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 # Exit status constants
  39 exit_failure = 1
  40 exit_success = 0
  41
  42 # Global constants
  43 const_video_url_str = 'http://www.youtube.com/watch?v=%s'
  44 const_video_url_re = re.compile(r'http://(?:www\.)?youtube\.com/(?:v/|(?:watch(?:\.php)?)?\?v=)([^&]+).*')
  45 const_login_url_str = 'http://www.youtube.com/signup?next=/'
  46 const_login_post_str = 'current_form=loginForm&next=%%2F&username=%s&password=%s&action_login=Log+In'
  47 const_age_url_str = 'http://www.youtube.com/verify_age?next_url=/watch%%3Fv%%3D%s'
  48 const_age_post_str = 'next_url=%%2Fwatch%%3Fv%%3D%s&action_confirm=Confirm'
  49 const_video_url_params_re = re.compile(r'player2\.swf\?([^"]+)"', re.M)
  50 const_video_url_real_str = 'http://www.youtube.com/get_video?%s'
  51 const_1k = 1024
  52 const_block_size = 10 * const_1k
  53
  54 # Print error message, followed by standard advice information, and then exit
  55 def error_advice_exit(error_text):
  56         global exit_failure
  57         sys.stderr.write('Error: %s.\n' % error_text)
  58         sys.stderr.write('Try again several times. It may be a temporal problem.\n')
  59         sys.stderr.write('Other typical problems:\n\n')
  60         sys.stderr.write('\tVideo no longer exists.\n')
  61         sys.stderr.write('\tVideo requires age confirmation but you did not provide an account.\n')
  62         sys.stderr.write('\tYou provided the account data, but it is not valid.\n')
  63         sys.stderr.write('\tThe connection was cut suddenly for some reason.\n')
  64         sys.stderr.write('\tYouTube changed their system, and the program no longer works.\n')
  65         sys.stderr.write('\nTry to confirm you are able to view the video using a web browser.\n')
  66         sys.stderr.write('Use the same video URL and account information, if needed, with this program.\n')
  67         sys.stderr.write('When using a proxy, make sure http_proxy has http://host:port format.\n')
  68         sys.stderr.write('Try again several times and contact me if the problem persists.\n')
  69         sys.exit(exit_failure)
  70
  71 # Wrapper to create custom requests with typical headers
  72 def request_create(url, data=None):
  73         retval = urllib2.Request(url)
  74         if not data is None:
  75                 retval.add_data(data)
  76         # Try to mimic Firefox, at least a little bit
  77         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')
  78         retval.add_header('Accept-Charset', 'ISO-8859-1,utf-8;q=0.7,*;q=0.7')
  79         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')
  80         retval.add_header('Accept-Language', 'en-us,en;q=0.5')
  81         return retval
  82
  83 # Perform a request, process headers and return response
  84 def perform_request(url, data=None):
  85         request = request_create(url, data)
  86         response = urllib2.urlopen(request)
  87         return response
  88
  89 # Convert bytes to KiB
  90 def to_k(bytes):
  91         global const_1k
  92         return bytes / const_1k
  93
  94 # Conditional print
  95 def cond_print(str):
  96         global cmdl_opts
  97         if not cmdl_opts.quiet:
  98                 sys.stdout.write(str)
  99                 sys.stdout.flush()
 100
 101 # Create the command line options parser and parse command line
 102 cmdl_usage = 'usage: %prog [options] video_url'
 103 cmdl_version = '2006.09.25'
 104 cmdl_parser = optparse.OptionParser(usage=cmdl_usage, version=cmdl_version, conflict_handler='resolve')
 105 cmdl_parser.add_option('-h', '--help', action='help', help='print this help text and exit')
 106 cmdl_parser.add_option('-v', '--version', action='version', help='print program version and exit')
 107 cmdl_parser.add_option('-u', '--username', dest='username', metavar='USERNAME', help='account username')
 108 cmdl_parser.add_option('-p', '--password', dest='password', metavar='PASSWORD', help='account password')
 109 cmdl_parser.add_option('-o', '--output', dest='outfile', metavar='FILE', help='output video file name')
 110 cmdl_parser.add_option('-q', '--quiet', action='store_true', dest='quiet', help='activates quiet mode')
 111 cmdl_parser.add_option('-s', '--simulate', action='store_true', dest='simulate', help='do not download video')
 112 (cmdl_opts, cmdl_args) = cmdl_parser.parse_args()
 113
 114 # Get video URL
 115 if len(cmdl_args) != 1:
 116         cmdl_parser.print_help()
 117         sys.exit(exit_failure)
 118 video_url_cmdl = cmdl_args[0]
 119
 120 # Verify video URL format and convert to "standard" format
 121 video_url_mo = const_video_url_re.match(video_url_cmdl)
 122 if video_url_mo is None:
 123         sys.exit('Error: URL does not seem to be a youtube video URL. If it is, report a bug.')
