2 # -*- coding: utf-8 -*-
3 # Author: Ricardo Garcia Gonzalez
4 # Author: Danny Colligan
5 # Author: Benjamin Johnson
6 # Author: Vasyl' Vavrychuk
7 # Author: Witold Baryluk
8 # Author: Paweł Paprota
9 # Author: Gergely Imreh
10 # Author: Philipp Hagemeister <phihag@phihag.de>
11 # License: Public domain code
12 from __future__ import with_statement
38 import cStringIO as StringIO
42 # parse_qs was moved from the cgi module to the urlparse module recently.
44 from urlparse import parse_qs
46 from cgi import parse_qs
50 except ImportError: # Python < 2.6
54 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:2.0b11) Gecko/20100101 Firefox/4.0b11',
55 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
56 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
57 'Accept-Encoding': 'gzip, deflate',
58 'Accept-Language': 'en-us,en;q=0.5',
61 simple_title_chars = string.ascii_letters.decode('ascii') + string.digits.decode('ascii')
65 except ImportError: # Python <2.6, use trivialjson (https://github.com/phihag/trivialjson):
71 def raiseError(msg, i):
72 raise ValueError(msg + ' at position ' + str(i) + ' of ' + repr(s) + ': ' + repr(s[i:]))
73 def skipSpace(i, expectMore=True):
74 while i < len(s) and s[i] in ' \t\r\n':
78 raiseError('Premature end', i)
80 def decodeEscape(match):
96 return unichr(int(esc[1:5], 16))
97 if len(esc) == 5+6 and esc[5:7] == '\\u':
98 hi = int(esc[1:5], 16)
99 low = int(esc[7:11], 16)
100 return unichr((hi - 0xd800) * 0x400 + low - 0xdc00 + 0x10000)
101 raise ValueError('Unknown escape ' + str(esc))
108 while s[e-bslashes-1] == '\\':
110 if bslashes % 2 == 1:
114 rexp = re.compile(r'\\(u[dD][89aAbB][0-9a-fA-F]{2}\\u[0-9a-fA-F]{4}|u[0-9a-fA-F]{4}|.|$)')
115 stri = rexp.sub(decodeEscape, s[i:e])
121 if s[i] == '}': # Empty dictionary
125 raiseError('Expected a string object key', i)
126 i,key = parseString(i)
128 if i >= len(s) or s[i] != ':':
129 raiseError('Expected a colon', i)
136 raiseError('Expected comma or closing curly brace', i)
141 if s[i] == ']': # Empty array
146 i = skipSpace(i) # Raise exception if premature end
150 raiseError('Expected a comma or closing bracket', i)
152 def parseDiscrete(i):
153 for k,v in {'true': True, 'false': False, 'null': None}.items():
154 if s.startswith(k, i):
156 raiseError('Not a boolean (or null)', i)
158 mobj = re.match('^(-?(0|[1-9][0-9]*)(\.[0-9]*)?([eE][+-]?[0-9]+)?)', s[i:])
160 raiseError('Not a number', i)
162 if '.' in nums or 'e' in nums or 'E' in nums:
163 return (i+len(nums), float(nums))
164 return (i+len(nums), int(nums))
165 CHARMAP = {'{': parseObj, '[': parseArray, '"': parseString, 't': parseDiscrete, 'f': parseDiscrete, 'n': parseDiscrete}
168 i,res = CHARMAP.get(s[i], parseNumber)(i)
169 i = skipSpace(i, False)
173 raise ValueError('Extra data at end of input (index ' + str(i) + ' of ' + repr(s) + ': ' + repr(s[i:]) + ')')
176 def preferredencoding():
177 """Get preferred encoding.
179 Returns the best encoding scheme for the system, based on
180 locale.getpreferredencoding() and some further tweaks.
182 def yield_preferredencoding():
184 pref = locale.getpreferredencoding()
190 return yield_preferredencoding().next()
192 def htmlentity_transform(matchobj):
193 """Transforms an HTML entity to a Unicode character.
195 This function receives a match object and is intended to be used with
196 the re.sub() function.
198 entity = matchobj.group(1)
200 # Known non-numeric HTML entity
201 if entity in htmlentitydefs.name2codepoint:
202 return unichr(htmlentitydefs.name2codepoint[entity])
205 mobj = re.match(ur'(?u)#(x?\d+)', entity)
207 numstr = mobj.group(1)
208 if numstr.startswith(u'x'):
210 numstr = u'0%s' % numstr
213 return unichr(long(numstr, base))
215 # Unknown entity in name, return its literal representation
216 return (u'&%s;' % entity)
218 def sanitize_title(utitle):
219 """Sanitizes a video title so it could be used as part of a filename."""
220 utitle = re.sub(ur'(?u)&(.+?);', htmlentity_transform, utitle)
221 return utitle.replace(unicode(os.sep), u'%')
223 def sanitize_open(filename, open_mode):
224 """Try to open the given filename, and slightly tweak it if this fails.
226 Attempts to open the given filename. If this fails, it tries to change
227 the filename slightly, step by step, until it's either able to open it
228 or it fails and raises a final exception, like the standard open()
231 It returns the tuple (stream, definitive_file_name).
235 if sys.platform == 'win32':
237 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
238 return (sys.stdout, filename)
239 stream = open(filename, open_mode)
240 return (stream, filename)
241 except (IOError, OSError), err:
242 # In case of error, try to remove win32 forbidden chars
243 filename = re.sub(ur'[/<>:"\|\?\*]', u'#', filename)
245 # An exception here should be caught in the caller
246 stream = open(filename, open_mode)
247 return (stream, filename)
249 def timeconvert(timestr):
250 """Convert RFC 2822 defined time string into system timestamp"""
252 timetuple = email.utils.parsedate_tz(timestr)
253 if timetuple is not None:
254 timestamp = email.utils.mktime_tz(timetuple)
257 class DownloadError(Exception):
258 """Download Error exception.
260 This exception may be thrown by FileDownloader objects if they are not
261 configured to continue on errors. They will contain the appropriate
266 class SameFileError(Exception):
267 """Same File exception.
269 This exception will be thrown by FileDownloader objects if they detect
270 multiple files would have to be downloaded to the same file on disk.
274 class PostProcessingError(Exception):
275 """Post Processing exception.
277 This exception may be raised by PostProcessor's .run() method to
278 indicate an error in the postprocessing task.
282 class UnavailableVideoError(Exception):
283 """Unavailable Format exception.
285 This exception will be thrown when a video is requested
286 in a format that is not available for that video.
290 class ContentTooShortError(Exception):
291 """Content Too Short exception.
293 This exception may be raised by FileDownloader objects when a file they
294 download is too small for what the server announced first, indicating
295 the connection was probably interrupted.
301 def __init__(self, downloaded, expected):
302 self.downloaded = downloaded
303 self.expected = expected
305 class YoutubeDLHandler(urllib2.HTTPHandler):
306 """Handler for HTTP requests and responses.
308 This class, when installed with an OpenerDirector, automatically adds
309 the standard headers to every HTTP request and handles gzipped and
310 deflated responses from web servers. If compression is to be avoided in
311 a particular request, the original request in the program code only has
312 to include the HTTP header "Youtubedl-No-Compression", which will be
313 removed before making the real request.
315 Part of this code was copied from:
317 http://techknack.net/python-urllib2-handlers/
319 Andrew Rowls, the author of that code, agreed to release it to the
326 return zlib.decompress(data, -zlib.MAX_WBITS)
328 return zlib.decompress(data)
331 def addinfourl_wrapper(stream, headers, url, code):
332 if hasattr(urllib2.addinfourl, 'getcode'):
333 return urllib2.addinfourl(stream, headers, url, code)
334 ret = urllib2.addinfourl(stream, headers, url)
338 def http_request(self, req):
339 for h in std_headers:
342 req.add_header(h, std_headers[h])
343 if 'Youtubedl-no-compression' in req.headers:
344 if 'Accept-encoding' in req.headers:
345 del req.headers['Accept-encoding']
346 del req.headers['Youtubedl-no-compression']
349 def http_response(self, req, resp):
352 if resp.headers.get('Content-encoding', '') == 'gzip':
353 gz = gzip.GzipFile(fileobj=StringIO.StringIO(resp.read()), mode='r')
354 resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
355 resp.msg = old_resp.msg
357 if resp.headers.get('Content-encoding', '') == 'deflate':
358 gz = StringIO.StringIO(self.deflate(resp.read()))
359 resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
360 resp.msg = old_resp.msg
363 class FileDownloader(object):
364 """File Downloader class.
366 File downloader objects are the ones responsible of downloading the
367 actual video file and writing it to disk if the user has requested
368 it, among some other tasks. In most cases there should be one per
369 program. As, given a video URL, the downloader doesn't know how to
370 extract all the needed information, task that InfoExtractors do, it
371 has to pass the URL to one of them.
373 For this, file downloader objects have a method that allows
374 InfoExtractors to be registered in a given order. When it is passed
375 a URL, the file downloader handles it to the first InfoExtractor it
376 finds that reports being able to handle it. The InfoExtractor extracts
377 all the information about the video or videos the URL refers to, and
378 asks the FileDownloader to process the video information, possibly
379 downloading the video.
381 File downloaders accept a lot of parameters. In order not to saturate
382 the object constructor with arguments, it receives a dictionary of
383 options instead. These options are available through the params
384 attribute for the InfoExtractors to use. The FileDownloader also
385 registers itself as the downloader in charge for the InfoExtractors
386 that are added to it, so this is a "mutual registration".
390 username: Username for authentication purposes.
391 password: Password for authentication purposes.
392 usenetrc: Use netrc for authentication instead.
393 quiet: Do not print messages to stdout.
394 forceurl: Force printing final URL.
395 forcetitle: Force printing title.
396 forcethumbnail: Force printing thumbnail URL.
397 forcedescription: Force printing description.
398 forcefilename: Force printing final filename.
399 simulate: Do not download the video files.
400 format: Video format code.
401 format_limit: Highest quality format to try.
402 outtmpl: Template for output names.
403 ignoreerrors: Do not stop on download errors.
404 ratelimit: Download speed limit, in bytes/sec.
405 nooverwrites: Prevent overwriting files.
406 retries: Number of times to retry for HTTP error 5xx
407 continuedl: Try to continue downloads if possible.
408 noprogress: Do not print the progress bar.
409 playliststart: Playlist item to start at.
410 playlistend: Playlist item to end at.
411 logtostderr: Log messages to stderr instead of stdout.
412 consoletitle: Display progress in console window's titlebar.
413 nopart: Do not use temporary .part files.
414 updatetime: Use the Last-modified header to set output file timestamps.
415 writedescription: Write the video description to a .description file
416 writeinfojson: Write the video description to a .info.json file
422 _download_retcode = None
423 _num_downloads = None
426 def __init__(self, params):
427 """Create a FileDownloader object with the given options."""
430 self._download_retcode = 0
431 self._num_downloads = 0
432 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
436 def pmkdir(filename):
437 """Create directory components in filename. Similar to Unix "mkdir -p"."""
438 components = filename.split(os.sep)
439 aggregate = [os.sep.join(components[0:x]) for x in xrange(1, len(components))]
440 aggregate = ['%s%s' % (x, os.sep) for x in aggregate] # Finish names with separator
441 for dir in aggregate:
442 if not os.path.exists(dir):
446 def format_bytes(bytes):
449 if type(bytes) is str:
454 exponent = long(math.log(bytes, 1024.0))
455 suffix = 'bkMGTPEZY'[exponent]
456 converted = float(bytes) / float(1024**exponent)
457 return '%.2f%s' % (converted, suffix)
460 def calc_percent(byte_counter, data_len):
463 return '%6s' % ('%3.1f%%' % (float(byte_counter) / float(data_len) * 100.0))
466 def calc_eta(start, now, total, current):
470 if current == 0 or dif < 0.001: # One millisecond
472 rate = float(current) / dif
473 eta = long((float(total) - float(current)) / rate)
474 (eta_mins, eta_secs) = divmod(eta, 60)
477 return '%02d:%02d' % (eta_mins, eta_secs)
480 def calc_speed(start, now, bytes):
482 if bytes == 0 or dif < 0.001: # One millisecond
483 return '%10s' % '---b/s'
484 return '%10s' % ('%s/s' % FileDownloader.format_bytes(float(bytes) / dif))
487 def best_block_size(elapsed_time, bytes):
488 new_min = max(bytes / 2.0, 1.0)
489 new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
490 if elapsed_time < 0.001:
492 rate = bytes / elapsed_time
500 def parse_bytes(bytestr):
501 """Parse a string indicating a byte quantity into a long integer."""
