hippopx网站图片下载-高清原图

funcdefmain

已于 2023-09-14 10:34:12 修改

阅读量1.5k

点赞数

分类专栏：爬虫文章标签： javascript python

于 2023-08-28 19:28:22 首次发布

本文链接：https://blog.csdn.net/qq_45444679/article/details/132545617

版权

爬虫专栏收录该内容

12 篇文章

订阅专栏

Hippopx 网站高清原图下载

关于Hippopx

https://www.hippopx.com/, Hippopx是一个免版权图库网站，收录超过20万张以上的免费CC0授权相片，图片的质量很高。高清摄影图片包含人物、动物、风景、名胜、美食、旅游等，都可以免费下载。

准备工作

在开始之前，确保你已经安装了Python和以下库：

requests
lxml
execjs

pip install requests lxml PyExecJS

分析目标网站

在首页输入关键词mountain进入搜索结果页，结果页点击图片进入图片详情页，

点击下载原图跳到了一个新的地址：

https://i0.hippopx.com/photos/174/613/190/mountain-landscape-mountains-landscape-steinweg-3c4a787b5f1473f26f31e3234a764984.jpg

在详情页匹配到预览图地址：

https://i0.hippopx.com/photos/174/613/190/mountain-landscape-mountains-landscape-steinweg-preview.jpg

发现唯一的区别：原图地址替换了"preview",怀疑是加密参数，发现下载按钮有id参数"dl"，

f12 在源代码页面“global.js”里面搜索dl,找到关键加密函数dld

function dld(f) {
    var e = g("s").innerHTML.split("x");
    var d = g("t").value;
    var c = g("i").src;
    var b = u(parseInt(d), parseInt(e[0]), parseInt(e[1]));
    var a = site + "/" + lang + "/download/" + location.href.split("-").pop() + "-" + f + "-" + b;
    location.href = a
}

打断点调试，发现b就是我们要找的加密参数，于是改写函数，扣js,一步步调试，完成该加密参数的构造。下面是完整加密的js

function k(d, a) {
    var c = (65535 & d) + (65535 & a);
    return (d >> 16) + (a >> 16) + (c >> 16) << 16 | 65535 & c
}

function z(c, a) {
    return c << a | c >>> 32 - a
}

function M(r, h, m, a, p, d) {
    return k(z(k(k(h, r), k(a, d)), p), m)
}

function E(s, d, e, m, a, p, h) {
    return M(d & e | ~d & m, s, d, a, p, h)
}

function j(s, d, e, m, a, p, h) {
    return M(d & m | e & ~m, s, d, a, p, h)
}

function O(s, d, e, m, a, p, h) {
    return M(d ^ e ^ m, s, d, a, p, h)
}

function L(s, d, e, m, a, p, h) {
    return M(e ^ (d | ~m), s, d, a, p, h)
}

function N(f) {
    var a, c = [];
    for (c[(f.length >> 2) - 1] = void 0,
             a = 0; a < c.length; a += 1) {
        c[a] = 0
    }
    var d = 8 * f.length;
    for (a = 0; a < d; a += 8) {
        c[a >> 5] |= (255 & f.charCodeAt(a / 8)) << a % 32
    }
    return c
}

