Add local self-test flow and fix fallback regressions

scripts/mock_searxng.py

@@ -0,0 +1,81 @@
+#!/usr/bin/env python3
+import argparse
+import json
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from urllib.parse import parse_qs, urlparse
+
+
+def build_results(engine: str, query: str):
+    lowered = (engine or "").lower()
+    if "google" in lowered and "video" in lowered:
+        return [
+            {
+                "title": f"{query.title()} cinematic b-roll",
+                "url": "https://www.youtube.com/watch?v=abcdefghijk",
+                "content": f"{query} stock footage edit reference",
+                "thumbnail": "https://i.ytimg.com/vi/abcdefghijk/hqdefault.jpg",
+                "engine": engine,
+            }
+        ]
+
+    return [
+        {
+            "title": f"{query.title()} envato clip",
+            "url": "https://elements.envato.com/city-rain-b-roll-AB12CD",
+            "content": f"{query} stock footage cinematic scene",
+            "thumbnail": "https://images.example.com/envato-city-rain.jpg",
+            "thumbnail_src": "https://images.example.com/envato-city-rain.jpg",
+            "engine": engine,
+        },
+        {
+            "title": f"{query.title()} artgrid clip",
+            "url": "https://artgrid.io/clip/123456/city-rain-night",
+            "content": f"{query} editorial footage establishing shot",
+            "thumbnail": "https://images.example.com/artgrid-city-rain.jpg",
+            "thumbnail_src": "https://images.example.com/artgrid-city-rain.jpg",
+            "img_src": "https://images.example.com/artgrid-city-rain.jpg",
+            "engine": engine,
+        },
+    ]
+
+
+class Handler(BaseHTTPRequestHandler):
+    def do_GET(self):
+        parsed = urlparse(self.path)
+        if parsed.path != "/search":
+            self.send_response(404)
+            self.end_headers()
+            return
+
+        params = parse_qs(parsed.query)
+        query = params.get("q", ["city rain"])[0]
+        engine = params.get("engines", [""])[0]
+        payload = {"results": build_results(engine, query)}
+
+        body = json.dumps(payload).encode("utf-8")
+        self.send_response(200)
+        self.send_header("Content-Type", "application/json")
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        self.wfile.write(body)
+
+    def log_message(self, format, *args):
+        return
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--port", type=int, default=18080)
+    args = parser.parse_args()
+
+    server = ThreadingHTTPServer(("127.0.0.1", args.port), Handler)
+    try:
+        server.serve_forever()
+    except KeyboardInterrupt:
+        pass
+    finally:
+        server.server_close()
+
+
+if __name__ == "__main__":
+    main()

scripts/selftest.sh

@@ -0,0 +1,106 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+ROOT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+TMP_DIR="$(mktemp -d)"
+MOCK_PORT="${SELFTEST_MOCK_PORT:-18080}"
+APP_PORT="${SELFTEST_APP_PORT:-18081}"
+MOCK_PID=""
+APP_PID=""
+
+cleanup() {
+  if [[ -n "${APP_PID}" ]] && kill -0 "${APP_PID}" >/dev/null 2>&1; then
+    kill "${APP_PID}" >/dev/null 2>&1 || true
+    wait "${APP_PID}" >/dev/null 2>&1 || true
+  fi
+  if [[ -n "${MOCK_PID}" ]] && kill -0 "${MOCK_PID}" >/dev/null 2>&1; then
+    kill "${MOCK_PID}" >/dev/null 2>&1 || true
+    wait "${MOCK_PID}" >/dev/null 2>&1 || true
+  fi
+  rm -rf "${TMP_DIR}"
+}
+trap cleanup EXIT
+
+cd "${ROOT_DIR}"
+
+echo "[selftest] gofmt"
+gofmt -w backend/main.go backend/handlers/api.go backend/models/db.go backend/services/cse.go backend/services/gemini.go backend/services/gemini_test.go
+
+echo "[selftest] python syntax"
+python3 -m py_compile worker/downloader.py scripts/mock_searxng.py
+
+echo "[selftest] go test"
+go test ./...
+
+echo "[selftest] go build"
+go build -o "${TMP_DIR}/ai-media-hub" ./backend
+
+echo "[selftest] start mock searxng"
+python3 scripts/mock_searxng.py --port "${MOCK_PORT}" >"${TMP_DIR}/mock-searxng.log" 2>&1 &
+MOCK_PID=$!
+
+echo "[selftest] start app"
+mkdir -p "${TMP_DIR}/downloads"
+APP_ROOT="${ROOT_DIR}" \
+APP_ADDR="127.0.0.1:${APP_PORT}" \
+SQLITE_PATH="${TMP_DIR}/media.db" \
+DOWNLOADS_DIR="${TMP_DIR}/downloads" \
+FRONTEND_DIR="${ROOT_DIR}/frontend" \
+WORKER_SCRIPT="${ROOT_DIR}/worker/downloader.py" \
+SEARXNG_BASE_URL="http://127.0.0.1:${MOCK_PORT}" \
+"${TMP_DIR}/ai-media-hub" >"${TMP_DIR}/app.log" 2>&1 &
+APP_PID=$!
+
+for _ in $(seq 1 30); do
+  if curl -fsS "http://127.0.0.1:${APP_PORT}/healthz" >/dev/null; then
+    break
+  fi
+  sleep 1
+done
+
+echo "[selftest] verify healthz"
+curl -fsS "http://127.0.0.1:${APP_PORT}/healthz" >"${TMP_DIR}/healthz.json"
+python3 - "${TMP_DIR}/healthz.json" <<'PY'
+import json
+import sys
+with open(sys.argv[1], "r", encoding="utf-8") as handle:
+    payload = json.load(handle)
+if payload.get("status") != "ok":
+    raise SystemExit(f"unexpected healthz payload: {payload}")
+PY
+
+echo "[selftest] verify search"
+curl -fsS \
+  -H "Content-Type: application/json" \
+  -d '{"query":"city rain","platforms":["envato","artgrid","google video"]}' \
+  "http://127.0.0.1:${APP_PORT}/api/search" >"${TMP_DIR}/search.json"
+python3 - "${TMP_DIR}/search.json" <<'PY'
+import json
+import sys
+with open(sys.argv[1], "r", encoding="utf-8") as handle:
+    payload = json.load(handle)
+results = payload.get("results") or []
+if len(results) < 2:
+    raise SystemExit(f"expected >= 2 search results, got {len(results)}: {payload}")
+if not all(item.get("link") for item in results):
+    raise SystemExit(f"search results must include links: {payload}")
+PY
+
+echo "[selftest] verify upload"
+printf 'selftest upload\n' > "${TMP_DIR}/sample.txt"
+curl -fsS -F "file=@${TMP_DIR}/sample.txt" "http://127.0.0.1:${APP_PORT}/api/upload" >"${TMP_DIR}/upload.json"
+python3 - "${TMP_DIR}/upload.json" "${TMP_DIR}/downloads" <<'PY'
+import json
+import os
+import sys
+with open(sys.argv[1], "r", encoding="utf-8") as handle:
+    payload = json.load(handle)
+filename = payload.get("filename")
+if not filename:
+    raise SystemExit(f"missing filename in upload payload: {payload}")
+target = os.path.join(sys.argv[2], filename)
+if not os.path.exists(target):
+    raise SystemExit(f"expected uploaded file at {target}")
+PY
+
+echo "[selftest] ok"