502 matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
505 number = float(matchobj.group(1))
506 multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
507 return long(round(number * multiplier))
509 def add_info_extractor(self, ie):
510 """Add an InfoExtractor object to the end of the list."""
512 ie.set_downloader(self)
514 def add_post_processor(self, pp):
515 """Add a PostProcessor object to the end of the chain."""
517 pp.set_downloader(self)
519 def to_screen(self, message, skip_eol=False, ignore_encoding_errors=False):
520 """Print message to stdout if not in quiet mode."""
522 if not self.params.get('quiet', False):
523 terminator = [u'\n', u''][skip_eol]
524 print >>self._screen_file, (u'%s%s' % (message, terminator)).encode(preferredencoding()),
525 self._screen_file.flush()
526 except (UnicodeEncodeError), err:
527 if not ignore_encoding_errors:
530 def to_stderr(self, message):
531 """Print message to stderr."""
532 print >>sys.stderr, message.encode(preferredencoding())
534 def to_cons_title(self, message):
535 """Set console/terminal window title to message."""
536 if not self.params.get('consoletitle', False):
538 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
539 # c_wchar_p() might not be necessary if `message` is
540 # already of type unicode()
541 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
542 elif 'TERM' in os.environ:
543 sys.stderr.write('\033]0;%s\007' % message.encode(preferredencoding()))
545 def fixed_template(self):
546 """Checks if the output template is fixed."""
547 return (re.search(ur'(?u)%\(.+?\)s', self.params['outtmpl']) is None)
549 def trouble(self, message=None):
550 """Determine action to take when a download problem appears.
552 Depending on if the downloader has been configured to ignore
553 download errors or not, this method may throw an exception or
554 not when errors are found, after printing the message.
556 if message is not None:
557 self.to_stderr(message)
558 if not self.params.get('ignoreerrors', False):
559 raise DownloadError(message)
560 self._download_retcode = 1
562 def slow_down(self, start_time, byte_counter):
563 """Sleep if the download speed is over the rate limit."""
564 rate_limit = self.params.get('ratelimit', None)
565 if rate_limit is None or byte_counter == 0:
568 elapsed = now - start_time
571 speed = float(byte_counter) / elapsed
572 if speed > rate_limit:
573 time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
575 def temp_name(self, filename):
576 """Returns a temporary filename for the given filename."""
577 if self.params.get('nopart', False) or filename == u'-' or \
578 (os.path.exists(filename) and not os.path.isfile(filename)):
580 return filename + u'.part'
582 def undo_temp_name(self, filename):
583 if filename.endswith(u'.part'):
584 return filename[:-len(u'.part')]
587 def try_rename(self, old_filename, new_filename):
589 if old_filename == new_filename:
591 os.rename(old_filename, new_filename)
592 except (IOError, OSError), err:
593 self.trouble(u'ERROR: unable to rename file')
595 def try_utime(self, filename, last_modified_hdr):
596 """Try to set the last-modified time of the given file."""
597 if last_modified_hdr is None:
599 if not os.path.isfile(filename):
601 timestr = last_modified_hdr
604 filetime = timeconvert(timestr)
608 os.utime(filename,(time.time(), filetime))
612 def report_writedescription(self, descfn):
613 """ Report that the description file is being written """
614 self.to_screen(u'[info] Writing video description to: %s' % descfn, ignore_encoding_errors=True)
616 def report_writeinfojson(self, infofn):
617 """ Report that the metadata file has been written """
618 self.to_screen(u'[info] Video description metadata as JSON to: %s' % infofn, ignore_encoding_errors=True)
620 def report_destination(self, filename):
621 """Report destination filename."""
622 self.to_screen(u'[download] Destination: %s' % filename, ignore_encoding_errors=True)
624 def report_progress(self, percent_str, data_len_str, speed_str, eta_str):
625 """Report download progress."""
626 if self.params.get('noprogress', False):
628 self.to_screen(u'\r[download] %s of %s at %s ETA %s' %
629 (percent_str, data_len_str, speed_str, eta_str), skip_eol=True)
630 self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' %
631 (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip()))
633 def report_resuming_byte(self, resume_len):
634 """Report attempt to resume at given byte."""
635 self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
637 def report_retry(self, count, retries):
638 """Report retry in case of HTTP error 5xx"""
639 self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
641 def report_file_already_downloaded(self, file_name):
642 """Report file has already been fully downloaded."""
644 self.to_screen(u'[download] %s has already been downloaded' % file_name)
645 except (UnicodeEncodeError), err:
646 self.to_screen(u'[download] The file has already been downloaded')
648 def report_unable_to_resume(self):
649 """Report it was impossible to resume download."""
650 self.to_screen(u'[download] Unable to resume')
652 def report_finish(self):
653 """Report download finished."""
654 if self.params.get('noprogress', False):
655 self.to_screen(u'[download] Download completed')
659 def increment_downloads(self):
660 """Increment the ordinal that assigns a number to each file."""
661 self._num_downloads += 1
663 def prepare_filename(self, info_dict):
664 """Generate the output filename."""
666 template_dict = dict(info_dict)
667 template_dict['epoch'] = unicode(long(time.time()))
668 template_dict['autonumber'] = unicode('%05d' % self._num_downloads)
669 filename = self.params['outtmpl'] % template_dict
671 except (ValueError, KeyError), err:
672 self.trouble(u'ERROR: invalid system charset or erroneous output template')
675 def process_info(self, info_dict):
676 """Process a single dictionary returned by an InfoExtractor."""
677 filename = self.prepare_filename(info_dict)
678 # Do nothing else if in simulate mode
679 if self.params.get('simulate', False):
681 if self.params.get('forcetitle', False):
682 print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace')
683 if self.params.get('forceurl', False):
684 print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace')
685 if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict:
686 print info_dict['thumbnail'].encode(preferredencoding(), 'xmlcharrefreplace')
687 if self.params.get('forcedescription', False) and 'description' in info_dict:
688 print info_dict['description'].encode(preferredencoding(), 'xmlcharrefreplace')
689 if self.params.get('forcefilename', False) and filename is not None:
690 print filename.encode(preferredencoding(), 'xmlcharrefreplace')
696 if self.params.get('nooverwrites', False) and os.path.exists(filename):
697 self.to_stderr(u'WARNING: file exists and will be skipped')
701 self.pmkdir(filename)
702 except (OSError, IOError), err:
703 self.trouble(u'ERROR: unable to create directories: %s' % str(err))
706 if self.params.get('writedescription', False):
708 descfn = filename + '.description'
709 self.report_writedescription(descfn)
710 with contextlib.closing(open(descfn, 'wb')) as descfile:
711 descfile.write(info_dict['description'].encode('utf-8'))
712 except (OSError, IOError):
713 self.trouble(u'ERROR: Cannot write description file: %s' % str(descfn))
716 if self.params.get('writeinfojson', False):
717 infofn = filename + '.info.json'
718 self.report_writeinfojson(infofn)
721 except (NameError,AttributeError):
722 self.trouble(u'ERROR: No JSON encoder found. Update to Python 2.6+, setup a json module, or leave out --write-info-json.')
725 with contextlib.closing(open(infofn, 'wb')) as infof:
726 json.dump(info_dict, infof)
727 except (OSError, IOError):
728 self.trouble(u'ERROR: Cannot write metadata to JSON file: %s' % str(infofn))
732 success = self._do_download(filename, info_dict['url'].encode('utf-8'), info_dict.get('player_url', None))
733 except (OSError, IOError), err:
734 raise UnavailableVideoError
735 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
736 self.trouble(u'ERROR: unable to download video data: %s' % str(err))
738 except (ContentTooShortError, ), err:
739 self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
744 self.post_process(filename, info_dict)
745 except (PostProcessingError), err:
746 self.trouble(u'ERROR: postprocessing: %s' % str(err))
749 def download(self, url_list):
750 """Download a given list of URLs."""
751 if len(url_list) > 1 and self.fixed_template():
752 raise SameFileError(self.params['outtmpl'])
755 suitable_found = False
757 # Go to next InfoExtractor if not suitable
758 if not ie.suitable(url):
761 # Suitable InfoExtractor found
762 suitable_found = True
764 # Extract information from URL and process it
767 # Suitable InfoExtractor had been found; go to next URL
770 if not suitable_found:
771 self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url)
773 return self._download_retcode
775 def post_process(self, filename, ie_info):
776 """Run the postprocessing chain on the given file."""
778 info['filepath'] = filename
784 def _download_with_rtmpdump(self, filename, url, player_url):
785 self.report_destination(filename)
786 tmpfilename = self.temp_name(filename)
788 # Check for rtmpdump first
790 subprocess.call(['rtmpdump', '-h'], stdout=(file(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
791 except (OSError, IOError):
792 self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run')
795 # Download using rtmpdump. rtmpdump returns exit code 2 when
796 # the connection was interrumpted and resuming appears to be
797 # possible. This is part of rtmpdump's normal usage, AFAIK.
798 basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename]
799 retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)])
800 while retval == 2 or retval == 1:
801 prevsize = os.path.getsize(tmpfilename)
802 self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True)
803 time.sleep(5.0) # This seems to be needed
804 retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
805 cursize = os.path.getsize(tmpfilename)
806 if prevsize == cursize and retval == 1:
809 self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(tmpfilename))
810 self.try_rename(tmpfilename, filename)
813 self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval)
816 def _do_download(self, filename, url, player_url):
817 # Check file already present
818 if self.params.get('continuedl', False) and os.path.isfile(filename) and not self.params.get('nopart', False):
819 self.report_file_already_downloaded(filename)
822 # Attempt to download using rtmpdump
823 if url.startswith('rtmp'):
824 return self._download_with_rtmpdump(filename, url, player_url)
826 tmpfilename = self.temp_name(filename)
830 # Do not include the Accept-Encoding header
831 headers = {'Youtubedl-no-compression': 'True'}
832 basic_request = urllib2.Request(url, None, headers)
833 request = urllib2.Request(url, None, headers)
835 # Establish possible resume length
836 if os.path.isfile(tmpfilename):
837 resume_len = os.path.getsize(tmpfilename)
841 # Request parameters in case of being able to resume
842 if self.params.get('continuedl', False) and resume_len != 0:
843 self.report_resuming_byte(resume_len)
844 request.add_header('Range','bytes=%d-' % resume_len)
848 retries = self.params.get('retries', 0)
849 while count <= retries:
850 # Establish connection
852 data = urllib2.urlopen(request)
854 except (urllib2.HTTPError, ), err:
855 if (err.code < 500 or err.code >= 600) and err.code != 416:
856 # Unexpected HTTP error
858 elif err.code == 416:
859 # Unable to resume (requested range not satisfiable)
861 # Open the connection again without the range header
862 data = urllib2.urlopen(basic_request)
863 content_length = data.info()['Content-Length']
864 except (urllib2.HTTPError, ), err:
865 if err.code < 500 or err.code >= 600:
868 # Examine the reported length
869 if (content_length is not None and
870 (resume_len - 100 < long(content_length) < resume_len + 100)):
871 # The file had already been fully downloaded.
872 # Explanation to the above condition: in issue #175 it was revealed that
873 # YouTube sometimes adds or removes a few bytes from the end of the file,
874 # changing the file size slightly and causing problems for some users. So
875 # I decided to implement a suggested change and consider the file
876 # completely downloaded if the file size differs less than 100 bytes from
877 # the one in the hard drive.
878 self.report_file_already_downloaded(filename)
879 self.try_rename(tmpfilename, filename)
882 # The length does not match, we start the download over
883 self.report_unable_to_resume()
889 self.report_retry(count, retries)
892 self.trouble(u'ERROR: giving up after %s retries' % retries)
895 data_len = data.info().get('Content-length', None)
896 if data_len is not None:
897 data_len = long(data_len) + resume_len
898 data_len_str = self.format_bytes(data_len)
899 byte_counter = 0 + resume_len
905 data_block = data.read(block_size)
907 if len(data_block) == 0:
909 byte_counter += len(data_block)
911 # Open file just in time
914 (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
915 filename = self.undo_temp_name(tmpfilename)
916 self.report_destination(filename)
917 except (OSError, IOError), err:
918 self.trouble(u'ERROR: unable to open for writing: %s' % str(err))
921 stream.write(data_block)
922 except (IOError, OSError), err:
923 self.trouble(u'\nERROR: unable to write data: %s' % str(err))
925 block_size = self.best_block_size(after - before, len(data_block))
928 percent_str = self.calc_percent(byte_counter, data_len)
929 eta_str = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
930 speed_str = self.calc_speed(start, time.time(), byte_counter - resume_len)
931 self.report_progress(percent_str, data_len_str, speed_str, eta_str)
934 self.slow_down(start, byte_counter - resume_len)
938 if data_len is not None and byte_counter != data_len:
939 raise ContentTooShortError(byte_counter, long(data_len))
940 self.try_rename(tmpfilename, filename)
942 # Update file modification time
943 if self.params.get('updatetime', True):
944 self.try_utime(filename, data.info().get('last-modified', None))
948 class InfoExtractor(object):
949 """Information Extractor class.
