Raphaël G. Git Repositories - youtubedl/blob - youtube-dl

   1 #!/usr/bin/env python
   2 #
   3 # Copyright (c) 2006 Ricardo Garcia Gonzalez
   4 #
   5 # Permission is hereby granted, free of charge, to any person obtaining a
   6 # copy of this software and associated documentation files (the "Software"),
   7 # to deal in the Software without restriction, including without limitation
   8 # the rights to use, copy, modify, merge, publish, distribute, sublicense,
   9 # and/or sell copies of the Software, and to permit persons to whom the
  10 # Software is furnished to do so, subject to the following conditions:
  11 #
  12 # The above copyright notice and this permission notice shall be included
  13 # in all copies or substantial portions of the Software.
  14 #
  15 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  16 # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  17 # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
  18 # THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
  19 # OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
  20 # ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  21 # OTHER DEALINGS IN THE SOFTWARE.
  22 #
  23 # Except as contained in this notice, the name(s) of the above copyright
  24 # holders shall not be used in advertising or otherwise to promote the
  25 # sale, use or other dealings in this Software without prior written
  26 # authorization.
  27 #
  28 import sys
  29 import optparse
  30 import httplib
  31 import urllib2
  32 import re
  33
  34 # First off, check Python and refuse to run
  35 if sys.hexversion < 0x020400f0:
  36         sys.exit('Error: Python 2.4 or later needed to run the program')
  37
  38 # Global constants
  39 const_video_url_str = 'http://www.youtube.com/watch?v=%s'
  40 const_video_url_re = re.compile(r'http://(?:www\d*\.)?youtube\.com/(?:v/|(?:watch(?:\.php)?)?\?v=)([^&]+).*')
  41 const_login_url_str = 'http://www.youtube.com/login?next=/watch%%3Fv%%3D%s'
  42 const_login_post_str = 'current_form=loginForm&next=%%2Fwatch%%3Fv%%3D%s&username=%s&password=%s&action_login=Log+In'
  43 const_age_url_str = 'http://www.youtube.com/verify_age?next_url=/watch%%3Fv%%3D%s'
  44 const_age_post_str = 'next_url=%%2Fwatch%%3Fv%%3D%s&action_confirm=Confirm'
  45 const_video_url_params_re = re.compile(r'player2\.swf\?([^"]+)"', re.M)
  46 const_video_url_real_str = 'http://www.youtube.com/get_video?%s'
  47 const_1k = 1024
  48 const_block_size = 10 * const_1k
  49
  50 # Print error message, followed by standard advice information, and then exit
  51 def error_advice_exit(error_text):
  52         sys.stderr.write('Error: %s.\n' % error_text)
  53         sys.stderr.write('Try again several times. It may be a temporal problem.\n')
  54         sys.stderr.write('Other typical problems:\n\n')
  55         sys.stderr.write('\tVideo no longer exists.\n')
  56         sys.stderr.write('\tVideo requires age confirmation but you did not provide an account.\n')
  57         sys.stderr.write('\tYou provided the account data, but it is not valid.\n')
  58         sys.stderr.write('\tThe connection was cut suddenly for some reason.\n')
  59         sys.stderr.write('\tYouTube changed their system, and the program no longer works.\n')
  60         sys.stderr.write('\nTry to confirm you are able to view the video using a web browser.\n')
  61         sys.stderr.write('Use the same video URL and account information, if needed, with this program.\n')
  62         sys.stderr.write('When using a proxy, make sure http_proxy has http://host:port format.\n')
  63         sys.stderr.write('Try again several times and contact me if the problem persists.\n')
  64         sys.exit('\n')
  65
  66 # Wrapper to create custom requests with typical headers
  67 def request_create(url, data=None):
  68         retval = urllib2.Request(url)
  69         if not data is None:
  70                 retval.add_data(data)
  71         # Try to mimic Firefox, at least a little bit
  72         retval.add_header('User-Agent', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0')
  73         retval.add_header('Accept-Charset', 'ISO-8859-1,utf-8;q=0.7,*;q=0.7')
  74         retval.add_header('Accept', 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5')
  75         retval.add_header('Accept-Language', 'en-us,en;q=0.5')
  76         return retval
  77
  78 # Perform a request, process headers and return response
  79 def perform_request(url, data=None):
  80         request = request_create(url, data)
  81         response = urllib2.urlopen(request)
  82         return response
  83
  84 # Convert bytes to KiB
  85 def to_k(bytes):
  86         global const_1k
  87         return bytes / const_1k
  88
  89 # Conditional print
  90 def cond_print(str):
  91         global cmdl_opts
  92         if not cmdl_opts.quiet:
  93                 sys.stdout.write(str)
  94                 sys.stdout.flush()
  95
  96 # Create the command line options parser and parse command line
  97 cmdl_usage = 'usage: %prog [options] video_url'
  98 cmdl_version = '2006.11.12'
  99 cmdl_parser = optparse.OptionParser(usage=cmdl_usage, version=cmdl_version, conflict_handler='resolve')
 100 cmdl_parser.add_option('-h', '--help', action='help', help='print this help text and exit')
 101 cmdl_parser.add_option('-v', '--version', action='version', help='print program version and exit')
 102 cmdl_parser.add_option('-u', '--username', dest='username', metavar='USERNAME', help='account username')
 103 cmdl_parser.add_option('-p', '--password', dest='password', metavar='PASSWORD', help='account password')
 104 cmdl_parser.add_option('-o', '--output', dest='outfile', metavar='FILE', help='output video file name')
 105 cmdl_parser.add_option('-q', '--quiet', action='store_true', dest='quiet', help='activates quiet mode')
 106 cmdl_parser.add_option('-s', '--simulate', action='store_true', dest='simulate', help='do not download video')
 107 (cmdl_opts, cmdl_args) = cmdl_parser.parse_args()
 108
 109 # Get video URL
 110 if len(cmdl_args) != 1:
 111         cmdl_parser.print_help()
 112         sys.exit('\n')
 113 video_url_cmdl = cmdl_args[0]
 114
 115 # Verify video URL format and convert to "standard" format
 116 video_url_mo = const_video_url_re.match(video_url_cmdl)
 117 if video_url_mo is None:
 118         sys.exit('Error: URL does not seem to be a youtube video URL. If it is, report a bug.')
