Compare commits

18 Commits
Author SHA1 Message Date
SumGuyV5 7f1f1e7b4f add hight var to chatrender for 720p videos 2026-08-22 13:13:54 +00:00
SumGuyV5 33ea2b7fcf update 2026-08-22 03:39:48 +00:00
SumGuyV5 33d78759fe putting it all togeather 2026-08-04 02:34:16 +00:00
SumGuyV5 d6503b7d92 database , chat vod and shorts update error messages 2026-07-31 16:56:40 +00:00
SumGuyV5 736fb47e7a formating changes 2026-07-29 21:36:08 +00:00
SumGuyV5 93b480ae8d update linux run_command to give more output when stderrs out 2026-07-29 04:00:43 +00:00
SumGuyV5 b6d9b56848 update Database to use default in init 2026-07-29 03:03:29 +00:00
SumGuyV5 500fe87614 update and make test for transcribe and chat_vod and shorts 2026-07-29 01:58:18 +00:00
SumGuyV5 bbac53d15d add build_videos file 2026-07-28 04:14:59 +00:00
SumGuyV5 9203b0f88c exparmental run_command in linux.py migrate_sql1.py 2026-07-28 03:27:26 +00:00
SumGuyV5 6882b7ae04 update files to change [""] to [''] 2026-07-24 12:22:35 +00:00
SumGuyV5 0f96eedc54 update requirements.txt 2026-07-24 12:04:02 +00:00
SumGuyV5 83f7bbead8 uncomment the mark_as_downloaded 2026-07-24 04:54:39 +00:00
SumGuyV5 1257a15eb9 update chmod 755 and addes twitch_download_videos.py 2026-07-24 04:09:12 +00:00
SumGuyV5 765eaf5026 file name change 2026-07-23 17:34:36 +00:00
SumGuyV5 4934ddb3f8 update and rename twitch_vod.py and use httpx for async 2026-07-23 17:34:12 +00:00
SumGuyV5 c8d12469e3 Update Database to use new table fromat and twitch api 2026-07-23 15:32:40 +00:00
SumGuyV5 81e76b25d0 Made some chanages and now Can't remmebr what I did 2026-07-23 04:55:36 +00:00
24 changed files with 1740 additions and 741 deletions
+5
View File
@@ -4,3 +4,8 @@
database.db
save
__pycache__
twitch_secrets.json
.vscode/settings.json
download
output.log
*.mp4
Executable
+74
View File
@@ -0,0 +1,74 @@
#!/usr/bin/env python3
import asyncio
import json
import os
import webbrowser
from twitchAPI.twitch import Twitch
from twitchAPI.oauth import UserAuthenticator
from twitchAPI.type import AuthScope
SECRETS_FILE = "twitch_secrets.json"
def load_credentials():
"""Loads existing Client ID and Secret from your JSON file."""
if not os.path.exists(SECRETS_FILE):
raise FileNotFoundError(f"Could not find {SECRETS_FILE} in this directory.")
with open(SECRETS_FILE, "r") as f:
data = json.load(f)
return data.get("client_id"), data.get("client_secret")
def save_token_to_json(token):
"""Saves the generated token into twitch_secrets.json under 'manual_token'."""
with open(SECRETS_FILE, "r") as f:
data = json.load(f)
# Inject the new token
data["manual_token"] = token
with open(SECRETS_FILE, "w") as f:
json.dump(data, f, indent=4)
print(f"\n[SUCCESS] Token saved inside '{SECRETS_FILE}' under 'manual_token'!")
async def main():
try:
client_id, client_secret = load_credentials()
if not client_id or not client_secret:
print("[ERROR] Please add your client_id and client_secret to the JSON file first.")
return
print("Initializing local connection loop...")
# Initialize official Twitch connection interface
twitch = await Twitch(client_id, client_secret)
# Scopes: We leave this empty [] since VOD collection only requires basic public clearance
scopes = []
# Create an authenticator that automatically sets up http://localhost:17563
auth = UserAuthenticator(twitch, scopes, url="http://localhost:17563")
# Request authentication URL
auth_url = auth.return_auth_url()
print(f"\nIf your browser does not open automatically, copy and paste this URL into your browser:\n{auth_url}\n")
# Open your system default browser to let you manually click "Authorize"
webbrowser.open(auth_url)
print("Waiting for you to click 'Authorize' in your web browser...")
# The script halts here, running a local background server until you click authorize
token, refresh_token = await auth.authenticate()
print(f"\nSuccessfully generated Token: {token}")
# Save it right back into your configuration file
save_token_to_json(token)
# Gracefully shut down the library connection
await twitch.close()
except Exception as e:
print(f"\n[ERROR] An error occurred: {e}")
print("Double-check that http://localhost:17563 is added to your Twitch Dev Console.")
if __name__ == "__main__":
# Run the asynchronous loop
asyncio.run(main())
+127
View File
@@ -0,0 +1,127 @@
#!/usr/bin/env python3
import database
import youtube_short
import transcribe_video
import twitch_chat_vod
DB = None
def build_chat_video():
chats = DB.get_unuploaded_chats()
for chat in chats:
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = chat
print("====================================================")
print(f"🚀 Chat Video: {title}")
print("====================================================")
twitch_chat_vod.combine_twitch_vod_and_chat(f"download/videos/{id}/{id}.mp4", "side-by-side")
print("====================================================")
print(f"✅ Processing: Chat Video Done.")
print("====================================================")
def build_transcribe():
unuploadeds = DB.get_unuploaded()
for unuploaded in unuploadeds:
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = unuploaded
if clip_is:
target_dir = f"download/clips"
else:
target_dir = f"download/videos"
print("====================================================")
print(f"🚀 Transcribe: {title}")
print("====================================================")
transcribe_video.transcribe_to_srt(f"{target_dir}/{id}/{id}.mp4")
print("====================================================")
print(f"✅ Processing: Transcribing Done.")
print("====================================================")
def build_shorts():
shorts = DB.get_unuploaded_shorts()
top_txt = "TeamPGP Live on Twitch...\n Every Friday and Sunday @7:30 ET.\n twitch.tv/teampgp"
for short in shorts:
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = short
target_dir = f"download/clips/{id}/{id}.mp4"
print("====================================================")
print(f"🚀 Processing: Clip to Youtube Short {title}")
print("====================================================")
youtube_short.fit_to_9_16_letterbox(target_dir, top_txt, f"Clipped By: {creator_name}.")
print("====================================================")
print(f"✅ Processing: Clip to Youtube Short Done.")
print("====================================================")
def main():
global DB
DB = database.Database()
build_shorts()
build_transcribe()
build_chat_video()
if __name__ == "__main__":
main()
Regular → Executable
+126 -112
View File
@@ -1,148 +1,162 @@
#!/usr/bin/env python3
import sqlite3
from datetime import datetime
from datetime import date as datetime_date
from pathlib import Path
from typing import Any
class Database:
def __init__(self, table: str):
self.table = table
self.columns = ""
def __init__(self, db_path: str = "database.db"):
self.columns = "id, title, created_at, view_count, duration, url, thumbnail_url, game_id, game_name, stream_id, creator_name, clip_is, downloaded, uploaded_yt, uploaded_yt_chats, uploaded_yt_shorts"
# IMPORTANT Must check if database.db exists before connecting to it.
file_exists = Path("database.db").is_file()
file_exists = Path(db_path).is_file()
self.CONN = sqlite3.connect("database.db")
self.CURSOR = self.CONN.cursor()
if self.table == "vods":
self.columns = "id, date, title, gamename, downloaded, uploaded_yt, chats_upload_yt"
elif self.table == "clips":
self.columns = "slug, date, title, gamename, clip_by, view_count, downloaded, uploaded_yt, shorts_upload_yt"
self.conn = sqlite3.connect(db_path)
self.cursor = self.conn.cursor()
if not file_exists:
self.create_database()
def __del__(self):
self.close_database()
def __exit__(self):
# Destructors are unpredictable in Python; explicitly close when done instead
try:
self.close_database()
except:
pass
def create_database(self):
"""Creates a table structured explicitly for Twitch clip properties."""
self.CURSOR.execute(
"""
CREATE TABLE IF NOT EXISTS clips (
slug TEXT PRIMARY KEY,
date TEXT NOT NULL,
title TEXT NOT NULL,
gamename TEXT NOT NULL,
clip_by TEXT NOT NULL,
view_count INTEGER NOT NULL,
downloaded INTEGER NOT NULL,
uploaded_yt INTEGER NOT NULL,
shorts_upload_yt INTEGER NOT NULL
)
"""
"""Creates a table structured explicitly for Twitch videos properties."""
self.cursor.execute("""
CREATE TABLE IF NOT EXISTS twitch_videos (
id TEXT PRIMARY KEY,
title TEXT NOT NULL,
created_at TEXT NOT NULL,
view_count INTEGER NOT NULL,
duration TEXT NOT NULL,
url TEXT NOT NULL,
thumbnail_url TEXT NOT NULL,
game_id TEXT NOT NULL,
game_name TEXT NOT NULL,
stream_id TEXT NOT NULL,
creator_name TEXT NOT NULL,
clip_is BOOLEAN NOT NULL DEFAULT 0,
downloaded BOOLEAN NOT NULL DEFAULT 0,
uploaded_yt BOOLEAN NOT NULL DEFAULT 0,
uploaded_yt_chats BOOLEAN NOT NULL DEFAULT 0,
uploaded_yt_shorts BOOLEAN NOT NULL DEFAULT 0
)
"""Creates a table structured explicitly for Twitch vods properties."""
self.CURSOR.execute(
"""
CREATE TABLE IF NOT EXISTS vods (
id INTEGER PRIMARY KEY,
date TEXT NOT NULL,
title TEXT NOT NULL,
gamename TEXT NOT NULL,
downloaded INTEGER NOT NULL,
uploaded_yt INTEGER NOT NULL,
chats_upload_yt INTEGER NOT NULL
)
"""
)
self.CONN.commit()
""")
self.conn.commit()
def close_database(self):
"""Commit before we close."""