function I(f, c) {
    f[c >> 5] |= 128 << c % 32,
        f[14 + (c + 64 >>> 9 << 4)] = c;
    var C, s, R, Q, t, p = 1732584193, A = -271733879, S = -1732584194, o = 271733878;
    for (C = 0; C < f.length; C += 16) {
        s = p,
            R = A,
            Q = S,
            t = o,
            A = L(A = L(A = L(A = L(A = O(A = O(A = O(A = O(A = j(A = j(A = j(A = j(A = E(A = E(A = E(A = E(A, S = E(S, o = E(o, p = E(p, A, S, o, f[C], 7, -680876936), A, S, f[C + 1], 12, -389564586), p, A, f[C + 2], 17, 606105819), o, p, f[C + 3], 22, -1044525330), S = E(S, o = E(o, p = E(p, A, S, o, f[C + 4], 7, -176418897), A, S, f[C + 5], 12, 1200080426), p, A, f[C + 6], 17, -1473231341), o, p, f[C + 7], 22, -45705983), S = E(S, o = E(o, p = E(p, A, S, o, f[C + 8], 7, 1770035416), A, S, f[C + 9], 12, -1958414417), p, A, f[C + 10], 17, -42063), o, p, f[C + 11], 22, -1990404162), S = E(S, o = E(o, p = E(p, A, S, o, f[C + 12], 7, 1804603682), A, S, f[C + 13], 12, -40341101), p, A, f[C + 14], 17, -1502002290), o, p, f[C + 15], 22, 1236535329), S = j(S, o = j(o, p = j(p, A, S, o, f[C + 1], 5, -165796510), A, S, f[C + 6], 9, -1069501632), p, A, f[C + 11], 14, 643717713), o, p, f[C], 20, -373897302), S = j(S, o = j(o, p = j(p, A, S, o, f[C + 5], 5, -701558691), A, S, f[C + 10], 9, 38016083), p, A, f[C + 15], 14, -660478335), o, p, f[C + 4], 20, -405537848), S = j(S, o = j(o, p = j(p, A, S, o, f[C + 9], 5, 568446438), A, S, f[C + 14], 9, -1019803690), p, A, f[C + 3], 14, -187363961), o, p, f[C + 8], 20, 1163531501), S = j(S, o = j(o, p = j(p, A, S, o, f[C + 13], 5, -1444681467), A, S, f[C + 2], 9, -51403784), p, A, f[C + 7], 14, 1735328473), o, p, f[C + 12], 20, -1926607734), S = O(S, o = O(o, p = O(p, A, S, o, f[C + 5], 4, -378558), A, S, f[C + 8], 11, -2022574463), p, A, f[C + 11], 16, 1839030562), o, p, f[C + 14], 23, -35309556), S = O(S, o = O(o, p = O(p, A, S, o, f[C + 1], 4, -1530992060), A, S, f[C + 4], 11, 1272893353), p, A, f[C + 7], 16, -155497632), o, p, f[C + 10], 23, -1094730640), S = O(S, o = O(o, p = O(p, A, S, o, f[C + 13], 4, 681279174), A, S, f[C], 11, -358537222), p, A, f[C + 3], 16, -722521979), o, p, f[C + 6], 23, 76029189), S = O(S, o = O(o, p = O(p, A, S, o, f[C + 9], 4, -640364487), A, S, f[C + 12], 11, -421815835), p, A, f[C + 15], 16, 530742520), o, p, f[C + 2], 23, -995338651), S = L(S, o = L(o, p = L(p, A, S, o, f[C], 6, -198630844), A, S, f[C + 7], 10, 1126891415), p, A, f[C + 14], 15, -1416354905), o, p, f[C + 5], 21, -57434055), S = L(S, o = L(o, p = L(p, A, S, o, f[C + 12], 6, 1700485571), A, S, f[C + 3], 10, -1894986606), p, A, f[C + 10], 15, -1051523), o, p, f[C + 1], 21, -2054922799), S = L(S, o = L(o, p = L(p, A, S, o, f[C + 8], 6, 1873313359), A, S, f[C + 15], 10, -30611744), p, A, f[C + 6], 15, -1560198380), o, p, f[C + 13], 21, 1309151649), S = L(S, o = L(o, p = L(p, A, S, o, f[C + 4], 6, -145523070), A, S, f[C + 11], 10, -1120210379), p, A, f[C + 2], 15, 718787259), o, p, f[C + 9], 21, -343485551),
            p = k(p, s),
            A = k(A, R),
            S = k(S, Q),
            o = k(o, t)
    }
    return [p, A, S, o]
}

