X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/c705a91c64059c7d2724e0c3ca2359afee2f592a..fc901acadf85100810b1201a612a570e48e891fd:/youtube-dl?ds=sidebyside diff --git a/youtube-dl b/youtube-dl index a4c8f24..818a3a7 100755 --- a/youtube-dl +++ b/youtube-dl @@ -1,18 +1,32 @@ #!/usr/bin/env python # -*- coding: utf-8 -*- -# Author: Ricardo Garcia Gonzalez -# Author: Danny Colligan -# Author: Benjamin Johnson -# Author: Vasyl' Vavrychuk -# Author: Witold Baryluk -# Author: PaweŠPaprota -# License: Public domain code + +__authors__ = ( + 'Ricardo Garcia Gonzalez', + 'Danny Colligan', + 'Benjamin Johnson', + 'Vasyl\' Vavrychuk', + 'Witold Baryluk', + 'PaweŠPaprota', + 'Gergely Imreh', + 'Rogério Brito', + 'Philipp Hagemeister', + 'Sören Schulze', + 'Kevin Ngo', + 'Ori Avtalion', + 'shizeeg', + ) + +__license__ = 'Public Domain' +__version__ = '2012.01.05' + +UPDATE_URL = 'https://raw.github.com/rg3/youtube-dl/master/youtube-dl' + import cookielib -import ctypes import datetime -import email.utils import gzip import htmlentitydefs +import HTMLParser import httplib import locale import math @@ -22,29 +36,162 @@ import os.path import re import socket import string -import StringIO import subprocess import sys import time import urllib import urllib2 +import warnings import zlib +if os.name == 'nt': + import ctypes + +try: + import email.utils +except ImportError: # Python 2.4 + import email.Utils +try: + import cStringIO as StringIO +except ImportError: + import StringIO + # parse_qs was moved from the cgi module to the urlparse module recently. try: from urlparse import parse_qs except ImportError: from cgi import parse_qs +try: + import lxml.etree +except ImportError: + pass # Handled below + +try: + import xml.etree.ElementTree +except ImportError: # Python<2.5: Not officially supported, but let it slip + warnings.warn('xml.etree.ElementTree support is missing. Consider upgrading to Python >= 2.5 if you get related errors.') + std_headers = { - 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:2.0b10) Gecko/20100101 Firefox/4.0b10', + 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:5.0.1) Gecko/20100101 Firefox/5.0.1', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', 'Accept-Encoding': 'gzip, deflate', 'Accept-Language': 'en-us,en;q=0.5', } -simple_title_chars = string.ascii_letters.decode('ascii') + string.digits.decode('ascii') +try: + import json +except ImportError: # Python <2.6, use trivialjson (https://github.com/phihag/trivialjson): + import re + class json(object): + @staticmethod + def loads(s): + s = s.decode('UTF-8') + def raiseError(msg, i): + raise ValueError(msg + ' at position ' + str(i) + ' of ' + repr(s) + ': ' + repr(s[i:])) + def skipSpace(i, expectMore=True): + while i < len(s) and s[i] in ' \t\r\n': + i += 1 + if expectMore: + if i >= len(s): + raiseError('Premature end', i) + return i + def decodeEscape(match): + esc = match.group(1) + _STATIC = { + '"': '"', + '\\': '\\', + '/': '/', + 'b': unichr(0x8), + 'f': unichr(0xc), + 'n': '\n', + 'r': '\r', + 't': '\t', + } + if esc in _STATIC: + return _STATIC[esc] + if esc[0] == 'u': + if len(esc) == 1+4: + return unichr(int(esc[1:5], 16)) + if len(esc) == 5+6 and esc[5:7] == '\\u': + hi = int(esc[1:5], 16) + low = int(esc[7:11], 16) + return unichr((hi - 0xd800) * 0x400 + low - 0xdc00 + 0x10000) + raise ValueError('Unknown escape ' + str(esc)) + def parseString(i): + i += 1 + e = i + while True: + e = s.index('"', e) + bslashes = 0 + while s[e-bslashes-1] == '\\': + bslashes += 1 + if bslashes % 2 == 1: + e += 1 + continue + break + rexp = re.compile(r'\\(u[dD][89aAbB][0-9a-fA-F]{2}\\u[0-9a-fA-F]{4}|u[0-9a-fA-F]{4}|.