951 Information extractors are the classes that, given a URL, extract
952 information from the video (or videos) the URL refers to. This
953 information includes the real video URL, the video title and simplified
954 title, author and others. The information is stored in a dictionary
955 which is then passed to the FileDownloader. The FileDownloader
956 processes this information possibly downloading the video to the file
957 system, among other possible outcomes. The dictionaries must include
958 the following fields:
960 id: Video identifier.
961 url: Final video URL.
962 uploader: Nickname of the video uploader.
963 title: Literal title.
964 stitle: Simplified title.
965 ext: Video filename extension.
966 format: Video format.
967 player_url: SWF Player URL (may be None).
969 The following fields are optional. Their primary purpose is to allow
970 youtube-dl to serve as the backend for a video search function, such
971 as the one in youtube2mp3. They are only used when their respective
972 forced printing functions are called:
974 thumbnail: Full URL to a video thumbnail image.
975 description: One-line video description.
977 Subclasses of this one should re-define the _real_initialize() and
978 _real_extract() methods, as well as the suitable() static method.
979 Probably, they should also be instantiated and added to the main
986 def __init__(self, downloader=None):
987 """Constructor. Receives an optional downloader."""
989 self.set_downloader(downloader)
993 """Receives a URL and returns True if suitable for this IE."""
996 def initialize(self):
997 """Initializes an instance (authentication, etc)."""
999 self._real_initialize()
1002 def extract(self, url):
1003 """Extracts URL information and returns it in list of dicts."""
1005 return self._real_extract(url)
1007 def set_downloader(self, downloader):
1008 """Sets the downloader for this IE."""
1009 self._downloader = downloader
1011 def _real_initialize(self):
1012 """Real initialization process. Redefine in subclasses."""
1015 def _real_extract(self, url):
1016 """Real extraction process. Redefine in subclasses."""
1019 class YoutubeIE(InfoExtractor):
1020 """Information extractor for youtube.com."""
1022 _VALID_URL = r'^((?:https?://)?(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/)(?:(?:(?:v|embed|e)/)|(?:(?:watch(?:_popup)?(?:\.php)?)?(?:\?|#!?)(?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$'
1023 _LANG_URL = r'http://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
1024 _LOGIN_URL = 'https://www.youtube.com/signup?next=/&gl=US&hl=en'
1025 _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
1026 _NETRC_MACHINE = 'youtube'
1027 # Listed in order of quality
1028 _available_formats = ['38', '37', '22', '45', '35', '34', '43', '18', '6', '5', '17', '13']
1029 _video_extensions = {
1035 '38': 'video', # You actually don't know if this will be MOV, AVI or whatever
1042 return (re.match(YoutubeIE._VALID_URL, url) is not None)
1044 def report_lang(self):
1045 """Report attempt to set language."""
1046 self._downloader.to_screen(u'[youtube] Setting language')
1048 def report_login(self):
1049 """Report attempt to log in."""
1050 self._downloader.to_screen(u'[youtube] Logging in')
1052 def report_age_confirmation(self):
1053 """Report attempt to confirm age."""
1054 self._downloader.to_screen(u'[youtube] Confirming age')
1056 def report_video_webpage_download(self, video_id):
1057 """Report attempt to download video webpage."""
1058 self._downloader.to_screen(u'[youtube] %s: Downloading video webpage' % video_id)
1060 def report_video_info_webpage_download(self, video_id):
1061 """Report attempt to download video info webpage."""
1062 self._downloader.to_screen(u'[youtube] %s: Downloading video info webpage' % video_id)
1064 def report_information_extraction(self, video_id):
1065 """Report attempt to extract video information."""
1066 self._downloader.to_screen(u'[youtube] %s: Extracting video information' % video_id)
1068 def report_unavailable_format(self, video_id, format):
1069 """Report extracted video URL."""
1070 self._downloader.to_screen(u'[youtube] %s: Format %s not available' % (video_id, format))
1072 def report_rtmp_download(self):
1073 """Indicate the download will use the RTMP protocol."""
1074 self._downloader.to_screen(u'[youtube] RTMP download detected')
1076 def _real_initialize(self):
1077 if self._downloader is None:
1082 downloader_params = self._downloader.params
1084 # Attempt to use provided username and password or .netrc data
1085 if downloader_params.get('username', None) is not None:
1086 username = downloader_params['username']
1087 password = downloader_params['password']
1088 elif downloader_params.get('usenetrc', False):
1090 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
1091 if info is not None:
1095 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
1096 except (IOError, netrc.NetrcParseError), err:
1097 self._downloader.to_stderr(u'WARNING: parsing .netrc: %s' % str(err))
1101 request = urllib2.Request(self._LANG_URL)
1104 urllib2.urlopen(request).read()
1105 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1106 self._downloader.to_stderr(u'WARNING: unable to set language: %s' % str(err))
1109 # No authentication to be performed
1110 if username is None:
1115 'current_form': 'loginForm',
1117 'action_login': 'Log In',
1118 'username': username,
1119 'password': password,
1121 request = urllib2.Request(self._LOGIN_URL, urllib.urlencode(login_form))
1124 login_results = urllib2.urlopen(request).read()
1125 if re.search(r'(?i)<form[^>]* name="loginForm"', login_results) is not None:
1126 self._downloader.to_stderr(u'WARNING: unable to log in: bad username or password')
1128 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1129 self._downloader.to_stderr(u'WARNING: unable to log in: %s' % str(err))
1135 'action_confirm': 'Confirm',
1137 request = urllib2.Request(self._AGE_URL, urllib.urlencode(age_form))
1139 self.report_age_confirmation()
1140 age_results = urllib2.urlopen(request).read()
1141 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1142 self._downloader.trouble(u'ERROR: unable to confirm age: %s' % str(err))
1145 def _real_extract(self, url):
1146 # Extract video id from URL
1147 mobj = re.match(self._VALID_URL, url)
1149 self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
1151 video_id = mobj.group(2)
1154 self.report_video_webpage_download(video_id)
1155 request = urllib2.Request('http://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id)
1157 video_webpage = urllib2.urlopen(request).read()
1158 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1159 self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err))
1162 # Attempt to extract SWF player URL
1163 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1164 if mobj is not None:
1165 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1170 self.report_video_info_webpage_download(video_id)
1171 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
1172 video_info_url = ('http://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1173 % (video_id, el_type))
1174 request = urllib2.Request(video_info_url)
1176 video_info_webpage = urllib2.urlopen(request).read()
1177 video_info = parse_qs(video_info_webpage)
1178 if 'token' in video_info:
1180 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1181 self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err))
1183 if 'token' not in video_info:
1184 if 'reason' in video_info:
1185 self._downloader.trouble(u'ERROR: YouTube said: %s' % video_info['reason'][0].decode('utf-8'))
1187 self._downloader.trouble(u'ERROR: "token" parameter not in video info for unknown reason')
1190 # Start extracting information
1191 self.report_information_extraction(video_id)
1194 if 'author' not in video_info:
1195 self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
1197 video_uploader = urllib.unquote_plus(video_info['author'][0])
1200 if 'title' not in video_info:
1201 self._downloader.trouble(u'ERROR: unable to extract video title')
1203 video_title = urllib.unquote_plus(video_info['title'][0])
1204 video_title = video_title.decode('utf-8')
1205 video_title = sanitize_title(video_title)
1208 simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
1209 simple_title = simple_title.strip(ur'_')
1212 if 'thumbnail_url' not in video_info:
1213 self._downloader.trouble(u'WARNING: unable to extract video thumbnail')
1214 video_thumbnail = ''
1215 else: # don't panic if we can't find it
1216 video_thumbnail = urllib.unquote_plus(video_info['thumbnail_url'][0])
1220 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
1221 if mobj is not None:
1222 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1223 format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y']
1224 for expression in format_expressions:
1226 upload_date = datetime.datetime.strptime(upload_date, expression).strftime('%Y%m%d')
1234 video_description = u'No description available.'
1235 if self._downloader.params.get('forcedescription', False) or self._downloader.params.get('writedescription', False):
1236 mobj = re.search(r'<meta name="description" content="(.*)"(?:\s*/)?>', video_webpage)
1237 if mobj is not None:
1238 video_description = mobj.group(1).decode('utf-8')
1240 html_parser = lxml.etree.HTMLParser(encoding='utf-8')
1241 vwebpage_doc = lxml.etree.parse(StringIO.StringIO(video_webpage), html_parser)
1242 video_description = u''.join(vwebpage_doc.xpath('id("eow-description")//text()'))
1243 # TODO use another parser
1246 video_token = urllib.unquote_plus(video_info['token'][0])
1248 # Decide which formats to download
1249 req_format = self._downloader.params.get('format', None)
1252 if 'fmt_url_map' in video_info and len(video_info['fmt_url_map']) >= 1:
1253 raw_map = video_info['fmt_url_map'][0]
1254 elif 'fmt_stream_map' in video_info and len(video_info['fmt_stream_map']) >= 1:
1255 raw_map = video_info['fmt_stream_map'][0]
1257 if raw_map is not None:
1258 url_map = dict(tuple(pair.split('|')[:2]) for pair in raw_map.split(','))
1259 format_limit = self._downloader.params.get('format_limit', None)
1260 if format_limit is not None and format_limit in self._available_formats:
1261 format_list = self._available_formats[self._available_formats.index(format_limit):]
1263 format_list = self._available_formats
1264 existing_formats = [x for x in format_list if x in url_map]
1265 if len(existing_formats) == 0:
1266 self._downloader.trouble(u'ERROR: no known formats available for video')
1268 if req_format is None:
1269 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
1270 elif req_format == '-1':
1271 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
1274 if req_format not in url_map:
1275 self._downloader.trouble(u'ERROR: requested format not available')
1277 video_url_list = [(req_format, url_map[req_format])] # Specific format
1279 elif 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1280 self.report_rtmp_download()
1281 video_url_list = [(None, video_info['conn'][0])]
1284 self._downloader.trouble(u'ERROR: no fmt_url_map or conn information found in video info')
1287 for format_param, video_real_url in video_url_list:
1288 # At this point we have a new video
1289 self._downloader.increment_downloads()
1292 video_extension = self._video_extensions.get(format_param, 'flv')
1294 # Find the video URL in fmt_url_map or conn paramters
1296 # Process video information
1297 self._downloader.process_info({
1298 'id': video_id.decode('utf-8'),
1299 'url': video_real_url.decode('utf-8'),
1300 'uploader': video_uploader.decode('utf-8'),
1301 'upload_date': upload_date,
1302 'title': video_title,
1303 'stitle': simple_title,
1304 'ext': video_extension.decode('utf-8'),
1305 'format': (format_param is None and u'NA' or format_param.decode('utf-8')),
1306 'thumbnail': video_thumbnail.decode('utf-8'),
1307 'description': video_description,
1308 'player_url': player_url,
1310 except UnavailableVideoError, err:
1311 self._downloader.trouble(u'\nERROR: unable to download video')
1314 class MetacafeIE(InfoExtractor):
1315 """Information Extractor for metacafe.com."""
1317 _VALID_URL = r'(?:http://)?(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
1318 _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
1319 _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
1322 def __init__(self, youtube_ie, downloader=None):
1323 InfoExtractor.__init__(self, downloader)
1324 self._youtube_ie = youtube_ie
1328 return (re.match(MetacafeIE._VALID_URL, url) is not None)
1330 def report_disclaimer(self):
1331 """Report disclaimer retrieval."""
1332 self._downloader.to_screen(u'[metacafe] Retrieving disclaimer')
1334 def report_age_confirmation(self):
1335 """Report attempt to confirm age."""
1336 self._downloader.to_screen(u'[metacafe] Confirming age')
1338 def report_download_webpage(self, video_id):
1339 """Report webpage download."""
1340 self._downloader.to_screen(u'[metacafe] %s: Downloading webpage' % video_id)
1342 def report_extraction(self, video_id):
1343 """Report information extraction."""
