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 absolute_import, division, print_function, unicode_literals, generators, with_statement, nested_scopes
  19 import re
  20 import os
  21 import sys
  22 import pywwwget
  23 import argparse
  24 import logging as log
  25
  26 __project__ = pywwwget.__project__
  27 __program_name__ = pywwwget.__program_name__
  28 __project_url__ = pywwwget.__project_url__
  29 __version_info__ = pywwwget.__version_info__
  30 __version_date_info__ = pywwwget.__version_date_info__
  31 __version_date__ = pywwwget.__version_date__
  32 __version_date_plusrc__ = pywwwget.__version_date_plusrc__
  33 __version__ = pywwwget.__version__
  34 __version_date_plusrc__ = pywwwget.__version_date_plusrc__
  35
  36 geturls_cj = pywwwget.geturls_cj
  37 geturls_ua = pywwwget.geturls_ua
  38 geturls_ua_firefox_windows7 = pywwwget.geturls_ua_firefox_windows7
  39 geturls_ua_seamonkey_windows7 = pywwwget.geturls_ua_seamonkey_windows7
  40 geturls_ua_chrome_windows7 = pywwwget.geturls_ua_chrome_windows7
  41 geturls_ua_chromium_windows7 = pywwwget.geturls_ua_chromium_windows7
  42 geturls_ua_palemoon_windows7 = pywwwget.geturls_ua_palemoon_windows7
  43 geturls_ua_opera_windows7 = pywwwget.geturls_ua_opera_windows7
  44 geturls_ua_vivaldi_windows7 = pywwwget.geturls_ua_chromium_windows7
  45 geturls_ua_internet_explorer_windows7 = pywwwget.geturls_ua_internet_explorer_windows7
  46 geturls_ua_microsoft_edge_windows7 = pywwwget.geturls_ua_microsoft_edge_windows7
  47 geturls_ua_pywwwget_python = pywwwget.geturls_ua_pywwwget_python
  48 geturls_ua_pywwwget_python_alt = pywwwget.geturls_ua_pywwwget_python_alt
  49 geturls_ua_googlebot_google = pywwwget.geturls_ua_googlebot_google
  50 geturls_ua_googlebot_google_old = pywwwget.geturls_ua_googlebot_google_old
  51 geturls_headers = pywwwget.geturls_headers
  52 geturls_headers_firefox_windows7 = pywwwget.geturls_headers_firefox_windows7
  53 geturls_headers_seamonkey_windows7 = pywwwget.geturls_headers_seamonkey_windows7
  54 geturls_headers_chrome_windows7 = pywwwget.geturls_headers_chrome_windows7
  55 geturls_headers_chromium_windows7 = pywwwget.geturls_headers_chromium_windows7
  56 geturls_headers_palemoon_windows7 = pywwwget.geturls_headers_palemoon_windows7
  57 geturls_headers_opera_windows7 = pywwwget.geturls_headers_opera_windows7
  58 geturls_headers_vivaldi_windows7 = pywwwget.geturls_headers_vivaldi_windows7
  59 geturls_headers_internet_explorer_windows7 = pywwwget.geturls_headers_internet_explorer_windows7
  60 geturls_headers_microsoft_edge_windows7 = pywwwget.geturls_headers_microsoft_edge_windows7
  61 geturls_headers_pywwwget_python = pywwwget.geturls_headers_pywwwget_python
  62 geturls_headers_pywwwget_python_alt = pywwwget.geturls_headers_pywwwget_python_alt
  63 geturls_headers_googlebot_google = pywwwget.geturls_headers_googlebot_google
  64 geturls_headers_googlebot_google_old = pywwwget.geturls_headers_googlebot_google_old
  65 geturls_download_sleep = pywwwget.geturls_download_sleep
  66
  67 parser = argparse.ArgumentParser(
  68     description="Python libary/module to download files.", conflict_handler="resolve", add_help=True)
  69 parser.add_argument("url", help="enter a url")
  70 parser.add_argument("-V", "--version", action="version",
  71                     version=__program_name__+" "+__version__)
  72 parser.add_argument("-u", "--update", action="store_true",
  73                     help="update this program to latest version. Make sure that you have sufficient permissions (run with sudo if needed)")
  74 parser.add_argument("-d", "--dump-user-agent", action="store_true",
  75                     help="display the current browser identification")
  76 parser.add_argument("-u", "--user-agent", default=geturls_ua_firefox_windows7,
  77                     help="specify a custom user agent")
  78 parser.add_argument("-r", "--referer", default="https://www.google.com/",
  79                     help="specify a custom referer, use if the video access")
  80 parser.add_argument("-O", "--output-document", default="-",
  81                     help="specify a file name for output")
  82 parser.add_argument("-o", "--output-directory", default=os.path.realpath(
  83     os.getcwd()), help="specify a directory to output file to")
  84 parser.add_argument("-l", "--use-httplib", default="urllib",
  85                     help="select library to download file can be urllib or requests or mechanize")
  86 parser.add_argument("-b", "--set-buffersize", default=524288, type=int,
  87                     help="set how big buffersize is in bytes. how much it will download")
  88 parser.add_argument("-t", "--timeout", default=10, type=int,
  89                     help="set timeout time for http request")
  90 parser.add_argument("-s", "--sleep", default=10, type=int,
  91                     help="set sleep time for request")
  92 parser.add_argument("-v", "--verbose", action="store_true",
  93                     help="print various debugging information")
  94 getargs = parser.parse_args()
  95
  96 if(not pywwwget.check_httplib_support(getargs.use_httplib)):
  97     getargs.use_httplib = "urllib"
  98
  99 getargs_cj = geturls_cj
 100 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",
 101                    '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"}
 102
 103 getargs.output_directory = os.path.realpath(getargs.output_directory)
 104
 105 if(getargs.verbose == True):
 106     log.basicConfig(format="%(levelname)s: %(message)s", level=log.DEBUG)
 107
 108 if(getargs.dump_user_agent == True):
 109     print(getargs.user_agent)
 110     sys.exit()
 111
 112 if(getargs.output_document == "-"):
 113     if(sys.version[0] == "2"):
 114         precontstr = pywwwget.download_from_url_to_file(getargs.url, getargs_headers, getargs.user_agent, getargs.referer, geturls_cj, httplibuse=getargs.use_httplib, buffersize=[
 115                                                         getargs.set_buffersize, getargs.set_buffersize], outfile=getargs.output_document, outpath=os.getcwd(), sleep=getargs.sleep, timeout=getargs.timeout)
 116         print(precontstr['Content'])
 117     if(sys.version[0] >= "3"):
 118         precontstr = pywwwget.download_from_url_to_file(getargs.url, getargs_headers, getargs.user_agent, getargs.referer, geturls_cj, httplibuse=getargs.use_httplib, buffersize=[
 119                                                         getargs.set_buffersize, getargs.set_buffersize], outfile=getargs.output_document, outpath=os.getcwd(), sleep=getargs.sleep, timeout=getargs.timeout)
 120         print(precontstr['Content'].decode('ascii', 'replace'))
 121
 122 if(getargs.output_document != "-"):
 123     pywwwget.download_from_url_to_file(getargs.url, getargs_headers, getargs.user_agent, getargs.referer, geturls_cj, httplibuse=getargs.use_httplib, buffersize=[
 124                                        getargs.set_buffersize, getargs.set_buffersize], outfile=getargs.output_document, outpath=getargs.output_directory, sleep=getargs.sleep, timeout=getargs.timeout)