纯 Python 实现的 Google 批量翻译

测试通过时间:2019-8-20

参阅:C#实现谷歌翻译APIPython之Google翻译爬虫

首先声明,没有什么不良动机,因为经常会用 translate.google.cn,就想着用 Python 模拟网页提交实现文档的批量翻译。据说有 API,可是要收费。

生成 Token

Google 为防爬虫而生成 token 的代码是 Javascript 的,且是根据网站的 TKK 值和提交的文本动态生成。更新规律未知,只好定时去取一下了。

网上能找到的 Python 代码大部分是去调用 PyExecJS 库,先不说执行效率的高低(大概是差一个数量级),首先是舍近求远,不纯粹,本人不喜欢。

好不容易找到了一段 Python 代码还有点小 Bug,且缺少动态获取 TKK 的步骤。最后还是对照 Javascript 代码自己改成 Python 了。方法很简单,先转成易懂的 Javascript,再转成 Python。Javascript 代码来自C#实现谷歌翻译API

原始(晦涩) Javascript 代码

var b = function (a, b) {
for (var d = 0; d < b.length - 2; d += 3) {
var c = b.charAt(d + 2),
c = "a" <= c ? c.charCodeAt(0) - 87 : Number(c),
c = "+" == b.charAt(d + 1) ? a >>> c : a << c;
a = "+" == b.charAt(d) ? a + c & 4294967295 : a ^ c
}
return a
} var tk = function (a,TKK) {
for (var e = TKK.split("."), h = Number(e[0]) || 0, g = [], d = 0, f = 0; f < a.length; f++) {
var c = a.charCodeAt(f);
128 > c ? g[d++] = c : (2048 > c ? g[d++] = c >> 6 | 192 : (55296 == (c & 64512) && f + 1 < a.length && 56320 == (a.charCodeAt(f + 1) & 64512) ? (c = 65536 + ((c & 1023) << 10) + (a.charCodeAt(++f) & 1023), g[d++] = c >> 18 | 240, g[d++] = c >> 12 & 63 | 128) : g[d++] = c >> 12 | 224, g[d++] = c >> 6 & 63 | 128), g[d++] = c & 63 | 128)
}
a = h;
for (d = 0; d < g.length; d++) a += g[d], a = b(a, "+-a^+6");
a = b(a, "+-3^+b+-f");
a ^= Number(e[1]) || 0;
0 > a && (a = (a & 2147483647) + 2147483648);
a %= 1E6;
return a.toString() + "." + (a ^ h)
}

易懂的 Javascript 代码

function RL(a, b) {
for (var d = 0; d < b.length - 2; d += 3) {
var c = b.charAt(d + 2);
c = "a" <= c ? c.charCodeAt(0) - 87 : Number(c);
c = "+" == b.charAt(d + 1) ? a >>> c : a << c;
a = "+" == b.charAt(d) ? a + c & 4294967295 : a ^ c;
}
return a
} function TL(a,TKK) {
var e = TKK.split(".");
var h = Number(e[0]) || 0;
var g = [];
var d = 0;
for (var f = 0; f < a.length; f++) {
var c = a.charCodeAt(f);
if (128 > c)
{
g[d++] = c;
}
else
{
if (2048 > c)
{
g[d++] = c >> 6 | 192;
}
else
{
if (55296 == (c & 64512) && f + 1 < a.length && 56320 == (a.charCodeAt(f + 1) & 64512))
{
c = 65536 + ((c & 1023) << 10) + (a.charCodeAt(++f) & 1023);
g[d++] = c >> 18 | 240;
g[d++] = c >> 12 & 63 | 128;
}
else
{
g[d++] = c >> 12 | 224;
g[d++] = c >> 6 & 63 | 128;
}
}
g[d++] = c & 63 | 128;
}
}
a = h;
for (var d = 0; d < g.length; d++) {
a += g[d];
a = b(a, "+-a^+6");
}
a = b(a, "+-3^+b+-f");
a ^= Number(e[1]) || 0;
0 > a && (a = (a & 2147483647) + 2147483648);
a %= 1E6;
return a.toString() + "." + (a ^ h)
}