1344 self._downloader.to_screen(u'[metacafe] %s: Extracting information' % video_id)
1346 def _real_initialize(self):
1347 # Retrieve disclaimer
1348 request = urllib2.Request(self._DISCLAIMER)
1350 self.report_disclaimer()
1351 disclaimer = urllib2.urlopen(request).read()
1352 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1353 self._downloader.trouble(u'ERROR: unable to retrieve disclaimer: %s' % str(err))
1359 'submit': "Continue - I'm over 18",
1361 request = urllib2.Request(self._FILTER_POST, urllib.urlencode(disclaimer_form))
1363 self.report_age_confirmation()
1364 disclaimer = urllib2.urlopen(request).read()
1365 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1366 self._downloader.trouble(u'ERROR: unable to confirm age: %s' % str(err))
1369 def _real_extract(self, url):
1370 # Extract id and simplified title from URL
1371 mobj = re.match(self._VALID_URL, url)
1373 self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
1376 video_id = mobj.group(1)
1378 # Check if video comes from YouTube
1379 mobj2 = re.match(r'^yt-(.*)$', video_id)
1380 if mobj2 is not None:
1381 self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % mobj2.group(1))
1384 # At this point we have a new video
1385 self._downloader.increment_downloads()
1387 simple_title = mobj.group(2).decode('utf-8')
1389 # Retrieve video webpage to extract further information
1390 request = urllib2.Request('http://www.metacafe.com/watch/%s/' % video_id)
1392 self.report_download_webpage(video_id)
1393 webpage = urllib2.urlopen(request).read()
1394 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1395 self._downloader.trouble(u'ERROR: unable retrieve video webpage: %s' % str(err))
1398 # Extract URL, uploader and title from webpage
1399 self.report_extraction(video_id)
1400 mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
1401 if mobj is not None:
1402 mediaURL = urllib.unquote(mobj.group(1))
1403 video_extension = mediaURL[-3:]
1405 # Extract gdaKey if available
1406 mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
1408 video_url = mediaURL
1410 gdaKey = mobj.group(1)
1411 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
1413 mobj = re.search(r' name="flashvars" value="(.*?)"', webpage)
1415 self._downloader.trouble(u'ERROR: unable to extract media URL')
1417 vardict = parse_qs(mobj.group(1))
1418 if 'mediaData' not in vardict:
1419 self._downloader.trouble(u'ERROR: unable to extract media URL')
1421 mobj = re.search(r'"mediaURL":"(http.*?)","key":"(.*?)"', vardict['mediaData'][0])
1423 self._downloader.trouble(u'ERROR: unable to extract media URL')
1425 mediaURL = mobj.group(1).replace('\\/', '/')
1426 video_extension = mediaURL[-3:]
1427 video_url = '%s?__gda__=%s' % (mediaURL, mobj.group(2))
1429 mobj = re.search(r'(?im)<title>(.*) - Video</title>', webpage)
1431 self._downloader.trouble(u'ERROR: unable to extract title')
1433 video_title = mobj.group(1).decode('utf-8')
1434 video_title = sanitize_title(video_title)
1436 mobj = re.search(r'(?ms)By:\s*<a .*?>(.+?)<', webpage)
1438 self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
1440 video_uploader = mobj.group(1)
1443 # Process video information
1444 self._downloader.process_info({
1445 'id': video_id.decode('utf-8'),
1446 'url': video_url.decode('utf-8'),
1447 'uploader': video_uploader.decode('utf-8'),
1448 'upload_date': u'NA',
1449 'title': video_title,
1450 'stitle': simple_title,
1451 'ext': video_extension.decode('utf-8'),
1455 except UnavailableVideoError:
1456 self._downloader.trouble(u'\nERROR: unable to download video')
1459 class DailymotionIE(InfoExtractor):
1460 """Information Extractor for Dailymotion"""
1462 _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^_/]+)_([^/]+)'
1464 def __init__(self, downloader=None):
1465 InfoExtractor.__init__(self, downloader)
1469 return (re.match(DailymotionIE._VALID_URL, url) is not None)
1471 def report_download_webpage(self, video_id):
1472 """Report webpage download."""
1473 self._downloader.to_screen(u'[dailymotion] %s: Downloading webpage' % video_id)
1475 def report_extraction(self, video_id):
1476 """Report information extraction."""
1477 self._downloader.to_screen(u'[dailymotion] %s: Extracting information' % video_id)
1479 def _real_initialize(self):
1482 def _real_extract(self, url):
1483 # Extract id and simplified title from URL
1484 mobj = re.match(self._VALID_URL, url)
1486 self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
1489 # At this point we have a new video
1490 self._downloader.increment_downloads()
1491 video_id = mobj.group(1)
1493 simple_title = mobj.group(2).decode('utf-8')
1494 video_extension = 'flv'
1496 # Retrieve video webpage to extract further information
1497 request = urllib2.Request(url)
1499 self.report_download_webpage(video_id)
1500 webpage = urllib2.urlopen(request).read()
1501 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1502 self._downloader.trouble(u'ERROR: unable retrieve video webpage: %s' % str(err))
1505 # Extract URL, uploader and title from webpage
1506 self.report_extraction(video_id)
1507 mobj = re.search(r'(?i)addVariable\(\"video\"\s*,\s*\"([^\"]*)\"\)', webpage)
1509 self._downloader.trouble(u'ERROR: unable to extract media URL')
1511 mediaURL = urllib.unquote(mobj.group(1))
1513 # if needed add http://www.dailymotion.com/ if relative URL
1515 video_url = mediaURL
1517 # '<meta\s+name="title"\s+content="Dailymotion\s*[:\-]\s*(.*?)"\s*\/\s*>'
1518 mobj = re.search(r'(?im)<title>Dailymotion\s*[\-:]\s*(.+?)</title>', webpage)
1520 self._downloader.trouble(u'ERROR: unable to extract title')
1522 video_title = mobj.group(1).decode('utf-8')
1523 video_title = sanitize_title(video_title)
1525 mobj = re.search(r'(?im)<Attribute name="owner">(.+?)</Attribute>', webpage)
1527 self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
1529 video_uploader = mobj.group(1)
1532 # Process video information
1533 self._downloader.process_info({
1534 'id': video_id.decode('utf-8'),
1535 'url': video_url.decode('utf-8'),
1536 'uploader': video_uploader.decode('utf-8'),
1537 'upload_date': u'NA',
1538 'title': video_title,
1539 'stitle': simple_title,
1540 'ext': video_extension.decode('utf-8'),
1544 except UnavailableVideoError:
1545 self._downloader.trouble(u'\nERROR: unable to download video')
1547 class GoogleIE(InfoExtractor):
1548 """Information extractor for video.google.com."""
1550 _VALID_URL = r'(?:http://)?video\.google\.(?:com(?:\.au)?|co\.(?:uk|jp|kr|cr)|ca|de|es|fr|it|nl|pl)/videoplay\?docid=([^\&]+).*'
1552 def __init__(self, downloader=None):
1553 InfoExtractor.__init__(self, downloader)
1557 return (re.match(GoogleIE._VALID_URL, url) is not None)
1559 def report_download_webpage(self, video_id):
1560 """Report webpage download."""
1561 self._downloader.to_screen(u'[video.google] %s: Downloading webpage' % video_id)
1563 def report_extraction(self, video_id):
1564 """Report information extraction."""
1565 self._downloader.to_screen(u'[video.google] %s: Extracting information' % video_id)
1567 def _real_initialize(self):
1570 def _real_extract(self, url):
1571 # Extract id from URL
1572 mobj = re.match(self._VALID_URL, url)
1574 self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
1577 # At this point we have a new video
1578 self._downloader.increment_downloads()
1579 video_id = mobj.group(1)
1581 video_extension = 'mp4'
1583 # Retrieve video webpage to extract further information
1584 request = urllib2.Request('http://video.google.com/videoplay?docid=%s&hl=en&oe=utf-8' % video_id)
1586 self.report_download_webpage(video_id)
1587 webpage = urllib2.urlopen(request).read()
1588 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1589 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1592 # Extract URL, uploader, and title from webpage
1593 self.report_extraction(video_id)
1594 mobj = re.search(r"download_url:'([^']+)'", webpage)
1596 video_extension = 'flv'
1597 mobj = re.search(r"(?i)videoUrl\\x3d(.+?)\\x26", webpage)
1599 self._downloader.trouble(u'ERROR: unable to extract media URL')
1601 mediaURL = urllib.unquote(mobj.group(1))
1602 mediaURL = mediaURL.replace('\\x3d', '\x3d')
1603 mediaURL = mediaURL.replace('\\x26', '\x26')
1605 video_url = mediaURL
1607 mobj = re.search(r'<title>(.*)</title>', webpage)
1609 self._downloader.trouble(u'ERROR: unable to extract title')
1611 video_title = mobj.group(1).decode('utf-8')
1612 video_title = sanitize_title(video_title)
1613 simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
1615 # Extract video description
1616 mobj = re.search(r'<span id=short-desc-content>([^<]*)</span>', webpage)
1618 self._downloader.trouble(u'ERROR: unable to extract video description')
1620 video_description = mobj.group(1).decode('utf-8')
1621 if not video_description:
1622 video_description = 'No description available.'
1624 # Extract video thumbnail
1625 if self._downloader.params.get('forcethumbnail', False):
1626 request = urllib2.Request('http://video.google.com/videosearch?q=%s+site:video.google.com&hl=en' % abs(int(video_id)))
1628 webpage = urllib2.urlopen(request).read()
1629 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1630 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1632 mobj = re.search(r'<img class=thumbnail-img (?:.* )?src=(http.*)>', webpage)
1634 self._downloader.trouble(u'ERROR: unable to extract video thumbnail')
1636 video_thumbnail = mobj.group(1)
1637 else: # we need something to pass to process_info
1638 video_thumbnail = ''
1642 # Process video information
1643 self._downloader.process_info({
1644 'id': video_id.decode('utf-8'),
1645 'url': video_url.decode('utf-8'),
1647 'upload_date': u'NA',
1648 'title': video_title,
1649 'stitle': simple_title,
1650 'ext': video_extension.decode('utf-8'),
1654 except UnavailableVideoError:
1655 self._downloader.trouble(u'\nERROR: unable to download video')
1658 class PhotobucketIE(InfoExtractor):
1659 """Information extractor for photobucket.com."""
1661 _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)'
1663 def __init__(self, downloader=None):
1664 InfoExtractor.__init__(self, downloader)
1668 return (re.match(PhotobucketIE._VALID_URL, url) is not None)
1670 def report_download_webpage(self, video_id):
1671 """Report webpage download."""
1672 self._downloader.to_screen(u'[photobucket] %s: Downloading webpage' % video_id)
1674 def report_extraction(self, video_id):
1675 """Report information extraction."""
1676 self._downloader.to_screen(u'[photobucket] %s: Extracting information' % video_id)
1678 def _real_initialize(self):
1681 def _real_extract(self, url):
1682 # Extract id from URL
1683 mobj = re.match(self._VALID_URL, url)
1685 self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
1688 # At this point we have a new video
1689 self._downloader.increment_downloads()
1690 video_id = mobj.group(1)
1692 video_extension = 'flv'
1694 # Retrieve video webpage to extract further information
1695 request = urllib2.Request(url)
1697 self.report_download_webpage(video_id)
1698 webpage = urllib2.urlopen(request).read()
1699 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1700 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1703 # Extract URL, uploader, and title from webpage
1704 self.report_extraction(video_id)
1705 mobj = re.search(r'<link rel="video_src" href=".*\?file=([^"]+)" />', webpage)
1707 self._downloader.trouble(u'ERROR: unable to extract media URL')
1709 mediaURL = urllib.unquote(mobj.group(1))
1711 video_url = mediaURL
1713 mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', webpage)
1715 self._downloader.trouble(u'ERROR: unable to extract title')
1717 video_title = mobj.group(1).decode('utf-8')
1718 video_title = sanitize_title(video_title)
1719 simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
1721 video_uploader = mobj.group(2).decode('utf-8')
1724 # Process video information
1725 self._downloader.process_info({
1726 'id': video_id.decode('utf-8'),
1727 'url': video_url.decode('utf-8'),
1728 'uploader': video_uploader,
1729 'upload_date': u'NA',
1730 'title': video_title,
1731 'stitle': simple_title,
1732 'ext': video_extension.decode('utf-8'),
1736 except UnavailableVideoError:
1737 self._downloader.trouble(u'\nERROR: unable to download video')
1740 class YahooIE(InfoExtractor):
1741 """Information extractor for video.yahoo.com."""
1743 # _VALID_URL matches all Yahoo! Video URLs
1744 # _VPAGE_URL matches only the extractable '/watch/' URLs
1745 _VALID_URL = r'(?:http://)?(?:[a-z]+\.)?video\.yahoo\.com/(?:watch|network)/([0-9]+)(?:/|\?v=)([0-9]+)(?:[#\?].*)?'
1746 _VPAGE_URL = r'(?:http://)?video\.yahoo\.com/watch/([0-9]+)/([0-9]+)(?:[#\?].*)?'