|$)') + stri = rexp.sub(decodeEscape, s[i:e]) + return (e+1,stri) + def parseObj(i): + i += 1 + res = {} + i = skipSpace(i) + if s[i] == '}': # Empty dictionary + return (i+1,res) + while True: + if s[i] != '"': + raiseError('Expected a string object key', i) + i,key = parseString(i) + i = skipSpace(i) + if i >= len(s) or s[i] != ':': + raiseError('Expected a colon', i) + i,val = parse(i+1) + res[key] = val + i = skipSpace(i) + if s[i] == '}': + return (i+1, res) + if s[i] != ',': + raiseError('Expected comma or closing curly brace', i) + i = skipSpace(i+1) + def parseArray(i): + res = [] + i = skipSpace(i+1) + if s[i] == ']': # Empty array + return (i+1,res) + while True: + i,val = parse(i) + res.append(val) + i = skipSpace(i) # Raise exception if premature end + if s[i] == ']': + return (i+1, res) + if s[i] != ',': + raiseError('Expected a comma or closing bracket', i) + i = skipSpace(i+1) + def parseDiscrete(i): + for k,v in {'true': True, 'false': False, 'null': None}.items(): + if s.startswith(k, i): + return (i+len(k), v) + raiseError('Not a boolean (or null)', i) + def parseNumber(i): + mobj = re.match('^(-?(0|[1-9][0-9]*)(\.[0-9]*)?([eE][+-]?[0-9]+)?)', s[i:]) + if mobj is None: + raiseError('Not a number', i) + nums = mobj.group(1) + if '.' in nums or 'e' in nums or 'E' in nums: + return (i+len(nums), float(nums)) + return (i+len(nums), int(nums)) + CHARMAP = {'{': parseObj, '[': parseArray, '"': parseString, 't': parseDiscrete, 'f': parseDiscrete, 'n': parseDiscrete} + def parse(i): + i = skipSpace(i) + i,res = CHARMAP.get(s[i], parseNumber)(i) + i = skipSpace(i, False) + return (i,res) + i,res = parse(0) + if i < len(s): + raise ValueError('Extra data at end of input (index ' + str(i) + ' of ' + repr(s) + ': ' + repr(s[i:]) + ')') + return res def preferredencoding(): """Get preferred encoding. @@ -62,6 +209,7 @@ def preferredencoding(): yield pref return yield_preferredencoding().next() + def htmlentity_transform(matchobj): """Transforms an HTML entity to a Unicode character. @@ -88,11 +236,13 @@ def htmlentity_transform(matchobj): # Unknown entity in name, return its literal representation return (u'&%s;' % entity) + def sanitize_title(utitle): """Sanitizes a video title so it could be used as part of a filename.""" utitle = re.sub(ur'(?u)&(.+?);', htmlentity_transform, utitle) return utitle.replace(unicode(os.sep), u'%') + def sanitize_open(filename, open_mode): """Try to open the given filename, and slightly tweak it if this fails. @@ -119,13 +269,35 @@ def sanitize_open(filename, open_mode): stream = open(filename, open_mode) return (stream, filename) + def timeconvert(timestr): - """Convert RFC 2822 defined time string into system timestamp""" - timestamp = None - timetuple = email.utils.parsedate_tz(timestr) - if timetuple is not None: - timestamp = email.utils.mktime_tz(timetuple) - return timestamp + """Convert RFC 2822 defined time string into system timestamp""" + timestamp = None + timetuple = email.utils.parsedate_tz(timestr) + if timetuple is not None: + timestamp = email.utils.mktime_tz(timetuple) + return timestamp + +def _simplify_title(title): + expr = re.compile(ur'[^\w\d_\-]+', flags=re.UNICODE) + return expr.sub(u'_', title).strip(u'_') + +def _orderedSet(iterable): + """ Remove all duplicates from the input iterable """ + res = [] + for el in iterable: + if el not in res: + res.append(el) + return res + +def _unescapeHTML(s): + """ + @param s a string (of type unicode) + """ + assert type(s) == type(u'') + + htmlParser = HTMLParser.HTMLParser() + return htmlParser.unescape(s) class DownloadError(Exception): """Download Error exception. @@ -136,6 +308,7 @@ class DownloadError(Exception): """ pass + class SameFileError(Exception): """Same File exception. @@ -144,6 +317,7 @@ class SameFileError(Exception): """ pass + class PostProcessingError(Exception): """Post Processing exception. @@ -152,6 +326,11 @@ class PostProcessingError(Exception): """ pass +class MaxDownloadsReached(Exception): + """ --max-downloads limit has been reached. """ + pass + + class UnavailableVideoError(Exception): """Unavailable Format exception. @@ -160,6 +339,7 @@ class UnavailableVideoError(Exception): """ pass + class ContentTooShortError(Exception): """Content Too Short exception. @@ -175,6 +355,7 @@ class ContentTooShortError(Exception): self.downloaded = downloaded self.expected = expected + class YoutubeDLHandler(urllib2.HTTPHandler): """Handler for HTTP requests and responses. @@ -184,11 +365,11 @@ class YoutubeDLHandler(urllib2.HTTPHandler): a particular request, the