f

main.py

@@ -0,0 +1,121 @@
+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
+
+class TranslateRequest(BaseModel):
+    text: str
+    targetLang: 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)
+    return match.group(1) if match else None
+
+@app.post("/api/transcript")
+async def get_transcript(req: TranscriptRequest):
+    video_id = extract_video_id(req.url)
+    if not video_id:
+        raise HTTPException(status_code=400, detail="Invalid YouTube URL")
+
+    # We will try multiple Invidious instances in case one is down
+    instances = [
+        "https://inv.nadeko.net",
+        "https://invidious.fdn.fr",
+        "https://invidious.protokolla.fi",
+        "https://iv.datura.network"
+    ]
+
+    for host in instances:
+        try:
+            print(f"Trying {host} for video {video_id}...")
+            # 1. Fetch available captions from Invidious
+            res = requests.get(f"{host}/api/v1/captions/{video_id}?label=English", timeout=5)
+            
+            if res.status_code != 200:
+                continue # Try next instance
+                
+            data = res.json()
+            if not data.get("captions"):
+                continue # Try next instance
+            
+            # 2. Find the English caption URL (or first available)
+            caption = next((c for c in data["captions"] if "English" in c.get("label", "")), data["captions"][0])
+            caption_url = caption.get("url")
+            
+            if not caption_url:
+                continue
+                
+            if caption_url.startswith("/"):
+                caption_url = f"{host}{caption_url}"
+
+            # 3. Fetch the actual caption XML
+            cap_res = requests.get(caption_url, timeout=5)
+            if cap_res.status_code != 200:
+                continue
+            
+            # 4. Parse XML to plain text
+            texts = re.findall(r'<text[^>]*>([^<]*)</text>', cap_res.text)
+            if not texts:
+                continue
+            
+            full_text = " ".join(texts)
+            full_text = full_text.replace("&amp;#39;", "'").replace("&amp;quot;", '"').replace("&amp;amp;", "&").replace("&#39;", "'").replace("\n", " ")
+            
+            print(f"Successfully fetched transcript from {host}. Length: {len(full_text)} chars")
+            return {"text": full_text}
+            
+        except Exception as e:
+            print(f"Failed on {host}: {e}")
+            continue # Try next instance
+            
+    raise HTTPException(status_code=500, detail="Failed to fetch transcript. All sources timed out or are blocked.")
+
+# ========================================================
+# TRANSLATE API
+# ========================================================
+@app.post("/api/translate")
+async def translate_text(req: TranslateRequest):
+    if not req.text or not req.targetLang:
+        raise HTTPException(status_code=400, detail="Missing text or target language")
+
+    try:
+        url = f"https://translate.googleapis.com/translate_a/single?client=gtx&sl=auto&tl={req.targetLang}&dt=t"
+        res = requests.post(url, 
+            headers={
+                "Content-Type": "application/x-www-form-urlencoded",
+                "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
+            },
+            data=f"q={requests.utils.quote(req.text)}",
+            timeout=10
+        )
+
+        if res.status_code != 200:
+            raise Exception(f"Google API returned {res.status_code}")
+
+        data = res.json()
+        translated = "".join([item[0] for item in data[0]])
+        return {"translatedText": translated}
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Translation failed: {str(e)}")
+
+# KEEP-ALIVE ENDPOINT
+@app.get("/api/health")
+async def health_check():
+    return {"status": "ok", "message": "Transcript API is running!"}