # needed for gh pr view (labels)
jobs:
cancel-unfinished-pr-runs:
runs-on: ubuntu-latest
steps:
- name: Install GitHub CLI
run: sudo apt-get install -y gh jq
- name: Cancel unfinished PR-associated runs
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
REPO: ${{ github.repository }}
WORKFLOWS: ${{ github.event.inputs.workflows }}
INCLUDE_HIGHEST_PRIORITY: ${{ github.event.inputs.include_highest_priority }}
INCLUDE_RERUN_TEST: ${{ github.event.inputs.include_rerun_test }}
shell: bash
run: |
set -euo pipefail
# Read the space-separated string from the input into a bash array
read -r -a WORKFLOW_FILES <<< "${WORKFLOWS}"
if [ "$INCLUDE_RERUN_TEST" = "true" ]; then
WORKFLOW_FILES+=("rerun-test.yml")
fi
echo "Targeting ${#WORKFLOW_FILES[@]} workflow(s): ${WORKFLOW_FILES[*]}"
echo "include_highest_priority=$INCLUDE_HIGHEST_PRIORITY, include_rerun_test=$INCLUDE_RERUN_TEST"
echo ""
# Decide whether to cancel run_id given a PR-lookup endpoint.
# $1 = run_id
# $2 = gh api path returning a list of PR objects (head=... or commits/<sha>/pulls)
# $3 = short label for log messages
maybe_cancel_for_pr() {
local run_id="$1"
local pr_query="$2"
local pr_label="$3"
local pr_info
pr_info=$(gh api -H "Accept: application/vnd.github+json" "$pr_query" \
--jq '.[0] | {number, state, merged_at}' 2>/dev/null || true)
if [ -z "$pr_info" ] || [ "$pr_info" = "null" ]; then
echo " ⚠️ No PR found ($pr_label), skipping"
return
fi
local pr_number pr_state
pr_number=$(echo "$pr_info" | jq -r '.number // empty')
pr_state=$(echo "$pr_info" | jq -r '.state // empty')
if [ -z "$pr_number" ]; then
echo " ⚠️ PR lookup returned empty number, skipping"
return
fi
local pr_url="https://github.com/$REPO/pull/$pr_number"
echo " PR: $pr_url ($pr_state)"
# Closed PR (merged or not): always cancel, skip label checks.
if [ "$pr_state" = "closed" ]; then
echo " 🚫 Cancelling (PR closed)..."
gh run cancel "$run_id" --repo "$REPO" || echo " ⚠️ Cancellation failed"
return
fi
# Open PR: apply label-based skip rules.
local labels
labels=$(gh pr view "$pr_number" --repo "$REPO" --json labels \
| jq -r '.labels[].name' 2>/dev/null || true)
if echo "$labels" | grep -Fxq "bypass-maintenance"; then
echo " 🛑 Skipping (bypass-maintenance label, never cancelled)"
return
fi
if echo "$labels" | grep -Fxq "highest-priority"; then
if [ "$INCLUDE_HIGHEST_PRIORITY" != "true" ]; then
echo " 🛑 Skipping (highest-priority label)"
return
fi
echo " ⚠️ highest-priority PR, but include_highest_priority is enabled"
fi
echo " 🚫 Cancelling..."
gh run cancel "$run_id" --repo "$REPO" || echo " ⚠️ Cancellation failed"
}
export -f maybe_cancel_for_pr
for workflow_file in "${WORKFLOW_FILES[@]}"; do
echo "========================================="
echo "Workflow: $workflow_file"
echo "========================================="
