1 import sys
   2 import os
   3 import string
   4 
   5 headers = [('JFIF', 6, 'jpg'), ('GIF', 0, 'gif'), ('PNG', 1, 'png')]
   6 # headers 中的offset为什么分别是6， 0， 1，是标识前面的数据长度吗？是什么数据？
   7 marker = []
   8 filename = '/path/to/a/file'
   9 
  10 try:
  11    fid = open(filename, 'rb')
  12 except:
  13    sys.exit(1)
  14 
  15 s = 0
  16 
  17 for line in fid:
  18    for flag, offset, ext in headers:
  19        index = string.find(line, flag)
  20        if index > 0:
  21            pos = s + index - offset
  22            marker.append((pos, ext))
  23    s += len(line)
  24 
  25 fid.seek(0)
  26 j = len(marker)
  27 imgnum = 0
  28 if j == 0:
  29    print 'No images included in this document'
  30    sys.exit(1)
  31 
  32 for i in range(j):
  33    info = marker[i]
  34    thispos = info[0]
  35    thisext = info[1]
  36    if i == j-1:
  37        nextpos = s
  38        gap = nextpos - thispos
  39    else:
  40        nextinfo = marker[i+1]
  41        nextpos = nextinfo[0]
  42        gap = nextpos - thispos
  43    fid.seek(thispos)
  44    data = fid.read(gap)
  45    imgname = 'imgname%02d.%s' % (i, thisext)
  46    fid1 = open(imgname, 'wb')
  47    fid1.write(data)
  48    fid1.close()
  49    imgnum += 1
  50 
  51 fid.close()
  52 print '%02d images have been extracted' % imgnum