LibreOffice WriterのファイルをPython＋DeepLで機械翻訳する 【Windows編】

& 'C:\Program Files\LibreOffice\program\soffice.exe'

& 'C:\Program Files\LibreOffice\program\soffice.exe' --accept="socket,host=localhost,port=2002,tcpNoDelay=1;urp;"

& 'C:\Program Files\LibreOffice\program\python.exe' mycode.py

import uno
from pythonscript import ScriptContext


class LoServer:
    UNO_RESOLVER = 'com.sun.star.bridge.UnoUrlResolver'
    UNO_DESKTOP = 'com.sun.star.frame.Desktop'
    PRESET_DIRS = ['00-REF', '01-ORIGINAL', '02-MAP', '03-TRANSLATION', '04-CHECK', '05-DONE']

    def __init__(self, host='localhost', port='2002', named_pipe=None):
        self.host = host
        self.port = port
        self.named_pipe = named_pipe
        self.script = None

    def connect_to_libreoffice(self):
        local_ctx = uno.getComponentContext()
        local_smgr = local_ctx.ServiceManager
        resolver = local_smgr.createInstanceWithContext(LoServer.UNO_RESOLVER, local_ctx)
        if self.named_pipe is None:
            uno_string = f'uno:socket,host={self.host},port={self.port};urp;StarOffice.ComponentContext'
        else:
            uno_string = f'uno:pipe,name={self.named_pipe};urp;StarOffice.ComponentContext'
        ctx = resolver.resolve(uno_string)
        smgr = ctx.ServiceManager
        script_context = ScriptContext(ctx,
                                       smgr.createInstanceWithContext(LoServer.UNO_DESKTOP, ctx),
                                       None)

        self.script = script_context

if __name__ == '__main__':
    lo = LoServer()
    lo.connect_to_libreoffice()

desktop = lo.script.getDesktop()
url = f'file:///{path_to_file}'
document = desktop.loadComponentFromURL(url, '_blank', 0, ())

# Example python script for the scripting framework

#
# This file is part of the LibreOffice project.
#
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.
#
# This file incorporates work covered by the following license notice:
#
#   Licensed to the Apache Software Foundation (ASF) under one or more
#   contributor license agreements. See the NOTICE file distributed
#   with this work for additional information regarding copyright
#   ownership. The ASF licenses this file to you under the Apache
#   License, Version 2.0 (the "License"); you may not use this file
#   except in compliance with the License. You may obtain a copy of
#   the License at http://www.apache.org/licenses/LICENSE-2.0 .
#

def InsertText(text):
    """Inserts the argument string into the current document.
    If there is a selection, the selection is replaced by it.
    """

    # Get the doc from the scripting context which is made available to
    # all scripts.
    desktop = XSCRIPTCONTEXT.getDesktop()
    model = desktop.getCurrentComponent()

    # Check whether there's already an opened document.
    if not hasattr(model, "Text"):
        return

    # The context variable is of type XScriptContext and is available to
    # all BeanShell scripts executed by the Script Framework
    xModel = XSCRIPTCONTEXT.getDocument()

    # The writer controller impl supports the css.view.XSelectionSupplier
    # interface.
    xSelectionSupplier = xModel.getCurrentController()

    # See section 7.5.1 of developers' guide
    xIndexAccess = xSelectionSupplier.getSelection()
    count = xIndexAccess.getCount()

    if count >= 1:  # ie we have a selection
        i = 0

    while i < count:
        xTextRange = xIndexAccess.getByIndex(i)
        theString = xTextRange.getString()

        if not len(theString):
            # Nothing really selected, just insert.
            xText = xTextRange.getText()
            xWordCursor = xText.createTextCursorByRange(xTextRange)
            xWordCursor.setString(text)
            xSelectionSupplier.select(xWordCursor)
        else:
            # Replace the selection.
            xTextRange.setString(text)
            xSelectionSupplier.select(xTextRange)

        i += 1

import json
import LoServer
from urllib import request, parse

class WriterMt:
    def __init__(self, key):
        self.lo = LoServer()
        self.lo.connect_to_libreoffice()
        # DeepL 無料版のAPIキーをセット
        self.endpoint = f'https://api-free.deepl.com/v2/translate?auth_key={key}'

    def open_and_mt(self, path_to_file):
        # ファイルを開く
        desktop = self.lo.script.getDesktop()
        url = f'file:///{path_to_file}'
        document = desktop.loadComponentFromURL(url, '_blank', 0, ())

        # インスタンスの準備
        selector = document.getCurrentController()
        selection = selector.getSelection()
        text_range = selection.getByIndex(0)
        cursor = text_range.getText().createTextCursorByRange(text_range)
        current_text = ''

        # ファイル終わりの目印
        END_MARK = '_@@_EOF_@@_'
        content = document.Text
        end = content.End
        end.String = END_MARK

        # 段落ごとに選択
        while current_text != END_MARK:
            if cursor.isEndOfParagraph():
                pass
            else:
                # goto～系のメソッドは、Trueを渡すと選択範囲を拡張、Falseを渡すと選択範囲を消去して移動する
                cursor.gotoEndOfParagraph(True)
                current_text = cursor.getString()
                if current_text == '':
                    pass
                elif current_text == END_MARK:
                    cursor.setString('')
                else:
                    mt = get_deepl(current_text)
                    if mt != '':
                        cursor.setString(mt)

                    # 次の段落へ移動する
                    cursor.gotoEndOfParagraph(False)
                    cursor.goRight(1, False)
                    selector.select(cursor)

        # 最後にファイルを上書きし、操作権を放棄する
        document.store()
        document.dispose()

    def get_deepl(text):
        headers = {
            'Content-Type': 'application/x-www-form-urlencoded'
        }
        params = {
            'text': text,
            'source_lang': 'JA',
            'target_lang': 'EN'
        }
        req = request.Request(self.endpoint, parse.urlencode(params).encode(), headers)
        try:
            with request.urlopen(req) as res:
                if res.status == 200:
                   data = json.loads(res.read().decode('utf-8'))
                   mt = data['translations'][0]['text'].strip()
                   return mt
                else:
                   return ''
            except:
                print(f'DeepL got an error')
                return ''

if __name__ == '__main__':
    writer = WriterMt('xxxxxxxxxxxxxxxxxxxxxxxxxxxx')
    writer.open_and_mt(path_to_file)
    print('MT Done!')