1748 def __init__(self, downloader=None):
1749 InfoExtractor.__init__(self, downloader)
1753 return (re.match(YahooIE._VALID_URL, url) is not None)
1755 def report_download_webpage(self, video_id):
1756 """Report webpage download."""
1757 self._downloader.to_screen(u'[video.yahoo] %s: Downloading webpage' % video_id)
1759 def report_extraction(self, video_id):
1760 """Report information extraction."""
1761 self._downloader.to_screen(u'[video.yahoo] %s: Extracting information' % video_id)
1763 def _real_initialize(self):
1766 def _real_extract(self, url, new_video=True):
1767 # Extract ID from URL
1768 mobj = re.match(self._VALID_URL, url)
1770 self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
1773 # At this point we have a new video
1774 self._downloader.increment_downloads()
1775 video_id = mobj.group(2)
1776 video_extension = 'flv'
1778 # Rewrite valid but non-extractable URLs as
1779 # extractable English language /watch/ URLs
1780 if re.match(self._VPAGE_URL, url) is None:
1781 request = urllib2.Request(url)
1783 webpage = urllib2.urlopen(request).read()
1784 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1785 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1788 mobj = re.search(r'\("id", "([0-9]+)"\);', webpage)
1790 self._downloader.trouble(u'ERROR: Unable to extract id field')
1792 yahoo_id = mobj.group(1)
1794 mobj = re.search(r'\("vid", "([0-9]+)"\);', webpage)
1796 self._downloader.trouble(u'ERROR: Unable to extract vid field')
1798 yahoo_vid = mobj.group(1)
1800 url = 'http://video.yahoo.com/watch/%s/%s' % (yahoo_vid, yahoo_id)
1801 return self._real_extract(url, new_video=False)
1803 # Retrieve video webpage to extract further information
1804 request = urllib2.Request(url)
1806 self.report_download_webpage(video_id)
1807 webpage = urllib2.urlopen(request).read()
1808 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1809 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1812 # Extract uploader and title from webpage
1813 self.report_extraction(video_id)
1814 mobj = re.search(r'<meta name="title" content="(.*)" />', webpage)
1816 self._downloader.trouble(u'ERROR: unable to extract video title')
1818 video_title = mobj.group(1).decode('utf-8')
1819 simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
1821 mobj = re.search(r'<h2 class="ti-5"><a href="http://video\.yahoo\.com/(people|profile)/[0-9]+" beacon=".*">(.*)</a></h2>', webpage)
1823 self._downloader.trouble(u'ERROR: unable to extract video uploader')
1825 video_uploader = mobj.group(1).decode('utf-8')
1827 # Extract video thumbnail
1828 mobj = re.search(r'<link rel="image_src" href="(.*)" />', webpage)
1830 self._downloader.trouble(u'ERROR: unable to extract video thumbnail')
1832 video_thumbnail = mobj.group(1).decode('utf-8')
1834 # Extract video description
1835 mobj = re.search(r'<meta name="description" content="(.*)" />', webpage)
1837 self._downloader.trouble(u'ERROR: unable to extract video description')
1839 video_description = mobj.group(1).decode('utf-8')
1840 if not video_description: video_description = 'No description available.'
1842 # Extract video height and width
1843 mobj = re.search(r'<meta name="video_height" content="([0-9]+)" />', webpage)
1845 self._downloader.trouble(u'ERROR: unable to extract video height')
1847 yv_video_height = mobj.group(1)
1849 mobj = re.search(r'<meta name="video_width" content="([0-9]+)" />', webpage)
1851 self._downloader.trouble(u'ERROR: unable to extract video width')
1853 yv_video_width = mobj.group(1)
1855 # Retrieve video playlist to extract media URL
1856 # I'm not completely sure what all these options are, but we
1857 # seem to need most of them, otherwise the server sends a 401.
1858 yv_lg = 'R0xx6idZnW2zlrKP8xxAIR' # not sure what this represents
1859 yv_bitrate = '700' # according to Wikipedia this is hard-coded
1860 request = urllib2.Request('http://cosmos.bcst.yahoo.com/up/yep/process/getPlaylistFOP.php?node_id=' + video_id +
1861 '&tech=flash&mode=playlist&lg=' + yv_lg + '&bitrate=' + yv_bitrate + '&vidH=' + yv_video_height +
1862 '&vidW=' + yv_video_width + '&swf=as3&rd=video.yahoo.com&tk=null&adsupported=v1,v2,&eventid=1301797')
1864 self.report_download_webpage(video_id)
1865 webpage = urllib2.urlopen(request).read()
1866 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1867 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1870 # Extract media URL from playlist XML
1871 mobj = re.search(r'<STREAM APP="(http://.*)" FULLPATH="/?(/.*\.flv\?[^"]*)"', webpage)
1873 self._downloader.trouble(u'ERROR: Unable to extract media URL')
1875 video_url = urllib.unquote(mobj.group(1) + mobj.group(2)).decode('utf-8')
1876 video_url = re.sub(r'(?u)&(.+?);', htmlentity_transform, video_url)
1879 # Process video information
1880 self._downloader.process_info({
1881 'id': video_id.decode('utf-8'),
1883 'uploader': video_uploader,
1884 'upload_date': u'NA',
1885 'title': video_title,
1886 'stitle': simple_title,
1887 'ext': video_extension.decode('utf-8'),
1888 'thumbnail': video_thumbnail.decode('utf-8'),
1889 'description': video_description,
1890 'thumbnail': video_thumbnail,
1891 'description': video_description,
1894 except UnavailableVideoError:
1895 self._downloader.trouble(u'\nERROR: unable to download video')
1898 class GenericIE(InfoExtractor):
1899 """Generic last-resort information extractor."""
1901 def __init__(self, downloader=None):
1902 InfoExtractor.__init__(self, downloader)
1908 def report_download_webpage(self, video_id):
1909 """Report webpage download."""
1910 self._downloader.to_screen(u'WARNING: Falling back on generic information extractor.')
1911 self._downloader.to_screen(u'[generic] %s: Downloading webpage' % video_id)
1913 def report_extraction(self, video_id):
1914 """Report information extraction."""
1915 self._downloader.to_screen(u'[generic] %s: Extracting information' % video_id)
1917 def _real_initialize(self):
1920 def _real_extract(self, url):
1921 # At this point we have a new video
1922 self._downloader.increment_downloads()
1924 video_id = url.split('/')[-1]
1925 request = urllib2.Request(url)
1927 self.report_download_webpage(video_id)
1928 webpage = urllib2.urlopen(request).read()
1929 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
1930 self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err))
1932 except ValueError, err:
1933 # since this is the last-resort InfoExtractor, if
1934 # this error is thrown, it'll be thrown here
1935 self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
1938 self.report_extraction(video_id)
1939 # Start with something easy: JW Player in SWFObject
1940 mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
1942 # Broaden the search a little bit
1943 mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
1945 self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
1948 # It's possible that one of the regexes
1949 # matched, but returned an empty group:
1950 if mobj.group(1) is None:
1951 self._downloader.trouble(u'ERROR: Invalid URL: %s' % url)
1954 video_url = urllib.unquote(mobj.group(1))
1955 video_id = os.path.basename(video_url)
1957 # here's a fun little line of code for you:
1958 video_extension = os.path.splitext(video_id)[1][1:]
1959 video_id = os.path.splitext(video_id)[0]
1961 # it's tempting to parse this further, but you would
1962 # have to take into account all the variations like
1963 # Video Title - Site Name
1964 # Site Name | Video Title
1965 # Video Title - Tagline | Site Name
1966 # and so on and so forth; it's just not practical
1967 mobj = re.search(r'<title>(.*)</title>', webpage)
1969 self._downloader.trouble(u'ERROR: unable to extract title')
1971 video_title = mobj.group(1).decode('utf-8')
1972 video_title = sanitize_title(video_title)
1973 simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
1975 # video uploader is domain name
1976 mobj = re.match(r'(?:https?://)?([^/]*)/.*', url)
1978 self._downloader.trouble(u'ERROR: unable to extract title')
1980 video_uploader = mobj.group(1).decode('utf-8')
1983 # Process video information
1984 self._downloader.process_info({
1985 'id': video_id.decode('utf-8'),
1986 'url': video_url.decode('utf-8'),
1987 'uploader': video_uploader,
1988 'upload_date': u'NA',
1989 'title': video_title,
1990 'stitle': simple_title,
1991 'ext': video_extension.decode('utf-8'),
1995 except UnavailableVideoError, err:
1996 self._downloader.trouble(u'\nERROR: unable to download video')
1999 class YoutubeSearchIE(InfoExtractor):
2000 """Information Extractor for YouTube search queries."""
2001 _VALID_QUERY = r'ytsearch(\d+|all)?:[\s\S]+'
2002 _TEMPLATE_URL = 'http://www.youtube.com/results?search_query=%s&page=%s&gl=US&hl=en'
2003 _VIDEO_INDICATOR = r'href="/watch\?v=.+?"'
2004 _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*</a>'
2006 _max_youtube_results = 1000
2008 def __init__(self, youtube_ie, downloader=None):
2009 InfoExtractor.__init__(self, downloader)
2010 self._youtube_ie = youtube_ie
2014 return (re.match(YoutubeSearchIE._VALID_QUERY, url) is not None)
2016 def report_download_page(self, query, pagenum):
2017 """Report attempt to download playlist page with given number."""
2018 query = query.decode(preferredencoding())
2019 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
2021 def _real_initialize(self):
2022 self._youtube_ie.initialize()
2024 def _real_extract(self, query):
2025 mobj = re.match(self._VALID_QUERY, query)
2027 self._downloader.trouble(u'ERROR: invalid search query "%s"' % query)
2030 prefix, query = query.split(':')
2032 query = query.encode('utf-8')
2034 self._download_n_results(query, 1)
2036 elif prefix == 'all':
2037 self._download_n_results(query, self._max_youtube_results)
2043 self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
2045 elif n > self._max_youtube_results:
2046 self._downloader.to_stderr(u'WARNING: ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n))
2047 n = self._max_youtube_results
2048 self._download_n_results(query, n)
2050 except ValueError: # parsing prefix as integer fails
2051 self._download_n_results(query, 1)
2054 def _download_n_results(self, query, n):
2055 """Downloads a specified number of results for a query"""
2058 already_seen = set()
2062 self.report_download_page(query, pagenum)
2063 result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
2064 request = urllib2.Request(result_url)
2066 page = urllib2.urlopen(request).read()
2067 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2068 self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
2071 # Extract video identifiers
2072 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
2073 video_id = page[mobj.span()[0]:mobj.span()[1]].split('=')[2][:-1]
2074 if video_id not in already_seen:
2075 video_ids.append(video_id)
2076 already_seen.add(video_id)
2077 if len(video_ids) == n:
2078 # Specified n videos reached
2079 for id in video_ids:
2080 self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
2083 if re.search(self._MORE_PAGES_INDICATOR, page) is None:
2084 for id in video_ids:
2085 self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
2088 pagenum = pagenum + 1
2090 class GoogleSearchIE(InfoExtractor):
2091 """Information Extractor for Google Video search queries."""
2092 _VALID_QUERY = r'gvsearch(\d+|all)?:[\s\S]+'
2093 _TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en'
2094 _VIDEO_INDICATOR = r'videoplay\?docid=([^\&>]+)\&'
2095 _MORE_PAGES_INDICATOR = r'<span>Next</span>'
2097 _max_google_results = 1000
2099 def __init__(self, google_ie, downloader=None):
2100 InfoExtractor.__init__(self, downloader)
2101 self._google_ie = google_ie
2105 return (re.match(GoogleSearchIE._VALID_QUERY, url) is not None)
2107 def report_download_page(self, query, pagenum):
2108 """Report attempt to download playlist page with given number."""
2109 query = query.decode(preferredencoding())
2110 self._downloader.to_screen(u'[video.google] query "%s": Downloading page %s' % (query, pagenum))
2112 def _real_initialize(self):
2113 self._google_ie.initialize()
2115 def _real_extract(self, query):
2116 mobj = re.match(self._VALID_QUERY, query)
2118 self._downloader.trouble(u'ERROR: invalid search query "%s"' % query)
2121 prefix, query = query.split(':')
2123 query = query.encode('utf-8')
2125 self._download_n_results(query, 1)
2127 elif prefix == 'all':
2128 self._download_n_results(query, self._max_google_results)
2134 self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
2136 elif n > self._max_google_results:
2137 self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n))
2138 n = self._max_google_results
2139 self._download_n_results(query, n)
2141 except ValueError: # parsing prefix as integer fails
2142 self._download_n_results(query, 1)
2145 def _download_n_results(self, query, n):
2146 """Downloads a specified number of results for a query"""
2149 already_seen = set()
2153 self.report_download_page(query, pagenum)
2154 result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
2155 request = urllib2.Request(result_url)
2157 page = urllib2.urlopen(request).read()
2158 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2159 self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
2162 # Extract video identifiers
2163 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
2164 video_id = mobj.group(1)
2165 if video_id not in already_seen:
2166 video_ids.append(video_id)
2167 already_seen.add(video_id)
2168 if len(video_ids) == n:
2169 # Specified n videos reached
2170 for id in video_ids:
2171 self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id)
2174 if re.search(self._MORE_PAGES_INDICATOR, page) is None:
2175 for id in video_ids:
2176 self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id)
2179 pagenum = pagenum + 1
2181 class YahooSearchIE(InfoExtractor):
2182 """Information Extractor for Yahoo! Video search queries."""
