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 urllib2
  31 import re
  32
  33 # Exit status constants
  34 exit_failure = 1
  35 exit_success = 0
  36
  37 # Global constants
  38 const_video_url_str = 'http://www.youtube.com/watch?v=%s'
  39 const_video_url_re = re.compile(r'http://(?:www\.)?youtube\.com/(?:v/|(?:watch)?\?v=)([^&]+).*')
  40 const_login_url_str = 'http://www.youtube.com/signup?next=/'
  41 const_login_post_str = 'current_form=loginForm&next=%%2F&username=%s&password=%s&action_login=Log+In'
  42 const_age_url_str = 'http://www.youtube.com/verify_age?next_url=/watch%%3Fv%%3D%s'
  43 const_age_post_str = 'next_url=%%2Fwatch%%3Fv%%3D%s&action_confirm=Confirm'
  44 const_video_url_params_re = re.compile(r'player2\.swf\?video_id=([^&]+)&.*t=([^&]+)&', re.M)
  45 const_video_url_real_str = 'http://www.youtube.com/get_video?video_id=%s&t=%s'
  46 const_1k = 1024
  47 const_block_size = 10 * const_1k
  48
  49 # Print error message, followed by standard advice information, and then exit
  50 def error_advice_exit(error_text):
  51         global exit_failure
  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('Try again several times and contact me if the problem persists.\n')
  63         sys.exit(exit_failure)
  64
  65 # Wrapper to create custom requests with typical headers
  66 def request_create(url, data=None):
  67         retval = urllib2.Request(url)
  68         if not data is None:
  69                 retval.add_data(data)
  70         # Try to mimic Firefox, at least a little bit
  71         retval.add_header('User-Agent', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.6) Gecko/20060728 Firefox/1.5.0.6')
  72         retval.add_header('Accept-Charset', 'ISO-8859-1,utf-8;q=0.7,*;q=0.7')
  73         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')
  74         retval.add_header('Accept-Language', 'en-us,en;q=0.5')
  75         return retval
  76
  77 # Perform a request, process headers and return response
  78 def perform_request(url, data=None):
  79         request = request_create(url, data)
  80         response = urllib2.urlopen(request)
  81         return response
  82
  83 # Convert bytes to KiB
  84 def to_k(bytes):
  85         global const_1k
  86         return bytes / const_1k
  87
  88 # Conditional print
  89 def cond_print(str):
  90         global cmdl_opts
  91         if not cmdl_opts.quiet:
  92                 sys.stdout.write(str)
  93                 sys.stdout.flush()
  94
  95 # Create the command line options parser and parse command line
  96 cmdl_usage = 'usage: %prog [options] video_url'
  97 cmdl_version = '2006.08.28'
  98 cmdl_parser = optparse.OptionParser(usage=cmdl_usage, version=cmdl_version, conflict_handler='resolve')
  99 cmdl_parser.add_option('-h', '--help', action='help', help='print this help text and exit')
 100 cmdl_parser.add_option('-v', '--version', action='version', help='print program version and exit')
 101 cmdl_parser.add_option('-u', '--username', dest='username', metavar='USERNAME', help='account username')
 102 cmdl_parser.add_option('-p', '--password', dest='password', metavar='PASSWORD', help='account password')
 103 cmdl_parser.add_option('-o', '--output', dest='outfile', metavar='FILE', help='output video file name')
 104 cmdl_parser.add_option('-q', '--quiet', action='store_true', dest='quiet', help='activates quiet mode')
 105 cmdl_parser.add_option('-s', '--simulate', action='store_true', dest='simulate', help='do not download video')
 106 (cmdl_opts, cmdl_args) = cmdl_parser.parse_args()
 107
 108 # Get video URL
 109 if len(cmdl_args) != 1:
 110         cmdl_parser.print_help()
 111         sys.exit(exit_failure)
 112 video_url_cmdl = cmdl_args[0]
 113
 114 # Verify video URL format and convert to "standard" format
 115 video_url_mo = const_video_url_re.match(video_url_cmdl)
 116 if video_url_mo is None:
 117         sys.exit('Error: URL does not seem to be a youtube video URL. If it is, report a bug.\n')
 118 video_url_id = video_url_mo.group(1)
 119 video_url = const_video_url_str % video_url_id
 120
 121 # Check conflicting options
 122 if not cmdl_opts.outfile is None and cmdl_opts.simulate:
 123         sys.stderr.write('Warning: video file name given but will not be used.\n')
 124
 125 # Get output file name
 126 if cmdl_opts.outfile is None:
 127         video_filename = '%s.flv' % video_url_id
 128 else:
 129         video_filename = cmdl_opts.outfile
 130
 131 # Check name
 132 if not video_filename.lower().endswith('.flv'):
 133         sys.stderr.write('Warning: video file name does not end in .flv\n')
 134
 135 # Verify both or none present
 136 if ((cmdl_opts.username is None and not cmdl_opts.password is None) or
 137     (not cmdl_opts.username is None and cmdl_opts.password is None)):
 138         sys.exit('Error: both username and password must be given, or none.')
