From 7b485abfbbc51c0816ecc6c061e42105bde78315 Mon Sep 17 00:00:00 2001
From: "Gan, Jimmy" <jimmy.gan@adidas.com>
Date: Wed, 24 Jun 2026 00:50:36 +0800
Subject: [PATCH] feat: DeepSeek API + transcript fixes; add OAuth captions
 scope

---
 backend/auth.py       |  8 +++++++-
 backend/summarizer.py | 31 ++++++++++++++-----------------
 2 files changed, 21 insertions(+), 18 deletions(-)

diff --git a/backend/auth.py b/backend/auth.py
index 3e33332..50d40cd 100644
--- a/backend/auth.py
+++ b/backend/auth.py
@@ -6,7 +6,10 @@ from google_auth_oauthlib.flow import InstalledAppFlow
 
 log = logging.getLogger(__name__)
 
-SCOPES = ["https://www.googleapis.com/auth/youtube.readonly"]
+SCOPES = [
+    "https://www.googleapis.com/auth/youtube.readonly",
+    "https://www.googleapis.com/auth/youtube.force-ssl"
+]
 CONFIG_DIR = os.environ.get("CONFIG_DIR", "/app/config")
 _local_dir = os.path.dirname(os.path.abspath(__file__))
 _local_config = os.path.join(os.path.dirname(_local_dir), "config")
@@ -34,6 +37,9 @@ def get_credentials():
                     "Download from Google Cloud Console."
                 )
             flow = InstalledAppFlow.from_client_secrets_file(CLIENT_SECRET, SCOPES)
+            # Print URL for manual opening
+            auth_url, _ = flow.authorization_url(prompt='consent', access_type='offline')
+            print(f"\nOpen this URL in your browser:\n{auth_url}\n")
             creds = flow.run_local_server(port=0)
 
         os.makedirs(os.path.dirname(TOKEN_FILE), exist_ok=True)
diff --git a/backend/summarizer.py b/backend/summarizer.py
index 75ef8fb..babc530 100644
--- a/backend/summarizer.py
+++ b/backend/summarizer.py
@@ -28,38 +28,35 @@ Keep the entire summary under 150 words. Be extremely direct so I can decide in
 Transcript:
 {transcript}"""
 
+COOKIES_PATH = None
+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):
+        COOKIES_PATH = p
+        break
+
 def download_transcript(video_id: str) -> str:
     try:
         session = requests.Session()
         session.headers.update({
-            "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",
+            "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/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
+        if COOKIES_PATH:
+            from http.cookiejar import MozillaCookieJar
+            cj = MozillaCookieJar(COOKIES_PATH)
+            cj.load(ignore_discard=True, ignore_expires=True)
+            session.cookies.update({c.name: c.value for c in cj})
         transcript_list = YouTubeTranscriptApi(http_client=session).list(video_id)
-
-        # Try to find manual transcripts in preferred languages
         try:
             transcript = transcript_list.find_manually_created_transcript(['en', 'zh-Hans', 'zh-Hant', 'zh', 'ja'])
         except Exception:
-            # Fallback to generated (auto-captions) in preferred languages
             try:
                 transcript = transcript_list.find_generated_transcript(['en', 'zh-Hans', 'zh-Hant', 'zh', 'ja'])
             except Exception:
-                # Fallback to any transcript
                 transcript = next(iter(transcript_list))
-
         parts = transcript.fetch()
-        text = " ".join([part["text"] for part in parts])
-        return text
+        return " ".join([part["text"] for part in parts])
     except Exception as e:
         log.warning(f"Failed to fetch transcript for {video_id}: {e}")
         return None