# Get all unfinished runs.
# Use server-side --status filter: without it, `gh run list --limit 1000`
# only sees the most recent 1000 runs by createdAt, which on busy workflows
# like pr-test.yml is < 3 days. Old stuck runs would be missed.
all_runs=""
for status in queued in_progress waiting; do
batch=$(gh run list \
--repo "$REPO" \
--workflow "$workflow_file" \
--status "$status" \
--json databaseId,status,event,url,createdAt,displayTitle \
--limit 1000 \
| jq -c '.[]')
if [ -n "$batch" ]; then
if [ -n "$all_runs" ]; then
all_runs="$all_runs"$'\n'"$batch"
else
all_runs="$batch"
fi
fi
done
if [ -z "$all_runs" ]; then
echo "✅ No unfinished runs found"
echo ""
continue
fi
# Count runs by event type
total_runs=$(echo "$all_runs" | wc -l)
pr_runs=$(echo "$all_runs" | jq -s '[.[] | select(.event=="pull_request")] | length')
dispatch_runs=$(echo "$all_runs" | jq -s '[.[] | select(.event=="workflow_dispatch")] | length')
other_runs=$(echo "$all_runs" | jq -s '[.[] | select(.event!="pull_request" and .event!="workflow_dispatch")] | length')
echo "📊 Summary: $total_runs unfinished ($pr_runs pull_request, $dispatch_runs workflow_dispatch, $other_runs other)"
echo ""
# Other runs: list only, do not cancel.
if [ "$other_runs" -gt 0 ]; then
echo "--- Other Runs (listed only, not cancelled) ---"
echo "$all_runs" | jq -c 'select(.event!="pull_request" and .event!="workflow_dispatch")' | while read -r run; do
run_url=$(echo "$run" | jq -r '.url')
run_event=$(echo "$run" | jq -r '.event')
run_status=$(echo "$run" | jq -r '.status')
echo " • $run_event ($run_status): $run_url"
done
echo ""
fi
# PR runs: resolve PR via head=owner:branch.
if [ "$pr_runs" -gt 0 ]; then
echo "--- PR Runs (resolving via head=owner:branch) ---"
echo "$all_runs" | jq -c 'select(.event=="pull_request")' | while read -r run; do
run_id=$(echo "$run" | jq -r '.databaseId')
run_url=$(echo "$run" | jq -r '.url')
run_status=$(echo "$run" | jq -r '.status')
echo ""
echo "Run ($run_status): $run_url"
run_details=$(gh api -H "Accept: application/vnd.github+json" \
"repos/$REPO/actions/runs/$run_id" 2>/dev/null || true)
if [ -z "$run_details" ]; then
echo " ⚠️ Could not fetch run details, skipping"
continue
fi
head_owner=$(echo "$run_details" | jq -r '.head_repository.owner.login // empty')
head_branch=$(echo "$run_details" | jq -r '.head_branch // empty')
if [ -z "$head_owner" ] || [ -z "$head_branch" ]; then
echo " ⚠️ Missing head info, skipping"
continue
fi
echo " Branch: ${head_owner}:${head_branch}"
maybe_cancel_for_pr "$run_id" \
"repos/$REPO/pulls?state=all&head=${head_owner}:${head_branch}" \
"head=${head_owner}:${head_branch}"
done
echo ""
fi
# workflow_dispatch runs (e.g. /rerun-test): resolve PR via pr_head_sha
# parsed from run-name (`[rerun-test] <cmd> <sha>`).
if [ "$dispatch_runs" -gt 0 ]; then
echo "--- Dispatch Runs (resolving via pr_head_sha in display_title) ---"
echo "$all_runs" | jq -c 'select(.event=="workflow_dispatch")' | while read -r run; do
run_id=$(echo "$run" | jq -r '.databaseId')
run_url=$(echo "$run" | jq -r '.url')
run_status=$(echo "$run" | jq -r '.status')
display_title=$(echo "$run" | jq -r '.displayTitle // empty')
echo ""
echo "Run ($run_status): $run_url"
echo " Title: $display_title"
# Last whitespace-delimited token if it is a 40-hex SHA.
last_token=$(echo "$display_title" | awk '{print $NF}')
if ! [[ "$last_token" =~ ^[0-9a-f]{40}$ ]]; then
echo " ⚠️ No pr_head_sha in title, skipping"
continue
fi
echo " pr_head_sha: $last_token"
maybe_cancel_for_pr "$run_id" \
"repos/$REPO/commits/$last_token/pulls" \
"sha=$last_token"
done
echo ""
fi
done
echo "========================================="
echo "✅ Processing complete"
echo "========================================="