self.CONN.commit()
self.CONN.close()
if self.conn:
self.conn.commit()
self.conn.close()
def __mark_as(self, record_id: int, set_sql: str):
id = "id"
if self.table == "clips":
id = "slug"
def __mark_as(self, record_id: str, set_row: str, mark: str = "1" ):
self.cursor.execute(
f"UPDATE twitch_videos SET {set_row} = ? WHERE id = ?",
(mark, record_id))
self.conn.commit()
self.CURSOR.execute(
f"UPDATE {self.table} SET {set_sql} = 1 WHERE {id} = ?",
(record_id,)
)
self.CONN.commit()
def mark_as_uploaded_shorts(self, record_id: str):
"""Flags a specific row record to uploaded_yt_shorts (1)."""
self.__mark_as(record_id, "uploaded_yt_shorts")
def mark_as_uploaded_shorts_chats(self, record_id: int):
"""Flags a specific row record to uploaded (1)."""
set_sql = "chats_upload_yt"
if self.table == "clips":
set_sql = "shorts_uploaded_yt"
self.__mark_as(record_id, set_sql)
def mark_as_uploaded(self, record_id: int):
"""Flags a specific row record to uploaded (1)."""
def mark_as_uploaded_chats(self, record_id: str):
"""Flags a specific row record to uploaded_yt_chats (1)."""
self.__mark_as(record_id, "uploaded_yt_chats")
def mark_as_uploaded_yt(self, record_id: str):
"""Flags a specific row record to uploaded_yt (1)."""
self.__mark_as(record_id, "uploaded_yt")
def mark_as_downloaded(self, record_id: int):
"""Updates the downloaded status to True (1) for a specific record ID."""
def mark_as_downloaded(self, record_id: str):
"""Flags a specific row record to downloaded (1)."""
self.__mark_as(record_id, "downloaded")
def get_unuploaded_shorts_chats(self):
"""Retrieves all clip rows remaining to be chat uploaded."""
def unmark_as_download(self, record_id: str):
"""Flags a specific row record to downloaded (0)."""
self.__mark_as(record_id, "downloaded", "0")
where_sql = "chats_upload_yt"
if self.table == "clips":
where_sql = "shorts_uploaded_yt"
def __get_unuploaded(self, set_row: str, also: str = "") -> list[Any]:
"""Retrieve all rows that were download but not uploaded"""
self.cursor.execute(f"SELECT {self.columns} FROM twitch_videos WHERE downloaded = 1 AND {set_row} = 0 {also}")
return self.cursor.fetchall()
self.CURSOR.execute(f"SELECT {self.columns} FROM {self.table} WHERE downloaded = 1 AND uploaded_yt = 1 AND {where_sql} = 0")
return self.CURSOR.fetchall()
def get_unuploaded_shorts(self) -> list[Any]:
"""Retrieve all rows that were download but not uploaded_yt_shorts"""
return self.__get_unuploaded("uploaded_yt_shorts", "AND clip_is = 1")
def get_unuploaded_chats(self) -> list[Any]:
"""Retrieve all rows that were download but not uploaded_yt_chats"""
return self.__get_unuploaded("uploaded_yt_chats", "AND clip_is = 0")
def get_unuploaded(self):
"""Retrieves all clip rows remaining to be uploaded."""
self.CURSOR.execute(f"SELECT {self.columns} FROM {self.table} WHERE downloaded = 1 AND uploaded_yt = 0")
return self.CURSOR.fetchall()
def get_unuploaded(self) -> list[Any]:
"""Retrieve all rows that were download but not uploaded_yt"""
return self.__get_unuploaded("uploaded_yt")
def get_undownloaded(self):
"""Retrieves all rows where downloaded status is False (0)."""
self.CURSOR.execute(f"SELECT {self.columns} FROM {self.table} WHERE downloaded = 0")
return self.CURSOR.fetchall()
def get_undownloaded(self) -> list[Any]:
"""Retrieves all rows that were not downloaded"""
self.cursor.execute(f"SELECT {self.columns} FROM twitch_videos WHERE downloaded = 0")
return self.cursor.fetchall()
def insert_vods_record(self, record_id: int, record_date_str: datetime_date, title: str, gamename: str):
"""Inserts a record with ID, date, gamename, and title into a SQLite database."""
def get_download(self) -> list[Any]:
"""Retrieves all rows that were downloaded"""
self.cursor.execute(f"SELECT {self.columns} FROM twitch_videos WHERE downloaded = 1")
return self.cursor.fetchall()
def get_clips(self) -> list[Any]:
"""Retrieves all rows that are clip_is"""
self.cursor.execute(f"SELECT {self.columns} FROM twitch_videos WHERE clip_is = 1")
return self.cursor.fetchall()
def get_vods(self) -> list[Any]:
self.cursor.execute(f"SELECT {self.columns} FROM twitch_videos WHERE clip_is = 0")
return self.cursor.fetchall()
def insert_video_record(
self, id: str, title: str, created_at: str, view_count: int, duration: str,
url: str, thumbnail_url: str, game_id: int, game_name: str, stream_id: str,
creator_name: str, clip_is: bool
):
try:
clean_date = datetime.strptime(record_date_str, "%Y-%m-%dT%H:%M:%SZ").date()
except ValueError:
clean_date = record_date_str
# Connects to database file (creates it if missing)
dt_obj = datetime.strptime(created_at, "%Y-%m-%dT%H:%M:%SZ")
clean_datetime = dt_obj.strftime("%Y-%m-%d %H:%M:%S")
except (ValueError, TypeError):
clean_datetime = created_at
# Inserts data using parameterized queries to prevent SQL injection
self.CURSOR.execute(
f"INSERT OR IGNORE INTO {self.table} ({self.columns}) VALUES (?, ?, ?, ?, ?, ?, ?)",
(record_id, str(clean_date), title, gamename, False, False, False),
# Explicitly defining columns removes the security risk and column-count bug
query = """
INSERT INTO twitch_videos (
id, title, created_at, view_count, duration, url, thumbnail_url,
game_id, game_name, stream_id, creator_name, clip_is
) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
ON CONFLICT(id) DO UPDATE SET
title = excluded.title,
created_at = excluded.created_at,
view_count = excluded.view_count,
duration = excluded.duration,
url = excluded.url,
thumbnail_url = excluded.thumbnail_url,
game_id = excluded.game_id,
game_name = excluded.game_name,
stream_id = excluded.stream_id,
creator_name = excluded.creator_name,
clip_is = excluded.clip_is
WHERE excluded.duration != twitch_videos.duration
"""
values = (
id, title, clean_datetime, view_count, duration, url, thumbnail_url,
game_id, game_name, stream_id, creator_name, clip_is
)
# Saves changes and closes the connection
self.CONN.commit()
def insert_clips_record(self, slug: str, record_date_str: str, title: str, gamename: str, clip_by: str, views: int):
"""Cleans up ISO-8601 strings into unified date structures for the database."""
try:
clean_date = datetime.strptime(record_date_str, "%Y-%m-%dT%H:%M:%SZ").date()
except ValueError:
clean_date = record_date_str
self.CURSOR.execute(
f"INSERT OR IGNORE INTO {self.table} ({self.columns}) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)",
(slug, str(clean_date), title, gamename, clip_by, views, False, False, False),
)
self.CONN.commit()
self.cursor.execute(query, values)
self.conn.commit()
except Exception as e:
# Prevent silent failures if the database connection drops
print(f"❌ Database insertion failed: {e}")
self.conn.rollback()
Regular → Executable
+6 -6
View File
@@ -38,8 +38,8 @@ def get_my_uploads_playlist_id(youtube):
request = youtube.channels().list(part="contentDetails", mine=True)
response = request.execute()
if "items" in response and len(response["items"]) > 0:
return response["items"][0]["contentDetails"]["relatedPlaylists"]["uploads"]
if "items" in response and len(response['items']) > 0:
return response['items'][0]['contentDetails']['relatedPlaylists']['uploads']
else:
print("No channel found for these credentials.")
return None
@@ -70,7 +70,7 @@ def scan_channel_videos_for_tag(youtube, uploads_playlist_id: str, target_tag: s
playlist_response = playlist_request.execute()
video_ids_batch = [
item["snippet"]["resourceId"]["videoId"]
item['snippet']['resourceId']['videoId']
for item in playlist_response.get("items", [])
]
@@ -87,10 +87,10 @@ def scan_channel_videos_for_tag(youtube, uploads_playlist_id: str, target_tag: s
video_response = video_request.execute()
for video in video_response.get("items", []):
title = video["snippet"]["title"]
video_id = video["id"]
title = video['snippet']['title']
video_id = video['id']
# Tags are optional fields on YouTube; default to an empty list if absent
tags = video["snippet"].get("tags", [])
tags = video['snippet'].get("tags", [])
# Normalize tags to lowercase for clean matching evaluation
tags_lower = [tag.lower() for tag in tags]
Executable
+56
View File
@@ -0,0 +1,56 @@
#!/usr/bin/env python3
import sys
import shlex
import subprocess
def run_command(cmd_str: str, progress_prefix: str = "Progress", look_for: list = ["frame=", "time=", "fps=", "Rendering frame"]) -> bool:
"""Runs a system command, streams its output live, and reports errors on failure."""
args = shlex.split(cmd_str)
# Redirect stderr to stdout to catch all logging/progress in one stream
process = subprocess.Popen(
args,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
text=True,
bufsize=1
)
print(f"Executing: {cmd_str[:90]}...")
