利用 Python 爬虫来对文本进行批量化翻译
很多翻译软件网站都提供了 API，这样可以自己编写应用程序，自动化翻译工作，但是免费的 API 越来越少，而且限制颇多，
有时候就只是懒得重复性复制粘贴，于是想用 Python 的爬虫软件包 selenium 调用网页数据进行进行自动化翻译操作。
网上类似开源软件找了一遍，发现基本上都被翻译软件网站也进行了反制措施后实效，页面使用了各种动态 JS，防止被程序化抓取调用。
但强大的 selenium 还是可以搞定，以下是代码：
from selenium import webdriver
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.common.action_chains import ActionChains
from selenium.common.exceptions import StaleElementReferenceException
from time import sleep
from selenium.webdriver.chrome.options import Options
import os
import re
import sys
import glob
from loguru import logger
import random
import argparse
parser = argparse.ArgumentParser()
parser.add_argument("--target", type=str, choices=["french", "english", "german", "spanish", "portuguese", "italian", "dutch", "polish", "russian"],
help="""Choose target language for translation {"french", "english", "german", "spanish", "portuguese", "italian", "dutch", "polish", "russian"}""")
args = parser.parse_args()
logger.add(sys.stderr, format="{time} {message}", filter="selenium_translate", level="INFO")
TEXTS_FOLDER_PATH = './texts_folder/'
TRANSLATED_TEXTS_PATH = './texts_folder/translated_texts/'
if os.path.exists(TEXTS_FOLDER_PATH):
text_paths = glob.glob(TEXTS_FOLDER_PATH + "*.txt")
if not os.path.exists(TRANSLATED_TEXTS_PATH):
logger.info(f"Creating new folder: '{TRANSLATED_TEXTS_PATH}'")
os.makedirs(TRANSLATED_TEXTS_PATH)
else:
raise AssertionError("Could not find './texts_folder/', folder in which you should add the texts files you want to translate")
options = Options()
options.add_argument('--headless')
options.add_argument('--disable-gpu')
WEBDRIVER_PATH = './webdrivers/chromedriver'
DEEPL_URL = 'https://www.deepl.com/translator#en/pt/'
HEADLESS = True
if HEADLESS:
driver = webdriver.Chrome(WEBDRIVER_PATH, options=options)
logger.info("Running headless...")
else:
driver = webdriver.Chrome(WEBDRIVER_PATH)
TARGET_LANGUAGES = ({"french":1,
"english":2,
"german":3,
"spanish":4,
"portuguese":5,
"italian":6,
"dutch":7,
"polish":8,
"russian":9})
def translate(text_to_translate, target_language=TARGET_LANGUAGES["portuguese"]):
driver.get(DEEPL_URL)
input_text_xpath = """//*[@id="panelTranslateText"]/div[1]/div[2]/section[1]/div[3]/div[2]/d-textarea/div"""
output_text_xpath = """//*[@id="panelTranslateText"]/div[1]/div[2]/section[2]/div[3]/div[1]/d-textarea/div/p[1]"""
language_select_xpath = """//*[@id="dl_translator"]/div[1]/div[3]/div[1]/div/button/div"""
target_language_xpath = f"""//*[@id="dl_translator"]/div[1]/div[3]/div[1]/div/div/button[{target_language}]"""
input_box = WebDriverWait(driver, 20).until(lambda driver: driver.find_element_by_xpath(input_text_xpath))
ActionChains(driver).click(input_box).send_keys(text_to_translate).perform()
sleep(random.uniform(2,3))
output_box = WebDriverWait(driver, 45).until(lambda driver: driver.find_element_by_xpath(output_text_xpath))
try:
translated_text = output_box.text
except e as StaleElementReferenceException:
sleep(random.uniform(3,5))
output_box = WebDriverWait(driver, 45).until(lambda driver: driver.find_element_by_xpath(output_text_xpath))
translated_text = output_box.text
logger.info("返回："+translated_text)
return translated_text
def main(target):
logger.info("enumerate files...")
translated_text = "";
for i, path in enumerate(text_paths):
logger.info(f"Text file:{path}")
with open(path) as f:
lines = f.readlines()
for text in lines:
translated_text += translate(text) +"\r\n"
sleep(random.uniform(2,3))
logger.info(f"Progress: Translated -> {text}")
output_fname = os.path.join(TRANSLATED_TEXTS_PATH, 'TRANSLATED_' + os.path.basename(path))
logger.info(f"Writing to file '{output_fname}'")
with open(output_fname, 'w') as out:
out.write(translated_text)
out.flush
if not HEADLESS:
sleep(10)
driver.close()
logger.info("Done.")
if __name__ == "__main__":
if args.target in TARGET_LANGUAGES.keys():
main(args.target)
else:
logger.info("""Could not understand the target language for translation. Target language mut be in this set {"french", "english", "german", "spanish", "portuguese", "italian", "dutch", "polish", "russian"}""")
用法如下：
先要用安装 python 的下列软件包
pip3 -i selenium
pip3 -i loguru
这两个软件包
在https://chromedriver.chromium.org/ 下载 chrome web driver
并放在当前目录下
将以上 python 代码保存为 trans_tool.py
建一个子目录 texts_folder， 将待翻译的文本(一次翻译一行)，放入目录内
在命令行执行 python3 trans_tool.py --target portuguese
我这里是英语翻译为葡萄亚语，你也可以选择其他目标语言。
希望对你有帮助！
我的微信
