This commit is contained in:
Leonardo Mortari
2025-08-01 01:30:12 -03:00
parent 7a4c6ea668
commit 711c28a894
2 changed files with 23 additions and 24 deletions

View File

@@ -12,9 +12,8 @@ services:
- .:/app - .:/app
environment: environment:
- PYTHONUNBUFFERED=1 - PYTHONUNBUFFERED=1
networks: # networks:
- dokploy-network # - dokploy-network
# networks:
networks: # dokploy-network:
dokploy-network: # external: true
external: true

36
main.py
View File

@@ -1,7 +1,8 @@
from typing import Optional from typing import Optional
from fastapi import FastAPI, HTTPException, Query from fastapi import FastAPI, HTTPException, Query
from youtube_transcript_api import YouTubeTranscriptApi, \ from youtube_transcript_api import YouTubeTranscriptApi
TranscriptsDisabled, NoTranscriptFound from youtube_transcript_api.formatters import SRTFormatter
from youtube_transcript_api._errors import TranscriptsDisabled, NoTranscriptFound
from yt_dlp import YoutubeDL from yt_dlp import YoutubeDL
app = FastAPI( app = FastAPI(
@@ -11,12 +12,9 @@ app = FastAPI(
def extract_video_id(url: str) -> str: def extract_video_id(url: str) -> str:
import re import re
match = re.search(r"(?:v=|youtu\.be/)([A-Za-z0-9_-]{11})", url) match = re.search(r"(?:v=|youtu\.be/)([A-Za-z0-9_-]{11})", url)
if not match: if not match:
raise ValueError("URL inválida do YouTube") raise ValueError("URL inválida do YouTube")
return match.group(1) return match.group(1)
@app.get("/get-transcript") @app.get("/get-transcript")
@@ -34,21 +32,25 @@ def get_transcript(
else: else:
video_id = videoId video_id = videoId
try:
transcript_list = YouTubeTranscriptApi.get_transcript(video_id, languages=['pt', 'en'])
except TranscriptsDisabled:
raise HTTPException(status_code=404, detail="Transcrição desativada para este vídeo")
except NoTranscriptFound:
raise HTTPException(status_code=404, detail="Nenhuma transcrição encontrada")
except Exception as e:
raise HTTPException(status_code=500, detail=f"Erro ao obter transcrição: {e}")
full_text = " ".join(item["text"] for item in transcript_list) try:
ytt_api = YouTubeTranscriptApi()
result = ytt_api.fetch(video_id, languages=['pt'])
formatter = SRTFormatter()
if not result:
raise NoTranscriptFound("Nenhuma transcrição encontrada para este vídeo")
except TranscriptsDisabled:
raise HTTPException(status_code=404, error="Transcrição desativada para este vídeo")
except NoTranscriptFound:
raise HTTPException(status_code=404, error="Nenhuma transcrição encontrada")
except Exception as e:
raise HTTPException(status_code=500, error=f"Erro ao obter transcrição: {e}")
return { return {
"video_id": video_id, "video_id": video_id,
"transcript": transcript_list, "transcript": formatter.format_transcript(result),
"full_text": full_text
} }
@app.get("/get-video-metadata") @app.get("/get-video-metadata")
@@ -79,5 +81,3 @@ def get_video_metadata(
raise HTTPException(status_code=500, detail=f"Erro ao extrair metadata: {e}") raise HTTPException(status_code=500, detail=f"Erro ao extrair metadata: {e}")
return info return info
# Para executar via: uvicorn main:app --reload --host 0.0.0.0 --port 8000