ok

app/api/server.py

@@ -11,7 +11,6 @@ from app.config import get_settings
 from app.core.database import pool_manager
 from app.core.logger import get_logger
 from app.api.v1.router import api_v1_router
-from app.api.verify import router as verify_router
 
 _logger = get_logger(__name__)
 _settings = get_settings()
@@ -66,7 +65,6 @@ def create_application() -> FastAPI:
     )
 
     app.include_router(api_v1_router, prefix="/api/v1")
-    app.include_router(verify_router, prefix="/verify")
 
     @app.get("/", include_in_schema=False)
     async def root():

app/api/v1/router.py

@@ -3,9 +3,11 @@ from __future__ import annotations
 from fastapi import APIRouter
 
 from app.api.v1 import batch, convert, database, system
+from app.api.verify import router as verify_router
 
 api_v1_router = APIRouter()
 api_v1_router.include_router(convert.router, tags=["Convert"])
 api_v1_router.include_router(batch.router, tags=["Batch"])
 api_v1_router.include_router(system.router, tags=["System"])
 api_v1_router.include_router(database.router, tags=["Database"])
+api_v1_router.include_router(verify_router, prefix="/verify", tags=["Verify"])

app/config.py

@@ -1,7 +1,6 @@
 from __future__ import annotations
 
 from functools import lru_cache
-from typing import Optional
 
 from pydantic_settings import BaseSettings, SettingsConfigDict
 
@@ -36,7 +35,6 @@ class Settings(BaseSettings):
 
     self_ping_url: str = "https://aetherbase-llm-ready-data.hf.space/ping"
     spacy_model: str = "en_core_web_sm"
-
     @property
     def max_upload_mb(self) -> int:
         return self.max_upload_bytes // (1024 * 1024)

app/services/converter_service.py

@@ -6,44 +6,27 @@ import mimetypes
 import re
 import time
 from pathlib import Path
-from typing import Optional
-from urllib.parse import urlparse, parse_qs
+from urllib.parse import urlparse
 
 import httpx
 from markitdown import MarkItDown
 
-from app.config import get_settings
 from app.core.constants import IMAGE_EXTENSIONS, IMAGE_MIME_PREFIXES
 from app.core.logger import get_logger
 from app.models.domain import ConversionError, ConversionResult
 from app.services.ocr_service import ocr_image, ocr_pdf
 
 _logger = get_logger(__name__)
-_settings = get_settings()
 
-_YOUTUBE_URL_PATTERN = re.compile(
-    r"(?:https?://)?"
-    r"(?:www\.|m\.)?"
-    r"(?:youtube\.com/(?:watch\?v=|embed/|v/|shorts/|live/)|youtu\.be/)"
-    r"([\w-]{11})"
-)
 
-_YOUTUBE_DOMAINS = {"youtube.com", "www.youtube.com", "m.youtube.com", "youtu.be", "www.youtu.be"}
-
-
-def _is_youtube_url(url: str) -> bool:
-    return _YOUTUBE_URL_PATTERN.match(url) is not None
-
-
-def _extract_youtube_video_id(url: str) -> str | None:
-    match = _YOUTUBE_URL_PATTERN.match(url)
+def _extract_youtube_video_id(url_or_id: str) -> str:
+    if len(url_or_id) == 11 and not url_or_id.startswith("http"):
+        return url_or_id
+    pattern = r"(?:v=|\/)([0-9A-Za-z_-]{11}).*"
+    match = re.search(pattern, url_or_id)
     if match:
         return match.group(1)
-    parsed = urlparse(url)
-    if parsed.hostname in _YOUTUBE_DOMAINS:
-        qs = parse_qs(parsed.query)
-        return qs.get("v", [None])[0]
-    return None
+    raise ValueError(f"Could not extract a valid YouTube ID from: {url_or_id}")
 
 
 def _fetch_youtube_oembed(video_id: str) -> dict | None:
@@ -59,18 +42,49 @@ def _fetch_youtube_oembed(video_id: str) -> dict | None:
 
 def _fetch_youtube_transcript(video_id: str) -> str | None:
     try:
+        import json as _json
+        import os as _os
+
         from youtube_transcript_api import YouTubeTranscriptApi
-        api = YouTubeTranscriptApi()
-        transcript = api.fetch(video_id, languages=["en"])
-        return " ".join(part.text for part in transcript)
+        from youtube_transcript_api.formatters import TextFormatter
+        from youtube_transcript_api._errors import (
+            TranscriptsDisabled,
+            NoTranscriptFound,
+            VideoUnavailable,
+        )
+
+        kwargs: dict = {}
+        cookies_raw = _os.environ.get("YOUTUBE_COOKIES", "").strip()
+        if cookies_raw:
+            try:
+                import requests as _requests
+                session = _requests.Session()
+                session.cookies.update(_json.loads(cookies_raw))
+                kwargs["http_client"] = session
+            except Exception as exc:
+                _logger.warning("Failed to apply YOUTUBE_COOKIES: %s", exc)
+
+        ytt_api = YouTubeTranscriptApi(**kwargs)
+        transcript_data = ytt_api.fetch(video_id, languages=["en"])
+        formatter = TextFormatter()
+        return formatter.format_transcript(transcript_data)
+    except TranscriptsDisabled:
+        _logger.warning("Transcripts disabled for video %s", video_id)
+    except NoTranscriptFound:
+        _logger.warning("No transcript found for video %s in language 'en'", video_id)
+    except VideoUnavailable:
+        _logger.warning("Video %s is unavailable, deleted, or private", video_id)
+    except ValueError as ve:
+        _logger.warning("Invalid video ID %s: %s", video_id, ve)
     except Exception as exc:
         _logger.warning("YouTube transcript failed for %s: %s", video_id, exc)
-        return None
+    return None
 
 
 def _convert_youtube(url: str) -> ConversionResult | None:
-    video_id = _extract_youtube_video_id(url)
-    if not video_id:
+    try:
+        video_id = _extract_youtube_video_id(url)
+    except ValueError:
         return None
 
     oembed = _fetch_youtube_oembed(video_id)
@@ -179,7 +193,11 @@ class ConverterService:
 
         start = time.perf_counter()
 
-        if _is_youtube_url(url):
+        try:
+            _extract_youtube_video_id(url)
+        except ValueError:
+            pass
+        else:
             result = _convert_youtube(url)
             if result is not None:
                 result = ConversionResult(