# Maintain a small buffer history to display context if a crash occurs
output_history = []
# Stream the output live to the terminal
while True:
line = process.stdout.readline()
if not line and process.poll() is not None:
break
if line:
clean_line = line.strip()
output_history.append(clean_line) # Keep history for error reporting
# Keep history slim by only keeping the last 20 lines
if len(output_history) > 20:
output_history.pop(0)
# Only print updates that show progress metrics to keep terminal clean
if any(metric in clean_line for metric in look_for):
sys.stdout.write(f"\r[{progress_prefix}] {clean_line}")
sys.stdout.flush()
elif "Error" in clean_line or "failed" in clean_line:
print(f"\n[Alert] {clean_line}")
print("\n") # New line after process finishes
# Evaluate success status
success = (process.returncode == 0)
if not success:
print(f"❌ Command failed with exit code: {process.returncode}")
print("--- Technical Error Details (Last 5 lines of output) ---")
# Print the last 5 captured lines to show the exact point of failure
for error_line in output_history[-5:]:
print(f" > {error_line}")
print("---------------------------------------------------------")
return success
+20
View File
@@ -0,0 +1,20 @@
#!/usr/bin/env python3
import asyncio
import twitch_video_info
import twitch_download_videos
import twitch_download_thumbnails
import build_videos
if __name__ == "__main__":
# Get Twitch Video Info
asyncio.run(twitch_video_info.main())
# Download Twitch Videos
twitch_download_videos.main()
twitch_download_thumbnails.main()
#Build
build_videos.main()
+67
View File
@@ -0,0 +1,67 @@
#!/usr/bin/env python3
import sqlite3
# Connect to your database file
db_name = "database.db" # Change to your actual file name
conn = sqlite3.connect(db_name)
cursor = conn.cursor()
try:
# 1. Create a temporary table with the new text-based game_id
cursor.execute(
"""
CREATE TABLE IF NOT EXISTS twitch_videos_temp (
id TEXT PRIMARY KEY,
title TEXT NOT NULL,
created_at TEXT NOT NULL,
view_count INTEGER NOT NULL,
duration TEXT NOT NULL,
url TEXT NOT NULL,
thumbnail_url TEXT NOT NULL,
game_id TEXT NOT NULL,
game_name TEXT NOT NULL,
stream_id TEXT NOT NULL,
creator_name TEXT NOT NULL,
clip_is BOOLEAN NOT NULL DEFAULT 0,
downloaded BOOLEAN NOT NULL DEFAULT 0,
uploaded_yt BOOLEAN NOT NULL DEFAULT 0,
uploaded_yt_chats BOOLEAN NOT NULL DEFAULT 0,
uploaded_yt_shorts BOOLEAN NOT NULL DEFAULT 0
)
"""
)
# 2. Copy and convert data to the temporary table
cursor.execute(
"""
INSERT INTO twitch_videos_temp (
id, title, created_at, view_count, duration, url, thumbnail_url,
game_id, game_name, stream_id, creator_name, clip_is, downloaded,
uploaded_yt, uploaded_yt_chats, uploaded_yt_shorts
)
SELECT
id, title, created_at, view_count, duration, url, thumbnail_url,
CAST(game_id AS TEXT), game_name, stream_id, creator_name, clip_is, downloaded,
uploaded_yt, uploaded_yt_chats, uploaded_yt_shorts
FROM twitch_videos
"""
)
# 3. Drop the old table configuration
cursor.execute("DROP TABLE twitch_videos")
# 4. Rename the temporary table to your exact original table name
cursor.execute("ALTER TABLE twitch_videos_temp RENAME TO twitch_videos")
# Commit changes if everything succeeded
conn.commit()
print("Migration successful! 'twitch_videos' table updated.")
except sqlite3.Error as e:
# Roll back changes if an error occurs
conn.rollback()
print(f"An error occurred: {e}")
finally:
# Close database connection
conn.close()
+35
View File
@@ -0,0 +1,35 @@
#!/usr/bin/env python3
import database
DB = None
def redownload_clips():
clips = DB.get_clips()
for clip in clips:
# Unpack variables clearly
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = clip
DB.unmark_as_download(id)
if __name__ == "__main__":
DB = database.Database()
redownload_clips()
+72
View File
@@ -0,0 +1,72 @@
import os
import database
DB = None
def remove_files():
datas = DB.get_download()
top_txt = "TeamPGP Live on Twitch...\n Every Friday and Sunday @7:30 ET.\n twitch.tv/teampgp"
for data in datas:
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = data
if clip_is == 0:
video_path = f"download/videos/{id}/{id}.mp4"
else:
video_path = f"download/clips/{id}/{id}.mp4"
temp_chat = video_path.replace(".mp4", "_temp_chat.mp4")
temp_with_chat = video_path.replace(".mp4", "_temp_with_chat.mp4")
video_chat = video_path.replace(".mp4", "_with_chat.mp4")
video_chatSS = video_path.replace(".mp4", "_SS_with_chat.mp4")
video_chat_over = video_path.replace(".mp4", "_over_with_chat.mp4")
mask_path = video_path.replace(".mp4", "_temp_chat_mask.mp4")
shorts_gaussian_9_16 = ""#video_path.replace(".mp4", "_gaussian_9_16.mp4")
shorts_9_16 = ""#video_path.replace(".mp4", "_9_16.mp4")
srt_file = ""#video_path.replace(".mp4", ".srt")
files = [temp_chat, temp_with_chat, video_chatSS, video_chat_over, mask_path, shorts_gaussian_9_16, shorts_9_16, srt_file, video_chat]
for file in files:
if os.path.exists(file):
if file == video_path:
print("❌ Error: trying to delete import file.")
return
else:
os.remove(file)
def redownload():
DB.unmark_as_download("2840110867")
def main():
global DB
DB = database.Database()
remove_files()
redownload()
if __name__ == "__main__":
main()
+41 -29
View File
@@ -1,39 +1,51 @@
apt-listchanges==4.8
aiohappyeyeballs==2.7.1
aiohttp==3.14.2
aiosignal==1.4.0
anyio==4.14.2
attrs==26.1.0
beautifulsoup4==4.14.3
certifi==2026.2.25
cffi==2.0.0
chardet==5.2.0
charset-normalizer==3.4.7
beautifulsoup4==4.15.0
certifi==2026.7.22
cffi==2.1.0
charset-normalizer==3.4.9
chat-downloader==0.2.8
click==8.4.2
colorama==0.4.6
colorlog==6.12.0
cryptography==49.0.0
cuda-bindings==13.3.1
cuda-pathfinder==1.5.6
cuda-pathfinder==1.6.0
cuda-toolkit==13.0.3.0
decorator==5.3.1
defusedxml==0.7.1
docopt==0.6.2
filelock==3.31.0
docstring_parser==0.18.0
enum-tools==0.13.0
filelock==3.32.0
frozenlist==1.8.0
fsspec==2026.6.0
google-api-core==2.31.0
git-filter-repo==2.47.0
google==3.0.0
google-api-core==2.32.0
google-api-python-client==2.198.0
google-auth==2.56.0
google-auth==2.56.2
google-auth-httplib2==0.4.0
google-auth-oauthlib==1.4.0
googleapis-common-protos==1.75.0
h11==0.16.0
httpcore==1.0.9
httplib2==0.32.0
idna==3.11
ImageIO==2.37.3
httpx==0.28.1
idna==3.18
ImageIO==2.37.4
imageio-ffmpeg==0.6.0
isodate==0.7.2
Jinja2==3.1.6
joblib==1.5.3
llvmlite==0.48.0
lxml==6.1.1
MarkupSafe==3.0.3
more-itertools==11.1.0
moviepy==2.2.1
mpmath==1.3.0
multidict==6.7.1
networkx==3.6.1
nltk==3.10.0
numba==0.66.0
@@ -56,42 +68,42 @@ nvidia-nvtx==13.0.85
oauthlib==3.3.1
openai-whisper==20250625
outcome==1.3.0.post0
packaging==26.2
pillow==11.3.0
pipreqs==0.4.13
pip_system_certs==5.3
proglog==0.1.12
propcache==0.5.2
proto-plus==1.28.1
protobuf==7.35.1
pyasn1==0.6.4
pyasn1_modules==0.4.2
pycountry==26.2.16
pycparser==3.0
pycryptodome==3.23.0
Pygments==2.20.0
pyparsing==3.3.2
PySocks==1.7.1
python-apt==3.0.0
python-debian==1.0.1
python-debianbts==4.1.1
python-dateutil==2.9.0.post0
python-dotenv==1.2.2
regex==2026.7.19
reportbug==13.2.0
requests==2.33.1
requests==2.34.2
requests-oauthlib==2.0.0
selenium==4.43.0
setuptools==83.0.0
six==1.17.0
sniffio==1.3.1
sortedcontainers==2.4.0
soupsieve==2.8.3
soupsieve==2.9.1
streamlink==8.4.0
sympy==1.14.0
tiktoken==0.13.0
torch==2.13.0
tqdm==4.67.3
tqdm==4.69.0
trio==0.33.0
trio-websocket==0.12.2
triton==3.7.1
typing_extensions==4.15.0
twitchAPI==4.5.0
typing_extensions==4.16.0
uritemplate==4.2.0
urllib3==2.6.3
webdriver-manager==4.0.2
urllib3==2.7.0
websocket-client==1.9.0
wheel==0.46.1
wsproto==1.3.2
yarg==0.1.10
yarl==1.24.5
Regular → Executable
+105 -53
View File
@@ -1,67 +1,119 @@
#!/usr/bin/env python3
import os
import sys
import shutil
import subprocess
import whisper
from moviepy import VideoFileClip
from whisper.utils import get_writer
from pathlib import Path
from faster_whisper import WhisperModel
from faster_whisper.utils import format_timestamp
model = whisper.load_model("base")
# Prevent OpenMP thread conflicts from crashing the script
os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"
def extract_audio(video_path: str, audio_temp_path: str):
# Create a temporary path for a sanitized copy of the video
sanitized_video_path = video_path.replace(".mp4", "_clean.mp4")
# 5 minutes per chunk (300 seconds) keeps RAM usage low and stable
CHUNK_DURATION_SEC = 300
print("🔧 Initializing C++ Engine...")
model = WhisperModel(
"base",
device="cpu",
compute_type="int8",
cpu_threads=0, # Let CTranslate2 auto-detect safe core counts
num_workers=1
)
print("✅ C++ Model loaded successfully.")
def extract_audio_and_chunk(video_path: str, output_dir: Path) -> list:
"""Extracts and splits audio into 5-minute chunks using a single FFmpeg pass."""
print("🚀 Extracting and chunking audio with FFmpeg...")
output_dir.mkdir(parents=True, exist_ok=True)
print("Sanitizing video metadata for MoviePy parser...")