 124 video_url_id = video_url_mo.group(1)
 125 video_url = const_video_url_str % video_url_id
 126
 127 # Check conflicting options
 128 if not cmdl_opts.outfile is None and cmdl_opts.simulate:
 129         sys.stderr.write('Warning: video file name given but will not be used.\n')
 130
 131 # Get output file name
 132 if cmdl_opts.outfile is None:
 133         video_filename = '%s.flv' % video_url_id
 134 else:
 135         video_filename = cmdl_opts.outfile
 136
 137 # Check name
 138 if not video_filename.lower().endswith('.flv'):
 139         sys.stderr.write('Warning: video file name does not end in .flv\n')
 140
 141 # Verify both or none present
 142 if ((cmdl_opts.username is None and not cmdl_opts.password is None) or
 143     (not cmdl_opts.username is None and cmdl_opts.password is None)):
 144         sys.exit('Error: both username and password must be given, or none.')
 145
 146 # Test writable file
 147 if not cmdl_opts.simulate:
 148         try:
 149                 disk_test = open(video_filename, 'wb')
 150                 disk_test.close()
 151
 152         except (OSError, IOError):
 153                 sys.exit('Error: unable to open %s for writing.' % video_filename)
 154
 155 # Install cookie and proxy handlers
 156 urllib2.install_opener(urllib2.build_opener(urllib2.HTTPCookieProcessor()))
 157 urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler()))
 158
 159 # Login and confirm age if needed
 160 if not cmdl_opts.username is None:
 161         try:
 162                 # Log in
 163                 cond_print('Logging in... ')
 164                 perform_request(const_login_url_str, const_login_post_str % (cmdl_opts.username, cmdl_opts.password)).read()
 165                 cond_print('done.\n')
 166
 167         except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 168                 cond_print('failed.\n')
 169                 error_advice_exit('unable to login')
 170
 171         except KeyboardInterrupt:
 172                 sys.exit('\n')
 173
 174         try:
 175                 # Get age confirmation cookie
 176                 cond_print('Confirming age... ')
 177                 perform_request(const_age_url_str % video_url_id, const_age_post_str % video_url_id).read()
 178                 cond_print('done.\n')
 179
 180         except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 181                 cond_print('failed.\n')
 182                 error_advice_exit('unable to confirm age')
 183
 184         except KeyboardInterrupt:
 185                 sys.exit('\n')
 186
 187 # Retrieve video webpage
 188 try:
 189         cond_print('Retrieving video webpage... ')
 190         video_webpage = perform_request(video_url).read()
 191         cond_print('done.\n')
 192
 193 except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 194         cond_print('failed.\n')
 195         error_advice_exit('unable to download video webpage')
 196
 197 except KeyboardInterrupt:
 198         sys.exit('\n')
 199
 200 # Extract needed video URL parameters
 201 try:
 202         cond_print('Extracting video URL parameters... ')
 203         video_url_params_mo = const_video_url_params_re.search(video_webpage)
 204
 205         if video_url_params_mo is None:
 206                 cond_print('failed.\n')
 207                 error_advice_exit('unable to extract URL parameters')
 208
 209         video_url_params = video_url_params_mo.group(1)
 210         video_url_real = const_video_url_real_str % video_url_params
 211         cond_print('done.\n')
 212
 213 except KeyboardInterrupt:
 214         sys.exit('\n')
 215
 216 # Retrieve video data
 217 try:
 218         video_data = perform_request(video_url_real)
 219         cond_print('Video data found at %s\n' % video_data.geturl())
 220
 221         # Abort here if in simulate mode
 222         if cmdl_opts.simulate:
 223                 sys.exit()
 224
 225         cond_print('Retrieving video data... ')
 226         video_file = open(video_filename, 'wb')
 227         try:
 228                 video_len_str = '%sk' % to_k(int(video_data.info()['Content-length']))
 229         except KeyError:
 230                 video_len_str = '(unknown)'
 231
 232         byte_counter = 0
 233         video_block = video_data.read(const_block_size)
 234         while len(video_block) != 0:
 235                 byte_counter += len(video_block)
 236                 video_file.write(video_block)
 237                 cond_print('\rRetrieving video data... %sk of %s ' % (to_k(byte_counter), video_len_str))
 238                 video_block = video_data.read(const_block_size)
 239
 240         video_file.close()
 241         cond_print('done.\n')
 242         cond_print('Video data saved to %s\n' % video_filename)
 243         sys.exit()
 244
 245 except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError):
 246         cond_print('failed.\n')
 247         error_advice_exit('unable to download video data')
 248
 249 except KeyboardInterrupt:
 250         sys.exit('\n')