Extracts terms from Apple lg glossary files, combining into a JSON file

extract-glossaries.py

#!/usr/bin/env python

import sys, os.path
import os
import json
from xml.dom.minidom import parse

rootdir = os.getcwd()
lgPaths = ["iOS/AuthKitUI.lg", "iOS/CalendarUIKit.lg", "iOS/iCloudDriveSettings.lg", "iOS/MobileNotes.lg", "macOS/CalendarUI.lg", "macOS/Finder_FE.lg", "macOS/Finder_FinderKit.lg", "macOS/iCloudPrefPane.lg", "macOS/Notes.lg", "macOS/Reminders.lg", "macOS/TextEdit.lg"]
languageDirs = ["Dutch", "French"]

resultDict = {}
for lang in languageDirs:
    for lgPath in lgPaths: 
        path = os.path.join(rootdir, lang, lgPath)
        dom = parse(path)
        proj = dom.documentElement
        files = proj.getElementsByTagName("File")
        for f in files:
            filepath = f.getElementsByTagName("Filepath")[0].childNodes[0].data
            items = f.getElementsByTagName("TextItem")
            for i in items:
                position = i.getElementsByTagName("Position")[0].childNodes[0].data
                idInGlossary = lgPath + "__" + filepath + "__" + position
                
                transset = i.getElementsByTagName("TranslationSet")[0]
                english = transset.getElementsByTagName("base")[0].childNodes[0].data
                foreign = transset.getElementsByTagName("tran")[0].childNodes[0].data
                langCode = transset.getElementsByTagName("tran")[0].getAttribute('loc')
                
                entry = resultDict.setdefault(idInGlossary, {})
                entry["en"] = english.encode('utf-8')
                entry[langCode] = foreign.encode('utf-8')
                
# Remove any entry that doesn't include all languages
for k in resultDict.keys():
    if len(resultDict[k]) != len(languageDirs)+1:
        del resultDict[k]
        
# Convert dictionary to array
jsonList = []
for k in resultDict:
    # Have to convert dictionary to list, because 
    # Swift Codable can't handle a dictionary where 
    # the keys are not strings
    langDict = resultDict[k]
    langList = []
    for l in langDict: 
        langList.append(l)
        langList.append(langDict[l])
    entry = {"id":k, "textByLanguage":langList}
    jsonList.append(entry)
                
with open('glossary.json', 'w') as outfile:
    json.dump(jsonList, outfile, sort_keys=True, indent=4)