function P(f) {
    var a, c = "", d = 32 * f.length;
    for (a = 0; a < d; a += 8) {
        c += String.fromCharCode(f[a >> 5] >>> a % 32 & 255)
    }
    return c
}

function b(a) {
    return unescape(encodeURIComponent(a))
}

function J(a) {
    return P(I(N(a), 8 * a.length))
}

function K(f) {
    var a, c, d = "";
    for (c = 0; c < f.length; c += 1) {
        a = f.charCodeAt(c),
            d += "0123456789abcdef".charAt(a >>> 4 & 15) + "0123456789abcdef".charAt(15 & a)
    }
    return d
}

function G(a) {
    return J(b(a))
}

function D(a) {
    return K(G(a))
}

function q(c, a) {
    return H(b(c), b(a))
}

function w(c, a) {
    return K(q(c, a))
}

function ec(d, a, c) {
    return a ? c ? q(a, d) : w(a, d) : c ? G(d) : D(d)
}

function u(a, f, c) {
    var d = "";
    if (a > 0 && f > 0 && c > 0) {
        var n = ec(a);
        var k = ec(f + c);
        var j = n.split("").reverse();
        var e = k.split("");
        for (var b = 0; b < 32; b++) {
            if (b % 2 == 0) {
                d += j[b]
            } else {
                d += e[b]
            }
        }
    }
    return d
}

function dld(f, d) {
    // // 图片尺寸
    // var f= "7599x3215"
    // // 上传日期后的隐藏数字 //span//input/@value
    // var d = "1510317657";
    var e = f.split("x");
    var b = u(parseInt(d), parseInt(e[0]), parseInt(e[1]));
    return b
}

python调用js

import execjs


def encrypt_download_url(preview_url, reso, hidden_value):
    with open('global.js', encoding='utf-8') as f:
        js_code = f.read()
    context = execjs.compile(js_code)
    ret = context.call("dld", reso, hidden_value)
    download_url = preview_url.replace('preview', ret)
    return download_url


if __name__ == '__main__':
    print(encrypt_download_url(
        preview_url="https://i0.hippopx.com/photos/88/296/962/clouds-dark-landscape-mountain-range-preview.jpg",
        reso="7599x3215",
        hidden_value="1510317657"
    ))

批量下载

import os.path
import traceback

import requests
from lxml import etree

from encrypt import encrypt_download_url
from spider import Spider


def run(query, page, save_dir):
    headers = {
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/115.0.0.0 Safari/537.36"
    }

    url = "https://www.hippopx.com/en/search"
    params = {
        'q': query,
        'page': page
    }
    resp = requests.get(url, params=params, headers=headers)
    html = etree.HTML(resp.text)
    src_list = html.xpath('//ul//a[@itemprop="url"]/@href')
    print(src_list)
    task_list = []
    for src in src_list:
        resp = requests.get(url=src, headers=headers)
        html = etree.HTML(resp.text)
        preview_url = html.xpath('//*[@id="i"]/@src')[0]
        reso = html.xpath('//*[@id="s"]/text()')[0]
        size_str = html.xpath('//ul[@class="info_list"]/li[2]/span[@class="info_detail"]/text()')[0].strip()
        hidden_value = html.xpath('//*[@id="t"]/@value')[0]
        download_url = encrypt_download_url(preview_url, reso, hidden_value)
        save_name = download_url.split('/')[-1]
        print(download_url, size_str)
        save_path = os.path.join(save_dir, save_name)
        task_list.append((download_url, save_path))
    Spider(task_list=task_list, thread_num=3).run()


if __name__ == '__main__':
    keyword = "mountain"
    for page in range(1, 334):
        print(f'下载第{page}页数据')
        try:
            run(
                query=keyword,
                page=page,
                save_dir=f'./{keyword}'
            )
        except:
            print(traceback.format_exc())
            continue