original request in the program code only has to include the HTTP header "Youtubedl-No-Compression", which will be removed before making the real request. - + Part of this code was copied from: - http://techknack.net/python-urllib2-handlers/ - + http://techknack.net/python-urllib2-handlers/ + Andrew Rowls, the author of that code, agreed to release it to the public domain. """ @@ -199,7 +380,7 @@ class YoutubeDLHandler(urllib2.HTTPHandler): return zlib.decompress(data, -zlib.MAX_WBITS) except zlib.error: return zlib.decompress(data) - + @staticmethod def addinfourl_wrapper(stream, headers, url, code): if hasattr(urllib2.addinfourl, 'getcode'): @@ -207,7 +388,7 @@ class YoutubeDLHandler(urllib2.HTTPHandler): ret = urllib2.addinfourl(stream, headers, url) ret.code = code return ret - + def http_request(self, req): for h in std_headers: if h in req.headers: @@ -233,6 +414,7 @@ class YoutubeDLHandler(urllib2.HTTPHandler): resp.msg = old_resp.msg return resp + class FileDownloader(object): """File Downloader class. @@ -281,10 +463,14 @@ class FileDownloader(object): noprogress: Do not print the progress bar. playliststart: Playlist item to start at. playlistend: Playlist item to end at. + matchtitle: Download only matching titles. + rejecttitle: Reject downloads for matching titles. logtostderr: Log messages to stderr instead of stdout. consoletitle: Display progress in console window's titlebar. nopart: Do not use temporary .part files. updatetime: Use the Last-modified header to set output file timestamps. + writedescription: Write the video description to a .description file + writeinfojson: Write the video description to a .info.json file """ params = None @@ -303,16 +489,6 @@ class FileDownloader(object): self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] self.params = params - @staticmethod - def pmkdir(filename): - """Create directory components in filename. Similar to Unix "mkdir -p".""" - components = filename.split(os.sep) - aggregate = [os.sep.join(components[0:x]) for x in xrange(1, len(components))] - aggregate = ['%s%s' % (x, os.sep) for x in aggregate] # Finish names with separator - for dir in aggregate: - if not os.path.exists(dir): - os.mkdir(dir) - @staticmethod def format_bytes(bytes): if bytes is None: @@ -324,7 +500,7 @@ class FileDownloader(object): else: exponent = long(math.log(bytes, 1024.0)) suffix = 'bkMGTPEZY'[exponent] - converted = float(bytes) / float(1024**exponent) + converted = float(bytes) / float(1024 ** exponent) return '%.2f%s' % (converted, suffix) @staticmethod @@ -462,7 +638,7 @@ class FileDownloader(object): os.rename(old_filename, new_filename) except (IOError, OSError), err: self.trouble(u'ERROR: unable to rename file') - + def try_utime(self, filename, last_modified_hdr): """Try to set the last-modified time of the given file.""" if last_modified_hdr is None: @@ -474,11 +650,20 @@ class FileDownloader(object): return filetime = timeconvert(timestr) if filetime is None: - return + return filetime try: - os.utime(filename,(time.time(), filetime)) + os.utime(filename, (time.time(), filetime)) except: pass + return filetime + + def report_writedescription(self, descfn): + """ Report that the description file is being written """ + self.to_screen(u'[info] Writing video description to: %s' % descfn, ignore_encoding_errors=True) + + def report_writeinfojson(self, infofn): + """ Report that the metadata file has been written """ + self.to_screen(u'[info] Video description metadata as JSON to: %s' % infofn, ignore_encoding_errors=True) def report_destination(self, filename): """Report destination filename.""" @@ -535,55 +720,116 @@ class FileDownloader(object): self.trouble(u'ERROR: invalid system charset or erroneous output template') return None + def _match_entry(self, info_dict): + """ Returns None iff the file should be downloaded """ + + title = info_dict['title'] + matchtitle = self.params.get('matchtitle', False) + if matchtitle and not re.search(matchtitle, title, re.IGNORECASE): + return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"' + rejecttitle = self.params.get('rejecttitle', False) + if