|
| 1 | +#!/usr/bin/env python |
| 2 | +# -*- coding: utf-8 -*- |
| 3 | +# @Author: loveNight |
| 4 | +# @Date: 2015-10-28 19:59:24 |
| 5 | +# @Last Modified by: loveNight |
| 6 | +# @Last Modified time: 2015-11-15 19:43:20 |
| 7 | + |
| 8 | + |
| 9 | +import urllib |
| 10 | +import requests |
| 11 | +import os |
| 12 | +import re |
| 13 | +import sys |
| 14 | +import time |
| 15 | +import threading |
| 16 | +from datetime import datetime as dt |
| 17 | +from multiprocessing.dummy import Pool |
| 18 | +from multiprocessing import Queue |
| 19 | + |
| 20 | + |
| 21 | +class BaiduImgDownloader(object): |
| 22 | + |
| 23 | + """百度图片搜索下载工具，目前只支持单个关键词""" |
| 24 | + |
| 25 | + # 解码网址用的映射表 |
| 26 | + str_table = { |
| 27 | + '_z2C$q': ':', |
| 28 | + '_z&e3B': '.', |
| 29 | + 'AzdH3F': '/' |
| 30 | + } |
| 31 | + |
| 32 | + char_table = { |
| 33 | + 'w': 'a', |
| 34 | + 'k': 'b', |
| 35 | + 'v': 'c', |
| 36 | + '1': 'd', |
| 37 | + 'j': 'e', |
| 38 | + 'u': 'f', |
| 39 | + '2': 'g', |
| 40 | + 'i': 'h', |
| 41 | + 't': 'i', |
| 42 | + '3': 'j', |
| 43 | + 'h': 'k', |
| 44 | + 's': 'l', |
| 45 | + '4': 'm', |
| 46 | + 'g': 'n', |
| 47 | + '5': 'o', |
| 48 | + 'r': 'p', |
| 49 | + 'q': 'q', |
| 50 | + '6': 'r', |
| 51 | + 'f': 's', |
| 52 | + 'p': 't', |
| 53 | + '7': 'u', |
| 54 | + 'e': 'v', |
| 55 | + 'o': 'w', |
| 56 | + '8': '1', |
| 57 | + 'd': '2', |
| 58 | + 'n': '3', |
| 59 | + '9': '4', |
| 60 | + 'c': '5', |
| 61 | + 'm': '6', |
| 62 | + '0': '7', |
| 63 | + 'b': '8', |
| 64 | + 'l': '9', |
| 65 | + 'a': '0' |
| 66 | + } |
| 67 | + |
| 68 | + re_objURL = re.compile(r'"objURL":"(.*?)".*?"type":"(.*?)"') |
| 69 | + re_downNum = re.compile(r"已下载\s(\d+)\s张图片") |
| 70 | + headers = { |
| 71 | + "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/46.0.2490.71 Safari/537.36", |
| 72 | + "Accept-Encoding": "gzip, deflate, sdch", |
| 73 | + } |
| 74 | + |
| 75 | + def __init__(self, word, dirpath=None, processNum=30): |
| 76 | + if " " in word: |
| 77 | + raise AttributeError("本脚本仅支持单个关键字") |
| 78 | + self.word = word |
| 79 | + self.char_table = {ord(key): ord(value) |
| 80 | + for key, value in BaiduImgDownloader.char_table.items()} |
| 81 | + if not dirpath: |
| 82 | + dirpath = os.path.join(sys.path[0], 'results') |
| 83 | + self.dirpath = dirpath |
| 84 | + self.jsonUrlFile = os.path.join(sys.path[0], 'jsonUrl.txt') |
| 85 | + self.logFile = os.path.join(sys.path[0], 'logInfo.txt') |
| 86 | + self.errorFile = os.path.join(sys.path[0], 'errorUrl.txt') |
| 87 | + if os.path.exists(self.errorFile): |
| 88 | + os.remove(self.errorFile) |
| 89 | + if not os.path.exists(self.dirpath): |
| 90 | + os.mkdir(self.dirpath) |
| 91 | + self.pool = Pool(30) |
| 92 | + self.session = requests.Session() |
| 93 | + self.session.headers = BaiduImgDownloader.headers |
| 94 | + self.queue = Queue() |
| 95 | + self.messageQueue = Queue() |
| 96 | + self.index = 0 # 图片起始编号，牵涉到计数，不要更改 |