2183 _VALID_QUERY = r'yvsearch(\d+|all)?:[\s\S]+'
2184 _TEMPLATE_URL = 'http://video.yahoo.com/search/?p=%s&o=%s'
2185 _VIDEO_INDICATOR = r'href="http://video\.yahoo\.com/watch/([0-9]+/[0-9]+)"'
2186 _MORE_PAGES_INDICATOR = r'\s*Next'
2188 _max_yahoo_results = 1000
2190 def __init__(self, yahoo_ie, downloader=None):
2191 InfoExtractor.__init__(self, downloader)
2192 self._yahoo_ie = yahoo_ie
2196 return (re.match(YahooSearchIE._VALID_QUERY, url) is not None)
2198 def report_download_page(self, query, pagenum):
2199 """Report attempt to download playlist page with given number."""
2200 query = query.decode(preferredencoding())
2201 self._downloader.to_screen(u'[video.yahoo] query "%s": Downloading page %s' % (query, pagenum))
2203 def _real_initialize(self):
2204 self._yahoo_ie.initialize()
2206 def _real_extract(self, query):
2207 mobj = re.match(self._VALID_QUERY, query)
2209 self._downloader.trouble(u'ERROR: invalid search query "%s"' % query)
2212 prefix, query = query.split(':')
2214 query = query.encode('utf-8')
2216 self._download_n_results(query, 1)
2218 elif prefix == 'all':
2219 self._download_n_results(query, self._max_yahoo_results)
2225 self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query))
2227 elif n > self._max_yahoo_results:
2228 self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n))
2229 n = self._max_yahoo_results
2230 self._download_n_results(query, n)
2232 except ValueError: # parsing prefix as integer fails
2233 self._download_n_results(query, 1)
2236 def _download_n_results(self, query, n):
2237 """Downloads a specified number of results for a query"""
2240 already_seen = set()
2244 self.report_download_page(query, pagenum)
2245 result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum)
2246 request = urllib2.Request(result_url)
2248 page = urllib2.urlopen(request).read()
2249 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2250 self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
2253 # Extract video identifiers
2254 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
2255 video_id = mobj.group(1)
2256 if video_id not in already_seen:
2257 video_ids.append(video_id)
2258 already_seen.add(video_id)
2259 if len(video_ids) == n:
2260 # Specified n videos reached
2261 for id in video_ids:
2262 self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id)
2265 if re.search(self._MORE_PAGES_INDICATOR, page) is None:
2266 for id in video_ids:
2267 self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id)
2270 pagenum = pagenum + 1
2272 class YoutubePlaylistIE(InfoExtractor):
2273 """Information Extractor for YouTube playlists."""
2275 _VALID_URL = r'(?:http://)?(?:\w+\.)?youtube.com/(?:(?:view_play_list|my_playlists|artist)\?.*?(p|a)=|user/.*?/user/|p/|user/.*?#[pg]/c/)([0-9A-Za-z]+)(?:/.*?/([0-9A-Za-z_-]+))?.*'
2276 _TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en'
2277 _VIDEO_INDICATOR = r'/watch\?v=(.+?)&'
2278 _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*</a>'
2281 def __init__(self, youtube_ie, downloader=None):
2282 InfoExtractor.__init__(self, downloader)
2283 self._youtube_ie = youtube_ie
2287 return (re.match(YoutubePlaylistIE._VALID_URL, url) is not None)
2289 def report_download_page(self, playlist_id, pagenum):
2290 """Report attempt to download playlist page with given number."""
2291 self._downloader.to_screen(u'[youtube] PL %s: Downloading page #%s' % (playlist_id, pagenum))
2293 def _real_initialize(self):
2294 self._youtube_ie.initialize()
2296 def _real_extract(self, url):
2297 # Extract playlist id
2298 mobj = re.match(self._VALID_URL, url)
2300 self._downloader.trouble(u'ERROR: invalid url: %s' % url)
2304 if mobj.group(3) is not None:
2305 self._youtube_ie.extract(mobj.group(3))
2308 # Download playlist pages
2309 # prefix is 'p' as default for playlists but there are other types that need extra care
2310 playlist_prefix = mobj.group(1)
2311 if playlist_prefix == 'a':
2312 playlist_access = 'artist'
2314 playlist_prefix = 'p'
2315 playlist_access = 'view_play_list'
2316 playlist_id = mobj.group(2)
2321 self.report_download_page(playlist_id, pagenum)
2322 request = urllib2.Request(self._TEMPLATE_URL % (playlist_access, playlist_prefix, playlist_id, pagenum))
2324 page = urllib2.urlopen(request).read()
2325 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2326 self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
2329 # Extract video identifiers
2331 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
2332 if mobj.group(1) not in ids_in_page:
2333 ids_in_page.append(mobj.group(1))
2334 video_ids.extend(ids_in_page)
2336 if re.search(self._MORE_PAGES_INDICATOR, page) is None:
2338 pagenum = pagenum + 1
2340 playliststart = self._downloader.params.get('playliststart', 1) - 1
2341 playlistend = self._downloader.params.get('playlistend', -1)
2342 video_ids = video_ids[playliststart:playlistend]
2344 for id in video_ids:
2345 self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id)
2348 class YoutubeUserIE(InfoExtractor):
2349 """Information Extractor for YouTube users."""
2351 _VALID_URL = r'(?:(?:(?:http://)?(?:\w+\.)?youtube.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
2352 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
2353 _GDATA_PAGE_SIZE = 50
2354 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
2355 _VIDEO_INDICATOR = r'/watch\?v=(.+?)&'
2358 def __init__(self, youtube_ie, downloader=None):
2359 InfoExtractor.__init__(self, downloader)
2360 self._youtube_ie = youtube_ie
2364 return (re.match(YoutubeUserIE._VALID_URL, url) is not None)
2366 def report_download_page(self, username, start_index):
2367 """Report attempt to download user page."""
2368 self._downloader.to_screen(u'[youtube] user %s: Downloading video ids from %d to %d' %
2369 (username, start_index, start_index + self._GDATA_PAGE_SIZE))
2371 def _real_initialize(self):
2372 self._youtube_ie.initialize()
2374 def _real_extract(self, url):
2376 mobj = re.match(self._VALID_URL, url)
2378 self._downloader.trouble(u'ERROR: invalid url: %s' % url)
2381 username = mobj.group(1)
2383 # Download video ids using YouTube Data API. Result size per
2384 # query is limited (currently to 50 videos) so we need to query
2385 # page by page until there are no video ids - it means we got
2392 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
2393 self.report_download_page(username, start_index)
2395 request = urllib2.Request(self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index))
2398 page = urllib2.urlopen(request).read()
2399 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2400 self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
2403 # Extract video identifiers
2406 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
2407 if mobj.group(1) not in ids_in_page:
2408 ids_in_page.append(mobj.group(1))
2410 video_ids.extend(ids_in_page)
2412 # A little optimization - if current page is not
2413 # "full", ie. does not contain PAGE_SIZE video ids then
2414 # we can assume that this page is the last one - there
2415 # are no more ids on further pages - no need to query
2418 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
2423 all_ids_count = len(video_ids)
2424 playliststart = self._downloader.params.get('playliststart', 1) - 1
2425 playlistend = self._downloader.params.get('playlistend', -1)
2427 if playlistend == -1:
2428 video_ids = video_ids[playliststart:]
2430 video_ids = video_ids[playliststart:playlistend]
2432 self._downloader.to_screen("[youtube] user %s: Collected %d video ids (downloading %d of them)" %
2433 (username, all_ids_count, len(video_ids)))
2435 for video_id in video_ids:
2436 self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % video_id)
2439 class DepositFilesIE(InfoExtractor):
2440 """Information extractor for depositfiles.com"""
2442 _VALID_URL = r'(?:http://)?(?:\w+\.)?depositfiles.com/(?:../(?#locale))?files/(.+)'
2444 def __init__(self, downloader=None):
2445 InfoExtractor.__init__(self, downloader)
2449 return (re.match(DepositFilesIE._VALID_URL, url) is not None)
2451 def report_download_webpage(self, file_id):
2452 """Report webpage download."""
2453 self._downloader.to_screen(u'[DepositFiles] %s: Downloading webpage' % file_id)
2455 def report_extraction(self, file_id):
2456 """Report information extraction."""
2457 self._downloader.to_screen(u'[DepositFiles] %s: Extracting information' % file_id)
2459 def _real_initialize(self):
2462 def _real_extract(self, url):
2463 # At this point we have a new file
2464 self._downloader.increment_downloads()
2466 file_id = url.split('/')[-1]
2467 # Rebuild url in english locale
2468 url = 'http://depositfiles.com/en/files/' + file_id
2470 # Retrieve file webpage with 'Free download' button pressed
2471 free_download_indication = { 'gateway_result' : '1' }
2472 request = urllib2.Request(url, urllib.urlencode(free_download_indication))
2474 self.report_download_webpage(file_id)
2475 webpage = urllib2.urlopen(request).read()
2476 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2477 self._downloader.trouble(u'ERROR: Unable to retrieve file webpage: %s' % str(err))
2480 # Search for the real file URL
2481 mobj = re.search(r'<form action="(http://fileshare.+?)"', webpage)
2482 if (mobj is None) or (mobj.group(1) is None):
2483 # Try to figure out reason of the error.
2484 mobj = re.search(r'<strong>(Attention.*?)</strong>', webpage, re.DOTALL)
2485 if (mobj is not None) and (mobj.group(1) is not None):
2486 restriction_message = re.sub('\s+', ' ', mobj.group(1)).strip()
2487 self._downloader.trouble(u'ERROR: %s' % restriction_message)
2489 self._downloader.trouble(u'ERROR: unable to extract download URL from: %s' % url)
2492 file_url = mobj.group(1)
2493 file_extension = os.path.splitext(file_url)[1][1:]
2495 # Search for file title
2496 mobj = re.search(r'<b title="(.*?)">', webpage)
2498 self._downloader.trouble(u'ERROR: unable to extract title')
2500 file_title = mobj.group(1).decode('utf-8')
2503 # Process file information
2504 self._downloader.process_info({
2505 'id': file_id.decode('utf-8'),
2506 'url': file_url.decode('utf-8'),
2508 'upload_date': u'NA',
2509 'title': file_title,
2510 'stitle': file_title,
2511 'ext': file_extension.decode('utf-8'),
2515 except UnavailableVideoError, err:
2516 self._downloader.trouble(u'ERROR: unable to download file')
2518 class FacebookIE(InfoExtractor):
2519 """Information Extractor for Facebook"""
2521 _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook.com/video/video.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
2522 _LOGIN_URL = 'https://login.facebook.com/login.php?m&next=http%3A%2F%2Fm.facebook.com%2Fhome.php&'
2523 _NETRC_MACHINE = 'facebook'
2524 _available_formats = ['highqual', 'lowqual']
2525 _video_extensions = {
2530 def __init__(self, downloader=None):
2531 InfoExtractor.__init__(self, downloader)
2535 return (re.match(FacebookIE._VALID_URL, url) is not None)
2537 def _reporter(self, message):
2538 """Add header and report message."""
2539 self._downloader.to_screen(u'[facebook] %s' % message)
2541 def report_login(self):
2542 """Report attempt to log in."""
2543 self._reporter(u'Logging in')
2545 def report_video_webpage_download(self, video_id):
2546 """Report attempt to download video webpage."""
2547 self._reporter(u'%s: Downloading video webpage' % video_id)
2549 def report_information_extraction(self, video_id):
2550 """Report attempt to extract video information."""
