feat: switch summarizer to DeepSeek API, add cookie support for transcripts

backend/main.py

@@ -37,6 +37,11 @@ async def startup():
 async def health():
     return {"status": "ok"}
 
+@app.get("/")
+async def root():
+    from fastapi.responses import FileResponse
+    return FileResponse(os.path.join(frontend_build_path, "index.html"))
+
 @app.get("/api/videos")
 async def get_videos(
     status: Optional[str] = "pending",

backend/summarizer.py

@@ -10,12 +10,9 @@ from db import DB_PATH
 log = logging.getLogger(__name__)
 
 # Fallback pattern to retrieve the correct bytecatcode credentials
-API_KEY = os.environ.get("ANTHROPIC_API_KEY")
+API_KEY = os.environ.get("DEEPSEEK_API_KEY", os.environ.get("OPENAI_API_KEY", ""))
-if not API_KEY:
+BASE_URL = os.environ.get("DEEPSEEK_BASE_URL", "https://api.deepseek.com")
-    API_KEY = os.environ.get("OPENAI_API_KEY", "sk-Kq5JbgmnEEkdzTqNAQ6bV6nztQ0ngeHTHF9Vg8nUglQvrOlx")
+MODEL = os.environ.get("DEEPSEEK_MODEL", "deepseek-chat")
-
-BASE_URL = os.environ.get("ANTHROPIC_BASE_URL", "https://www.bytecatcode.org")
-MODEL = "claude-haiku-4-5-20251001"
 
 SYSTEM_PROMPT = "You are a professional video summarizing assistant. Create highly concise, structured summaries from video transcripts. Your output must be in Chinese if the transcript is Chinese, otherwise in English."
 
@@ -38,6 +35,15 @@ def download_transcript(video_id: str) -> str:
             "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
             "Accept-Language": "en-US,en;q=0.9"
         })
+        # Load cookies to bypass YouTube IP block
+        for p in [os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "config/cookies.txt"),
+                   "/app/config/cookies.txt"]:
+            if os.path.exists(p):
+                from http.cookiejar import MozillaCookieJar
+                cj = MozillaCookieJar(p)
+                cj.load(ignore_discard=True, ignore_expires=True)
+                session.cookies.update({c.name: c.value for c in cj})
+                break
         transcript_list = YouTubeTranscriptApi(http_client=session).list(video_id)
 
         # Try to find manual transcripts in preferred languages
@@ -71,20 +77,18 @@ async def summarize_video(video_id: str) -> str:
             await db.commit()
         return None
 
-    # 2. Call Claude API
+    # 2. Call DeepSeek API (OpenAI-compatible)
     headers = {
-        "x-api-key": API_KEY,
+        "Authorization": f"Bearer {API_KEY}",
-        "anthropic-version": "2023-06-01",
+        "Content-Type": "application/json",
-        "content-type": "application/json",
+        "User-Agent": "t-youtube/0.1"
-        "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
     }
 
-    # Standard Anthropic payload format
     payload = {
         "model": MODEL,
         "max_tokens": 1000,
-        "system": SYSTEM_PROMPT,
         "messages": [
+            {"role": "system", "content": SYSTEM_PROMPT},
             {"role": "user", "content": PROMPT.format(transcript=transcript[:50000])}
         ]
     }
@@ -92,15 +96,15 @@ async def summarize_video(video_id: str) -> str:
     async with httpx.AsyncClient(timeout=120) as client:
         try:
             resp = await client.post(
-                f"{BASE_URL}/v1/messages",
+                f"{BASE_URL}/v1/chat/completions",
                 headers=headers,
                 json=payload
             )
             resp.raise_for_status()
             data = resp.json()
-            summary = data["content"][0]["text"]
+            summary = data["choices"][0]["message"]["content"]
         except Exception as e:
-            log.error(f"Claude API request failed for {video_id}: {e}")
+            log.error(f"DeepSeek API request failed for {video_id}: {e}")
             summary = f"Error generating summary: {e}"
 
     # 3. Save transcript & summary to db