import os
import glob
from lxml import etree as ET

# Directory path
input_directory = os.path.join(os.getcwd(), 'input')
output_directory = os.path.join(os.getcwd(), 'output')

# Create output directory if it doesn't exist
os.makedirs(output_directory, exist_ok=True)

# Remove all existing files in output directory
for filename in os.listdir(output_directory):
    os.remove(os.path.join(output_directory, filename))

# Get all XML files in the input directory
files = glob.glob(os.path.join(input_directory, '*.xml'))

ns = {'ns': 'http://openlyrics.info/namespace/2009/song'}

for file in files:
    # Parse XML file
    parser = ET.XMLParser(remove_blank_text=True)
    tree = ET.parse(file, parser)
    root = tree.getroot()

    # Prepare output
    output = "Title(s):\n"
    titles = root.findall('.//ns:title', ns)
    for title in titles:
        output += (title.text or '').strip() + "\n"
        
    copyright = root.find('.//ns:copyright', ns)
    if copyright is not None and copyright.text:
        output += "Copyright: " + copyright.text.strip() + "\n"

    verseOrder = root.find('.//ns:verseOrder', ns)
    if verseOrder is not None and verseOrder.text:
        output += "Verse Order: " + verseOrder.text.strip() + "\n"

    ccliNo = root.find('.//ns:ccliNo', ns)
    if ccliNo is not None and ccliNo.text:
        output += "CCLI No.: " + ccliNo.text.strip() + "\n"
        
    authors = root.findall('.//ns:author', ns)
    if authors:
        output += "Author(s): "
        for i, author in enumerate(authors):
            output += (author.text or '').strip()
            if i < len(authors) - 1:
                output += ", "
            else:
                output += "\n"

    output += "Lyrics:\n"
    verses = root.findall('.//ns:verse', ns)
    for verse in verses:
        lines_in_verse = verse.findall('.//ns:lines', ns)
        for lines in lines_in_verse:
            if lines is not None:
                lines_text = ET.tostring(lines, method='text', encoding='utf-8').decode('utf-8')
                lines_text = lines_text.replace('<br/>', '\n').strip()
                lines_text = ' '.join(lines_text.split())
                output += lines_text + "\n"
        output += "\n"



    # Write output to a text file in output directory
    txt_file_name = os.path.join(output_directory, os.path.splitext(os.path.basename(file))[0] + '.txt')
    with open(txt_file_name, 'w') as f:
        f.write(output)

print("All files have been processed and saved!")