# -map_chapters -1 removes chapter layouts that break the parser.
# -sn strips text/subtitle streams that crash MoviePy.
# -c copy copies video and audio instantly without quality loss.
cleanup_cmd = [
"ffmpeg", "-y", "-i", video_path,
"-map_chapters", "-1", "-sn",
"-c", "copy", sanitized_video_path
# Segment format output: chunk_000.wav, chunk_001.wav, etc.
chunk_pattern = str(output_dir / "chunk_%03d.wav")
command = [
"ffmpeg", "-y", "-i", video_path,
"-vn", "-ac", "1", "-ar", "16000",
"-acodec", "pcm_s16le", "-sn", "-map_chapters", "-1",
"-f", "segment", "-segment_time", str(CHUNK_DURATION_SEC),
chunk_pattern
]
# Run the sanitization process silently
subprocess.run(cleanup_cmd, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
result = subprocess.run(command, stdout=subprocess.DEVNULL, stderr=subprocess.PIPE, text=True)
if result.returncode != 0:
print(f"❌ FFmpeg Error Output:\n{result.stderr}")
raise RuntimeError("FFmpeg extraction and chunking failed.")
# Return sorted list of generated chunk files
return sorted(list(output_dir.glob("chunk_*.wav")))
def transcribe_to_srt(video_path: str, force: bool = False):
video_path_obj = Path(video_path)
srt_path = video_path_obj.with_suffix(".srt")
temp_dir = video_path_obj.parent / f"temp_chunks_{video_path_obj.stem}"
if srt_path.exists():
if not force:
print(f"❌ Error: Transcription file already exists: {srt_path}")
return
else:
srt_path.unlink()
print("Extracting uncompressed WAV audio...")
try:
# Load the sanitized file instead of the raw Twitch clip
with VideoFileClip(sanitized_video_path) as video:
video.audio.write_audiofile(
audio_temp_path,
fps=16000,
codec="pcm_s16le",
ffmpeg_params=["-ac", "1"]
)
# Step 1: Split audio into bite-sized pieces
audio_chunks = extract_audio_and_chunk(str(video_path_obj), temp_dir)
if not audio_chunks:
print("❌ Error: No audio chunks were generated.")
return
print(f"📦 Successfully split audio into {len(audio_chunks)} chunks.")
print("🎙️ Starting safe chunk-by-chunk transcription...")
global_segment_index = 1
with open(srt_path, "w", encoding="utf-8") as srt_file:
for chunk_idx, chunk_path in enumerate(audio_chunks):
# Calculate the time offset for the current chunk
time_offset = chunk_idx * CHUNK_DURATION_SEC
print(f"\n⏳ Processing chunk {chunk_idx + 1}/{len(audio_chunks)} ({chunk_path.name})...")
segments_generator, info = model.transcribe(
str(chunk_path),
beam_size=1,
vad_filter=True,
temperature=0.0
)
# Consume chunk generator and shift timestamps instantly
for segment in segments_generator:
# Shift timestamps relative to the original video timeline
actual_start = segment.start + time_offset
actual_end = segment.end + time_offset
start_str = format_timestamp(actual_start, always_include_hours=True)
end_str = format_timestamp(actual_end, always_include_hours=True)
srt_file.write(f"{global_segment_index}\n{start_str} --> {end_str}\n{segment.text.strip()}\n\n")
global_segment_index += 1
# Free up space as we go by deleting the processed chunk
chunk_path.unlink()
print(f"\n✅ All chunks combined! SRT subtitle file saved in: {srt_path}")
except Exception as e:
print(f"\n❌ Execution Error: {e}")
finally:
# Always clean up the temporary sanitized video on Windows 11
if os.path.exists(sanitized_video_path):
os.remove(sanitized_video_path)
def transcribe_to_srt(audio_path: str, output_directory: str, output_filename: str):
print("Transcribing audio...")
result = model.transcribe(audio_path)
print("Creating SRT file...")
srt_writer = get_writer("srt", output_directory)
srt_writer(result, output_filename, {})
print(f"SRT subtitle file saved in: {output_directory}")
if os.path.exists(audio_path):
os.remove(audio_path)
# Clean up the temporary folder entirely
if temp_dir.exists():
shutil.rmtree(temp_dir)
if __name__ == "__main__":
video_path = "my_video.mp4"
audio_temp_path = "temp_audio.wav" # Changed extension to .wav
output_dir = os.getcwd()
output_prefix = "my_video_subtitles"
extract_audio(video_path, audio_temp_path)
transcribe_to_srt(audio_temp_path, output_dir, output_prefix)
if os.path.exists(audio_temp_path):
os.remove(audio_temp_path)
target_video = "download/videos/2813112936/2813112936.mp4"
if not os.path.exists(target_video):
print(f"❌ System Error: Target video file does not exist at path: {target_video}")
else:
transcribe_to_srt(target_video, force=True)
+150
View File
@@ -0,0 +1,150 @@
#!/usr/bin/env python3
import os
import cv2
import linux
def get_video_height(video_path: str) -> int:
# Open the video file
video = cv2.VideoCapture(video_path)
# Get the height property
height = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT))
# Always release the video object
video.release()
return height
def combine_twitch_vod_and_chat(vod_path: str, layout: str = "side-by-side", force: bool = False) -> bool:
"""
Renders Twitch chat JSON to video and combines it with the source VOD.
Provides real-time terminal feedback for all processing steps.
"""
threads = 8
video_height = 1080
video_path = vod_path
video_chat = ""
chat_path = video_path.replace(".mp4", "_chat.json")
temp_with_chat = video_path.replace(".mp4", "_temp_with_chat.mp4")
temp_chat = video_path.replace(".mp4", "_temp_chat.mp4")
video_chat_SS = video_path.replace(".mp4", "_SS_with_chat.mp4")
video_chat_over = video_path.replace(".mp4", "_over_with_chat.mp4")
mask_path = video_path.replace(".mp4", "_temp_chat_mask.mp4")
# Step 1: Pre-flight checks
if not os.path.exists(video_path):
print(f"❌ Error: Source video not found: {video_path}")
return False
if not os.path.exists(chat_path):
print(f"❌ Error: Chat file not found: {chat_path}")
return False
if os.path.exists(temp_with_chat):
os.remove(temp_with_chat)
if layout == "side-by-side":
video_chat = video_chat_SS
elif layout == "overlay":
video_chat = video_chat_over
else:
return False
if os.path.exists(video_chat):
if not force:
print(f"❌ Error: Chat video already exists: {video_chat}")
return False
else:
os.remove(video_chat)
if os.path.exists(temp_chat):
os.remove(temp_chat)
if os.path.exists(mask_path):
os.remove(mask_path)
# Step 2: Render Chat to Video
print("====================================================")
print("🚀 STEP 1: Rendering Chat JSON to Video Layer")
print("====================================================")
video_height = get_video_height(vod_path)
chat_cmd = (
f"TwitchDownloaderCLI chatrender "
f"-i {chat_path} "
f"-w 400 -h {video_height} "
f"--collision Overwrite "
f"--temp-path download/temp "
f"--font-size 20 "
f"--background-color #00000000 "
f"-o {temp_chat}"
)
if layout == "overlay":
chat_cmd = (f"{chat_cmd} --generate-mask ")
if not os.path.exists(temp_chat):
chat_success = linux.run_command(chat_cmd, look_for=["[STATUS]"])
if chat_success:
print("✅ Success: TwitchDownloaderCLI render chat video.")
else:
print("❌ Error: TwitchDownloaderCLI failed to render chat video.")
os.remove(temp_chat)
os.remove(mask_path)
return False
# Step 3: Combine Video and Chat using FFmpeg
print("====================================================")
print(f"🚀 STEP 2: Merging VOD and Chat Layout ({layout})")
print("====================================================")
# -preset superfast speeds up the 3+ hour encoding process significantly
# -map 0:a? safely includes audio if it exists, without breaking on silent VODs
if layout == "side-by-side":
ffmpeg_cmd = (
f"ffmpeg -y -i {video_path} -i {temp_chat} "
f"-filter_complex '[1:v]scale=-1:ih[scaled_chat];[0:v][scaled_chat]hstack=inputs=2[v]' "
f"-map '[v]' -map 0:a? -c:v libx264 -crf 18 -preset slow -c:a copy "
f"-threads {threads} {temp_with_chat}"
)
elif layout == "overlay":
ffmpeg_cmd = (
f"ffmpeg -y -i {video_path} -i {temp_chat} -i {mask_path} "
f"-filter_complex '[1:v][2:v]alphamerge[masked_chat];[0:v][masked_chat]overlay=x=0:y=10[v]' "
f"-map '[v]' -map 0:a? -c:v libx264 -crf 18 -preset slow -c:a copy "
f"-threads {threads} {temp_with_chat}"
)
else:
raise ValueError("Invalid layout choice. Choose 'side-by-side' or 'overlay'.")
ffmpeg_success = linux.run_command(ffmpeg_cmd, progress_prefix="FFmpeg Merge")
# Step 4: Final verification and cleanup
if ffmpeg_success and os.path.exists(temp_with_chat):
os.rename(temp_with_chat, video_chat)
print("====================================================")
print(f"🎉 SUCCESS: Video processing complete!")
print(f"📁 Output Saved: {video_chat}")
print("====================================================")
# Clean up the massive temporary chat video to save storage space
if os.path.exists(temp_chat):
print("🧹 Cleaning up temporary chat render video...")
os.remove(temp_chat)
if os.path.exists(mask_path):
os.remove(mask_path)
if os.path.exists(temp_with_chat):
os.remove(temp_with_chat)
return True
else:
print("❌ Error: FFmpeg failed to merge the video streams.")
if os.path.exists(video_chat):
os.remove(video_chat)
if os.path.exists(temp_with_chat):
os.remove(temp_with_chat)
return False
if __name__ == "__main__":
combine_twitch_vod_and_chat("download/videos/2813112936/2813112936.mp4", "overlay", True)
Regular → Executable
+140 -107
View File
@@ -1,7 +1,8 @@
#!/usr/bin/env python3
import requests
import os
import subprocess
import linux
import time
from database import Database
@@ -14,21 +15,6 @@ 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(id: str):
"""Transcribes the Video File."""
video_file = f"save/{DB.table}/{id}/{id}.mp4"
@@ -61,34 +47,36 @@ def top_hashtags(id: str):
return tags
def download():
"""Find all undownload vods and download them."""