rejecttitle and re.search(rejecttitle, title, re.IGNORECASE): + return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' + return None + def process_info(self, info_dict): """Process a single dictionary returned by an InfoExtractor.""" + + reason = self._match_entry(info_dict) + if reason is not None: + self.to_screen(u'[download] ' + reason) + return + + max_downloads = self.params.get('max_downloads') + if max_downloads is not None: + if self._num_downloads > int(max_downloads): + raise MaxDownloadsReached() + filename = self.prepare_filename(info_dict) + + # Forced printings + if self.params.get('forcetitle', False): + print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forceurl', False): + print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict: + print info_dict['thumbnail'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forcedescription', False) and 'description' in info_dict: + print info_dict['description'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forcefilename', False) and filename is not None: + print filename.encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forceformat', False): + print info_dict['format'].encode(preferredencoding(), 'xmlcharrefreplace') + # Do nothing else if in simulate mode if self.params.get('simulate', False): - # Forced printings - if self.params.get('forcetitle', False): - print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace') - if self.params.get('forceurl', False): - print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace') - if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict: - print info_dict['thumbnail'].encode(preferredencoding(), 'xmlcharrefreplace') - if self.params.get('forcedescription', False) and 'description' in info_dict: - print info_dict['description'].encode(preferredencoding(), 'xmlcharrefreplace') - if self.params.get('forcefilename', False) and filename is not None: - print filename.encode(preferredencoding(), 'xmlcharrefreplace') - return if filename is None: return - if self.params.get('nooverwrites', False) and os.path.exists(filename): - self.to_stderr(u'WARNING: file exists and will be skipped') - return try: - self.pmkdir(filename) + dn = os.path.dirname(filename) + if dn != '' and not os.path.exists(dn): + os.makedirs(dn) except (OSError, IOError), err: - self.trouble(u'ERROR: unable to create directories: %s' % str(err)) + self.trouble(u'ERROR: unable to create directory ' + unicode(err)) return - try: - success = self._do_download(filename, info_dict['url'].encode('utf-8'), info_dict.get('player_url', None)) - except (OSError, IOError), err: - raise UnavailableVideoError - except (urllib2.URLError, httplib.HTTPException, socket.error), err: - self.trouble(u'ERROR: unable to download video data: %s' % str(err)) - return - except (ContentTooShortError, ), err: - self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) - return + if self.params.get('writedescription', False): + try: + descfn = filename + '.description' + self.report_writedescription(descfn) + descfile = open(descfn, 'wb') + try: + descfile.write(info_dict['description'].encode('utf-8')) + finally: + descfile.close() + except (OSError, IOError): + self.trouble(u'ERROR: Cannot write description file ' + descfn) + return - if success: + if self.params.get('writeinfojson', False): + infofn = filename + '.info.json' + self.report_writeinfojson(infofn) + try: + json.dump + except (NameError,AttributeError): + self.trouble(u'ERROR: No JSON encoder found. Update to Python 2.6+, setup a json module, or leave out --write-info-json.') + return try: - self.post_process(filename, info_dict) - except (PostProcessingError), err: - self.trouble(u'ERROR: postprocessing: %s' % str(err)) + infof = open(infofn, 'wb') + try: + json_info_dict = dict((k,v) for k,v in info_dict.iteritems() if not k in ('urlhandle',)) + json.dump(json_info_dict, infof) + finally: + infof.close() + except (OSError, IOError): + self.trouble(u'ERROR: Cannot write metadata to JSON file ' + infofn) return + if not self.params.get('skip_download', False): + if self.params.get('nooverwrites', False) and