GarminHomeAssistant/translate.py

####################################################################################
#
# Distributed under MIT Licence
#   See https://github.com/house-of-abbey/GarminHomeAssistant/blob/main/LICENSE.
#
####################################################################################
#
# GarminHomeAssistant is a Garmin IQ application written in Monkey C and routinely
# tested on a Venu 2 device. The source code is provided at:
#            https://github.com/house-of-abbey/GarminHomeAssistant.
#
# J D Abbey & P A Abbey, 28 December 2022
#
#
# Description:
#
# Python script to automatically translate the strings.xml file to each supported
# language using Google Translate.
#
# Python installation:
#   pip install beautifulsoup4
#   pip install deep-translator
# NB. For XML formatting:
#   pip install lxml
#
# References:
#  * https://www.crummy.com/software/BeautifulSoup/bs4/doc/
#  * https://realpython.com/beautiful-soup-web-scraper-python/
#  * https://www.crummy.com/software/BeautifulSoup/bs4/doc/#parsing-xml
#  * https://www.crummy.com/software/BeautifulSoup/bs4/doc/#xml
#
####################################################################################
import os

from bs4 import BeautifulSoup
from bs4 import Comment
from deep_translator import GoogleTranslator

# List of tuples in the form os:
#  * Garmin IQ language three letter mnemonic,
#  * Google Translate language mnemonic,
#  * Language familiar name (mainly for reference)
languages: list[tuple[str, str, str]] = [
    ("ara", "ar", "Arabic"),
    ("bul", "bg", "Bulgarian"),
    ("zhs", "zh-CN", "Chinese (Simplified)"),
    ("zht", "zh-TW", "Chinese (Traditional)"),
    ("hrv", "hr", "Croatian"),
    ("ces", "cs", "Czech"),
    ("dan", "da", "Danish"),
    ("dut", "nl", "Dutch"),
    ("deu", "de", "German"),
    ("gre", "el", "Greek"),
    # ("eng", "en", "English"),
    ("est", "et", "Estonian"),
    ("fin", "fi", "Finnish"),
    ("fre", "fr", "French"),
    ("heb", "iw", "Hebrew"),
    ("hun", "hu", "Hungarian"),
    ("ind", "id", "Indonesian"),
    ("ita", "it", "Italian"),
    ("jpn", "ja", "Japanese"),
    ("kor", "ko", "Korean"),
    ("lav", "lv", "Latvian"),
    ("lit", "lt", "Lithuanian"),
    ("zsm", "ms", "Standard (Bahasa) Malay"),
    ("nob", "no", "Norwegian"),
    ("pol", "pl", "Polish"),
    ("por", "pt", "Portuguese"),
    ("ron", "ro", "Romanian"),
    # ("rus", "ru", "Russian"),
    ("slo", "sk", "Slovak"),
    ("slv", "sl", "Slovenian"),
    ("spa", "es", "Spanish"),
    ("swe", "sv", "Swedish"),
    ("tha", "th", "Thai"),
    ("tur", "tr", "Turkish"),
    ("ukr", "uk", "Ukrainian"),
    ("vie", "vi", "Vietnamese"),
]

langLength = len(languages)

exceptionIds: list[str] = ["AppName", "AppVersionTitle"]
titleIds: list[str] = []

# def merge(curr: BeautifulSoup, prev: BeautifulSoup) -> BeautifulSoup:
#     """
#     Merge the current strings.xml with the previous one, overwriting
#     the previous strings with the current ones if they exist.
#     """
#     out = prev.__copy__()
#     for s in curr.find(name="strings").findAll(name="string"):
#         s_prev = out.find(name="string", attrs={"id": s["id"]})
#         if s_prev:
#             s_prev.string = s.string
#         else:
#             out.find(name="strings").append(s)
#     return out

i = 1
with open("./resources/strings/strings.xml", "r") as f:
    c = f.read().replace("\r", "")
    for l in languages:
        os.makedirs(f"./resources-{l[0]}/strings/", exist_ok=True)
        # Old translations will not be automatically updated/removed, use removeTranslations.py
        try:
            with open(f"./resources-{l[0]}/strings/strings.xml", "r", encoding="utf-8") as r:
                prev = BeautifulSoup(r.read().replace("\r", ""), features="xml")
        except FileNotFoundError:
            prev = BeautifulSoup("", features="xml")
        try:
            with open(f"./resources-{l[0]}/strings/corrections.xml", "r", encoding="utf-8") as r:
                curr = BeautifulSoup(r.read().replace("\r", ""), features="xml")
        except FileNotFoundError:
            curr = BeautifulSoup("", features=["xml"])
        print(f"{i} of {langLength}: Translating English to {l[2]}")
        soup = BeautifulSoup(c, features="xml")
        translator = GoogleTranslator(source="en", target=l[1])
        soup.find(name="strings").insert_before("\n\n")
        soup.find(name="strings").insert_before(
            Comment(
                f"\n  Generated by Google Translate: English to {l[2]}\n  " +
                    translator.translate("Generated by Google Translate from English") + "\n"))
        soup.find(name="strings").insert_before("\n\n")

        for s in soup.find(name="strings").findAll(name="string"):
            s.insert_before("  ")
            if s["id"] in exceptionIds:
                continue

            s_curr = curr.find(name="string", attrs={"id": s["id"]})
            if s_curr:
                s.string = s_curr.string
            else:
                s_prev = prev.find(name="string", attrs={"id": s["id"]})
                if s_prev:
                    s.string = s_prev.string
                else:
                    a = translator.translate(s.string)
                    if s["id"] in titleIds:
                        s.string = a.title()
                    else:
                        s.string = a
        for s in soup.find(name="strings").findAll(
                string=lambda text: isinstance(text, Comment)):
            s.insert_before("  ")
            s.replace_with(Comment(" " + translator.translate(s) + " "))

        # print(str(soup))
        with open(f"./resources-{l[0]}/strings/strings.xml", "wb") as w:
            w.write(soup.encode("utf-8") + b"\n")
        i += 1