"""Find all undownload videos and download them."""
undownloaded = DB.get_undownloaded()
print(f"Download {DB.table}...")
if DB.table == "vods":
if DB.table == "videos":
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:
output = linux.run_command(f"TwitchDownloaderCLI videodownload --id {record_id} -o save/videos/{record_id}/{record_id}.mp4 --collision Overwrite")
output2 = linux.run_command(f"TwitchDownloaderCLI chatdownload --id {record_id} -o save/videos/{record_id}/{record_id}_chat.json -E --collision Overwrite")
time.sleep(1)
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"]}")
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")
output = linux.run_command(f"TwitchDownloaderCLI clipdownload --id {slug} -o save/clips/{slug}/{slug}.mp4 --collision Overwrite")
time.sleep(1)
if output["success"] is True:
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"Slug: {slug} | Download Process failed. {output['stdout']}. Error: {output['stderr']}")
print(f"Finished Downloading {DB.table}...")
@@ -109,11 +97,11 @@ def upload():
upload_queue = []
if DB.table == "vods":
if DB.table == "videos":
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"
file_path = f"save/videos/{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'])
@@ -161,10 +149,14 @@ def get_vod_ids_simplified(channel_name: str):
"Content-Type": "text/plain"
}
vods_query_string = """
query GetChannelVideos($login: String!, $limit: Int!) {
videos_query_string = """
query GetChannelVideos($login: String!, $limit: Int!, $after: Cursor) {
user(login: $login) {
videos(first: $limit, types: [ARCHIVE]) {
videos(first: $limit, types: [ARCHIVE], after: $after) {
pageInfo {
hasNextPage
endCursor
}
edges {
node {
id
@@ -181,10 +173,15 @@ def get_vod_ids_simplified(channel_name: str):
"""
clips_query_string = """
query GetChannelClips($login: String!, $limit: Int!) {
query GetChannelClips($login: String!, $limit: Int!, $after: Cursor) {
user(login: $login) {
clips(first: $limit, criteria: { period: ALL_TIME }) {
clips(first: $limit, criteria: { period: ALL_TIME }, after: $after) {
pageInfo {
hasNextPage
endCursor
}
edges {
cursor
node {
slug
title
@@ -202,93 +199,129 @@ def get_vod_ids_simplified(channel_name: str):
}
}
"""
video_ids = []
has_next_page = True
cursor = None
limit = 50
query_string = ""
operation_name = ""
if DB.table == "vods":
query_string = vods_query_string
limit = 50
if DB.table == "videos":
query_string = videos_query_string
limit = 100
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 []
while has_next_page:
time.sleep(1)
# 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,
"after": cursor
}
}]
try:
# Prepping ensures Python does not rewrite the Client-ID header case
req = requests.Request('POST', url, json=payload)
#req = requests.Request('POST', url, data=json.dumps(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.get('data', {}).get('user', {})
if not user_data:
print(f"Channel '{channel_name}' not found.")
return []
edges = user_data.get(DB.table, {}).get('edges', [])
# --- BREAK CONDITION 1: Stop if Twitch returns no more data items ---
if not edges or len(edges) == 0:
print("No more items returned by the server. Ending pagination loop.")
break
last_edge_cursor = None
print(f"--- Processing {DB.table} for {channel_name} ---")
for edge in edges:
last_edge_cursor = edge.get("cursor")
node = edge.get('node', {})
if not node:
continue
game_info = node.get('game')
game_name = game_info.get('displayName') if game_info else "Unknown/No Category"
# FIXED: Switched fields to use safe .get() metrics to completely prevent KeyErrors
node_id = node.get('id')
node_title = node.get('title', 'No Title')
if DB.table == "videos":
published_at = node.get('publishedAt')
print(f"ID: {node_id} | Date: {published_at} | Game: {game_name} | Title: {node_title}")
DB.insert_videos_record(node_id, published_at, node_title, game_name)
if node_id:
video_ids.append(node_id)
elif DB.table == "clips":
slug = node.get('slug')
created_at = node.get('createdAt')
view_count = node.get('viewCount', 0)
curator_info = node.get('curator')
clip_by = curator_info.get('login') if curator_info else "Unknown Creator"
print(f"Slug: {slug} | Date: {created_at} | Game: {game_name} | By: {clip_by} | Views: {view_count} | Title: {node_title}")
DB.insert_clips_record(slug, created_at, node_title, game_name, clip_by, int(view_count))
if slug:
video_ids.append(slug)
# --- CORRECTED PAGINATION ENGINE FOR BOTH TABLES ---
page_info = user_data.get(DB.table, {}).get('pageInfo', {})
has_next_page = page_info.get("hasNextPage", False)
next_cursor = page_info.get("endCursor") or last_edge_cursor
if not next_cursor or next_cursor == cursor:
print("Cursor did not advance or is null. Safely terminating loop.")
break
cursor = next_cursor
except Exception as e:
print(f"An unexpected error occurred: {e}")
if 'response' in locals():
print(f"Server Response Text: {response.text}")
return []
return video_ids
if __name__ == "__main__":
tables = ["vods", "clips"]
tables = ["clips"]
for table in tables:
DB = Database(table)
get_vod_ids_simplified(CHANNEL_NAME)
download()
upload()
DB.close_database()
#download()
#upload()
#DB.close_database()
Regular → Executable
+3 -4
View File
@@ -1,6 +1,5 @@
#!/usr/bin/env python3
import requests
import sqlite3
import subprocess
from datetime import datetime
@@ -67,11 +66,11 @@ def download_clips():
# Uses standard clipdownload directive
output = run_linux_command(f"TwitchDownloaderCLI clipdownload --id {slug} -o save/clips/{slug}/{slug}.mp4")
if output["success"] is True:
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"Slug: {slug} | Download Process failed. {output['stdout']}. Error: {output['stderr']}")
print("Finished Downloading Clips...")
@@ -95,7 +94,7 @@ def upload_clips():
tags.extend(top_hashtags({slug}))
#output = uploader.upload_video(file_path, title, categoryId, description, privatcyStatus, tags)
output = uploader.upload_video(file_path, title, categoryId, description, privatcyStatus, tags)
if output is True:
print(f"Slug: {slug} | Was successfully uploaded.")
+162
View File
@@ -0,0 +1,162 @@
import os
import json
import asyncio
import requests
from twitchAPI.twitch import Twitch
from twitchAPI.helper import first
import database
# 1. Fill in your credentials from the Twitch Developer Console
SECRETS = None
TWITCH = None
USER = None
CHANNEL_NAME = "teampgp"
async def get_twitch():
global SECRETS, TWITCH, USER
if SECRETS is None:
with open('twitch_secrets.json', 'r') as f:
SECRETS = json.load(f)
if TWITCH is None:
TWITCH = await Twitch(SECRETS['client_id'], SECRETS['client_secret'])
if USER is None:
USER = await first(TWITCH.get_users(logins=[CHANNEL_NAME]))
if not USER:
print("User not found.")
async def download_live_thumbnail(twitch_client, streamer_username: str, w: int = 1920, h: int = 1080):
"""Fetches and saves the live stream thumbnail for an active broadcast."""
print(f"🔎 Checking live status for: {streamer_username}...")
# Query the live streams endpoint
stream_generator = twitch_client.get_streams(user_logins=[streamer_username])
stream_data = await first(stream_generator)
if not stream_data:
print(f"❌ User '{streamer_username}' is offline. Live thumbnails require an active stream.")
return
# Twitch API live streams use the {width} and {height} format
raw_url = stream_data.thumbnail_url
clean_url = raw_url.replace('{width}', str(w)).replace('{height}', str(h))
filename = f"live_{streamer_username}_{w}x{h}.jpg"
save_image(clean_url, filename)
async def download_vod_thumbnail(twitch_client, vod_id: str, w: int = 1920, h: int = 1080):
"""Fetches and saves a thumbnail from a past broadcast VOD ID."""
print(f"🔎 Searching for VOD ID: {vod_id}...")