 119 video_url_id = video_url_mo.group(1)
 120 video_url = const_video_url_str % video_url_id
 121
 122 # Check conflicting options
 123 if not cmdl_opts.outfile is None and cmdl_opts.simulate:
 124         sys.stderr.write('Warning: video file name given but will not be used.\n')
 125
 126 # Get output file name
 127 if cmdl_opts.outfile is None:
 128         video_filename = '%s.flv' % video_url_id
 129 else:
 130         video_filename = cmdl_opts.outfile
 131
 132 # Check name
 133 if not video_filename.lower().endswith('.flv'):
 134         sys.stderr.write('Warning: video file name does not end in .flv\n')
 135
 136 # Verify both or none present
 137 if ((cmdl_opts.username is None and not cmdl_opts.password is None) or
 138     (not cmdl_opts.username is None and cmdl_opts.password is None)):
 139         sys.exit('Error: both username and password must be given, or none.')
 140
 141 # Test writable file
 142 if not cmdl_opts.simulate:
 143         try:
 144                 disk_test = open(video_filename, 'wb')
 145                 disk_test.close()
 146
 147         except (OSError, IOError):
 148                 sys.exit('Error: unable to open %s for writing.' % video_filename)
 149
 150 # Install cookie and proxy handlers
 151 urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler()))
 152 urllib2.install_opener(urllib2.build_opener(urllib2.HTTPCookieProcessor()))
 153
 154 # Login and confirm age if needed
 155 if not cmdl_opts.username is None:
 156         try:
 157                 # Log in
 158                 cond_print('Logging in... ')
 159                 perform_request(const_login_url_str % video_url_id, const_login_post_str % (video_url_id, cmdl_opts.username, cmdl_opts.password)).read()
 160                 cond_print('done.\n')
 161
 162         except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 163                 cond_print('failed.\n')
 164                 error_advice_exit('unable to login')
 165
 166         except KeyboardInterrupt:
 167                 sys.exit('\n')
 168
 169         try:
 170                 # Get age confirmation cookie
 171                 cond_print('Confirming age... ')
 172                 perform_request(const_age_url_str % video_url_id, const_age_post_str % video_url_id).read()
 173                 cond_print('done.\n')
 174
 175         except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 176                 cond_print('failed.\n')
 177                 error_advice_exit('unable to confirm age')
 178
 179         except KeyboardInterrupt:
 180                 sys.exit('\n')
 181
 182 # Retrieve video webpage
 183 try:
 184         cond_print('Retrieving video webpage... ')
 185         video_webpage = perform_request(video_url).read()
 186         cond_print('done.\n')
 187
 188 except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 189         cond_print('failed.\n')
 190         error_advice_exit('unable to download video webpage')
 191
 192 except KeyboardInterrupt:
 193         sys.exit('\n')
 194
 195 # Extract needed video URL parameters
 196 try:
 197         cond_print('Extracting video URL parameters... ')
 198         video_url_params_mo = const_video_url_params_re.search(video_webpage)
 199
 200         if video_url_params_mo is None:
 201                 cond_print('failed.\n')
 202                 error_advice_exit('unable to extract URL parameters')
 203
 204         video_url_params = video_url_params_mo.group(1)
 205         video_url_real = const_video_url_real_str % video_url_params
 206         cond_print('done.\n')
 207
 208 except KeyboardInterrupt:
 209         sys.exit('\n')
 210
 211 # Retrieve video data
 212 try:
 213         video_data = perform_request(video_url_real)
 214         cond_print('Video data found at %s\n' % video_data.geturl())
 215
 216         # Abort here if in simulate mode
 217         if cmdl_opts.simulate:
 218                 sys.exit()
 219
 220         cond_print('Retrieving video data... ')
 221         video_file = open(video_filename, 'wb')
 222         try:
 223                 video_len_str = '%sk' % to_k(int(video_data.info()['Content-length']))
 224         except KeyError:
 225                 video_len_str = '(unknown)'
 226
 227         byte_counter = 0
 228         video_block = video_data.read(const_block_size)
 229         while len(video_block) != 0:
 230                 byte_counter += len(video_block)
 231                 video_file.write(video_block)
 232                 cond_print('\rRetrieving video data... %sk of %s ' % (to_k(byte_counter), video_len_str))
 233                 video_block = video_data.read(const_block_size)
 234
 235         video_file.close()
 236         cond_print('done.\n')
 237         cond_print('Video data saved to %s\n' % video_filename)
 238         sys.exit()
 239
 240 except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 241         cond_print('failed.\n')
 242         error_advice_exit('unable to download video data')
 243
 244 except KeyboardInterrupt:
 245         sys.exit('\n')