update and test twitch clip downloading
This commit is contained in:
@@ -0,0 +1,294 @@
|
||||
#!/usr/bin/env python3
|
||||
import requests
|
||||
import os
|
||||
import subprocess
|
||||
|
||||
from database import Database
|
||||
|
||||
import uploader
|
||||
from uploader import CategoryId
|
||||
|
||||
import youtube_hashtags
|
||||
import transcribe_video
|
||||
|
||||
CHANNEL_NAME = 'teampgp' # Replace with the streamer's username
|
||||
DB = None
|
||||
|
||||
def run_linux_command(command: str):
|
||||
"""Executes a Linux command, waits for completion, and returns output."""
|
||||
try:
|
||||
# shell=True allows running full command strings with pipes/wildcards
|
||||
# text=True returns strings instead of bytes
|
||||
result = subprocess.run(
|
||||
command, shell=True, check=True, capture_output=True, text=True
|
||||
)
|
||||
|
||||
return {"success": True, "stdout": result.stdout, "stderr": result.stderr}
|
||||
|
||||
except subprocess.CalledProcessError as e:
|
||||
# Handles errors if the Linux command returns a non-zero exit code
|
||||
return {"success": False, "stdout": e.stdout, "stderr": e.stderr}
|
||||
|
||||
def transcribe(slug: str):
|
||||
"""Transcribes the Video File."""
|
||||
video_file = f"save/{DB.table}/{slug}/{slug}.mp4"
|
||||
|
||||
# Check if the video file exists
|
||||
if not os.path.exists(video_file):
|
||||
print(f"Error: File not found: {video_file}")
|
||||
return False
|
||||
|
||||
# no need to continue if srt transcribe file already exists
|
||||
if os.path.exists(f"save/{DB.table}/{slug}/transcribe_{slug}.srt"):
|
||||
print(f"video already transcribed:")
|
||||
return True
|
||||
|
||||
transcribe_video.extract_audio(video_file, f"save/{DB.table}/{slug}/temp_{slug}_audio.wav")
|
||||
transcribe_video.transcribe_to_srt(f"save/{DB.table}/{slug}/temp_{slug}_audio.wav", f"save/{DB.table}/{slug}/", f"transcribe_{slug}")
|
||||
|
||||
return True
|
||||
|
||||
def top_hashtags(slug: str):
|
||||
""""Hashtags from transcribed SRT file."""
|
||||
file_srt = f"save/{DB.table}/{slug}/transcribe_{slug}.srt"
|
||||
|
||||
# if srt transcribe file not exists
|
||||
if not os.path.exists(file_srt):
|
||||
print(f"Transcribe file not found {file_srt}")
|
||||
return []
|
||||
|
||||
tags = youtube_hashtags.get_top_hashtags(file_srt)
|
||||
return tags
|
||||
|
||||
def download():
|
||||
"""Find all undownload vods and download them."""
|
||||
undownloaded = DB.get_undownloaded()
|
||||
|
||||
print(f"Download {DB.table}...")
|
||||
|
||||
if DB.table == "vods":
|
||||
for record_id, record_date, title, game_name, downloaded, uploaded_yt, chat_upload_yt in undownloaded:
|
||||
print(f"ID: {record_id} | Date: {record_date} | Game: {game_name} | Title: {title}")
|
||||
output = run_linux_command(f"TwitchDownloaderCLI videodownload --id {record_id} -o save/vods/{record_id}/{record_id}.mp4 --collision Overwrite")
|
||||
output2 = run_linux_command(f"TwitchDownloaderCLI chatdownload --id {record_id} -o save/vods/{record_id}/{record_id}_chat.json -E --collision Overwrite")
|
||||
if output["success"] is True:
|
||||
print(f"ID: {record_id} | Date: {record_date} | Game: {game_name} | Title: {title} | was successful")
|
||||
DB.mark_as_downloaded(record_id)
|
||||
else:
|
||||
print(f"ID: {record_id} | Download Process failed. {output["stdout"]}. Error: {output["stderr"]}")
|
||||
|
||||
elif DB.table == "clips":
|
||||
for slug, record_date, title, game_name, clip_by, views, downloaded, uploaded_yt, uploaded_shorts_yt in undownloaded:
|
||||
print(f"Slug: {slug} | Date: {record_date} | Game: {game_name} | By: {clip_by} | Views: {views} | Title: {title}")
|
||||
|
||||
# Uses standard clipdownload directive
|
||||
output = run_linux_command(f"TwitchDownloaderCLI clipdownload --id {slug} -o save/clips/{slug}/{slug}.mp4 --collision Overwrite")
|
||||
|
||||
if output["success"] is True:
|
||||
print(f"Slug: {slug} | Was successfully downloaded.")
