# PYTHON 2.7   Ubuntu 14.04
nuswide = "$NUS-WIDE-urls_ROOT" #the location of your nus-wide-urls.txt
imagepath = "$IMAGE_ROOT" # path of dataset you want to download in
f = open(nuswide, 'r')
url = f.readlines()
import re
import urllib
import os
reg = r"ImageData.+?jpg"
location_re = re.compile(reg)
reg = r"(ImageData.+?)/0"
direction_re = re.compile(reg)
reg = r"http.+?jpg"
image_re = re.compile(reg)
for i in url:
  filename = re.findall(location_re, i)
  direction = re.findall(direction_re, i)
  image = re.findall(image_re, i)
  if image:
    path = imagepath+filename[0]
    path_n = imagepath+direction[0]
    print path_n
    if os.path.exists(path_n):
      urllib.urlretrieve(image[1], path)
    else:
      os.makedirs(path_n)
      urllib.urlretrieve(image[1], path)

#coding=utf-8

#urllib模塊提供了讀取Web頁面數(shù)據(jù)的接口
import urllib
#re模塊主要包含了正則表達(dá)式
import re
#定義一個(gè)getHtml()函數(shù)
def getHtml(url):
  page = urllib.urlopen(url) #urllib.urlopen()方法用于打開一個(gè)URL地址
  html = page.read() #read()方法用于讀取URL上的數(shù)據(jù)
  return html

def getImg(html):
  reg = r'src="(.+?\.jpg)" pic_ext'  #正則表達(dá)式，得到圖片地址
  imgre = re.compile(reg)   #re.compile() 可以把正則表達(dá)式編譯成一個(gè)正則表達(dá)式對(duì)象.
  imglist = re.findall(imgre,html)   #re.findall() 方法讀取html 中包含 imgre（正則表達(dá)式）的  數(shù)據(jù)
  #把篩選的圖片地址通過for循環(huán)遍歷并保存到本地
  #核心是urllib.urlretrieve()方法,直接將遠(yuǎn)程數(shù)據(jù)下載到本地，圖片通過x依次遞增命名
  x = 0

  for imgurl in imglist:
  urllib.urlretrieve(imgurl,'D:\E\%s.jpg' % x)
      x+=1


html = getHtml("http://tieba.baidu.com/p/xxxx")
print getImg(html)