| 97 | + self.promptNum = 10 # 下载几张图片提示一次 |
| 98 | + self.lock = threading.Lock() |
| 99 | + self.delay = 1.5 # 网络请求太频繁会被封 |
| 100 | + self.QUIT = "QUIT" # Queue中表示任务结束 |
| 101 | + self.printPrefix = "**" # 用于指定在控制台输出 |
| 102 | + |
| 103 | + def start(self): |
| 104 | + # 控制台输出线程 |
| 105 | + t = threading.Thread(target=self.__log) |
| 106 | + t.setDaemon(True) |
| 107 | + t.start() |
| 108 | + self.messageQueue.put(self.printPrefix + "脚本开始执行") |
| 109 | + start_time = dt.now() |
| 110 | + urls = self.__buildUrls() |
| 111 | + self.messageQueue.put(self.printPrefix + "已获取 %s 个Json请求网址" % len(urls)) |
| 112 | + # 解析出所有图片网址，该方法会阻塞直到任务完成 |
| 113 | + self.pool.map(self.__resolveImgUrl, urls) |
| 114 | + while self.queue.qsize(): |
| 115 | + imgs = self.queue.get() |
| 116 | + self.pool.map_async(self.__downImg, imgs) |
| 117 | + self.pool.close() |
| 118 | + self.pool.join() |
| 119 | + self.messageQueue.put(self.printPrefix + "下载完成！已下载 %s 张图片，总用时 %s" % |
| 120 | + (self.index, dt.now() - start_time)) |
| 121 | + self.messageQueue.put(self.printPrefix + "请到 %s 查看结果！" % self.dirpath) |
| 122 | + self.messageQueue.put(self.printPrefix + "错误信息保存在 %s" % self.errorFile) |
| 123 | + self.messageQueue.put(self.QUIT) |
| 124 | + |
| 125 | + |
| 126 | + def __log(self): |
| 127 | + """控制台输出，加锁以免被多线程打乱""" |
| 128 | + with open(self.logFile, "w", encoding = "utf-8") as f: |
| 129 | + while True: |
| 130 | + message = self.messageQueue.get() |
| 131 | + if message == self.QUIT: |
| 132 | + break |
| 133 | + message = str(dt.now()) + " " + message |
| 134 | + if self.printPrefix in message: |
| 135 | + print(message) |
| 136 | + elif "已下载" in message: |
| 137 | + # 下载N张图片提示一次 |
| 138 | + downNum = self.re_downNum.findall(message) |
| 139 | + if downNum and int(downNum[0]) % self.promptNum == 0: |
| 140 | + print(message) |
| 141 | + f.write(message + '\n') |
| 142 | + f.flush() |
| 143 | + |
| 144 | + def __getIndex(self): |
| 145 | + """获取文件编号""" |
| 146 | + self.lock.acquire() |
| 147 | + try: |
| 148 | + return self.index |
| 149 | + finally: |
| 150 | + self.index += 1 |
| 151 | + self.lock.release() |
| 152 | + |
| 153 | + def decode(self, url): |
| 154 | + """解码图片URL |
| 155 | + 解码前： |
| 156 | + ippr_z2C$qAzdH3FAzdH3Ffl_z&e3Bftgwt42_z&e3BvgAzdH3F4omlaAzdH3Faa8W3ZyEpymRmx3Y1p7bb&mla |
| 157 | + 解码后： |
| 158 | + http://s9.sinaimg.cn/mw690/001WjZyEty6R6xjYdtu88&690 |
| 159 | + """ |
| 160 | + # 先替换字符串 |
| 161 | + for key, value in self.str_table.items(): |
| 162 | + url = url.replace(key, value) |
| 163 | + # 再替换剩下的字符 |
| 164 | + return url.translate(self.char_table) |
| 165 | + |
| 166 | + def __buildUrls(self): |
| 167 | + """json请求网址生成器""" |
| 168 | + word = urllib.parse.quote(self.word) |
| 169 | + url = r"http://image.baidu.com/search/acjson?tn=resultjson_com&ipn=rj&ct=201326592&fp=result&queryWord={word}&cl=2&lm=-1&ie=utf-8&oe=utf-8&st=-1&ic=0&word={word}&face=0&istype=2nc=1&pn={pn}&rn=60" |