|
||||
DB.mark_as_downloaded(slug)
|
||||
else:
|
||||
print(f"Slug: {slug} | Download Process failed. {output["stdout"]}. Error: {output["stderr"]}")
|
||||
|
||||
|
||||
print(f"Finished Downloading {DB.table}...")
|
||||
|
||||
def upload():
|
||||
"""Loops over the downloaded videos entries and uploaded them to youtube."""
|
||||
print(f"Uploading {DB.table}...")
|
||||
|
||||
unuploaded = DB.get_unuploaded()
|
||||
|
||||
twitch_datetime = " #Twitch Every Friday and Sunday @7:30 EST https://twitch.tv/teampgp"
|
||||
file_path = ""
|
||||
title = ""
|
||||
description = ""
|
||||
categoryId = CategoryId.GAMING
|
||||
privatcyStatus = 'private'
|
||||
base_tags = ['gaming', 'TeamPGP', 'twitch', 'Level1Techs']
|
||||
|
||||
upload_queue = []
|
||||
|
||||
if DB.table == "vods":
|
||||
for record_id, record_date, title, game_name, downloaded, uploaded_yt, chats_upload_yt in unuploaded:
|
||||
print(f"ID: {record_id} | Date: {record_date} | Game: {game_name} | Title: {title}")
|
||||
|
||||
file_path = f"save/vods/{record_id}/{record_id}.mp4"
|
||||
description = f"Game: {game_name}, on {record_date}, #VODS {twitch_datetime}"
|
||||
tags = list(base_tags)
|
||||
tags.extend([f'{game_name}', 'twitch_vods', 'vods'])
|
||||
tags.extend(top_hashtags(record_id))
|
||||
|
||||
upload_queue.append([record_id, file_path, title, categoryId, description, privatcyStatus, tags])
|
||||
|
||||
elif DB.table == "clips":
|
||||
for slug, record_date, title, game_name, clip_by, views, downloaded, uploaded_yt, shorts_uploaded_yt in unuploaded:
|
||||
print(f"Slug: {slug} | Date: {record_date} | Game: {game_name} | By: {clip_by} | Views: {views} | Title: {title}")
|
||||
|
||||
transcribe(slug)
|
||||
|
||||
file_path = f"save/clips/{slug}/{slug}.mp4"
|
||||
description = f"Game: {game_name}, Cliped By: {clip_by}, on {record_date}, #Shorts #Clips {twitch_datetime}"
|
||||
tags = list(base_tags)
|
||||
tags.extend([f'{game_name}', 'twitch_clips', 'clips', f'{clip_by}', 'shorts'])
|
||||
tags.extend(top_hashtags(slug))
|
||||
|
||||
upload_queue.append([slug, file_path, title, categoryId, description, privatcyStatus, tags])
|
||||
|
||||
for db_id, file_path, title, categoryId, description, privatcyStatus, tags in upload_queue:
|
||||
output = uploader.upload_video(file_path, title, categoryId, description, privatcyStatus, tags)
|
||||
if output is True:
|
||||
print(f"Title: {title} | Was successfully uploaded.")
|
||||
DB.mark_as_uploaded(db_id)
|
||||
else:
|
||||
print(f"Title: {title} | Download Process failed.")
|
||||
|
||||
def create_chats():
|
||||
pass
|
||||
|
||||
def create_shorts():
|
||||
pass
|
||||
|
||||
def get_vod_ids_simplified(channel_name: str):
|
||||
"""Queries Twitch's public endpoint directly for trending clips."""