Python 代码

def getGoogleToken(a, TKK):
def RL(a, b):
for d in range(0, len(b)-2, 3):
c = b[d + 2]
c = ord(c[0]) - 87 if 'a' <= c else int(c)
c = a >> c if '+' == b[d + 1] else a << c
a = a + c & 4294967295 if '+' == b[d] else a ^ c
return a g = []
f = 0
while f < len(a):
c = ord(a[f])
if 128 > c:
g.append(c)
else:
if 2048 > c:
g.append((c >> 6) | 192)
else:
if (55296 == (c & 64512)) and (f + 1 < len(a)) and (56320 == (ord(a[f+1]) & 64512)):
f += 1
c = 65536 + ((c & 1023) << 10) + (ord(a[f]) & 1023)
g.append((c >> 18) | 240)
g.append((c >> 12) & 63 | 128)
else:
g.append((c >> 12) | 224)
g.append((c >> 6) & 63 | 128)
g.append((c & 63) | 128)
f += 1 e = TKK.split('.')
h = int(e[0]) or 0
t = h
for item in g:
t += item
t = RL(t, '+-a^+6')
t = RL(t, '+-3^+b+-f')
t ^= int(e[1]) or 0
if 0 > t:
t = (t & 2147483647) + 2147483648
result = t % 1000000
return str(result) + '.' + str(result ^ h)

获取 Token Key

Google 的 TKK 可以通过访问网站 https://translate.google.cn 获取,里面有段脚本里包含了“tkk:('xxxxxx.xxxxxx')”,用正则表达式截取即可。

    res = requests.get('https://translate.google.cn', timeout = 3)
res.raise_for_status()
result = re.search(r'tkk\:\'(\d+\.\d+)?\'', res.text).group(1)

划分文章段落

因为常从 PDF 里复制文本翻译,这样就不能依赖换行符来划分段落了。只能判断空行,作为段落的分界。

另外 Google 返回的结果 Json 里,会以英文句点作为分隔符,每一句译文均作为数组的一项分开。所以最后得合并一下,成为一个段落。

完整代码

代码不长,全文黏贴如下。

GoogleTranslator.py:

