js逆向爬虫3
定位需要的返回数据如上图,查看它的请求参数,api为v2transapi开头
搜索上面几个参数,其中最容易定位的是simple_means_flag,只搜索到一个结果
在源代码中打开,搜索simple_means_flag得到两个结果,都打上断点刷新调试;发现跳转到了第二个,说明这个才是关键加密函数
通过多次刷新发现,除了sign其他参数均为固定值,单步进入b函数,将整个函数复制下来
创建了个获取请求参数的函数,返回结果为一个字典变量
运行查看结果:41行的r未定义
找到源码中对应的位置打断点,查看r值,发现r值为空
故在41行前添加如下代码后再次运行
var r = null;
报错说window变量未定义,光标移动到window变量上,在弹窗中找到gtk的值(d变量的值为gtk)
在40行前添加如下代码:
var window = {'gtk': '320305.131321201'}
如图,获取了全部参数
完整代码:
baidu.py
import requests
import subprocess
from functools import partial
subprocess.Popen = partial(subprocess.Popen, encoding="utf-8")
import execjs
translate_words = input('请输入要翻译的内容:')
ctx = execjs.compile(open('baidu.js', encoding='utf-8').read())
params = ctx.call('get_req_data', translate_words)
# 获取输入语言类型
detect_lang = requests.post('https://fanyi.baidu.com/langdetect', data={'query': translate_words}).json()
if detect_lang['lan'] == 'en':
params['from'] = 'en'
params['to'] = 'zh'
headers = {
'Cookie': 'your cookie',
}
results = requests.post('https://fanyi.baidu.com/v2transapi?from=zh&to=en', data=params, headers=headers).json()
print(results['trans_result']['data'][0]['dst'])
baidu.js
function n(t, e) {
for (var n = 0; n < e.length - 2; n += 3) {
var r = e.charAt(n + 2);
r = "a" <= r ? r.charCodeAt(0) - 87 : Number(r),
r = "+" === e.charAt(n + 1) ? t >>> r : t << r,
t = "+" === e.charAt(n) ? t + r & 4294967295 : t ^ r
}
return t
}
function b(t) {
var o, i = t.match(/[\uD800-\uDBFF][\uDC00-\uDFFF]/g);
if (null === i) {
var a = t.length;
a > 30 && (t = "".concat(t.substr(0, 10)).concat(t.substr(Math.floor(a / 2) - 5, 10)).concat(t.substr(-10, 10)))
}
else {
for (var s = t.split(/[\uD800-\uDBFF][\uDC00-\uDFFF]/), c = 0, u = s.length, l = []; c < u; c++)
"" !== s[c] && l.push.apply(l, function(t) {
if (Array.isArray(t))
return e(t)
}(o = s[c].split("")) || function(t) {
if ("undefined" != typeof Symbol && null != t[Symbol.iterator] || null != t["@@iterator"])
return Array.from(t)
}(o) || function(t, n) {
if (t) {
if ("string" == typeof t)
return e(t, n);
var r = Object.prototype.toString.call(t).slice(8, -1);
return "Object" === r && t.constructor && (r = t.constructor.name),
"Map" === r || "Set" === r ? Array.from(t) : "Arguments" === r || /^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(r) ? e(t, n) : void 0
}
}(o) || function() {
throw new TypeError("Invalid attempt to spread non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method.")
}()),
c !== u - 1 && l.push(i[c]);
var p = l.length;
p > 30 && (t = l.slice(0, 10).join("") + l.slice(Math.floor(p / 2) - 5, Math.floor(p / 2) + 5).join("") + l.slice(-10).join(""))
}
var window = {'gtk': '320305.131321201'}
var r = null;
for (var d = "".concat(String.fromCharCode(103)).concat(String.fromCharCode(116)).concat(String.fromCharCode(107)), h = (null !== r ? r : (r = window[d] || "") || "").split("."), f = Number(h[0]) || 0, m = Number(h[1]) || 0, g = [], y = 0, v = 0; v < t.length; v++) {
var _ = t.charCodeAt(v);
_ < 128 ? g[y++] = _ : (_ < 2048 ? g[y++] = _ >> 6 | 192 : (55296 == (64512 & _) && v + 1 < t.length && 56320 == (64512 & t.charCodeAt(v + 1)) ? (_ = 65536 + ((1023 & _) << 10) + (1023 & t.charCodeAt(++v)),
g[y++] = _ >> 18 | 240,
g[y++] = _ >> 12 & 63 | 128) : g[y++] = _ >> 12 | 224,
g[y++] = _ >> 6 & 63 | 128),
g[y++] = 63 & _ | 128)
}
for (var b = f, w = "".concat(String.fromCharCode(43)).concat(String.fromCharCode(45)).concat(String.fromCharCode(97)) + "".concat(String.fromCharCode(94)).concat(String.fromCharCode(43)).concat(String.fromCharCode(54)), k = "".concat(String.fromCharCode(43)).concat(String.fromCharCode(45)).concat(String.fromCharCode(51)) + "".concat(String.fromCharCode(94)).concat(String.fromCharCode(43)).concat(String.fromCharCode(98)) + "".concat(String.fromCharCode(43)).concat(String.fromCharCode(45)).concat(String.fromCharCode(102)), x = 0; x < g.length; x++)
b = n(b += g[x], w);
return b = n(b, k),
(b ^= m) < 0 && (b = 2147483648 + (2147483647 & b)),
"".concat((b %= 1e6).toString(), ".").concat(b ^ f)
}
function get_req_data(query){
const w = {
from: 'zh',
to: 'en',
query: query,
transtype: '',
simple_means_flag: 3,
sign: b(query),
token: "d21ee97178386951469ceda3f0d7caf7",
domain: 'common'
}
return w;
}
console.log(get_req_data('今天是周三'))
小结:
- 百度翻译接口破解也很简单,和有道翻译相比要简单些,大部分参数都是固定的,只加密了一个;
- 翻译上来看貌似有道更生动些:smile:,百度甚至把must be直译成了必须是