pywwwget-dl.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 2016-2023 Game Maker 2k - https://github.com/GameMaker2k
  13     Copyright 2016-2023 Kazuki Przyborowski - https://github.com/KazukiPrzyborowski
  14
  15     $FileInfo: pywwwget-dl.py - Last Update: 10/5/2023 Ver. 2.0.2 RC 1 - Author: cooldude2k $
  16 '''
  17
  18 from __future__ import division, absolute_import, print_function;
  19 import re, os, sys, pywwwget, argparse;
  20 import logging as log;
  21
  22 __project__ = pywwwget.__project__;
  23 __program_name__ = pywwwget.__program_name__;
  24 __project_url__ = pywwwget.__project_url__;
  25 __version_info__ = pywwwget.__version_info__;
  26 __version_date_info__ = pywwwget.__version_date_info__;
  27 __version_date__ = pywwwget.__version_date__;
  28 __version_date_plusrc__ = pywwwget.__version_date_plusrc__
  29 __version__ = pywwwget.__version__;
  30 __version_date_plusrc__ = pywwwget.__version_date_plusrc__;
  31
  32 geturls_cj = pywwwget.geturls_cj;
  33 geturls_ua = pywwwget.geturls_ua;
  34 geturls_ua_firefox_windows7 = pywwwget.geturls_ua_firefox_windows7;
  35 geturls_ua_seamonkey_windows7 = pywwwget.geturls_ua_seamonkey_windows7;
  36 geturls_ua_chrome_windows7 = pywwwget.geturls_ua_chrome_windows7;
  37 geturls_ua_chromium_windows7 = pywwwget.geturls_ua_chromium_windows7;
  38 geturls_ua_palemoon_windows7 = pywwwget.geturls_ua_palemoon_windows7;
  39 geturls_ua_opera_windows7 = pywwwget.geturls_ua_opera_windows7;
  40 geturls_ua_vivaldi_windows7 = pywwwget.geturls_ua_chromium_windows7;
  41 geturls_ua_internet_explorer_windows7 = pywwwget.geturls_ua_internet_explorer_windows7;
  42 geturls_ua_microsoft_edge_windows7 = pywwwget.geturls_ua_microsoft_edge_windows7;
  43 geturls_ua_pywwwget_python = pywwwget.geturls_ua_pywwwget_python;
  44 geturls_ua_pywwwget_python_alt = pywwwget.geturls_ua_pywwwget_python_alt;
  45 geturls_ua_googlebot_google = pywwwget.geturls_ua_googlebot_google;
  46 geturls_ua_googlebot_google_old = pywwwget.geturls_ua_googlebot_google_old;
  47 geturls_headers = pywwwget.geturls_headers;
  48 geturls_headers_firefox_windows7 = pywwwget.geturls_headers_firefox_windows7;
  49 geturls_headers_seamonkey_windows7 = pywwwget.geturls_headers_seamonkey_windows7;
  50 geturls_headers_chrome_windows7 = pywwwget.geturls_headers_chrome_windows7;
  51 geturls_headers_chromium_windows7 = pywwwget.geturls_headers_chromium_windows7;
  52 geturls_headers_palemoon_windows7 = pywwwget.geturls_headers_palemoon_windows7;
  53 geturls_headers_opera_windows7 = pywwwget.geturls_headers_opera_windows7;
  54 geturls_headers_vivaldi_windows7 = pywwwget.geturls_headers_vivaldi_windows7;
  55 geturls_headers_internet_explorer_windows7 = pywwwget.geturls_headers_internet_explorer_windows7;
  56 geturls_headers_microsoft_edge_windows7 = pywwwget.geturls_headers_microsoft_edge_windows7;
  57 geturls_headers_pywwwget_python = pywwwget.geturls_headers_pywwwget_python;
  58 geturls_headers_pywwwget_python_alt = pywwwget.geturls_headers_pywwwget_python_alt;
  59 geturls_headers_googlebot_google = pywwwget.geturls_headers_googlebot_google;
  60 geturls_headers_googlebot_google_old = pywwwget.geturls_headers_googlebot_google_old;
  61 geturls_download_sleep = pywwwget.geturls_download_sleep;
  62
  63 parser = argparse.ArgumentParser(description="Python libary/module to download files.", conflict_handler="resolve", add_help=True);
  64 parser.add_argument("url", help="enter a url");