 139
 140 # Test writable file
 141 if not cmdl_opts.simulate:
 142         try:
 143                 disk_test = open(video_filename, 'wb')
 144                 disk_test.close()
 145
 146         except (OSError, IOError):
 147                 sys.exit('Error: unable to open %s for writing.' % video_filename)
 148
 149 # Install cookie handler
 150 urllib2.install_opener(urllib2.build_opener(urllib2.HTTPCookieProcessor()))
 151
 152 # Login and confirm age if needed
 153 if not cmdl_opts.username is None:
 154         try:
 155                 # Log in
 156                 cond_print('Logging in... ')
 157                 perform_request(const_login_url_str, const_login_post_str % (cmdl_opts.username, cmdl_opts.password)).read()
 158                 cond_print('done.\n')
 159
 160         except (urllib2.URLError, ValueError):
 161                 cond_print('failed.\n')
 162                 error_advice_exit('unable to login')
 163
 164         except KeyboardInterrupt:
 165                 sys.exit(exit_failure)
 166
 167         try:
 168                 # Get age confirmation cookie
 169                 cond_print('Confirming age... ')
 170                 perform_request(const_age_url_str % video_url_id, const_age_post_str % video_url_id).read()
 171                 cond_print('done.\n')
 172
 173         except (urllib2.URLError, ValueError):
 174                 cond_print('failed.\n')
 175                 error_advice_exit('unable to confirm age')
 176
 177         except KeyboardInterrupt:
 178                 sys.exit(exit_failure)
 179
 180 # Retrieve video webpage
 181 try:
 182         cond_print('Retrieving video webpage... ')
 183         video_webpage = perform_request(video_url).read()
 184
 185 except (urllib2.URLError, ValueError):
 186         cond_print('failed.\n')
 187         error_advice_exit('unable to download video webpage')
 188
 189 except KeyboardInterrupt:
 190         sys.exit(exit_failure)
 191
 192 cond_print('done.\n')
 193
 194 # Extract needed video URL parameters
 195 try:
 196         cond_print('Extracting video URL parameters... ')
 197         video_url_params_mo = const_video_url_params_re.search(video_webpage)
 198
 199         if video_url_params_mo is None:
 200                 cond_print('failed.\n')
 201                 error_advice_exit('unable to extract URL parameters')
 202
 203         video_real_id = video_url_params_mo.group(1)
 204         video_t_param = video_url_params_mo.group(2)
 205         video_url_real = const_video_url_real_str % (video_real_id, video_t_param)
 206         cond_print('done.\n')
 207
 208 except KeyboardInterrupt:
 209         sys.exit(exit_failure)
 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
 237 except (urllib2.URLError, ValueError):
 238         cond_print('failed.\n')
 239         error_advice_exit('unable to download video data')
 240
 241 except KeyboardInterrupt:
 242         sys.exit(exit_failure)
 243
 244 cond_print('done.\n')
 245 cond_print('Video data saved to %s\n' % video_filename)
 246 sys.exit()