2551 self._reporter(u'%s: Extracting video information' % video_id)
2553 def _parse_page(self, video_webpage):
2554 """Extract video information from page"""
2556 data = {'title': r'class="video_title datawrap">(.*?)</',
2557 'description': r'<div class="datawrap">(.*?)</div>',
2558 'owner': r'\("video_owner_name", "(.*?)"\)',
2559 'upload_date': r'data-date="(.*?)"',
2560 'thumbnail': r'\("thumb_url", "(?P<THUMB>.*?)"\)',
2563 for piece in data.keys():
2564 mobj = re.search(data[piece], video_webpage)
2565 if mobj is not None:
2566 video_info[piece] = urllib.unquote_plus(mobj.group(1).decode("unicode_escape"))
2570 for fmt in self._available_formats:
2571 mobj = re.search(r'\("%s_src\", "(.+?)"\)' % fmt, video_webpage)
2572 if mobj is not None:
2573 # URL is in a Javascript segment inside an escaped Unicode format within
2574 # the generally utf-8 page
2575 video_urls[fmt] = urllib.unquote_plus(mobj.group(1).decode("unicode_escape"))
2576 video_info['video_urls'] = video_urls
2580 def _real_initialize(self):
2581 if self._downloader is None:
2586 downloader_params = self._downloader.params
2588 # Attempt to use provided username and password or .netrc data
2589 if downloader_params.get('username', None) is not None:
2590 useremail = downloader_params['username']
2591 password = downloader_params['password']
2592 elif downloader_params.get('usenetrc', False):
2594 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
2595 if info is not None:
2599 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
2600 except (IOError, netrc.NetrcParseError), err:
2601 self._downloader.to_stderr(u'WARNING: parsing .netrc: %s' % str(err))
2604 if useremail is None:
2613 request = urllib2.Request(self._LOGIN_URL, urllib.urlencode(login_form))
2616 login_results = urllib2.urlopen(request).read()
2617 if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
2618 self._downloader.to_stderr(u'WARNING: unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
2620 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2621 self._downloader.to_stderr(u'WARNING: unable to log in: %s' % str(err))
2624 def _real_extract(self, url):
2625 mobj = re.match(self._VALID_URL, url)
2627 self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
2629 video_id = mobj.group('ID')
2632 self.report_video_webpage_download(video_id)
2633 request = urllib2.Request('https://www.facebook.com/video/video.php?v=%s' % video_id)
2635 page = urllib2.urlopen(request)
2636 video_webpage = page.read()
2637 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2638 self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err))
2641 # Start extracting information
2642 self.report_information_extraction(video_id)
2644 # Extract information
2645 video_info = self._parse_page(video_webpage)
2648 if 'owner' not in video_info:
2649 self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
2651 video_uploader = video_info['owner']
2654 if 'title' not in video_info:
2655 self._downloader.trouble(u'ERROR: unable to extract video title')
2657 video_title = video_info['title']
2658 video_title = video_title.decode('utf-8')
2659 video_title = sanitize_title(video_title)
2662 simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title)
2663 simple_title = simple_title.strip(ur'_')
2666 if 'thumbnail' not in video_info:
2667 self._downloader.trouble(u'WARNING: unable to extract video thumbnail')
2668 video_thumbnail = ''
2670 video_thumbnail = video_info['thumbnail']
2674 if 'upload_date' in video_info:
2675 upload_time = video_info['upload_date']
2676 timetuple = email.utils.parsedate_tz(upload_time)
2677 if timetuple is not None:
2679 upload_date = time.strftime('%Y%m%d', timetuple[0:9])
2684 video_description = video_info.get('description', 'No description available.')
2686 url_map = video_info['video_urls']
2687 if len(url_map.keys()) > 0:
2688 # Decide which formats to download
2689 req_format = self._downloader.params.get('format', None)
2690 format_limit = self._downloader.params.get('format_limit', None)
2692 if format_limit is not None and format_limit in self._available_formats:
2693 format_list = self._available_formats[self._available_formats.index(format_limit):]
2695 format_list = self._available_formats
2696 existing_formats = [x for x in format_list if x in url_map]
2697 if len(existing_formats) == 0:
2698 self._downloader.trouble(u'ERROR: no known formats available for video')
2700 if req_format is None:
2701 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
2702 elif req_format == '-1':
2703 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
2706 if req_format not in url_map:
2707 self._downloader.trouble(u'ERROR: requested format not available')
2709 video_url_list = [(req_format, url_map[req_format])] # Specific format
2711 for format_param, video_real_url in video_url_list:
2713 # At this point we have a new video
2714 self._downloader.increment_downloads()
2717 video_extension = self._video_extensions.get(format_param, 'mp4')
2719 # Find the video URL in fmt_url_map or conn paramters
2721 # Process video information
2722 self._downloader.process_info({
2723 'id': video_id.decode('utf-8'),
2724 'url': video_real_url.decode('utf-8'),
2725 'uploader': video_uploader.decode('utf-8'),
2726 'upload_date': upload_date,
2727 'title': video_title,
2728 'stitle': simple_title,
2729 'ext': video_extension.decode('utf-8'),
2730 'format': (format_param is None and u'NA' or format_param.decode('utf-8')),
2731 'thumbnail': video_thumbnail.decode('utf-8'),
2732 'description': video_description.decode('utf-8'),
2735 except UnavailableVideoError, err:
2736 self._downloader.trouble(u'\nERROR: unable to download video')
2738 class BlipTVIE(InfoExtractor):
2739 """Information extractor for blip.tv"""
2741 _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip.tv(/.+)$'
2742 _URL_EXT = r'^.*\.([a-z0-9]+)$'
2746 return (re.match(BlipTVIE._VALID_URL, url) is not None)
2748 def report_extraction(self, file_id):
2749 """Report information extraction."""
2750 self._downloader.to_screen(u'[blip.tv] %s: Extracting information' % file_id)
2752 def _simplify_title(self, title):
2753 res = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', title)
2754 res = res.strip(ur'_')
2757 def _real_extract(self, url):
2758 mobj = re.match(self._VALID_URL, url)
2760 self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
2763 json_url = url + ('&' if '?' in url else '?') + 'skin=json&version=2&no_wrap=1'
2764 request = urllib2.Request(json_url)
2765 self.report_extraction(mobj.group(1))
2767 json_code = urllib2.urlopen(request).read()
2768 except (urllib2.URLError, httplib.HTTPException, socket.error), err:
2769 self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err))
2772 json_data = json.loads(json_code)
2773 data = json_data['Post'] if 'Post' in json_data else json_data
2775 upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
2776 video_url = data['media']['url']
2777 umobj = re.match(self._URL_EXT, video_url)
2779 raise ValueError('Can not determine filename extension')
2780 ext = umobj.group(1)
2782 self._downloader.increment_downloads()
2785 'id': data['item_id'],
2787 'uploader': data['display_name'],
2788 'upload_date': upload_date,
2789 'title': data['title'],
2790 'stitle': self._simplify_title(data['title']),
2792 'format': data['media']['mimeType'],
2793 'thumbnail': data['thumbnailUrl'],
2794 'description': data['description'],
2795 'player_url': data['embedUrl']
2797 except (ValueError,KeyError), err:
2798 self._downloader.trouble(u'ERROR: unable to parse video information: %s' % repr(err))
2802 self._downloader.process_info(info)
2803 except UnavailableVideoError, err:
2804 self._downloader.trouble(u'\nERROR: unable to download video')
2807 class PostProcessor(object):
2808 """Post Processor class.
2810 PostProcessor objects can be added to downloaders with their
2811 add_post_processor() method. When the downloader has finished a
2812 successful download, it will take its internal chain of PostProcessors
2813 and start calling the run() method on each one of them, first with
2814 an initial argument and then with the returned value of the previous
2817 The chain will be stopped if one of them ever returns None or the end
2818 of the chain is reached.
2820 PostProcessor objects follow a "mutual registration" process similar
2821 to InfoExtractor objects.
2826 def __init__(self, downloader=None):
2827 self._downloader = downloader
2829 def set_downloader(self, downloader):
2830 """Sets the downloader for this PP."""
2831 self._downloader = downloader
2833 def run(self, information):
2834 """Run the PostProcessor.
2836 The "information" argument is a dictionary like the ones
2837 composed by InfoExtractors. The only difference is that this
2838 one has an extra field called "filepath" that points to the
2841 When this method returns None, the postprocessing chain is
2842 stopped. However, this method may return an information
2843 dictionary that will be passed to the next postprocessing
2844 object in the chain. It can be the one it received after
2845 changing some fields.
2847 In addition, this method may raise a PostProcessingError
2848 exception that will be taken into account by the downloader
2851 return information # by default, do nothing
2853 class FFmpegExtractAudioPP(PostProcessor):
2855 def __init__(self, downloader=None, preferredcodec=None):
2856 PostProcessor.__init__(self, downloader)
2857 if preferredcodec is None:
2858 preferredcodec = 'best'
2859 self._preferredcodec = preferredcodec
2862 def get_audio_codec(path):
2864 cmd = ['ffprobe', '-show_streams', '--', path]
2865 handle = subprocess.Popen(cmd, stderr=file(os.path.devnull, 'w'), stdout=subprocess.PIPE)
2866 output = handle.communicate()[0]
2867 if handle.wait() != 0:
2869 except (IOError, OSError):
2872 for line in output.split('\n'):
2873 if line.startswith('codec_name='):
2874 audio_codec = line.split('=')[1].strip()
2875 elif line.strip() == 'codec_type=audio' and audio_codec is not None:
2880 def run_ffmpeg(path, out_path, codec, more_opts):
2882 cmd = ['ffmpeg', '-y', '-i', path, '-vn', '-acodec', codec] + more_opts + ['--', out_path]
2883 ret = subprocess.call(cmd, stdout=file(os.path.devnull, 'w'), stderr=subprocess.STDOUT)
2885 except (IOError, OSError):
2888 def run(self, information):
2889 path = information['filepath']
2891 filecodec = self.get_audio_codec(path)
2892 if filecodec is None:
2893 self._downloader.to_stderr(u'WARNING: unable to obtain file audio codec with ffprobe')
2897 if self._preferredcodec == 'best' or self._preferredcodec == filecodec:
2898 if filecodec == 'aac' or filecodec == 'mp3':
2899 # Lossless if possible
2901 extension = filecodec
2902 if filecodec == 'aac':
2903 more_opts = ['-f', 'adts']
2906 acodec = 'libmp3lame'
2908 more_opts = ['-ab', '128k']
2910 # We convert the audio (lossy)
2911 acodec = {'mp3': 'libmp3lame', 'aac': 'aac'}[self._preferredcodec]
2912 extension = self._preferredcodec
2913 more_opts = ['-ab', '128k']
2914 if self._preferredcodec == 'aac':
2915 more_opts += ['-f', 'adts']
2917 (prefix, ext) = os.path.splitext(path)
2918 new_path = prefix + '.' + extension
2919 self._downloader.to_screen(u'[ffmpeg] Destination: %s' % new_path)
2920 status = self.run_ffmpeg(path, new_path, acodec, more_opts)
2923 self._downloader.to_stderr(u'WARNING: error running ffmpeg')
2928 except (IOError, OSError):
2929 self._downloader.to_stderr(u'WARNING: Unable to remove downloaded video file')
2932 information['filepath'] = new_path
2935 ### MAIN PROGRAM ###
2936 if __name__ == '__main__':
2938 # Modules needed only when running the main program
2942 # Function to update the program file with the latest version from the repository.
2943 def update_self(downloader, filename):
2944 # Note: downloader only used for options
2945 if not os.access(filename, os.W_OK):
2946 sys.exit('ERROR: no write permissions on %s' % filename)
2948 downloader.to_screen('Updating to latest stable version...')
