MiniScripts/miss-bone.py

   1 #!/usr/bin/env python
   2
   3 '''
   4 This program is free software; you can redistribute it and/or modify
   5 it under the terms of the Revised BSD License.
   6
   7 This program is distributed in the hope that it will be useful,
   8 but WITHOUT ANY WARRANTY; without even the implied warranty of
   9 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  10 Revised BSD License for more details.
  11
  12 Copyright 2014 Cool Dude 2k - http://idb.berlios.de/
  13 Copyright 2014 Game Maker 2k - http://intdb.sourceforge.net/
  14 Copyright 2014 Kazuki Przyborowski - https://github.com/KazukiPrzyborowski
  15
  16 $FileInfo: miss-bone.py - Last Update: 6/12/2014 Ver. 1.0.0 RC 2 - Author: cooldude2k $
  17 '''
  18
  19 import re, os, sys, urllib, urllib2, cookielib, StringIO, gzip, time, datetime, argparse, urlparse;
  20
  21 fakeua = "Mozilla/5.0 (Windows NT 6.1; rv:24.0) Gecko/20100101 Firefox/24.0";
  22 geturls_cj = cookielib.CookieJar();
  23 geturls_opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(geturls_cj));
  24 geturls_opener.addheaders = [("Referer", "http://www.emoticonplus.com/miss-bone/"), ("User-Agent", fakeua), ("Accept-Encoding", "gzip, deflate"), ("Accept-Language", "en-US,en-CA,en-GB,en-UK,en-AU,en-NZ,en-ZA,en;q=0.5"), ("Accept-Charset", "ISO-8859-1,ISO-8859-15,utf-8;q=0.7,*;q=0.7"), ("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"), ("Connection", "close")];
  25 sfcountsub=0;
  26 sfcountfull=0;
  27 i = 1;
  28 il = 15;
  29 while(i < il):
  30  print(str(i)+" Reading URL: http://www.emoticonplus.com/miss-bone/page/"+str(i));
  31  geturls_text = geturls_opener.open("http://www.emoticonplus.com/miss-bone/page/"+str(i));
  32  if(geturls_text.info().get("Content-Encoding")=="gzip" or geturls_text.info().get("Content-Encoding")=="deflate"):
  33   strbuf = StringIO.StringIO(geturls_text.read());
  34   gzstrbuf = gzip.GzipFile(fileobj=strbuf);
  35   out_text = gzstrbuf.read()[:];
  36  if(geturls_text.info().get("Content-Encoding")!="gzip" and geturls_text.info().get("Content-Encoding")!="deflate"):
  37   out_text = geturls_text.read()[:];
  38  regex_text = re.escape("<a class=\"emoticon-list\" href=\"#\" data=\"")+"(.*?)"+re.escape("\" im=\"")+"([0-9]+)"+re.escape("\" category=\"miss-bone\">");
  39  post_text = re.findall(regex_text, out_text);
  40  isub=0;
  41  ilsub=len(post_text);
  42  print(str(i)+" Found "+str(ilsub)+" GIF Images.");
  43  while(isub < ilsub):
  44   getsub2xurls_cj = geturls_cj;
  45   getsub2xurls_opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(getsub2xurls_cj));
  46   getsub2xurls_opener.addheaders = [("Referer", "http://www.emoticonplus.com/miss-bone/page/"+str(i)), ("User-Agent", fakeua), ("Accept-Encoding", "gzip, deflate"), ("Accept-Language", "en-US,en-CA,en-GB,en-UK,en-AU,en-NZ,en-ZA,en;q=0.5"), ("Accept-Charset", "ISO-8859-1,ISO-8859-15,utf-8;q=0.7,*;q=0.7"), ("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"), ("Connection", "close")];
  47   print(str(i)+","+str(isub+1)+" Start Downloading Image File: "+str(isub+1)+" of "+str(ilsub));
  48   print(str(i)+","+str(isub+1)+" Downloading Image: "+post_text[isub][0]);
  49   getsub2xurls_text = getsub2xurls_opener.open(post_text[isub][0]);
  50   gif_file_name = os.path.basename(urlparse.urlparse(post_text[isub][0]).path);
  51   print(str(i)+","+str(isub+1)+" Finished Downloading Image: "+post_text[isub][0]);
  52   print(str(i)+","+str(isub+1)+" Saving File: ./"+gif_file_name);
  53   gifsf = open("./"+gif_file_name, "wb");
  54   gifsf.write(getsub2xurls_text.read());
  55   gifsf.close();
  56   isub=isub+1;
  57   sfcountsub=sfcountsub+1;
  58  print(str(i)+" Downloaded "+str(sfcountsub)+" GIF Images");
  59  sfcountfull=sfcountfull+sfcountsub;
  60  sfcountsub=0;
  61  i=i+1;
  62 print("Downloaded "+str(sfcountfull)+" GIF Images");
  63 sfcountfull=0;