  65 parser.add_argument("-V", "--version", action="version", version=__program_name__+" "+__version__);
  66 parser.add_argument("-u", "--update", action="store_true", help="update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)");
  67 parser.add_argument("-d", "--dump-user-agent", action="store_true", help="display the current browser identification");
  68 parser.add_argument("-u", "--user-agent", default=geturls_ua_firefox_windows7, help="specify a custom user agent");
  69 parser.add_argument("-r", "--referer", default="https://www.google.com/", help="specify a custom referer, use if the video access");
  70 parser.add_argument("-O", "--output-document", default="-", help="specify a file name for output");
  71 parser.add_argument("-o", "--output-directory", default=os.path.realpath(os.getcwd()), help="specify a directory to output file to");
  72 parser.add_argument("-l", "--use-httplib", default="urllib", help="select library to download file can be urllib or requests or mechanize");
  73 parser.add_argument("-b", "--set-buffersize", default=524288, type=int, help="set how big buffersize is in bytes. how much it will download");
  74 parser.add_argument("-t", "--timeout", default=10, type=int, help="set timeout time for http request");
  75 parser.add_argument("-s", "--sleep", default=10, type=int, help="set sleep time for request");
  76 parser.add_argument("-v", "--verbose", action="store_true", help="print various debugging information");
  77 getargs = parser.parse_args();
  78
  79 if(not pywwwget.check_httplib_support(getargs.use_httplib)):
  80  getargs.use_httplib = "urllib";
  81
  82 getargs_cj = geturls_cj;
  83 getargs_headers = {'Referer': getargs.referer, 'User-Agent': getargs.user_agent, 'Accept-Encoding': "gzip, deflate", 'Accept-Language': "en-US,en;q=0.8,en-CA,en-GB;q=0.6", '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"};
  84
  85 getargs.output_directory = os.path.realpath(getargs.output_directory);
  86
  87 if(getargs.verbose==True):
  88  log.basicConfig(format="%(levelname)s: %(message)s", level=log.DEBUG);
  89
  90 if(getargs.dump_user_agent==True):
  91  print(getargs.user_agent);
  92  sys.exit();
  93
  94 if(getargs.output_document=="-"):
  95  if(sys.version[0]=="2"):
  96   precontstr = pywwwget.download_from_url_to_file(getargs.url, getargs_headers, getargs.user_agent, getargs.referer, geturls_cj, httplibuse=getargs.use_httplib, buffersize=[getargs.set_buffersize, getargs.set_buffersize], outfile=getargs.output_document, outpath=os.getcwd(), sleep=getargs.sleep, timeout=getargs.timeout);
  97   print(precontstr['Content']);
  98  if(sys.version[0]>="3"):
  99   precontstr = pywwwget.download_from_url_to_file(getargs.url, getargs_headers, getargs.user_agent, getargs.referer, geturls_cj, httplibuse=getargs.use_httplib, buffersize=[getargs.set_buffersize, getargs.set_buffersize], outfile=getargs.output_document, outpath=os.getcwd(), sleep=getargs.sleep, timeout=getargs.timeout);
 100   print(precontstr['Content'].decode('ascii', 'replace'));
 101
 102 if(getargs.output_document!="-"):
 103  pywwwget.download_from_url_to_file(getargs.url, getargs_headers, getargs.user_agent, getargs.referer, geturls_cj, httplibuse=getargs.use_httplib, buffersize=[getargs.set_buffersize, getargs.set_buffersize], outfile=getargs.output_document, outpath=getargs.output_directory, sleep=getargs.sleep, timeout=getargs.timeout);