os.path.exists(filename): + success = True + else: + try: + success = self._do_download(filename, info_dict) + except (OSError, IOError), err: + raise UnavailableVideoError + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self.trouble(u'ERROR: unable to download video data: %s' % str(err)) + return + except (ContentTooShortError, ), err: + self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) + return + + if success: + try: + self.post_process(filename, info_dict) + except (PostProcessingError), err: + self.trouble(u'ERROR: postprocessing: %s' % str(err)) + return + def download(self, url_list): """Download a given list of URLs.""" if len(url_list) > 1 and self.fixed_template(): @@ -643,6 +889,11 @@ class FileDownloader(object): cursize = os.path.getsize(tmpfilename) if prevsize == cursize and retval == 1: break + # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those + if prevsize == cursize and retval == 2 and cursize > 1024: + self.to_screen(u'\r[rtmpdump] Could not download the whole video. This can happen for some advertisements.') + retval = 0 + break if retval == 0: self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(tmpfilename)) self.try_rename(tmpfilename, filename) @@ -651,7 +902,10 @@ class FileDownloader(object): self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval) return False - def _do_download(self, filename, url, player_url): + def _do_download(self, filename, info_dict): + url = info_dict['url'] + player_url = info_dict.get('player_url', None) + # Check file already present if self.params.get('continuedl', False) and os.path.isfile(filename) and not self.params.get('nopart', False): self.report_file_already_downloaded(filename) @@ -663,7 +917,6 @@ class FileDownloader(object): tmpfilename = self.temp_name(filename) stream = None - open_mode = 'wb' # Do not include the Accept-Encoding header headers = {'Youtubedl-no-compression': 'True'} @@ -676,17 +929,22 @@ class FileDownloader(object): else: resume_len = 0 - # Request parameters in case of being able to resume - if self.params.get('continuedl', False) and resume_len != 0: - self.report_resuming_byte(resume_len) - request.add_header('Range','bytes=%d-' % resume_len) - open_mode = 'ab' + open_mode = 'wb' + if resume_len != 0: + if self.params.get('continuedl', False): + self.report_resuming_byte(resume_len) + request.add_header('Range','bytes=%d-' % resume_len) + open_mode = 'ab' + else: + resume_len = 0 count = 0 retries = self.params.get('retries', 0) while count <= retries: # Establish connection try: + if count == 0 and 'urlhandle' in info_dict: + data = info_dict['urlhandle'] data = urllib2.urlopen(request) break except (urllib2.HTTPError, ), err: @@ -705,7 +963,7 @@ class FileDownloader(object): else: # Examine the reported length if (content_length is not None and - (resume_len - 100 < long(content_length) < resume_len + 100)): + (resume_len - 100 < long(content_length) < resume_len + 100)): # The file had already been fully downloaded. # Explanation to the above condition: in issue #175 it was revealed that # YouTube sometimes adds or removes a few bytes from the end of the file, @@ -750,6 +1008,7 @@ class FileDownloader(object): if stream is None: try: (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode) + assert stream is not None filename = self.undo_temp_name(tmpfilename) self.report_destination(filename) except (OSError, IOError), err: @@ -763,14 +1022,20 @@ class FileDownloader(object): block_size = self.best_block_size(after - before, len(data_block)) # Progress message - percent_str = self.calc_percent(byte_counter, data_len) - eta_str = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len) speed_str = self.calc_speed(start, time.time(), byte_counter - resume_len) - self.report_progress(percent_str, data_len_str, speed_str, eta_str) + if data_len is None: + self.report_progress('Unknown %', data_len_str, speed_str, 'Unknown ETA') + else: + percent_str = self.calc_percent(byte_counter, data_len) + eta_str = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len) + self.report_progress(percent_str, data_len_str, speed_str, eta_str) # Apply