Automated Translation ffrom jp to en

basic-rep.json

{
    "「":"\"",
    "」":"\"",
    "『":"\"",
    "』":"\"",
    "―":"-"
}

make_orgs.py

class OrgFile():
    def __init__(self,filename):
        self.filename = filename
        self.chapters = []

    def add_chapter(self,chapname=None,chapcontent=None):
        with open(self.filename,'a') as writer:
            writer.write(f'\n* {chapname}\n')
            writer.write(chapcontent.replace('\n','\n\n'))
        self.chapters.append(chapname)
    

ncode.py

import requests
from bs4 import BeautifulSoup
import make_orgs
import re
from string import Template

# first_chap = 114,189 # v4
chapters_rng = range(503,505) #arc 7
# rezero = n2267be
chap_url  = Template("http://ncode.syosetu.com/n2267be/${chapter}/")
headers = {
'User-Agent' : "Mozilla/5.0 (X11; Linux x86_64; rv:81.0) Gecko/20100101 Firefox/81.0"
}

def get_soup(url):
    print(f'connecting...{url}')
    r = requests.get(url, headers = headers)
    print(':Connected:')
    return BeautifulSoup(r.text,'html.parser')


def add_chapter(org_file,chap_no):
    soup = get_soup(chap_url.substitute(chapter=chap_no))
    entry = soup.find('div',{'id':'novel_contents'})
    chapname = entry.find('p',{'class':'novel_subtitle'})
    content = entry.find('div',{'id':'novel_honbun'})

    chaptitle = chapname.text
    contents = content.text

    org_file.add_chapter(chaptitle,contents)
    print(f'Completed::{chap_no}-{chaptitle}')


org = make_orgs.OrgFile('./rezero/arc7/arc-7-jp.org')

for chap in chapters_rng:
    add_chapter(org,chap)

translations.py

import pyautogui as auto
import pyperclip as pc
import time
import math
import json
import re

copy, paste = pc.determine_clipboard()

try:
    delay_sec = int(input('Enter Delay:<10>'))
except ValueError:
    delay_sec = 10

wrd = "./rezero/arc7/"
input_file = f'{wrd}/arc-7-jp.org'

replaced_file = f'{wrd}/arc7-rep.org'
output_file = f'{wrd}/arc7-deepl-en.org'

replacements = './rezero/replacements.json'

input(f"Continue with input file: {input_file}")

def replace_words():
    with open(input_file,'r') as r:
        l = r.read()
    
    if replacements=='':
        rep = dict()
    else:
        with open(replacements,'r') as r:
            rep = json.load(r)

    l = re.sub(r'\n\n+','\n\n',l)
    for k,v in rep.items():
        l = l.replace(k,v)
    with open(replaced_file,'w') as w:
        w.write(l)


def automate(delay_sec = 10, start_line = 0, end_line=math.inf):
    time.sleep(delay_sec)
    r = open(replaced_file,'r')
    lines = ''
    for i in range(start_line):
        next(r)
    i = start_line
    while i<end_line:
        try:
            new_line =  f'{next(r)}\n'
        except StopIteration:
            end_line = i
            new_line = ''
        # python's character count and deepl's count seems to be different. 
        if (len(lines)+len(new_line))>4500 or i==end_line:
            process_line(lines, comment=f'{i}-{len(lines)}-{time.time()}')
            print(f'{i} lines processed')
            lines = new_line
        else:
            lines += new_line
        i+=1
    r.close()

def process_line(lines, comment=None):
    copy(lines)
    time.sleep(0.1)
    try:
        auto.click(300,350)
        time.sleep(0.5)
        auto.hotkey('ctrl','a')
        time.sleep(0.5)
        auto.hotkey('ctrl','v')
        time.sleep(20)
        auto.scroll(10)
        time.sleep(10)
        #didn't work
        # while True:
        #     time.sleep(1)
        #     if not auto.locateOnScreen('./screenshot.png'):
        #         break
        auto.click(1000,350)
        time.sleep(0.5)
        auto.hotkey('ctrl','a')
        translated = paste(primary=True)
        if comment:
            save_translated(f'# c:{comment}\n{translated}')
        else:
            save_translated(translated)
    except auto.FailSafeException:
        raise

def save_translated(lines):
    with open(output_file,'a') as w:
        w.write(lines)

replace_words()
automate(delay_sec)