# Query the videos endpoint
video_generator = twitch_client.get_videos(vod_id)
video_data = await first(video_generator)
filename = f"download/videos/{vod_id}/{vod_id}_{w}x{h}.jpg"
if os.path.exists(filename):
return
if not video_data:
print(f"❌ VOD ID {vod_id} could not be found.")
return
# Twitch VOD endpoints typically format string tokens as %{width} and %{height}
raw_url = video_data.thumbnail_url
if not raw_url:
print("❌ This VOD does not have an available thumbnail.")
return
clean_url = raw_url.replace('%{width}', str(w)).replace('%{height}', str(h))
#filename = f"vod_{vod_id}_{w}x{h}.jpg"
save_image(clean_url, filename)
async def download_clip_thumbnail(clip_id: str, url: str):
print(f"🔎 Searching for Clip ID: {clip_id}...")
filename = f"download/clips/{clip_id}/{clip_id}.jpg"
if os.path.exists(filename):
return
save_image(url, filename, False)
def save_image(url: str, filename: str, stream: bool = True):
"""Helper function to stream image bytes directly to a file."""
try:
response = requests.get(url, stream)
if response.status_code == 200:
with open(filename, 'wb') as file:
if stream is True:
for chunk in response.iter_content(1024):
file.write(chunk)
else:
file.write(response.content)
print(f"✅ Success! Saved as: {filename}")
else:
print(f"❌ Download failed. HTTP Status: {response.status_code}")
except Exception as e:
print(f"❌ An error occurred during file writing: {e}")
async def main():
# Initialize connection & automatically authorize the App token
await get_twitch()
twitch = TWITCH # Twitch(APP_ID, APP_SECRET)
# --- OPTION A: Download Live Thumbnail ---
# Target user must be streaming live right now
target_streamer = CHANNEL_NAME
#await download_live_thumbnail(twitch, target_streamer, 1920, 1080)
# --- OPTION B: Download Past Broadcast VOD Thumbnail ---
# Extract the ID sequence from your target video link
# target_vod = "2145678901"
db = database.Database()
vods = db.get_vods()
for vod in vods:
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = vod
await download_vod_thumbnail(twitch, id, 1920, 1080)
clips = db.get_clips()
for clip in clips:
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = clip
await download_clip_thumbnail(id, thumbnail_url)
if __name__ == '__main__':
asyncio.run(main())
+108
View File
@@ -0,0 +1,108 @@
#!/usr/bin/env python3
import csv
import subprocess
from datetime import datetime, timedelta, timezone
import linux
from database import Database
CHANNEL_NAME = 'teampgp'
DB = None
def write_csv(data, file_name):
# Open file with newline='' to prevent extra blank rows across platforms
with open(file_name, "w", newline="", encoding="utf-8") as file:
writer = csv.writer(file)
# Write all rows at once
writer.writerows(data)
def download():
"""Find all undownloaded videos and clips, and download them safely."""
undownloaded = DB.get_undownloaded()
print("Starting downloads...")
for row in undownloaded:
# Unpack variables clearly
(
id,
title,
created_at,
view_count,
duration,
url,
thumbnail_url,
game_id,
game_name,
stream_id,
creator_name,
clip_is,
downloaded,
uploaded_yt,
uploaded_yt_chats,
uploaded_yt_shorts,
) = row
# 1. Define your data's timestamp (Example: April 10, 2026, at 10:00 AM)
#datetime.strptime(created_at, "%Y-%m-%dT%H:%M:%SZ")
data_timestamp = datetime.fromisoformat(created_at)
# 2. Get the exact current date and time
current_time = datetime.now(timezone.utc)
# 3. Calculate the difference between the two times
time_difference = current_time - data_timestamp
# 4. Check if the difference is greater than 24 hours
if time_difference > timedelta(hours=24):
print("The data is more than 24 hours old.")
else:
#lets wait 24 hours befor downloading
print("The data is less than 24 hours old.")
continue
data = [list(row)]
print(
f"ID: {id} | Date: {created_at} | Game: {game_name} | Title: {title}"
)
# 1. Define paths and isolate base directory
if clip_is:
target_dir = f"download/clips/{id}"
cmd = f"TwitchDownloaderCLI clipdownload --id {id} -o {target_dir}/{id}.mp4 --collision Overwrite --temp-path download/temp"
else:
target_dir = f"download/videos/{id}"
# FIXED: Removed the duplicated command string combined with '&&'
cmd = f"TwitchDownloaderCLI videodownload --id {id} -o {target_dir}/{id}.mp4 --collision Overwrite --threads 2 --temp-path download/temp"
csv_file = f"{target_dir}/{id}.csv"
# 3. FIXED: Use the live streaming function to prevent Out-Of-Memory crashes
output = linux.run_command(cmd, look_for=["[STATUS]"])
output2 = linux.run_command(f"TwitchDownloaderCLI chatdownload --id {id} -o {target_dir}/{id}_chat.json -E --collision Overwrite --temp-path download/temp", look_for=["[STATUS]"])
if output:
# Only write CSV and update database if download actually completed
write_csv(data, csv_file)
print(f"✅ Success: TwitchDownloaderCLI video. {target_dir}")
DB.mark_as_downloaded(id)
else:
print(f"ID: {id} | Download Process failed.")
#print(f"❌ Reason/Error: {output.get('error', 'Unknown Error')}")
#if output.get("stderr"):
# print(f"Details: {output['stderr']}")
# Clear temporary chunk clutter immediately if a VOD crashes out
if not clip_is:
print("Flushing temporary crash chunks...")
subprocess.run("rm -rf download/temp/*", shell=True)
print("Finished Downloading Pipeline.")
def main():
global DB
DB = Database()
download()
if __name__ == "__main__":
main()
Regular → Executable
+87 -271
View File
@@ -1,286 +1,102 @@
#!/usr/bin/env python3
import json
import os
import requests
import sqlite3
import subprocess
# The target Twitch streamer username
TWITCH_USERNAME = "SumGuyV5"
from database import Database
def load_secrets(filepath="twitch_secrets.json"):
"""Loads client credentials and potential manual token from JSON file."""
if not os.path.exists(filepath):
raise FileNotFoundError(f"Missing credential file: '{filepath}'")
with open(filepath, "r") as file:
secrets = json.load(file)
if "client_id" not in secrets or "client_secret" not in secrets:
raise KeyError("JSON file must contain 'client_id' and 'client_secret'.")
return secrets["client_id"], secrets["client_secret"], secrets.get("manual_token")
import uploader
from uploader import CategoryId
def get_app_access_token(client_id, client_secret):
"""Generates an App Access Token using the correct Twitch ID server."""
auth_url = "https://twitch.tv" # FIXED: Correct auth endpoint
payload = {
"client_id": client_id,
"client_secret": client_secret,
"grant_type": "client_credentials"
}
headers = {"Content-Type": "application/x-www-form-urlencoded"}
response = requests.post(auth_url, data=payload, headers=headers)
response.raise_for_status()
return response.json()["access_token"]
CHANNEL_NAME = 'teampgp' # Replace with the streamer's username
DB = Database("vods")
def get_user_id(username, headers):
"""Retrieves the unique numerical Twitch User ID from Helix."""
url = f"https://twitch.tv{username}" # FIXED: Endpoint & parameter
response = requests.get(url, headers=headers)
response.raise_for_status()
data = response.json().get("data")
if data and len(data) > 0:
return data[0]["id"] # FIXED: Helix data array returns user dictionaries
else:
raise ValueError(f"Twitch user '{username}' not found.")
def run_linux_command(command: str):
"""Executes a Linux command, waits for completion, and returns output."""
def get_channel_vods(user_id, headers, limit=10):
"""Fetches past broadcasts (VODs) using valid Helix syntax."""
# FIXED: Restructured URL to use correct endpoint and standard query parameters
url = f"https://twitch.tv{user_id}&type=archive&first={limit}"
response = requests.get(url, headers=headers)
response.raise_for_status()
return response.json().get("data", [])
def main():
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/clips/{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/clips/{slug}/transcribe_{slug}.srt"):
print(f"video already transcribed:")
return True
import transcribe_video
transcribe_video.extract_audio(video_file, f"save/clips/{slug}/temp_{slug}_audio.wav")
transcribe_video.transcribe_to_srt(f"save/clips/{slug}/temp_{slug}_audio.wav", f"save/clips/{slug}/", f"transcribe_{slug}")
return True
def top_hashtags(slug: str):
file_srt = f"save/clips/{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 []
import youtube_hashtags
return youtube_hashtags.get_top_hashtags(file_srt)
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}/{title}.mp4")
output2 = run_linux_command(f"TwitchDownloaderCLI chatdownload --id {record_id} -o save/vods/{record_id}/{title}_chat.json -E")
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}/{title}.mp4")
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', 'twitch']
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}/{title}.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 in unuploaded:
print(f"Slug: {slug} | Date: {record_date} | Game: {game_name} | By: {clip_by} | Views: {views} | Title: {title}")
file_path = f"save/clips/{slug}/{title}.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)
# 1. Load credentials from external JSON file
client_id, client_secret, manual_token = load_secrets("twitch_secrets.json")
# 2. Assign or generate OAuth Access Token
if manual_token:
print("Using manual access token from JSON config file...")
access_token = manual_token
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: $limit, types: [ARCHIVE]) {
edges {
node {
id
title
publishedAt
game {
displayName
}
}
}
}
print("No manual token found. Attempting to contact Twitch Auth Server...")
access_token = get_app_access_token(client_id, client_secret)
# 3. Setup Headers required by Twitch Helix API
headers = {
"Client-ID": client_id,
"Authorization": f"Bearer {access_token}"
}
}
"""
# 4. Translate Username to User ID
user_id = get_user_id(TWITCH_USERNAME, headers)
print(f"Successfully retrieved ID for {TWITCH_USERNAME}: {user_id}\n")
# 5. Fetch and Print VOD details
vods = get_channel_vods(user_id, headers, limit=5)
if not vods:
print(f"No VODs found for {TWITCH_USERNAME}.")
return
clips_query_string = """
query GetChannelClips($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 = ""
if DB.table == "vods":
query_string = vods_query_string
limit = 50
elif DB.table == "clips":
query_string = clips_query_string
limit = 40
# A simplified query string that hardcodes the ARCHIVE type filter into the request structure
payload = [{
"operationName": "GetChannelVideos",
"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 []
print(f"--- Latest VODs for {TWITCH_USERNAME} ---")
for vod in vods:
print(f"Title: {vod['title']}")
print(f"URL: {vod['url']}")
print(f"Published At: {vod['published_at']}")
print(f"Duration: {vod['duration']}")
print(f"Views: {vod['view_count']}")
print("-" * 40)
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)
#insert_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 (FileNotFoundError, KeyError) as config_err:
print(f"Configuration Error: {config_err}")
except requests.exceptions.HTTPError as err:
print(f"HTTP Error detail: {err.response.text if err.response else err}")
except Exception as e:
print(f"An unexpected error occurred: {e}")
if 'response' in locals():
print(f"Server Response Text: {response.text}")
return []
print(f"An error occurred: {e}")
if __name__ == "__main__":
get_vod_ids_simplified(CHANNEL_NAME)
download_vods()
upload_vods()
DB.close_database()
main()
+185
View File
@@ -0,0 +1,185 @@
#!/usr/bin/env python3
import asyncio
import json
import httpx # Switched from requests to prevent async loop freezing
import database
from twitchAPI.twitch import Twitch
from twitchAPI.helper import first
from twitchAPI.type import VideoType
SECRETS = None
TWITCH = None
USER = None
GAME_CACHE = {} # Local cache dictionary to store game_id -> game_name mapping
CHANNEL_NAME = "teampgp"
async def get_twitch():
global SECRETS, TWITCH, USER
if SECRETS is None:
with open('twitch_secrets.json', 'r') as f:
SECRETS = json.load(f)
if TWITCH is None:
TWITCH = await Twitch(SECRETS['client_id'], SECRETS['client_secret'])
if USER is None:
USER = await first(TWITCH.get_users(logins=[CHANNEL_NAME]))
if not USER:
print("User not found.")
