daily data update and add ci

.gitea/workflows/daily-asmr-data.yml

@@ -0,0 +1,62 @@
+name: Update ASMR data
+
+on:
+  schedule:
+    # 00:30 JST. Scheduled actions may run a little later depending on runner load.
+    - cron: "30 15 * * *"
+  push:
+    branches:
+      - main
+  workflow_dispatch:
+
+jobs:
+  update:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          ref: main
+          fetch-depth: 0
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+
+      - name: Update ASMR JSONL data
+        run: |
+          if [ "${GITHUB_EVENT_NAME:-}" = "push" ]; then
+            last_commit_message="$(git log -1 --pretty=%B)"
+            case "$last_commit_message" in
+              *"[skip ci]"*)
+                echo "Skipping data update for CI-generated commit."
+                touch .skip-asmr-data-update
+                exit 0
+                ;;
+            esac
+
+            python scripts/update_daily_asmr_data.py --skip-web-data --force
+          else
+            python scripts/update_daily_asmr_data.py --skip-web-data
+          fi
+
+      - name: Commit and push changes
+        run: |
+          if [ -f .skip-asmr-data-update ]; then
+            echo "Data update skipped."
+            exit 0
+          fi
+
+          if git diff --quiet -- asmr_works.jsonl asmr_works.filtered.jsonl; then
+            echo "No ASMR data changes to commit."
+            exit 0
+          fi
+
+          git config user.name "gitea-actions[bot]"
+          git config user.email "gitea-actions[bot]@users.noreply.local"
+          git add asmr_works.jsonl asmr_works.filtered.jsonl
+          git commit -m "daily data update [skip ci]"
+          git pull --rebase --autostash origin main
+          git push origin HEAD:main