2950 latest_url = 'http://github.com/rg3/youtube-dl/raw/master/LATEST_VERSION'
2951 latest_version = urllib.urlopen(latest_url).read().strip()
2952 prog_url = 'http://github.com/rg3/youtube-dl/raw/%s/youtube-dl' % latest_version
2953 newcontent = urllib.urlopen(prog_url).read()
2954 except (IOError, OSError), err:
2955 sys.exit('ERROR: unable to download latest version')
2957 stream = open(filename, 'w')
2958 stream.write(newcontent)
2960 except (IOError, OSError), err:
2961 sys.exit('ERROR: unable to overwrite current version')
2962 downloader.to_screen('Updated to version %s' % latest_version)
2964 # Parse command line
2965 parser = optparse.OptionParser(
2966 usage='Usage: %prog [options] url...',
2967 version='2011.07.09-phihag',
2968 conflict_handler='resolve',
2971 parser.add_option('-h', '--help',
2972 action='help', help='print this help text and exit')
2973 parser.add_option('-v', '--version',
2974 action='version', help='print program version and exit')
2975 parser.add_option('-U', '--update',
2976 action='store_true', dest='update_self', help='update this program to latest stable version')
2977 parser.add_option('-i', '--ignore-errors',
2978 action='store_true', dest='ignoreerrors', help='continue on download errors', default=False)
2979 parser.add_option('-r', '--rate-limit',
2980 dest='ratelimit', metavar='LIMIT', help='download rate limit (e.g. 50k or 44.6m)')
2981 parser.add_option('-R', '--retries',
2982 dest='retries', metavar='RETRIES', help='number of retries (default is 10)', default=10)
2983 parser.add_option('--playlist-start',
2984 dest='playliststart', metavar='NUMBER', help='playlist video to start at (default is 1)', default=1)
2985 parser.add_option('--playlist-end',
2986 dest='playlistend', metavar='NUMBER', help='playlist video to end at (default is last)', default=-1)
2987 parser.add_option('--dump-user-agent',
2988 action='store_true', dest='dump_user_agent',
2989 help='display the current browser identification', default=False)
2991 authentication = optparse.OptionGroup(parser, 'Authentication Options')
2992 authentication.add_option('-u', '--username',
2993 dest='username', metavar='USERNAME', help='account username')
2994 authentication.add_option('-p', '--password',
2995 dest='password', metavar='PASSWORD', help='account password')
2996 authentication.add_option('-n', '--netrc',
2997 action='store_true', dest='usenetrc', help='use .netrc authentication data', default=False)
2998 parser.add_option_group(authentication)
3000 video_format = optparse.OptionGroup(parser, 'Video Format Options')
3001 video_format.add_option('-f', '--format',
3002 action='store', dest='format', metavar='FORMAT', help='video format code')
3003 video_format.add_option('--all-formats',
3004 action='store_const', dest='format', help='download all available video formats', const='-1')
3005 video_format.add_option('--max-quality',
3006 action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download')
3007 parser.add_option_group(video_format)
3009 verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options')
3010 verbosity.add_option('-q', '--quiet',
3011 action='store_true', dest='quiet', help='activates quiet mode', default=False)
3012 verbosity.add_option('-s', '--simulate',
3013 action='store_true', dest='simulate', help='do not download video', default=False)
3014 verbosity.add_option('-g', '--get-url',
3015 action='store_true', dest='geturl', help='simulate, quiet but print URL', default=False)
3016 verbosity.add_option('-e', '--get-title',
3017 action='store_true', dest='gettitle', help='simulate, quiet but print title', default=False)
3018 verbosity.add_option('--get-thumbnail',
3019 action='store_true', dest='getthumbnail',
3020 help='simulate, quiet but print thumbnail URL', default=False)
3021 verbosity.add_option('--get-description',
3022 action='store_true', dest='getdescription',
3023 help='simulate, quiet but print video description', default=False)
3024 verbosity.add_option('--get-filename',
3025 action='store_true', dest='getfilename',
3026 help='simulate, quiet but print output filename', default=False)
3027 verbosity.add_option('--no-progress',
3028 action='store_true', dest='noprogress', help='do not print progress bar', default=False)
3029 verbosity.add_option('--console-title',
3030 action='store_true', dest='consoletitle',
3031 help='display progress in console titlebar', default=False)
3032 parser.add_option_group(verbosity)
3034 filesystem = optparse.OptionGroup(parser, 'Filesystem Options')
3035 filesystem.add_option('-t', '--title',
3036 action='store_true', dest='usetitle', help='use title in file name', default=False)
3037 filesystem.add_option('-l', '--literal',
3038 action='store_true', dest='useliteral', help='use literal title in file name', default=False)
3039 filesystem.add_option('-A', '--auto-number',
3040 action='store_true', dest='autonumber',
3041 help='number downloaded files starting from 00000', default=False)
3042 filesystem.add_option('-o', '--output',
3043 dest='outtmpl', metavar='TEMPLATE', help='output filename template')
3044 filesystem.add_option('-a', '--batch-file',
3045 dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)')
3046 filesystem.add_option('-w', '--no-overwrites',
3047 action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
3048 filesystem.add_option('-c', '--continue',
3049 action='store_true', dest='continue_dl', help='resume partially downloaded files', default=False)
3050 filesystem.add_option('--cookies',
3051 dest='cookiefile', metavar='FILE', help='file to dump cookie jar to')
3052 filesystem.add_option('--no-part',
3053 action='store_true', dest='nopart', help='do not use .part files', default=False)
3054 filesystem.add_option('--no-mtime',
3055 action='store_false', dest='updatetime',
3056 help='do not use the Last-modified header to set the file modification time', default=True)
3057 filesystem.add_option('--write-description',
3058 action='store_true', dest='writedescription',
3059 help='write video description to a .description file', default=False)
3060 filesystem.add_option('--write-info-json',
3061 action='store_true', dest='writeinfojson',
3062 help='write video metadata to a .info.json file', default=False)
3063 parser.add_option_group(filesystem)
3065 postproc = optparse.OptionGroup(parser, 'Post-processing Options')
3066 postproc.add_option('--extract-audio', action='store_true', dest='extractaudio', default=False,
3067 help='convert video files to audio-only files (requires ffmpeg and ffprobe)')
3068 postproc.add_option('--audio-format', metavar='FORMAT', dest='audioformat', default='best',
3069 help='"best", "aac" or "mp3"; best by default')
3070 parser.add_option_group(postproc)
3072 (opts, args) = parser.parse_args()
3074 # Open appropriate CookieJar
3075 if opts.cookiefile is None:
3076 jar = cookielib.CookieJar()
3079 jar = cookielib.MozillaCookieJar(opts.cookiefile)
3080 if os.path.isfile(opts.cookiefile) and os.access(opts.cookiefile, os.R_OK):
3082 except (IOError, OSError), err:
3083 sys.exit(u'ERROR: unable to open cookie file')
3086 if opts.dump_user_agent:
3087 print std_headers['User-Agent']
3090 # General configuration
3091 cookie_processor = urllib2.HTTPCookieProcessor(jar)
3092 urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler(), cookie_processor, YoutubeDLHandler()))
3093 socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words)
3095 # Batch file verification
3097 if opts.batchfile is not None:
3099 if opts.batchfile == '-':
3102 batchfd = open(opts.batchfile, 'r')
3103 batchurls = batchfd.readlines()
3104 batchurls = [x.strip() for x in batchurls]
3105 batchurls = [x for x in batchurls if len(x) > 0 and not re.search(r'^[#/;]', x)]
3107 sys.exit(u'ERROR: batch file could not be read')
3108 all_urls = batchurls + args
3110 # Conflicting, missing and erroneous options
3111 if opts.usenetrc and (opts.username is not None or opts.password is not None):
3112 parser.error(u'using .netrc conflicts with giving username/password')
3113 if opts.password is not None and opts.username is None:
3114 parser.error(u'account username missing')
3115 if opts.outtmpl is not None and (opts.useliteral or opts.usetitle or opts.autonumber):
3116 parser.error(u'using output template conflicts with using title, literal title or auto number')
3117 if opts.usetitle and opts.useliteral:
3118 parser.error(u'using title conflicts with using literal title')
3119 if opts.username is not None and opts.password is None:
3120 opts.password = getpass.getpass(u'Type account password and press return:')
3121 if opts.ratelimit is not None:
3122 numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
3123 if numeric_limit is None:
3124 parser.error(u'invalid rate limit specified')
3125 opts.ratelimit = numeric_limit
3126 if opts.retries is not None:
3128 opts.retries = long(opts.retries)
3129 except (TypeError, ValueError), err:
3130 parser.error(u'invalid retry count specified')
3132 opts.playliststart = long(opts.playliststart)
3133 if opts.playliststart <= 0:
3135 except (TypeError, ValueError), err:
3136 parser.error(u'invalid playlist start number specified')
3138 opts.playlistend = long(opts.playlistend)
3139 if opts.playlistend != -1 and (opts.playlistend <= 0 or opts.playlistend < opts.playliststart):
3141 except (TypeError, ValueError), err:
3142 parser.error(u'invalid playlist end number specified')
3143 if opts.extractaudio:
3144 if opts.audioformat not in ['best', 'aac', 'mp3']:
3145 parser.error(u'invalid audio format specified')
3147 # Information extractors
3148 youtube_ie = YoutubeIE()
3149 metacafe_ie = MetacafeIE(youtube_ie)
3150 dailymotion_ie = DailymotionIE()
3151 youtube_pl_ie = YoutubePlaylistIE(youtube_ie)
3152 youtube_user_ie = YoutubeUserIE(youtube_ie)
3153 youtube_search_ie = YoutubeSearchIE(youtube_ie)
3154 google_ie = GoogleIE()
3155 google_search_ie = GoogleSearchIE(google_ie)
3156 photobucket_ie = PhotobucketIE()
3157 yahoo_ie = YahooIE()
3158 yahoo_search_ie = YahooSearchIE(yahoo_ie)
3159 deposit_files_ie = DepositFilesIE()
3160 facebook_ie = FacebookIE()
3161 bliptv_ie = BlipTVIE()
3162 generic_ie = GenericIE()
3165 fd = FileDownloader({
3166 'usenetrc': opts.usenetrc,
3167 'username': opts.username,
3168 'password': opts.password,
3169 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription or opts.getfilename),
3170 'forceurl': opts.geturl,
3171 'forcetitle': opts.gettitle,
3172 'forcethumbnail': opts.getthumbnail,
3173 'forcedescription': opts.getdescription,
3174 'forcefilename': opts.getfilename,
3175 'simulate': (opts.simulate or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription or opts.getfilename),
3176 'format': opts.format,
3177 'format_limit': opts.format_limit,
3178 'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding()))
3179 or (opts.format == '-1' and opts.usetitle and u'%(stitle)s-%(id)s-%(format)s.%(ext)s')
3180 or (opts.format == '-1' and opts.useliteral and u'%(title)s-%(id)s-%(format)s.%(ext)s')
3181 or (opts.format == '-1' and u'%(id)s-%(format)s.%(ext)s')
3182 or (opts.usetitle and opts.autonumber and u'%(autonumber)s-%(stitle)s-%(id)s.%(ext)s')
3183 or (opts.useliteral and opts.autonumber and u'%(autonumber)s-%(title)s-%(id)s.%(ext)s')
3184 or (opts.usetitle and u'%(stitle)s-%(id)s.%(ext)s')
3185 or (opts.useliteral and u'%(title)s-%(id)s.%(ext)s')
3186 or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s')
3187 or u'%(id)s.%(ext)s'),
3188 'ignoreerrors': opts.ignoreerrors,
3189 'ratelimit': opts.ratelimit,
3190 'nooverwrites': opts.nooverwrites,
3191 'retries': opts.retries,
3192 'continuedl': opts.continue_dl,
3193 'noprogress': opts.noprogress,
3194 'playliststart': opts.playliststart,
3195 'playlistend': opts.playlistend,
3196 'logtostderr': opts.outtmpl == '-',
3197 'consoletitle': opts.consoletitle,
3198 'nopart': opts.nopart,
3199 'updatetime': opts.updatetime,
3200 'writedescription': opts.writedescription,
3201 'writeinfojson': opts.writeinfojson,
3203 fd.add_info_extractor(youtube_search_ie)
3204 fd.add_info_extractor(youtube_pl_ie)
3205 fd.add_info_extractor(youtube_user_ie)
3206 fd.add_info_extractor(metacafe_ie)
3207 fd.add_info_extractor(dailymotion_ie)
3208 fd.add_info_extractor(youtube_ie)
3209 fd.add_info_extractor(google_ie)
3210 fd.add_info_extractor(google_search_ie)
3211 fd.add_info_extractor(photobucket_ie)
3212 fd.add_info_extractor(yahoo_ie)
3213 fd.add_info_extractor(yahoo_search_ie)
3214 fd.add_info_extractor(deposit_files_ie)
3215 fd.add_info_extractor(facebook_ie)
3216 fd.add_info_extractor(bliptv_ie)
3218 # This must come last since it's the
3219 # fallback if none of the others work
3220 fd.add_info_extractor(generic_ie)
3223 if opts.extractaudio:
3224 fd.add_post_processor(FFmpegExtractAudioPP(preferredcodec=opts.audioformat))
3227 if opts.update_self:
3228 update_self(fd, sys.argv[0])
3231 if len(all_urls) < 1:
3232 if not opts.update_self:
3233 parser.error(u'you must provide at least one URL')
3236 retcode = fd.download(all_urls)
3238 # Dump cookie jar if requested
3239 if opts.cookiefile is not None:
3242 except (IOError, OSError), err:
3243 sys.exit(u'ERROR: unable to save cookie jar')
3247 except DownloadError:
3249 except SameFileError:
3250 sys.exit(u'ERROR: fixed output name but more than one file to download')
3251 except KeyboardInterrupt:
3252 sys.exit(u'\nERROR: Interrupted by user')