async def get_game_name_by_id(game_id: str) -> str:
"""Helper function to fetch game names and cache them locally."""
if not game_id:
return "Unknown / No Category"
if game_id in GAME_CACHE:
return GAME_CACHE[game_id]
try:
game_generator = TWITCH.get_games(game_ids=[game_id])
game = await first(game_generator)
if game:
GAME_CACHE[game_id] = game.name
return game.name
except Exception:
pass
return "Unknown Game"
async def get_vod_game_name(vod_id: str):
"""Asynchronously query Twitch GQL endpoint for VOD game metadata."""
game_id = "0"
game_name = "Unknown Game"
url = "https://gql.twitch.tv/gql"
headers = {
"Client-ID": "kimne78kx3ncx6brgo4mv6wki5h1ko",
"Content-Type": "application/json"
}
payload = [{
"operationName": "VideoMetadata",
"variables": {
"channelLogin": "",
"videoID": str(vod_id)
},
"extensions": {
"persistedQuery": {
"version": 1,
"sha256Hash": "45111672eea2e507f8ba44d101a61862f9c56b11dee09a15634cb75cb9b9084d"
}
}
}]
# Using httpx async client to prevent blocking the asyncio event loop
async with httpx.AsyncClient() as client:
try:
response = await client.post(url, headers=headers, json=payload)
if response.status_code == 200:
data = response.json()
video_info = data[0].get('data', {}).get('video')
if video_info and video_info.get('game'):
game_id = str(video_info['game']['id'])
game_name = video_info['game']['displayName']
print(f"GQL Found: {game_name} (ID: {game_id})")
else:
print(f"No game information found in GQL for VOD {vod_id}.")
except Exception as e:
print(f"Error fetching GQL metadata for VOD {vod_id}: {e}")
return game_id, game_name
async def get_streamer_vods():
await get_twitch()
print(f"Starting VOD extraction for {USER.display_name}...")
vod_generator = TWITCH.get_videos(user_id=USER.id, first=100, video_type=VideoType.ALL)
all_vods = []
async for v in vod_generator:
# Resolving game category safely without freezing the event loop
game_id, game_name = await get_vod_game_name(v.id)
# Safely convert game_id to integer if possible, otherwise default to 0
try:
clean_game_id = game_id
except ValueError:
clean_game_id = 0
vod_data = {
"id": v.id,
"title": v.title,
"created_at": str(v.published_at),
"view_count": int(v.view_count) if v.view_count else 0,
"duration": v.duration,
"url": v.url,
"thumbnail_url": v.thumbnail_url,
"game_id": clean_game_id,
"game_name": game_name,
"stream_id": str(v.stream_id) if v.stream_id else "0",
"creator_name": CHANNEL_NAME,
"clip_is": False,
}
all_vods.append(vod_data)
print(f"Collected VOD: {v.title} | Category: {game_name} ({v.duration})")
print(f"\nFinished extracting VODs. Total gathered: {len(all_vods)}")
return all_vods
async def get_streamer_clips():
await get_twitch()
print(f"Starting clip extraction for {USER.display_name}...")
clip_generator = TWITCH.get_clips(broadcaster_id=USER.id, first=100)
all_clips = []
async for c in clip_generator:
game_name = await get_game_name_by_id(c.game_id)
clip_data = {
"id": c.id,
"title": c.title,
"created_at": str(c.created_at),
"view_count": int(c.view_count) if c.view_count else 0,
"duration": c.duration,
"url": c.url,
"thumbnail_url": c.thumbnail_url,
"game_id": c.game_id,
"game_name": game_name,
"stream_id": "0",
"creator_name": c.creator_name,
"clip_is": True,
}
all_clips.append(clip_data)
print(f"Collected clip: {c.title} | Category: {game_name} ({c.view_count} views)")
print(f"\nFinished extracting clips. Total gathered: {len(all_clips)}")
return all_clips
async def main():
print("--- Script Started ---")
db = database.Database()
vods_list = await get_streamer_vods()
for v in vods_list:
db.insert_video_record(
v['id'], v['title'], v['created_at'], v['view_count'], v['duration'],
v['url'], v['thumbnail_url'], v['game_id'], v['game_name'],
v['stream_id'], v['creator_name'], v['clip_is']
)
if vods_list:
print(f"Recent VOD: '{vods_list[0]['title']}'")
clips_list = await get_streamer_clips()
for v in clips_list:
db.insert_video_record(
v['id'], v['title'], v['created_at'], v['view_count'], v['duration'],
v['url'], v['thumbnail_url'], v['game_id'], v['game_name'],
v['stream_id'], v['creator_name'], v['clip_is']
)
if clips_list:
print(f"Recent VOD: '{clips_list[0]['title']}'")
if __name__ == "__main__":
asyncio.run(main())
Regular → Executable
+1 -1
View File
@@ -78,7 +78,7 @@ def upload_video(file_path: str, title: str, category: CategoryId, description:
try:
# Check if file exists
if not os.path.exists(file_path):
print(f"Error: File not found: {file_path}")
print(f"Error: File not found: {file_path}")
return False
# Load credentials
Regular → Executable
+1 -1
View File
@@ -183,7 +183,7 @@ def get_clip_slugs(channel_name):
return clip_slugs
except Exception as e:
print(f"An unexpected error occurred: {e}")
print(f"An unexpected error occurred: {e}")
return []
if __name__ == "__main__":
Regular → Executable
View File
Regular → Executable
+48 -22
View File
@@ -1,15 +1,18 @@
#!/usr/bin/env python3
import re
import json
import nltk
import string
from collections import Counter
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
import nltk
# Download necessary NLTK data
# Download necessary NLTK data modules
nltk.download('punkt', quiet=True)
nltk.download('stopwords', quiet=True)
nltk.download('punkt_tab', quiet=True)
nltk.download('averaged_perceptron_tagger', quiet=True) # Required for POS tagging
nltk.download('averaged_perceptron_tagger_eng', quiet=True)
def extract_text_from_srt(file_path: str):
with open(file_path, 'r', encoding='utf-8') as file:
@@ -19,32 +22,55 @@ def extract_text_from_srt(file_path: str):
clean_text = re.sub(r'\d+', '', clean_text)
return clean_text
def get_top_hashtags(srt_file_path: str, top_n: int = 10):
raw_text = extract_text_from_srt(srt_file_path)
def extract_text_from_json(file_path: str):
with open(file_path, "r", encoding="utf-8") as file:
data = json.load(file)
messages = []
for comment in data.get("comments", []):
message_text = comment.get("message", {}).get("body", "")
messages.append(message_text)
# Lowercase and remove punctuation
raw_text = raw_text.lower()
raw_text = raw_text.translate(str.maketrans('', '', string.punctuation))
# Return a single merged string of all chat text
return " ".join(messages)
def get_top_nouns(file_path: str, top_n: int = 10):
# 1. Extract raw text
if file_path.endswith(".srt"):
raw_text = extract_text_from_srt(file_path)
else:
raw_text = extract_text_from_json(file_path)
# 2. Basic cleanup (Keep original case for proper noun accuracy)
# Strip basic punctuation but leave words intact
clean_text = raw_text.translate(str.maketrans('', '', string.punctuation))
# Tokenize and remove stopwords
words = word_tokenize(raw_text)
# 3. Tokenize words
words = word_tokenize(clean_text)
# 4. Part-of-Speech Tagging
tagged_words = nltk.pos_tag(words)
# 5. Filter for Nouns (NN = Singular Noun, NNP = Proper Noun, NNS = Plural Noun)
stop_words = set(stopwords.words('english'))
nouns = []
# Filter for alphabetical words longer than 3 characters that aren't stop words
filtered_words = [
word for word in words
if word.isalpha() and word not in stop_words and len(word) > 3
]
for word, tag in tagged_words:
word_lower = word.lower()
# Filter out short fragments and standard stopwords
if tag in ['NN', 'NNP', 'NNS'] and len(word_lower) > 2 and word_lower not in stop_words:
nouns.append(word_lower)
# 6. Count frequencies
noun_counts = Counter(nouns)
return [noun for noun, count in noun_counts.most_common(top_n)]
# Get frequency and create hashtags
word_counts = Counter(filtered_words)
top_words = word_counts.most_common(top_n)
hashtags = [f"{word[0]}" for word in top_words]
return hashtags
if __name__ == "__main__":
# Example usage
# Replace 'your_video.srt' with the path to your file
results = get_top_hashtags('your_video.srt', top_n=10)
print("Trending Hashtags:", results)
results = get_top_nouns('download/videos/2813112936/2813112936.srt', top_n=10)
print("Trending Hashtags SRT:", results)
results = get_top_nouns("download/videos/2813112936/2813112936_chat.json", top_n=10)
print("Trending Hashtags JSON:", results)
Regular → Executable
+121 -135
View File
@@ -1,7 +1,7 @@
#!/usr/bin/env python3
import argparse
import sys
import subprocess
import os
import json
import linux
import tempfile
from pathlib import Path
from moviepy import VideoFileClip, ColorClip, CompositeVideoClip
@@ -11,147 +11,133 @@ from pathlib import Path
import numpy as np
from PIL import Image, ImageFilter
def apply_gaussian_blur(frame, radius: int = 30):
"""
Transforms a single NumPy array frame using PIL's true GaussianBlur filter.