|
||||
session = requests.Session()
|
||||
url = "https://gql.twitch.tv/gql"
|
||||
|
||||
# Case-preserved headers to prevent 400 Bad Request errors
|
||||
session.headers = {
|
||||
"Client-ID": "kimne78kx3ncx6brgo4mv6wki5h1ko",
|
||||
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
|
||||
"Content-Type": "text/plain"
|
||||
}
|
||||
|
||||
vods_query_string = """
|
||||
query GetChannelVideos($login: String!, $limit: Int!) {
|
||||
user(login: $login) {
|
||||
videos(first: $limit, types: [ARCHIVE]) {
|
||||
edges {
|
||||
node {
|
||||
id
|
||||
title
|
||||
publishedAt
|
||||
game {
|
||||
displayName
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
"""
|
||||
|
||||
clips_query_string = """
|
||||
query GetChannelClips($login: String!, $limit: Int!) {
|
||||
user(login: $login) {
|
||||
clips(first: $limit, criteria: { period: ALL_TIME }) {
|
||||
edges {
|
||||
node {
|
||||
slug
|
||||
title
|
||||
createdAt
|
||||
viewCount
|
||||
game {
|
||||
displayName
|
||||
}
|
||||
curator {
|
||||
login
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
"""
|
||||
|
||||
limit = 50
|
||||
query_string = ""
|
||||
operation_name = ""
|
||||
if DB.table == "vods":
|
||||
query_string = vods_query_string
|
||||
limit = 50
|
||||
operation_name = "GetChannelVideos"
|
||||
elif DB.table == "clips":
|
||||
query_string = clips_query_string
|
||||
limit = 40
|
||||
operation_name = "GetChannelClips"
|
||||
# A simplified query string that hardcodes the ARCHIVE type filter into the request structure
|
||||
payload = [{
|
||||
"operationName": operation_name,
|
||||
"query": query_string,
|
||||
"variables": {
|
||||
"login": channel_name.lower(),
|
||||
"limit": limit
|
||||
}
|
||||
}]
|
||||
|
||||
try:
|
||||
# Prepping ensures Python does not rewrite the Client-ID header case
|
||||
req = requests.Request('POST', url, json=payload)
|
||||
prepped = session.prepare_request(req)
|
||||
|
||||
response = session.send(prepped)
|
||||
response.raise_for_status()
|
||||
|
||||
data = response.json()
|
||||
|
||||
# Pull out the target index array dictionary object
|
||||
result = data[0] if isinstance(data, list) else data
|
||||
|
||||
if "errors" in result:
|
||||
print(f"Twitch GraphQL Error: {result['errors']}")
|
||||
return []
|
||||
|
||||
user_data = result['data']['user']
|
||||
if not user_data:
|
||||
print(f"Channel '{channel_name}' not found.")
|
||||
return []
|
||||
|
||||
edges = None
|
||||
if DB.table == "vods":
|
||||
edges = user_data['videos']['edges']
|
||||
elif DB.table == "clips":
|
||||
edges = user_data['clips']['edges']
|
||||
video_ids = []
|
||||
|
||||
print(f"--- Latest VODs for {channel_name} ---")
|
||||
for edge in edges:
|
||||
node = edge['node']
|
||||
|
||||
# Safe extraction in case a VOD has no category set (Just Chatting, Uncategorized, etc.)
|
||||
game_info = node.get('game')
|
||||
game_name = game_info.get('displayName') if game_info else "Unknown/No Category"
|
||||
if DB.table == "vods":
|
||||
print(f"ID: {node['id']} | Date: {node['publishedAt']} | Game: {game_name} | Title: {node['title']}")
|
||||
|
||||
# Pass game_name to your database logic
|
||||
DB.insert_vods_record(node['id'], node['publishedAt'], node['title'], game_name)
|
||||
video_ids.append(node['id'])
|
||||
elif DB.table == "clips":
|
||||
# Safe extraction in case the curator account was deleted/missing
|
||||
curator_info = node.get('curator')
|
||||
clip_by = curator_info.get('login') if curator_info else "Unknown Creator"
|
||||
|
||||
print(f"Slug: {node['slug']} | Date: {node['createdAt']} | Game: {game_name} | By: {clip_by} | Views: {node['viewCount']} | Title: {node['title']}")
|
||||
|
||||
DB.insert_clips_record(node['slug'], node['createdAt'], node['title'], game_name, clip_by, int(node['viewCount']))
|
||||
video_ids.append(node['slug'])
|
||||
|
||||
return video_ids
|
||||
|
||||
except Exception as e:
|
||||
print(f"An unexpected error occurred: {e}")
|
||||
if 'response' in locals():
|
||||
print(f"Server Response Text: {response.text}")
|
||||
return []
|
||||
|
||||
if __name__ == "__main__":
|
||||
tables = ["vods", "clips"]
|
||||
for table in tables:
|
||||
DB = Database(table)
|
||||
get_vod_ids_simplified(CHANNEL_NAME)
|
||||
download()
|
||||
upload()
|
||||
DB.close_database()
|
||||
Reference in New Issue
Block a user