acnh_jp_parser.py

# -*- coding: utf-8 -*-
from os import walk, path
from lxml import etree
from csv import writer
from html import unescape

## Get all files
eng_path = 'path\\to\\acnh1.1msgen'
jpn_path = 'path\\to\\acnh1.1msgjp'

eng_files = []
jpn_files = []

for r, _, f in walk(eng_path):
    for file in f:
        eng_files.append(path.join(r, file))

for r, _, f in walk(jpn_path):
    for file in f:
        jpn_files.append(path.join(r, file))

## Set up CSV writing
output_file = open('output.csv', 'w+', newline='', encoding='utf-8')
writer = writer(output_file)
writer.writerow(["File", "name", "Version", "English", "Japanese"])

parser = etree.XMLParser(recover=True)

for file_num in range(len(eng_files)):
    if file_num < len(jpn_files):
        eng_offset = 0
        eng_filename = eng_files[file_num + eng_offset].split('\\')[-1:][0]
        jpn_filename = jpn_files[file_num].split('\\')[-1:][0]
        if eng_filename != jpn_filename: # In case files don't match up, try searching around the file_num index for a match
            recovered = False
            for new_offset in range(-50,50):
                if file_num + eng_offset + new_offset < len(eng_files):
                    if eng_files[file_num + eng_offset + new_offset].split('\\')[-1:][0] == jpn_files[file_num].split('\\')[-1:][0]:
                        eng_offset += new_offset
                        recovered = True
                        eng_filename = eng_files[file_num + eng_offset].split('\\')[-1:][0]
                        jpn_filename = jpn_files[file_num].split('\\')[-1:][0]
            if not recovered:continue

        ## Read each file and parse it from XML formatting
        eng_f = open(eng_files[file_num + eng_offset], "r", encoding='utf-8')
        eng_lines = eng_f.readlines()
        eng_f.close()
        eng_root = etree.fromstringlist(eng_lines[1:], parser=parser)

        jpn_f = open(jpn_files[file_num], "r", encoding='utf-8')
        jpn_lines = jpn_f.readlines()
        jpn_f.close()
        jpn_root = etree.fromstringlist(jpn_lines[1:], parser=parser)

        ## Extract desired data from each entry
        for entry in range(len(eng_root[2])):
            name = eng_root[2][entry].attrib['name']
            if eng_root[2][entry][0].text != None:
                eng_text_original = unescape(eng_root[2][entry][0].text)
            else: eng_text_original = ''
            if eng_root[2][entry][1].text != None:
                eng_text_edited = unescape(eng_root[2][entry][1].text)
            else: eng_text_edited = ''
            if jpn_root[2][entry][0].text != None:
                jpn_text_original = unescape(jpn_root[2][entry][0].text)
            else: jpn_text_original = ''
            if jpn_root[2][entry][1].text != None:
                jpn_text_edited = unescape(jpn_root[2][entry][1].text)
            else: jpn_text_edited = ''
            ## Write data to CSV file
            writer.writerow([eng_filename, name, "Original", eng_text_original, jpn_text_original])
            writer.writerow([eng_filename, name, "Edited", eng_text_edited, jpn_text_edited])