公司最近有翻譯的需求，發現Google的Translate API只能FQ調，多方查詢彙集眾家之長，找到一個感覺最好用的，稍稍改進了一下，原址：

https://blog.csdn.net/andeyeluguo/article/details/78581590

很多部落格有提到tk值要js處理，不多說，直接上乾貨

首先，是要pip處理js的模組  

pip install PyExecJS

建立HandleJs.py

 1 import execjs  
 2   
 3 class Py4Js():  
 4       
 5     def __init__(self):  
 
 6         self.ctx = execjs.compile(""" 
 7         function TL(a) { 
 8         var k = ""; 
 9         var b = 406644; 
10         var b1 = 3293161072; 
11          
12         var jd = "."; 
13         var $b = "+-a^+6"; 
14         var Zb = "+-3^+b+-f"; 
15      
16         for (var e = [], f = 0, g = 0
 
; g < a.length; g++) { 
17             var m = a.charCodeAt(g); 
18             128 > m ? e[f++] = m : (2048 > m ? e[f++] = m >> 6 | 192 : (55296 == (m & 64512) && g + 1 < a.length && 56320 == (a.charCodeAt(g + 1) & 64512) ? (m = 65536 + ((m & 1023) << 10
 
) + (a.charCodeAt(++g) & 1023), 
19             e[f++] = m >> 18 | 240, 
20             e[f++] = m >> 12 & 63 | 128) : e[f++] = m >> 12 | 224, 
21             e[f++] = m >> 6 & 63 | 128), 
22             e[f++] = m & 63 | 128) 
23         } 
24         a = b; 
25         for (f = 0; f < e.length; f++) a += e[f], 
26         a = RL(a, $b); 
27         a = RL(a, Zb); 
28         a ^= b1 || 0; 
29         0 > a && (a = (a & 2147483647) + 2147483648); 
30         a %= 1E6; 
31         return a.toString() + jd + (a ^ b) 
32     }; 
33      
34     function RL(a, b) { 
35         var t = "a"; 
36         var Yb = "+"; 
37         for (var c = 0; c < b.length - 2; c += 3) { 
38             var d = b.charAt(c + 2), 
39             d = d >= t ? d.charCodeAt(0) - 87 : Number(d), 
40             d = b.charAt(c + 1) == Yb ? a >>> d: a << d; 
41             a = b.charAt(c) == Yb ? a + d & 4294967295 : a ^ d 
42         } 
43         return a 
44     } 
45     """)  
46           
47     def getTk(self,text):  
48         return self.ctx.call("TL",text)

建立main.py

 1 import urllib.request
 2 import urllib.parse
 3 from HandleJs import Py4Js
 4 
 5 
 6 def open_url(url):
 7     headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:23.0) Gecko/20100101 Firefox/23.0'}
 8     req = urllib.request.Request(url=url, headers=headers)
 9     response = urllib.request.urlopen(req)
10     data = response.read().decode('utf-8')
11     return data
12 
13 
14 def buildUrl(content, tk, tl):
15     baseUrl = 'http://translate.google.cn/translate_a/single'
16     baseUrl += '?client=t&'
17     baseUrl += 'sl=auto&'
18     baseUrl += 'tl=' + str(tl) + '&'
19     baseUrl += 'hl=zh-CN&'
20     baseUrl += 'dt=at&'
21     baseUrl += 'dt=bd&'
22     baseUrl += 'dt=ex&'
23     baseUrl += 'dt=ld&'
24     baseUrl += 'dt=md&'
25     baseUrl += 'dt=qca&'
26     baseUrl += 'dt=rw&'
27     baseUrl += 'dt=rm&'
28     baseUrl += 'dt=ss&'
29     baseUrl += 'dt=t&'
30     baseUrl += 'ie=UTF-8&'
31     baseUrl += 'oe=UTF-8&'
32     baseUrl += 'clearbtn=1&'
33     baseUrl += 'otf=1&'
34     baseUrl += 'pc=1&'
35     baseUrl += 'srcrom=0&'
36     baseUrl += 'ssel=0&'
37     baseUrl += 'tsel=0&'
38     baseUrl += 'kc=2&'
39     baseUrl += 'tk=' + str(tk) + '&'
40     baseUrl += 'q=' + content
41     return baseUrl
42 
43 
44 def translate(content, tk, tl):
45     if len(content) > 4891:
46         print("翻譯的長度超過限制！！！")
47         return
48 
49     content = urllib.parse.quote(content)
50     url = buildUrl(content, tk, tl)
51 
52     # 返回值是一個多層巢狀列表的字串形式，解析起來還相當費勁，寫了幾個正則，發現也很不理想，
53     # 後來感覺，使用正則簡直就是把簡單的事情複雜化，這裡直接切片就Ok了
54     result = open_url(url)
55 
56     end = result.find("\",")
57     if end > 4:
58         print(result[4:end])
59 
60 
61 def main():
62     js = Py4Js()
63     #content是要翻譯的內容
64     #tl是要翻譯的目標語種，值參照ISO 639-1標準，如果翻譯成中文"zh/zh-CN簡體中文"
65     content = "hello"
66     tl = "zh"
67     tk = js.getTk(content)
68     translate(content, tk, tl)
69 
70 
71 if __name__ == "__main__":
72     main()