Combine files, with filename/dir hints, to load into AI

combine_files.sh

#!/bin/bash

# ===========================================================================
# File Combiner - Combines multiple files into a single file with path headers
# ===========================================================================
#
# Description:
#   This script recursively combines files from a directory into a single file,
#   adding the file path as a comment before each file's content. Useful for
#   preparing codebases for AI analysis or documentation.
#
# Features:
#   - Adds file paths as comments (# path/to/file.ext)
#   - Skips binary files automatically
#   - Can filter by file extensions
#   - Excludes common dirs (.git, node_modules, etc.) by default
#   - Handles spaces in filenames
#   - Maintains relative paths
#
# Usage:
#   ./combine_files.sh [options] source_directory output_file
#
# Options:
#   -e 'ext1 ext2 ...'    Only include files with these extensions
#   -x 'dir1 dir2 ...'    Additional directories to exclude
#   -h                    Show help message
#
# Examples:
#   # Basic usage - combine all text files:
#   ./combine_files.sh ./my_project combined_files.txt
#
#   # Only combine Python and JavaScript files:
#   ./combine_files.sh -e "py js" ./my_project combined_files.txt
#
#   # Exclude additional directories:
#   ./combine_files.sh -x "build dist" ./my_project combined_files.txt
#
#   # Combine specific files and exclude dirs:
#   ./combine_files.sh -e "py js jsx" -x "build dist tests" ./my_project combined_files.txt
#
# Default excluded directories:
#   .git, node_modules, __pycache__, venv, .env

# Usage function
usage() {
    echo "Usage: $0 [options] directory output_file"
    echo "Options:"
    echo "  -e 'ext1 ext2 ...'    Only include files with these extensions (space-separated)"
    echo "  -x 'dir1 dir2 ...'    Exclude these directories (space-separated)"
    echo "  -h                    Show this help message"
    echo
    echo "Example:"
    echo "  $0 -e 'py js txt' -x 'node_modules .git' /path/to/source output.txt"
    exit 1
}

# Default excluded directories
EXCLUDE_DIRS=".git node_modules __pycache__ venv .env"
EXTENSIONS=""

# Parse command line options
while getopts "e:x:h" opt; do
    case $opt in
        e) EXTENSIONS="$OPTARG" ;;
        x) EXCLUDE_DIRS="$EXCLUDE_DIRS $OPTARG" ;;
        h) usage ;;
        ?) usage ;;
    esac
done

# Shift past the options
shift $((OPTIND-1))

# Check for required arguments
if [ $# -ne 2 ]; then
    usage
fi

SOURCE_DIR="$1"
OUTPUT_FILE="$2"

# Create exclude pattern for find command
EXCLUDE_PATTERN=""
for dir in $EXCLUDE_DIRS; do
    EXCLUDE_PATTERN="$EXCLUDE_PATTERN -not -path '*/$dir/*'"
done

# Create extension pattern for find command
EXT_PATTERN=""
if [ ! -z "$EXTENSIONS" ]; then
    for ext in $EXTENSIONS; do
        if [ -z "$EXT_PATTERN" ]; then
            EXT_PATTERN="-name '*.$ext'"
        else
            EXT_PATTERN="$EXT_PATTERN -o -name '*.$ext'"
        fi
    done
    EXT_PATTERN="\\( $EXT_PATTERN \\)"
fi

# Clear or create the output file
> "$OUTPUT_FILE"

# Main processing
eval "find '$SOURCE_DIR' -type f $EXCLUDE_PATTERN $EXT_PATTERN" | while read -r file; do
    # Get relative path
    rel_path=${file#$SOURCE_DIR/}
    
    # Check if file is binary
    if file "$file" | grep -q "text"; then
        echo "# $rel_path" >> "$OUTPUT_FILE"
        cat "$file" >> "$OUTPUT_FILE"
        echo "" >> "$OUTPUT_FILE"
    else
        echo "Skipping binary file: $rel_path" >&2
    fi
done

echo "Files combined successfully into $OUTPUT_FILE"