Skip to content

Documentation Freshness Check #46

Documentation Freshness Check

Documentation Freshness Check #46

name: Documentation Freshness Check
on:
schedule:
- cron: '0 0 * * 0'
workflow_dispatch:
jobs:
check_freshness:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
with:
fetch-depth: 0
- uses: actions/setup-python@v4
with:
python-version: '3.x'
- name: Create scripts directory
run: mkdir -p .github/scripts
- name: Copy Python script
run: |
cat > .github/scripts/check_freshness.py << 'EOL'
import os
import subprocess
from datetime import datetime, timedelta
FRESHNESS_PERIOD = timedelta(days=180)
IGNORED_FILES = ["_index.md", "support.md"]
now = datetime.now()
def get_last_modified_time(file_path):
result = subprocess.run(
['git', 'log', '-1', '--format=%ct', file_path],
stdout=subprocess.PIPE,
text=True
)
timestamp = int(result.stdout.strip())
return datetime.fromtimestamp(timestamp)
def check_freshness(directory):
stale_files = []
for root, _, files in os.walk(directory):
for file in files:
if file.endswith(".md"):
relative_path = os.path.relpath(os.path.join(root, file), start=directory)
if file in IGNORED_FILES:
continue
last_modified = get_last_modified_time(os.path.join(root, file))
time_diff = now - last_modified
if time_diff > FRESHNESS_PERIOD:
stale_files.append((relative_path, last_modified))
return stale_files
docs_directory = "content"
stale_docs = check_freshness(docs_directory)
sorted_stale_docs = sorted(stale_docs, key=lambda x: x[1])
if sorted_stale_docs:
with open("stale_docs.txt", "w") as f:
for doc, mod_time in sorted_stale_docs:
f.write(f"- *{doc}* (Last Modified: {mod_time.strftime('%Y-%m-%d')})\n")
else:
with open("stale_docs.txt", "w") as f:
f.write("All documents are fresh!\n")
EOL
- name: Run freshness check
id: check
run: |
python3 .github/scripts/check_freshness.py
if [ -s stale_docs.txt ]; then
echo "HAS_STALE=true" >> $GITHUB_ENV
else
echo "HAS_STALE=false" >> $GITHUB_ENV
fi
- name: Read stale docs output
if: env.HAS_STALE == 'true'
id: read_stale_docs
run: |
thread_message=$(cat stale_docs.txt | sed ':a;N;$!ba;s/\n/\\n/g')
echo "THREAD_MESSAGE=${thread_message}" >> $GITHUB_ENV
- name: Post initial Slack message
id: slack_message
run: |
RESPONSE=$(curl -X POST -H "Authorization: Bearer ${{ secrets.SLACK_BOT_TOKEN }}" \
-H "Content-Type: application/json" \
-d '{
"channel": "${{ secrets.SLACK_CHANNEL_ID }}",
"text": "📌 *Documentation Freshness Check Results*"
}' \
https://slack.com/api/chat.postMessage)
echo "$RESPONSE" > slack_response.json
cat slack_response.json # Debugging: print the response
THREAD_TS=$(echo "$RESPONSE" | jq -r '.ts')
if [ -z "$THREAD_TS" ] || [ "$THREAD_TS" == "null" ]; then
echo "Error: Failed to retrieve Slack thread timestamp."
exit 1
fi
echo "THREAD_TS=$THREAD_TS" >> $GITHUB_ENV
echo "Extracted thread_ts: $THREAD_TS"
- name: Post thread reply with formatted list
if: env.HAS_STALE == 'true'
run: |
curl -X POST -H "Authorization: Bearer ${{ secrets.SLACK_BOT_TOKEN }}" \
-H "Content-Type: application/json" \
-d '{
"channel": "${{ secrets.SLACK_CHANNEL_ID }}",
"thread_ts": "'"${{ env.THREAD_TS }}"'",
"text": ":warning: *The following docs have not been modified for at least 6 months:*\n\n${{ env.THREAD_MESSAGE }}"
}' \
https://slack.com/api/chat.postMessage