concat.sh

#!/usr/bin/env python3

import os
import argparse
import sys

def combine_markdown_files(folder_path, output_file, max_size):
    current_size = 0

    try:
        with open(output_file, 'w', encoding='utf-8') as outfile:
            for root, dirs, files in os.walk(folder_path, followlinks=False):
                for file in files:
                    if file.endswith('.md'):
                        filepath = os.path.join(root, file)
                        outfile.write(f"<file path=\"{filepath}\">\n\n")
                        try:
                            with open(filepath, 'r', encoding='utf-8') as infile:
                                content = infile.read()
                                outfile.write(content)
                                outfile.write('\n\n')
                            outfile.write(f"</file>\n\n")
                            current_size += len(content)
                            if current_size > max_size:
                                print(f"Reached size limit at {filepath}")
                                return
                        except IOError as e:
                            print(f"Error reading file {filepath}: {e}", file=sys.stderr)

        print(f"Combined Markdown files written to {output_file}")
    except IOError as e:
        print(f"Error writing to output file {output_file}: {e}", file=sys.stderr)

def main():
    parser = argparse.ArgumentParser(description="Combine Markdown files from a folder into a single file with a size limit.")
    parser.add_argument("folder_path", help="Path to the folder containing Markdown files")
    parser.add_argument("-o", "--output", default="combined.md", help="Output file name (default: combined.md)")
    parser.add_argument("-s", "--size", type=int, default=5, help="Maximum size limit in GB (default: 5)")
    args = parser.parse_args()

    if not os.path.isdir(args.folder_path):
        print(f"Error: '{args.folder_path}' is not a valid directory.", file=sys.stderr)
        sys.exit(1)

    max_size = 1024 * 1024 * 1024 * args.size  # Convert GB to bytes
    combine_markdown_files(args.folder_path, args.output, max_size)

if __name__ == "__main__":
    main()