Spaces:
Running
Running
File size: 3,368 Bytes
bcd4ece c3a8356 bcd4ece c3a8356 bcd4ece b0b8612 bcd4ece c3a8356 bcd4ece c3a8356 bcd4ece c3a8356 bcd4ece c3a8356 bcd4ece c3a8356 bcd4ece d147352 bcd4ece |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 |
from youtube_transcript_api import YouTubeTranscriptApi
from youtube_transcript_api.proxies import GenericProxyConfig
import re
import os
from agents import function_tool
@function_tool
def fetch_video_transcript(url: str) -> str:
"""
Extract transcript with timestamps from a YouTube video URL and format it for LLM consumption
Args:
url (str): YouTube video URL
Returns:
str: Formatted transcript with timestamps, where each entry is on a new line
in the format: "[MM:SS] Text"
"""
# Extract video ID from URL
video_id_pattern = r'(?:v=|\/)([0-9A-Za-z_-]{11}).*'
video_id_match = re.search(video_id_pattern, url)
if not video_id_match:
raise ValueError("Invalid YouTube URL")
video_id = video_id_match.group(1)
def format_transcript(transcript):
"""Format transcript entries with timestamps"""
formatted_entries = []
for entry in transcript:
# Convert seconds to MM:SS format
minutes = int(entry.start // 60)
seconds = int(entry.start % 60)
timestamp = f"[{minutes:02d}:{seconds:02d}]"
formatted_entry = f"{timestamp} {entry.text}"
formatted_entries.append(formatted_entry)
# Join all entries with newlines
return "\n".join(formatted_entries)
# First attempt: Try without proxy
try:
ytt_api = YouTubeTranscriptApi()
transcript = ytt_api.fetch(video_id)
return format_transcript(transcript)
except Exception as e:
# Second attempt: Try with proxy if credentials are available
proxy_username = os.getenv("PROXY_USERNAME")
proxy_password = os.getenv("PROXY_PASSWORD")
proxy_url_base = os.getenv("PROXY_URL")
if proxy_username and proxy_password:
try:
# Construct proxy URLs with credentials
http_proxy_url = f"http://{proxy_username}:{proxy_password}@{proxy_url_base}"
https_proxy_url = f"https://{proxy_username}:{proxy_password}@{proxy_url_base}"
proxy_config = GenericProxyConfig(
http_url=http_proxy_url,
https_url=https_proxy_url
)
ytt_api_with_proxy = YouTubeTranscriptApi(proxy_config=proxy_config)
transcript = ytt_api_with_proxy.fetch(video_id)
return format_transcript(transcript)
except Exception as proxy_error:
raise Exception(f"Error fetching transcript (tried with and without proxy): {str(proxy_error)}")
else:
raise Exception(f"Error fetching transcript: {str(e)}. Proxy credentials not available.")
@function_tool
def fetch_intstructions(prompt_name: str) -> str:
"""
Fetch instructions for a given prompt name from the prompts/ directory
Args:
prompt_name (str): Name of the prompt to fetch instructions for
Available prompts:
- write_blog_post
- write_social_post
- write_video_chapters
Returns:
str: Instructions for the given prompt
"""
import os
script_dir = os.path.dirname(__file__)
prompt_path = os.path.join(script_dir, "prompts", f"{prompt_name}.md")
with open(prompt_path, "r") as f:
return f.read() |