import requests
import re
import json
import time class GoogleTranslator ():
_host = 'translate.google.cn' _headers = {
'Host': _host,
'User-Agent': 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/76.0.3809.100 Mobile Safari/537.36',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'zh-CN,zh;q=0.8,en-US;q=0.5,en;q=0.3',
'Accept-Encoding': 'gzip, deflate, br',
'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
'Referer': 'https://' + _host,
'Connection': 'keep-alive',
'Cache-Control': 'max-age=0'
} _language = {
'afrikaans': 'af',
'arabic': 'ar',
'belarusian': 'be',
'bulgarian': 'bg',
'catalan': 'ca',
'czech': 'cs',
'welsh': 'cy',
'danish': 'da',
'german': 'de',
'greek': 'el',
'english': 'en',
'esperanto': 'eo',
'spanish': 'es',
'estonian': 'et',
'persian': 'fa',
'finnish': 'fi',
'french': 'fr',
'irish': 'ga',
'galician': 'gl',
'hindi': 'hi',
'croatian': 'hr',
'hungarian': 'hu',
'indonesian': 'id',
'icelandic': 'is',
'italian': 'it',
'hebrew': 'iw',
'japanese': 'ja',
'korean': 'ko',
'latin': 'la',
'lithuanian': 'lt',
'latvian': 'lv',
'macedonian': 'mk',
'malay': 'ms',
'maltese': 'mt',
'dutch': 'nl',
'norwegian': 'no',
'polish': 'pl',
'portuguese': 'pt',
'romanian': 'ro',
'russian': 'ru',
'slovak': 'sk',
'slovenian': 'sl',
'albanian': 'sq',
'serbian': 'sr',
'swedish': 'sv',
'swahili': 'sw',
'thai': 'th',
'filipino': 'tl',
'turkish': 'tr',
'ukrainian': 'uk',
'vietnamese': 'vi',
'yiddish': 'yi',
'chinese_simplified': 'zh-CN',
'chinese_traditional': 'zh-TW',
'auto': 'auto'
}
_url = 'https://' + _host + '/translate_a/single'
_params = {
'client': 'webapp',
'sl': 'en',
'tl': 'zh-CN',
'hl': 'zh-CN',
'dt': 'at',
'dt': 'bd',
'dt': 'ex',
'dt': 'ld',
'dt': 'md',
'dt': 'qca',
'dt': 'rw',
'dt': 'rm',
'dt': 'ss',
'dt': 't',
'otf': '1',
'ssel': '0',
'tsel': '0',
'kc': '1'
} __cookies = None __googleTokenKey = '376032.257956'
__googleTokenKeyUpdataTime = 600.0
__googleTokenKeyRetireTime = time.time() + 600.0 def __init__(self, src = 'en', dest = 'zh-CN', tkkUpdataTime = 600.0):
if src not in self._language and src not in self._language.values():
src = 'auto'
if dest not in self._language and dest not in self._language.values():
dest = 'auto'
self._params['sl'] = src
self._params['tl'] = dest
self.googleTokenKeyUpdataTime = tkkUpdataTime
self.__updateGoogleTokenKey() def __updateGoogleTokenKey(self):
self.__googleTokenKey = self.__getGoogleTokenKey()
self.__googleTokenKeyRetireTime = time.time() + self.__googleTokenKeyUpdataTime def __getGoogleTokenKey(self):
"""Get the Google TKK from https://translate.google.cn"""
# TKK example: '435075.3634891900'
result = ''
try:
res = requests.get('https://' + self._host, timeout = 3)
res.raise_for_status()
self.__cookies = res.cookies
result = re.search(r'tkk\:\'(\d+\.\d+)?\'', res.text).group(1)
except requests.exceptions.ReadTimeout as ex:
print('ERROR: ' + str(ex))
time.sleep(1)
return result def __getGoogleToken(self, a, TKK):
"""Calculate Google tk from TKK """
# https://www.cnblogs.com/chicsky/p/7443830.html
# if text = 'Tablet Developer' and TKK = '435102.3120524463', then tk = '315066.159012' def RL(a, b):
for d in range(0, len(b)-2, 3):
c = b[d + 2]
c = ord(c[0]) - 87 if 'a' <= c else int(c)
c = a >> c if '+' == b[d + 1] else a << c
a = a + c & 4294967295 if '+' == b[d] else a ^ c
return a g = []
f = 0
while f < len(a):
c = ord(a[f])
if 128 > c:
g.append(c)
else:
if 2048 > c:
g.append((c >> 6) | 192)
else:
if (55296 == (c & 64512)) and (f + 1 < len(a)) and (56320 == (ord(a[f+1]) & 64512)):
f += 1
c = 65536 + ((c & 1023) << 10) + (ord(a[f]) & 1023)
g.append((c >> 18) | 240)
g.append((c >> 12) & 63 | 128)
else:
g.append((c >> 12) | 224)
g.append((c >> 6) & 63 | 128)
g.append((c & 63) | 128)
f += 1 e = TKK.split('.')
h = int(e[0]) or 0
t = h
for item in g:
t += item
t = RL(t, '+-a^+6')
t = RL(t, '+-3^+b+-f')
t ^= int(e[1]) or 0
if 0 > t:
t = (t & 2147483647) + 2147483648
result = t % 1000000
return str(result) + '.' + str(result ^ h) def translate(self, text):
if time.time() > self.__googleTokenKeyRetireTime:
self.__updateGoogleTokenKey()
data = {'q': text}
self._params['tk'] = self.__getGoogleToken(text, self.__googleTokenKey)
result = ''
try:
res = requests.post(self._url,
headers = self._headers,
cookies = self.__cookies,
data = data,
params = self._params,
timeout = 6)
res.raise_for_status()
jsonText = res.text
if len(jsonText)>0:
jsonResult = json.loads(jsonText)
if len(jsonResult[0])>0:
for item in jsonResult[0]:
result += item[0]
return result
except Exception as ex:
print('ERROR: ' + str(ex))
return '' import time
from GoogleTranslator import GoogleTranslator def readFile(fileName):
with open(fileName, 'r') as f:
paragraph = ''
for line in f:
if line[0]!='\n':
paragraph += line.strip('\n')
else:
if len(paragraph)>0:
yield paragraph
paragraph = ''
if len(paragraph)>0:
yield paragraph

main.py:

def main():
translator = GoogleTranslator()
count = 0
with open('C:\\dx\\python\\d.txt', 'w', encoding='utf-8') as df:
for line in readFile('C:\\dx\\python\\s.txt'):
if len(line) > 1:
count += 1
print('\r' + str(count), end = '', flush = True)
df.write(line.strip() + "\n")
result = translator.translate(line)
df.write(result.strip() + "\n\n") if __name__ == "__main__":
startTime = time.time()
main()
print()
print('%.2f seconds' % (time.time() - startTime))

结束语

求人不如求己。不能怕烦,代码都是人敲出来的,找不到现成的还得靠自己编。

补充

高版本的word(至少2016可行)能直接打开PDF文件并自动转换格式,多个换行能够自动识别合并为段落。所以以上代码可以自行修改简化。

上一篇:使用python的Flask实现一个RESTful API服务器端[翻译]


下一篇:简单实现Python调用有道API接口(最新的)