rate limit self.slow_down(start, byte_counter - resume_len) + if stream is None: + self.trouble(u'\nERROR: Did not get any data blocks') + return False stream.close() self.report_finish() if data_len is not None and byte_counter != data_len: @@ -779,10 +1044,11 @@ class FileDownloader(object): # Update file modification time if self.params.get('updatetime', True): - self.try_utime(filename, data.info().get('last-modified', None)) + info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None)) return True + class InfoExtractor(object): """Information Extractor class. @@ -813,9 +1079,8 @@ class InfoExtractor(object): description: One-line video description. Subclasses of this one should re-define the _real_initialize() and - _real_extract() methods, as well as the suitable() static method. - Probably, they should also be instantiated and added to the main - downloader. + _real_extract() methods and define a _VALID_URL regexp. + Probably, they should also be added to the list of extractors. """ _ready = False @@ -826,10 +1091,9 @@ class InfoExtractor(object): self._ready = False self.set_downloader(downloader) - @staticmethod - def suitable(url): + def suitable(self, url): """Receives a URL and returns True if suitable for this IE.""" - return False + return re.match(self._VALID_URL, url) is not None def initialize(self): """Initializes an instance (authentication, etc).""" @@ -854,16 +1118,18 @@ class InfoExtractor(object): """Real extraction process. Redefine in subclasses.""" pass + class YoutubeIE(InfoExtractor): """Information extractor for youtube.com.""" - _VALID_URL = r'^((?:https?://)?(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/)(?:(?:(?:v|embed)/)|(?:(?:watch(?:_popup)?(?:\.php)?)?(?:\?|#!?)(?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' + _VALID_URL = r'^((?:https?://)?(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/)(?!view_play_list|my_playlists|artist|playlist)(?:(?:(?:v|embed|e)/)|(?:(?:watch(?:_popup)?(?:\.php)?)?(?:\?|#!?)(?:.+&)?v=))?)?([0-9A-Za-z_-]+)(?(1).+)?$' _LANG_URL = r'http://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' _LOGIN_URL = 'https://www.youtube.com/signup?next=/&gl=US&hl=en' _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NETRC_MACHINE = 'youtube' # Listed in order of quality - _available_formats = ['38', '37', '22', '45', '35', '34', '43', '18', '6', '5', '17', '13'] + _available_formats = ['38', '37', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] + _available_formats_prefer_free = ['38', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] _video_extensions = { '13': '3gp', '17': 'mp4', @@ -872,12 +1138,25 @@ class YoutubeIE(InfoExtractor): '37': 'mp4', '38': 'video', # You actually don't know if this will be MOV, AVI or whatever '43': 'webm', + '44': 'webm', '45': 'webm', } - - @staticmethod - def suitable(url): - return (re.match(YoutubeIE._VALID_URL, url) is not None) + _video_dimensions = { + '5': '240x400', + '6': '???', + '13': '???', + '17': '144x176', + '18': '360x640', + '22': '720x1280', + '34': '360x640', + '35': '480x854', + '37': '1080x1920', + '38': '3072x4096', + '43': '360x640', + '44': '480x854', + '45': '720x1280', + } + IE_NAME = u'youtube' def report_lang(self): """Report attempt to set language.""" @@ -911,6 +1190,11 @@ class YoutubeIE(InfoExtractor): """Indicate the download will use the RTMP protocol.""" self._downloader.to_screen(u'[youtube] RTMP download detected') + def _print_formats(self, formats): + print 'Available formats:' + for x in formats: + print '%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')) + def _real_initialize(self): if self._downloader is None: return @@ -990,7 +1274,7 @@ class YoutubeIE(InfoExtractor): # Get video webpage self.report_video_webpage_download(video_id) - request = urllib2.Request('http://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id) + request = urllib2.Request('http://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id) try: video_webpage = urllib2.urlopen(request).read() except (urllib2.URLError, httplib.HTTPException, socket.error), err: @@ -1008,7 +1292,7 @@ class YoutubeIE(InfoExtractor): self.report_video_info_webpage_download(video_id) for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: video_info_url = ('http://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' - % (video_id, el_type)) + % (video_id, el_type)) request = urllib2.Request(video_info_url) try: video_info_webpage = urllib2.urlopen(request).read() @@ -1043,8 +1327,7 @@ class YoutubeIE(InfoExtractor): video_title = sanitize_title(video_title) # simplified title - simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) - simple_title = simple_title.strip(ur'_') + simple_title = _simplify_title(video_title) # thumbnail image if 'thumbnail_url' not in video_info: @@ -1055,10 +1338,10 @@ class YoutubeIE(InfoExtractor): # upload date upload_date = u'NA' - mobj = re.search(r'id="eow-date".*?>(.*?)', video_webpage, re.DOTALL) + mobj = re.search(r'id="eow-date.*?>(.*?)', video_webpage, re.DOTALL) if mobj is not None: upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) - format_expressions = ['%d %B %Y', '%B %d %Y'] + format_expressions = ['%d %B %Y', '%B %d %Y', '%b %d %Y'] for expression in format_expressions: try: upload_date = datetime.datetime.strptime(upload_date, expression).strftime('%Y%m%d') @@ -1066,11 +1349,19 @@ class YoutubeIE(InfoExtractor): pass # description - video_description = 'No description available.' - if self._downloader.params.get('forcedescription', False): - mobj = re.search(r'', video_webpage) - if mobj is not None: - video_description = mobj.group(1) + try: + lxml.etree + except NameError: + video_description = u'No description available.' + if self._downloader.params.get('forcedescription', False) or self._downloader.params.get('writedescription', False): + mobj = re.search(r'', video_webpage) + if mobj is not None: + video_description = mobj.group(1).decode('utf-8') + else: + html_parser = lxml.etree.HTMLParser(encoding='utf-8') + vwebpage_doc = lxml.etree.parse(StringIO.StringIO(video_webpage), html_parser) + video_description = u''.join(vwebpage_doc.xpath('id("eow-description")//text()')) + # TODO use another parser # token video_token = urllib.unquote_plus(video_info['token'][0]) @@ -1078,34 +1369,48 @@ class YoutubeIE(InfoExtractor): # Decide which formats to download req_format = self._downloader.params.get('format', None) - if 'fmt_url_map' in video_info: - url_map = dict(tuple(pair.split('|')) for pair in video_info['fmt_url_map'][0].split(',')) + if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): + self.report_rtmp_download() + video_url_list = [(None, video_info['conn'][0])] + elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: + url_data_strs = video_info['url_encoded_fmt_stream_map'][0].split(',') + url_data = [parse_qs(uds) for uds in url_data_strs] + url_data = filter(lambda ud: 'itag' in ud and 'url' in ud, url_data) + url_map = dict((ud['itag'][0], ud['url'][0]) for ud in url_data) + format_limit = self._downloader.params.get('format_limit', None) - if format_limit is not None and format_limit in self._available_formats: - format_list = self._available_formats[self._available_formats.index(format_limit):] + available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats + if format_limit is not None and format_limit in available_formats: + format_list = available_formats[available_formats.index(format_limit):] else: - format_list = self._available_formats + format_list = available_formats existing_formats = [x for x in format_list if x in url_map] if len(existing_formats) == 0: self._downloader.trouble(u'ERROR: no known formats available for video') return - if req_format is None: + if self._downloader.params.get('listformats', None): + self._print_formats(existing_formats) + return + if req_format is None or req_format == 'best': video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality - elif req_format == '-1': + elif req_format == 'worst': + video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality + elif req_format in ('-1', 'all'): video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats else: - # Specific format - if req_format not