Plone po file to csv file

po2csv.py

"""
po file to CSV file for untranselated
"""
from collections import defaultdict
import csv

EMPTY_ONLY = True
CSV_HEADERS = ("uses", "default", "msgid", "fuzzy", "msgstr")


def po2block(filename):
    """po filename to msg block list (generator)
    """
    with open(filename, "r") as f:
        block_number = 0
        block = defaultdict(list)
        for i, line in enumerate(f):
            # if i > 50:
            #     break
            if not line.strip():
                if block_number:
                    yield block
                block_number += 1
                block = defaultdict(list)
                continue
            elif block_number == 0:
                continue
            if line.startswith("#. Default: "):
                block["default"].append(line.split("#. Default: ", 1)[1].strip())
            elif line.startswith("#: "):
                block["uses"].append(line.split("#: ", 1)[1].strip())
            elif line.startswith("#, fuzzy"):
                block["fuzzy"].append("1")
            elif line.startswith("msgid "):
                block["msgid"].append(line.split("msgid ", 1)[1].strip()[1:-1])
            elif line.startswith("msgstr "):
                block["msgstr"].append(line.split("msgstr ", 1)[1].strip()[1:-1])


def _block_line(block):
    for key in CSV_HEADERS:
        # print(key, block.get(key))
        yield " ".join(block.get(key, []))


def block2lines(blocks, output):
    """msg blocks to csv lines
    """
    with open(output, "w", newline="") as f:
        writer = csv.writer(f, delimiter=",", quotechar='"', quoting=csv.QUOTE_MINIMAL)
        for block in blocks:
            if not EMPTY_ONLY:
                writer.writerow(_block_line(block))
            elif (
                block.get("fuzzy", [""])[0] == "1" or block.get("msgstr", [""])[0] == ""
            ):
                writer.writerow(_block_line(block))


def main(input, output):
    blocks = po2block(input)
    block2lines(blocks, output)


if __name__ == "__main__":
    import sys

    args = sys.argv
    if len(args) < 3:
        print("Nedd args for po filename and output filename")
    else:
        main(args[1], args[2])