The highly-customizable fast-and-simple bash script that will do various text-processing operations on files in a certain directory. It is useful for code formatting at the repository level and can be used in continuous integration and deployment (CI/CD) pipelines.

clean.sh

# CL_FILE_EXTENSIONS: The script will apply on files with given extensions
if [ -z ${CL_FILE_EXTENSIONS+x} ]; then
    CL_FILE_EXTENSIONS=(txt md json xml yaml yml sh bat)
fi

# CL_IGNORED_SUB_DIR: The script will ignore files in given directories
if [ -z ${CL_IGNORED_SUB_DIR+x} ]; then
    CL_IGNORED_SUB_DIR=(.git)
fi

# CL_TARGET_DIR: The directory that the scripts will work on.
if [ -z ${CL_TARGET_DIR+x} ]; then
    CL_TARGET_DIR="${1:-$(pwd)}"
fi

# CL_VERBOSE: Set to 1 to print log
if [ -z ${CL_VERBOSE+x} ]; then
    CL_VERBOSE=0
fi

# Prepare parameters for find command to ignore directories in CL_IGNORED_SUB_DIR
CL_PRUNE_DIRS=()
for cl_dir in "${CL_IGNORED_SUB_DIR[@]}"; do
    CL_PRUNE_DIRS+=("-type" "f" "-name" "$cl_dir" "-prune" "-o")
done

# Prepare parameters for find command to include given extensions in CL_FILE_EXTENSIONS
CL_INCLUDE_FILES=()
for cl_ext in "${CL_FILE_EXTENSIONS[@]}"; do
    CL_INCLUDE_FILES+=("-type" "f" "-name" "*.${cl_ext}" "-o")
done


for cl_file in $(find "$CL_TARGET_DIR" "${CL_PRUNE_DIRS[@]}" "${CL_INCLUDE_FILES[@]}" -false); do
    
    # Verbose
    if [ ${CL_VERBOSE} == 1 ]; then
        echo "${cl_file}"
    fi
        
    # Remove trailing whitespaces and tab characters
    sed -i 's/[ \t]*$//' "$cl_file"

    # Remove empty lines at the beginning
    sed -i '/./,$!d' "$cl_file"

    # Remove empty lines at the end
    sed -i -Ez '$ s/\n+$//' "$cl_file"
done