in url_map: + # Specific formats. We pick the first in a slash-delimeted sequence. + # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. + req_formats = req_format.split('/') + video_url_list = None + for rf in req_formats: + if rf in url_map: + video_url_list = [(rf, url_map[rf])] + break + if video_url_list is None: self._downloader.trouble(u'ERROR: requested format not available') return - video_url_list = [(req_format, url_map[req_format])] # Specific format - - elif 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): - self.report_rtmp_download() - video_url_list = [(None, video_info['conn'][0])] - else: - self._downloader.trouble(u'ERROR: no fmt_url_map or conn information found in video info') + self._downloader.trouble(u'ERROR: no conn or url_encoded_fmt_stream_map information found in video info') return for format_param, video_real_url in video_url_list: @@ -1115,7 +1420,6 @@ class YoutubeIE(InfoExtractor): # Extension video_extension = self._video_extensions.get(format_param, 'flv') - # Find the video URL in fmt_url_map or conn paramters try: # Process video information self._downloader.process_info({ @@ -1128,7 +1432,7 @@ class YoutubeIE(InfoExtractor): 'ext': video_extension.decode('utf-8'), 'format': (format_param is None and u'NA' or format_param.decode('utf-8')), 'thumbnail': video_thumbnail.decode('utf-8'), - 'description': video_description.decode('utf-8'), + 'description': video_description, 'player_url': player_url, }) except UnavailableVideoError, err: @@ -1142,15 +1446,12 @@ class MetacafeIE(InfoExtractor): _DISCLAIMER = 'http://www.metacafe.com/family_filter/' _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user' _youtube_ie = None + IE_NAME = u'metacafe' def __init__(self, youtube_ie, downloader=None): InfoExtractor.__init__(self, downloader) self._youtube_ie = youtube_ie - @staticmethod - def suitable(url): - return (re.match(MetacafeIE._VALID_URL, url) is not None) - def report_disclaimer(self): """Report disclaimer retrieval.""" self._downloader.to_screen(u'[metacafe] Retrieving disclaimer') @@ -1284,14 +1585,11 @@ class DailymotionIE(InfoExtractor): """Information Extractor for Dailymotion""" _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^_/]+)_([^/]+)' + IE_NAME = u'dailymotion' def __init__(self, downloader=None): InfoExtractor.__init__(self, downloader) - @staticmethod - def suitable(url): - return (re.match(DailymotionIE._VALID_URL, url) is not None) - def report_download_webpage(self, video_id): """Report webpage download.""" self._downloader.to_screen(u'[dailymotion] %s: Downloading webpage' % video_id) @@ -1300,9 +1598,6 @@ class DailymotionIE(InfoExtractor): """Report information extraction.""" self._downloader.to_screen(u'[dailymotion] %s: Extracting information' % video_id) - def _real_initialize(self): - return - def _real_extract(self, url): # Extract id and simplified title from URL mobj = re.match(self._VALID_URL, url) @@ -1314,11 +1609,11 @@ class DailymotionIE(InfoExtractor): self._downloader.increment_downloads() video_id = mobj.group(1) - simple_title = mobj.group(2).decode('utf-8') video_extension = 'flv' # Retrieve video webpage to extract further information request = urllib2.Request(url) + request.add_header('Cookie', 'family_filter=off') try: self.report_download_webpage(video_id) webpage = urllib2.urlopen(request).read() @@ -1328,25 +1623,30 @@ class DailymotionIE(InfoExtractor): # Extract URL, uploader and title from webpage self.report_extraction(video_id) - mobj = re.search(r'(?i)addVariable\(\"video\"\s*,\s*\"([^\"]*)\"\)', webpage) + mobj = re.search(r'(?i)addVariable\(\"sequence\"\s*,\s*\"([^\"]+?)\"\)', webpage) if mobj is None: self._downloader.trouble(u'ERROR: unable to extract media URL') return - mediaURL = urllib.unquote(mobj.group(1)) + sequence = urllib.unquote(mobj.group(1)) + mobj = re.search(r',\"sdURL\"\:\"([^\"]+?)\",', sequence) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + mediaURL = urllib.unquote(mobj.group(1)).replace('\\', '') # if needed add http://www.dailymotion.com/ if relative URL video_url = mediaURL - # '' - mobj = re.search(r'(?im)