d

main.py

@@ -1,134 +0,0 @@
-from fastapi import FastAPI, HTTPException
-from fastapi.middleware.cors import CORSMiddleware
-import requests
-import re
-from pydantic import BaseModel
-
-app = FastAPI()
-
-# Enable CORS so your Vercel frontend can call this API
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"], 
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-class TranscriptRequest(BaseModel):
-    url: str
-
-def extract_video_id(url: str) -> str:
-    pattern = r'(?:youtube\.com\/(?:[^\/]+\/.+\/|(?:v|e(?:mbed)?)\/|.*[?&]v=)|youtu\.be\/)([^"&?\/\s]{11})'
-    match = re.search(pattern, url)
-    if match:
-        return match.group(1)
-    return None
-
-def parse_caption_xml(xml_string: str) -> str:
-    # Extract text content from the XML safely using regex
-    text_matches = re.findall(r'<text[^>]*>([^<]*)</text>', xml_string)
-    if not text_matches:
-        return ""
-    
-    full_text = " ".join(text_matches)
-    # Clean up HTML entities
-    full_text = full_text.replace("&amp;#39;", "'").replace("&amp;quot;", '"').replace("&amp;amp;", "&").replace("&#39;", "'").replace("&lt;", "<").replace("&gt;", ">")
-    return full_text
-
-@app.post("/api/transcript")
-async def get_transcript(req: TranscriptRequest):
-    url = req.url
-    
-    video_id = extract_video_id(url)
-    if not video_id:
-        raise HTTPException(status_code=400, detail="Invalid YouTube URL")
-
-    try:
-        print(f"Fetching InnerTube data for video_id: {video_id}")
-        
-        # YouTube's internal API endpoint
-        innertube_url = "https://www.youtube.com/youtubei/v1/player?prettyPrint=false"
-        
-        payload = {
-            "context": {
-                "client": {
-                    "clientName": "WEB",
-                    "clientVersion": "2.20241210.00.00",
-                    "hl": "en",
-                    "gl": "US"
-                }
-            },
-            "videoId": video_id
-        }
-        
-        headers = {
-            "Content-Type": "application/json",
-            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
-            "Origin": "https://www.youtube.com",
-            "Referer": "https://www.youtube.com/"
-        }
-        
-        # 1. Fetch the video player data
-        response = requests.post(innertube_url, json=payload, headers=headers, timeout=5)
-        
-        if response.status_code != 200:
-            raise HTTPException(status_code=500, detail=f"YouTube API returned status {response.status_code}")
-            
-        data = response.json()
-        
-        # Check if video is playable
-        playability = data.get("playabilityStatus", {})
-        if playability.get("status") == "ERROR":
-            raise HTTPException(status_code=404, detail=playability.get("reason", "Video unavailable"))
-            
-        # 2. Extract caption tracks
-        caption_tracks = data.get("captions", {}).get("playerCaptionsTracklistRenderer", {}).get("captionTracks", [])
-        
-        if not caption_tracks:
-            raise HTTPException(status_code=404, detail="No captions found for this video.")
-            
-        # Prefer English, fallback to first available
-        selected_track = None
-        for track in caption_tracks:
-            if track.get("languageCode") == "en":
-                selected_track = track
-                break
-                
-        if not selected_track:
-            selected_track = caption_tracks[0]
-            
-        caption_url = selected_track.get("baseUrl")
-        if not caption_url:
-            raise HTTPException(status_code=404, detail="Caption URL not found.")
-            
-        print(f"Fetching captions from URL...")
-        
-        # 3. Fetch the actual caption XML
-        caption_response = requests.get(caption_url, headers={"User-Agent": headers["User-Agent"]}, timeout=5)
-        
-        if caption_response.status_code != 200:
-            raise HTTPException(status_code=500, detail=f"Caption fetch failed with status {caption_response.status_code}")
-            
-        # 4. Parse the XML into plain text
-        full_text = parse_caption_xml(caption_response.text)
-        
-        if not full_text.strip():
-            raise HTTPException(status_code=404, detail="Caption XML was empty.")
-            
-        print(f"Successfully fetched transcript. Length: {len(full_text)} chars")
-        return {"text": full_text}
-
-    except HTTPException:
-        raise 
-    except requests.exceptions.Timeout:
-        raise HTTPException(status_code=504, detail="Request to YouTube timed out.")
-    except Exception as e:
-        error_name = type(e).__name__
-        print(f"UNHANDLED ERROR for {video_id}: {error_name} - {str(e)}")
-        raise HTTPException(status_code=500, detail=f"Failed to fetch transcript: {error_name} - {str(e)}")
-
-# KEEP-ALIVE ENDPOINT
-@app.get("/api/health")
-async def health_check():
-    return {"status": "ok", "message": "Transcript API is running!"}