| 170 | + time.sleep(self.delay) |
| 171 | + html = self.session.get(url.format(word=word, pn=0), timeout = 15).content.decode('utf-8') |
| 172 | + results = re.findall(r'"displayNum":(\d+),', html) |
| 173 | + maxNum = int(results[0]) if results else 0 |
| 174 | + urls = [url.format(word=word, pn=x) |
| 175 | + for x in range(0, maxNum + 1, 60)] |
| 176 | + with open(self.jsonUrlFile, "w", encoding="utf-8") as f: |
| 177 | + for url in urls: |
| 178 | + f.write(url + "\n") |
| 179 | + return urls |
| 180 | + |
| 181 | + def __resolveImgUrl(self, url): |
| 182 | + """从指定网页中解析出图片URL""" |
| 183 | + time.sleep(self.delay) |
| 184 | + html = self.session.get(url, timeout = 15).content.decode('utf-8') |
| 185 | + datas = self.re_objURL.findall(html) |
| 186 | + imgs = [Image(self.decode(x[0]), x[1]) for x in datas] |
| 187 | + self.messageQueue.put(self.printPrefix + "已解析出 %s 个图片网址" % len(imgs)) |
| 188 | + self.queue.put(imgs) |
| 189 | + |
| 190 | + def __downImg(self, img): |
| 191 | + """下载单张图片，传入的是Image对象""" |
| 192 | + imgUrl = img.url |
| 193 | + # self.messageQueue.put("线程 %s 正在下载 %s " % |
| 194 | + # (threading.current_thread().name, imgUrl)) |
| 195 | + try: |
| 196 | + time.sleep(self.delay) |
| 197 | + res = self.session.get(imgUrl, timeout = 15) |
| 198 | + message = None |
| 199 | + if str(res.status_code)[0] == "4": |
| 200 | + message = "\n%s： %s" % (res.status_code, imgUrl) |
| 201 | + elif "text/html" in res.headers["Content-Type"]: |
| 202 | + message = "\n无法打开图片： %s" % imgUrl |
| 203 | + except Exception as e: |
| 204 | + message = "\n抛出异常： %s\n%s" % (imgUrl, str(e)) |
| 205 | + finally: |
| 206 | + if message: |
| 207 | + self.messageQueue.put(message) |
| 208 | + self.__saveError(message) |
| 209 | + return |
| 210 | + index = self.__getIndex() |
| 211 | + # index从0开始 |
| 212 | + self.messageQueue.put("已下载 %s 张图片：%s" % (index + 1, imgUrl)) |
| 213 | + filename = os.path.join(self.dirpath, str(index) + "." + img.type) |
| 214 | + with open(filename, "wb") as f: |
| 215 | + f.write(res.content) |
| 216 | + |
| 217 | + def __saveError(self, message): |
| 218 | + self.lock.acquire() |
| 219 | + try: |
| 220 | + with open(self.errorFile, "a", encoding="utf-8") as f: |
| 221 | + f.write(message) |
| 222 | + finally: |
| 223 | + self.lock.release() |
| 224 | + |
| 225 | + |
| 226 | +class Image(object): |
| 227 | + |
| 228 | + """图片类，保存图片信息""" |
| 229 | + |
| 230 | + def __init__(self, url, type): |
| 231 | + super(Image, self).__init__() |
| 232 | + self.url = url |
| 233 | + self.type = type |
| 234 | + |
| 235 | + |
| 236 | +if __name__ == '__main__': |
| 237 | + print("欢迎使用百度图片下载脚本！\n目前仅支持单个关键词。") |
| 238 | + print("下载结果保存在脚本目录下的results文件夹中。") |
| 239 | + print("=" * 50) |
| 240 | + word = input("请输入你要下载的图片关键词：\n") |
| 241 | + down = BaiduImgDownloader(word) |
| 242 | + down.start() |
0 commit comments