zw2pp/convert.py

#!/usr/bin/env python3

import argparse
import os
import csv
import sys
import xml.etree.ElementTree as ET

# Mapping from ZionWorx verse types to labels understood by ProPresenter.
# NOTE: Some of these don't have colons after them.
VT_MAP = {
    "stVerse": "Verse:",
    "stChorus": "Chorus:",
    "stBridge": "Bridge:",
    "stPreChorus": "Pre Chorus",
    "stEnding": "Outro"
}

# Default verse type
DEFAULT_VT = "stVerse"

cli_parser = argparse.ArgumentParser(
    description="ZionWorx to ProPresenter conversion tool"
)
cli_parser.add_argument("csv_file", type=str,
                        help="CSV file exported using TurboDB")
cli_parser.add_argument("-d", "--delimiter", type=str,
                        default=";", help="CSV delimiter character (default: ;)")
cli_parser.add_argument("-q", "--quote", type=str,
                        default="\"", help="CSV quote character (default: \")")
cli_parser.add_argument("-o", "--output", type=str,
                        default=os.getcwd(), help="Folder to write song files in (default: current working directory)")

cli_args = cli_parser.parse_args()
csv_file = cli_args.csv_file
delim_char = cli_args.delimiter
quote_char = cli_args.quote
output_path = cli_args.output

print(f"Converting {csv_file} to path {output_path}...", file=sys.stderr)

with open(csv_file) as csv_handle:
    processed_songs = set()
    reader = csv.reader(csv_handle, delimiter=delim_char, quotechar=quote_char)

    i = 0
    for song in reader:
        i += 1
        if i == 1:
            continue

        if len(song) < 8:
            print("Row does not contain seven columns: " +
                  delim_char.join(song), file=sys.stderr)
            continue

        song_id = song[0]
        title = song[1]

        print(f"Processing #{song_id}: \"{title}\"...", file=sys.stderr)

        # Ensure broken whitespace doesn't destroy the conversion
        lyrics_data = song[4].replace("\v", "\n")
        lyrics_xml = ET.fromstring(lyrics_data)
        # copy_str = song[7]

        lyrics = []
        for section in lyrics_xml.findall("section"):
            if section.text is not None:
                text = section.text
                label = VT_MAP.get(section.get("type", DEFAULT_VT), DEFAULT_VT)
                lyrics.append(f"{label}\n{text}")

        lyrics_str = "\n\n".join(lyrics)

        # Sadly I don't know how to get any copyright information into ProPresenter. :(
        file_contents = f"""Title: {title}

{lyrics_str}
"""

        file_name = title.replace("/", "_")
        if file_name in processed_songs:
            orig_file_name = file_name
            r = 0
            while file_name in processed_songs:
                r += 1
                file_name = f"{orig_file_name}_{r}"

        target_file = f"{output_path}/{file_name}.txt"
        with open(target_file, mode="w") as target:
            target.write(file_contents)
            print(f"Wrote {target_file}", file=sys.stderr)
            processed_songs.add(file_name)

print("\nDone!", file=sys.stderr)