yt-dlp-dags/ytops_client/task_generator_tool.py

204 lines
8.9 KiB
Python

#!/usr/bin/env python3
"""
CLI tool to generate granular download task files from a directory of info.json files.
"""
import argparse
import json
import logging
import os
import re
import sys
from pathlib import Path
# Configure logging
logging.basicConfig(
level=logging.INFO,
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
)
logger = logging.getLogger(__name__)
def sanitize_format_for_filename(format_str: str) -> str:
"""Sanitizes a format selector string to be filesystem-friendly."""
# Replace common problematic characters with underscores
sanitized = re.sub(r'[\\/+:\[\]\s]', '_', format_str)
# Remove any trailing characters that might be problematic
sanitized = sanitized.strip('._-')
return sanitized
def add_task_generator_parser(subparsers):
"""Adds the parser for the 'task-generator' command."""
parser = subparsers.add_parser(
'task-generator',
description="Generate granular download task files from info.jsons.",
formatter_class=argparse.RawTextHelpFormatter,
help="Generate granular download task files."
)
# All functionality is under subcommands for extensibility.
generate_subparsers = parser.add_subparsers(dest='task_generator_command', help='Action to perform', required=True)
gen_parser = generate_subparsers.add_parser(
'generate',
help='Generate task files from a source directory.',
description='Reads info.json files from a source directory and creates one task file per format in an output directory.'
)
gen_parser.add_argument('--source-dir', required=True, help='Directory containing the source info.json files.')
gen_parser.add_argument('--output-dir', required=True, help='Directory where the generated task files will be saved.')
gen_parser.add_argument('--formats', required=True, help='A comma-separated list of format IDs or selectors to generate tasks for (e.g., "18,140,bestvideo").')
gen_parser.add_argument('--verbose', action='store_true', help='Enable verbose logging.')
reset_parser = generate_subparsers.add_parser(
'reset',
help='Reset processed source files.',
description='Finds all *.processed files in the source directory and renames them back to *.json to allow re-generation.'
)
reset_parser.add_argument('--source-dir', required=True, help='Directory containing the source info.json files to reset.')
reset_parser.add_argument('--verbose', action='store_true', help='Enable verbose logging.')
def _main_task_generator_reset(args):
"""Main logic for the 'reset' command."""
source_dir = Path(args.source_dir)
if not source_dir.is_dir():
logger.error(f"Source directory does not exist or is not a directory: {source_dir}")
return 1
logger.info(f"Scanning for *.processed and *.LOCKED.* files in '{source_dir}' (recursively) to reset...")
# Use rglob for recursive search
processed_files = list(source_dir.rglob('*.json.processed'))
locked_files = list(source_dir.rglob('*.json.LOCKED.*'))
files_to_reset = processed_files + locked_files
if not files_to_reset:
logger.info("No processed or locked files found to reset.")
return 0
reset_count = 0
for file_to_reset in files_to_reset:
original_path = None
if file_to_reset.name.endswith('.processed'):
# Handles cases like file.json.processed
original_path_str = str(file_to_reset).removesuffix('.processed')
original_path = Path(original_path_str)
elif '.LOCKED.' in file_to_reset.name:
# Handles cases like file.json.LOCKED.0
original_path_str = str(file_to_reset).split('.LOCKED.')[0]
original_path = Path(original_path_str)
if original_path:
try:
if original_path.exists():
logger.warning(f"Original file '{original_path.name}' already exists. Deleting '{file_to_reset.name}' instead of renaming.")
file_to_reset.unlink()
else:
file_to_reset.rename(original_path)
logger.debug(f"Reset '{file_to_reset.name}' to '{original_path.name}'")
reset_count += 1
except (IOError, OSError) as e:
logger.error(f"Failed to reset '{file_to_reset.name}': {e}")
else:
logger.warning(f"Could not determine original filename for '{file_to_reset.name}'. Skipping.")
logger.info(f"Successfully reset {reset_count} file(s).")
return 0
def main_task_generator(args):
"""Main logic for the 'task-generator' tool."""
if args.task_generator_command == 'generate':
return _main_task_generator_generate(args)
elif args.task_generator_command == 'reset':
return _main_task_generator_reset(args)
return 1
def _main_task_generator_generate(args):
if args.verbose:
logging.getLogger().setLevel(logging.DEBUG)
source_dir = Path(args.source_dir)
output_dir = Path(args.output_dir)
formats_to_generate = [f.strip() for f in args.formats.split(',') if f.strip()]
if not source_dir.is_dir():
logger.error(f"Source directory does not exist or is not a directory: {source_dir}")
return 1
try:
output_dir.mkdir(parents=True, exist_ok=True)
except OSError as e:
logger.error(f"Could not create output directory '{output_dir}': {e}")
return 1
logger.info(f"Scanning for info.json files in '{source_dir}' (recursively)...")
source_files = list(source_dir.rglob('*.json'))
if not source_files:
logger.info(f"No .json files found in '{source_dir}'. Nothing to do.")
return 0
logger.info(f"Found {len(source_files)} source file(s). Generating tasks for formats: {', '.join(formats_to_generate)}...")
total_tasks_generated = 0
for source_file in source_files:
try:
with open(source_file, 'r', encoding='utf-8') as f:
info_json_content = json.load(f)
except (IOError, json.JSONDecodeError) as e:
logger.warning(f"Skipping file '{source_file.name}' due to read/parse error: {e}")
continue
try:
tasks_generated_this_run = 0
# Use metadata to create a profile-specific subdirectory for better organization.
profile_name_from_meta = info_json_content.get('_ytops_metadata', {}).get('profile_name')
final_output_dir = output_dir
if profile_name_from_meta:
final_output_dir = output_dir / profile_name_from_meta
# Ensure subdirectory exists. This is done once per source file.
try:
final_output_dir.mkdir(parents=True, exist_ok=True)
except OSError as e:
logger.error(f"Could not create profile subdirectory '{final_output_dir}': {e}. Skipping tasks for this source file.")
continue
for format_str in formats_to_generate:
task_data = info_json_content.copy()
# Add the target format to the task data itself. This makes the task file self-contained.
task_data['_ytops_download_format'] = format_str
# Create a unique filename for the task
original_stem = source_file.stem
safe_format_str = sanitize_format_for_filename(format_str)
task_filename = f"{original_stem}-format-{safe_format_str}.json"
output_path = final_output_dir / task_filename
# Check if this specific task file already exists to avoid re-writing
if output_path.exists():
logger.debug(f"Task file already exists, skipping generation: {output_path}")
continue
with open(output_path, 'w', encoding='utf-8') as f:
json.dump(task_data, f, indent=2)
logger.debug(f"Generated task file: {output_path}")
tasks_generated_this_run += 1
if tasks_generated_this_run > 0:
total_tasks_generated += tasks_generated_this_run
# Mark source file as processed by renaming
try:
processed_path = source_file.with_suffix(f"{source_file.suffix}.processed")
source_file.rename(processed_path)
logger.debug(f"Marked '{source_file.name}' as processed.")
except (IOError, OSError) as e:
logger.error(f"Failed to mark source file '{source_file.name}' as processed: {e}")
except IOError as e:
logger.error(f"An I/O error occurred while generating tasks for '{source_file.name}': {e}. It will be retried on the next run.")
# The file is not renamed, so it will be picked up again
logger.info(f"Successfully generated {total_tasks_generated} new task file(s) in '{output_dir}'.")
return 0