"""
# Convert numpy array to PIL Image
image = Image.fromarray(frame)
# Apply high-quality true Gaussian Blur
blurred_image = image.filter(ImageFilter.GaussianBlur(radius=radius))
# Return back as a numpy array for MoviePy
return np.array(blurred_image)
def get_video_info(input_path: Path) -> tuple:
"""Uses ffprobe to instantly read input video dimensions and frame rate."""
cmd = f"ffprobe -v error -select_streams v:0 -show_entries stream=width,height,r_frame_rate -of json {input_path}"
# Run command and capture output (assumes linux.run_command prints or you use subprocess)
import subprocess
result = subprocess.run(cmd, shell=True, capture_output=True, text=True)
#linux.run_command(cmd)
try:
data = json.loads(result.stdout)
stream = data['streams'][0]
w = int(stream['width'])
h = int(stream['height'])
# Convert fractional FPS string (e.g. "60/1" or "30000/1001") to float
fps_parts = stream['r_frame_rate'].split('/')
fps = float(fps_parts[0]) / float(fps_parts[1]) if len(fps_parts) > 1 else float(fps_parts[0])
return w, h, fps
except Exception:
return 1920, 1080, 60.0 # Safe defaults if probe fails
def fit_to_9_16_letterbox(input_path: str, top_text: str = "TOP TEXT", bottom_text: str = "BOTTOM TEXT", use_blur: bool = True):
def fit_to_9_16_letterbox(input_path: str, top_text: str = "TOP TEXT", bottom_text: str = "BOTTOM TEXT", use_blur: bool = True, force: bool = False):
threads = "8"
input_file = Path(input_path)
output_suffix = "gaussian_9_16" if use_blur else "black_9_16"
output_path = input_file.parent / f"{input_file.stem}_{output_suffix}{input_file.suffix}"
print("🧼 Sanitizing video metadata streams inside an automated safe context...")
with tempfile.NamedTemporaryFile(suffix=input_file.suffix, delete=False) as temp_file:
temp_path = temp_file.name
if os.path.exists(output_path):
if not force:
print(f"❌ Error: Short video already exists: {output_path}")
return
else:
os.remove(output_path)
bg_scaled = None
bg_cropped = None
background_layer = None
# 1. Probe input metadata instantly
orig_w, orig_h, fps = get_video_info(input_file)
canvas_w = 1080
canvas_h = 1920
print("✍️ Generating text overlay graphics via MoviePy...")
# Render static images for text instead of running a video context
title_clip = TextClip(
text=top_text, font_size=55, color="white", font="DejaVuSans-Bold",
text_align="center", size=(canvas_w - 100, 300), method="caption"
)
bottom_clip = TextClip(
text=bottom_text, font_size=55, color="white", font="DejaVuSans-Bold",
text_align="center", size=(canvas_w - 100, 300), method="caption"
)
# Save text layers to temporary PNGs
top_png = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
bottom_png = tempfile.NamedTemporaryFile(suffix=".png", delete=False).name
title_clip.save_frame(top_png)
bottom_clip.save_frame(bottom_png)
title_clip.close()
bottom_clip.close()
print("🎬 Dispatching compilation workload to FFmpeg filtergraph...")
# 2. Build the complex FFmpeg filtergraph
# [0:v] is the raw input video stream
filter_complex = []
if use_blur:
# Scale height to 1920, crop center 1080x1920, apply fast boxblur (power of 3 approximates Gaussian)
filter_complex.append(
f"[0:v]scale=-1:{canvas_h},crop={canvas_w}:{canvas_h}:(iw-{canvas_w})/2:0,boxblur=luma_radius=35:luma_power=3[bg];"
)
else:
# Generate a pure black background canvas matching video frame specs
filter_complex.append(
f"color=c=black:s={canvas_w}x{canvas_h}:r={fps}[bg];"
)
# Scale the foreground video to a clean 1080 width, keeping aspect ratio
filter_complex.append(
f"[0:v]scale={canvas_w}:-1[fg];"
)
# Layer composition chain:
# Overlay 1: Put scaled foreground onto background (centered vertically)
filter_complex.append(
f"[bg][fg]overlay=0:(H-h)/2[tmp1];"
)
# Overlay 2: Drop top text asset onto position Y=180
filter_complex.append(
f"[tmp1][1:v]overlay=(W-w)/2:180[tmp2];"
)
# Overlay 3: Drop bottom text asset onto position Y=1430
filter_complex.append(
f"[tmp2][2:v]overlay=(W-w)/2:1430[finalv]"
)
filter_graph = "".join(filter_complex)
# 3. Execute the native assembly command
# -map_chapters -1 -sn: Strips unnecessary metadata chunks instantly
# -c:a copy: Safely pulls original digital audio directly without decompression cycles
# -threads 0: Forces FFmpeg to auto-consume all available processing cores
ffmpeg_cmd = (
f'ffmpeg -y -v error -i "{input_file}" -i "{top_png}" -i "{bottom_png}" '
f'-filter_complex "{filter_graph}" '
f'-map "[finalv]" -map 0:a? -c:v libx264 -crf 18 -preset slow -pix_fmt yuv420p '
f'-c:a copy -map_chapters -1 -sn -threads {threads} "{output_path}"'
)
try:
cleanup_cmd = [
"ffmpeg", "-y", "-i", str(input_file),
"-map_chapters", "-1", "-sn",
"-c", "copy", temp_path
]
subprocess.run(cleanup_cmd, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
# Load video
clip = VideoFileClip(temp_path)
# Standard vertical 9:16 canvas sizes
canvas_w = 1080
canvas_h = 1920
# Background logic
if use_blur:
print("📐 Scaling, cropping, and blurring background layer...")
bg_scaled = clip.resized(height=canvas_h)
bg_cropped = bg_scaled.cropped(width=canvas_w, x_center=bg_scaled.w / 2)
background_layer = bg_cropped.transform(lambda gf, t: apply_gaussian_blur(gf(t), radius=35))
else:
print("⚫ Creating solid black background canvas...")
background_layer = ColorClip(size=(canvas_w, canvas_h), color=(0, 0, 0), duration=clip.duration)
print("📐 Shrinking foreground video width to fit the 1080 wide canvas...")
foreground_clip = clip.resized(width=canvas_w)
print("✍️ Creating multi-line top title text clip...")
# FIXED: Added 'method="caption"' and increased vertical size to 300
title_clip = TextClip(
text=top_text,
font_size=55, # Slightly smaller to accommodate paragraphs comfortably
color="white",
font="DejaVuSans-Bold",
text_align="center",
size=(canvas_w - 100, 300), # Subtracted 100px for safety margins on left/right edges
method="caption", # Forces text to wrap cleanly onto a new line
duration=clip.duration
)
# Position adjusted to center the taller 300px box in the upper section
positioned_top_text = title_clip.with_position(("center", 180))
print("✍️ Creating multi-line bottom text clip...")
# FIXED: Added 'method="caption"' and increased vertical size to 300
bottom_clip = TextClip(
text=bottom_text,
font_size=55,
color="white",
font="DejaVuSans-Bold",
text_align="center",
size=(canvas_w - 100, 300), # Left/right margins included
method="caption", # Forces text to wrap cleanly onto a new line
duration=clip.duration
)
# Position adjusted to center the taller 300px box in the lower section
positioned_bottom_text = bottom_clip.with_position(("center", 1430))
# Composite layers
final_clip = CompositeVideoClip(
[
background_layer,
foreground_clip.with_position("center"),
positioned_top_text,
positioned_bottom_text
]
).with_audio(clip.audio)
print("🎬 Rendering final vertical composition...")
final_clip.write_videofile(
str(output_path),
codec="libx264",
audio_codec="aac",
fps=clip.fps
)
# Clean up file locks safely
clip.close()
if bg_scaled: bg_scaled.close()
if bg_cropped: bg_cropped.close()
background_layer.close()
foreground_clip.close()
title_clip.close()
bottom_clip.close()
final_clip.close()
linux.run_command(ffmpeg_cmd)
print(f"🎉 High-speed processing complete! Video saved to: {output_path}")
finally:
temp_file_path = Path(temp_path)
if temp_file_path.exists():
temp_file_path.unlink()
print(f"🎉 Text overlay video saved to: {output_path}")
# Clean up temporary PNG picture files safely
for path in (top_png, bottom_png):
if os.path.exists(path):
os.unlink(path)
def main():
# Set up the command-line argument parser
parser = argparse.ArgumentParser(description="TeamPGP Clip Processing and Upload Pipeline")
# Add optional arguments
parser.add_argument('--convert', type=str, metavar='CLIP_PATH', help='Path to a video file to convert to a 9:16 Short')
parser.add_argument('--upload', action='store_true', help='Process and upload pending Shorts in the database to YouTube')
args = parser.parse_args()
# If no flags are provided, show help text and exit
if not args.convert and not args.upload:
parser.print_help()
sys.exit("\n❌ Error: You must provide at least one action flag (--convert or --upload).")
# Execute conversion step if path is provided
if args.convert:
clip_by = "joelmckinney"
#convert_to_short(args.convert)
fit_to_9_16_letterbox(args.convert, "TeamPGP Live on Twitch...\n Every Friday and Sunday @7:30 ET.\n twitch.tv/teampgp", f"Clipped By: {clip_by}.", True)
# Execute database upload step if flag is provided
if args.upload:
upload_shorts()
if __name__ == "__main__":
main()
import time
start_time = time.perf_counter()
creator_name = "greenskiesbluegrass"
top_txt = "TeamPGP Live on Twitch...\n Every Friday and Sunday @7:30 ET.\n twitch.tv/teampgp"
bottom_txt = f"Clipped By: {creator_name}."
fit_to_9_16_letterbox("download/clips/AbnegateAgitatedGrassPJSalt/AbnegateAgitatedGrassPJSalt.mp4", top_txt, bottom_txt, True, True)
end_time = time.perf_counter()
execution_time = end_time - start_time
print(f"The function took {execution_time:.6f} seconds to complete.")