X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=inline;f=youtube-dl;h=fe8a591606a38c243cb1214372d90b505f41e207;hb=bd3cdf6dc42d48d010fad03d8d601b3920c7a8a7;hp=becf1d25de31ccbdbaaef76b517bb23490a9c001;hpb=2a04438c7c6de2dc9ef2a0bbaaa8ebc1dc654f44;p=youtube-dl.git diff --git a/youtube-dl b/youtube-dl index becf1d25d..fe8a59160 100755 --- a/youtube-dl +++ b/youtube-dl @@ -2,6 +2,7 @@ # -*- coding: utf-8 -*- # Author: Ricardo Garcia Gonzalez # Author: Danny Colligan +# Author: Benjamin Johnson # License: Public domain code import htmlentitydefs import httplib @@ -13,13 +14,20 @@ import os.path import re import socket import string +import subprocess import sys import time import urllib import urllib2 +# 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 + std_headers = { - 'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.1.2) Gecko/20090729 Firefox/3.5.2', + 'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Accept': 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5', 'Accept-Language': 'en-us,en;q=0.5', @@ -27,6 +35,75 @@ std_headers = { simple_title_chars = string.ascii_letters.decode('ascii') + string.digits.decode('ascii') +def preferredencoding(): + """Get preferred encoding. + + Returns the best encoding scheme for the system, based on + locale.getpreferredencoding() and some further tweaks. + """ + def yield_preferredencoding(): + try: + pref = locale.getpreferredencoding() + u'TEST'.encode(pref) + except: + pref = 'UTF-8' + while True: + yield pref + return yield_preferredencoding().next() + +def htmlentity_transform(matchobj): + """Transforms an HTML entity to a Unicode character. + + This function receives a match object and is intended to be used with + the re.sub() function. + """ + entity = matchobj.group(1) + + # Known non-numeric HTML entity + if entity in htmlentitydefs.name2codepoint: + return unichr(htmlentitydefs.name2codepoint[entity]) + + # Unicode character + mobj = re.match(ur'(?u)#(x?\d+)', entity) + if mobj is not None: + numstr = mobj.group(1) + if numstr.startswith(u'x'): + base = 16 + numstr = u'0%s' % numstr + else: + base = 10 + return unichr(long(numstr, base)) + + # 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. + + Attempts to open the given filename. If this fails, it tries to change + the filename slightly, step by step, until it's either able to open it + or it fails and raises a final exception, like the standard open() + function. + + It returns the tuple (stream, definitive_file_name). + """ + try: + stream = open(filename, open_mode) + return (stream, filename) + except (IOError, OSError), err: + # In case of error, try to remove win32 forbidden chars + filename = re.sub(ur'[<>:"\|\?\*]', u'#', filename) + + # An exception here should be caught in the caller + stream = open(filename, open_mode) + return (stream, filename) + + class DownloadError(Exception): """Download Error exception. @@ -205,11 +282,13 @@ class FileDownloader(object): @staticmethod def verify_url(url): - """Verify a URL is valid and data could be downloaded.""" + """Verify a URL is valid and data could be downloaded. Return real data URL.""" request = urllib2.Request(url, None, std_headers) data = urllib2.urlopen(request) data.read(1) + url = data.geturl() data.close() + return url def add_info_extractor(self, ie): """Add an InfoExtractor object to the end of the list.""" @@ -224,12 +303,12 @@ class FileDownloader(object): def to_stdout(self, message, skip_eol=False): """Print message to stdout if not in quiet mode.""" if not self.params.get('quiet', False): - print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(locale.getpreferredencoding()), + print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(preferredencoding()), sys.stdout.flush() def to_stderr(self, message): """Print message to stderr.""" - print >>sys.stderr, message.encode(locale.getpreferredencoding()) + print >>sys.stderr, message.encode(preferredencoding()) def fixed_template(self): """Checks if the output template is fixed.""" @@ -290,16 +369,18 @@ class FileDownloader(object): """Process a single dictionary returned by an InfoExtractor.""" # Do nothing else if in simulate mode if self.params.get('simulate', False): - try: - self.verify_url(info_dict['url']) - except (OSError, IOError, urllib2.URLError, httplib.HTTPException, socket.error), err: - raise UnavailableFormatError + # Verify URL if it's an HTTP one + if info_dict['url'].startswith('http'): + try: + info_dict['url'] = self.verify_url(info_dict['url'].encode('utf-8')).decode('utf-8') + except (OSError, IOError, urllib2.URLError, httplib.HTTPException, socket.error), err: + raise UnavailableFormatError # Forced printings if self.params.get('forcetitle', False): - print info_dict['title'].encode(locale.getpreferredencoding()) + print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace') if self.params.get('forceurl', False): - print info_dict['url'].encode(locale.getpreferredencoding()) + print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace') return @@ -309,7 +390,7 @@ class FileDownloader(object): filename = self.params['outtmpl'] % template_dict except (ValueError, KeyError), err: self.trouble('ERROR: invalid output template or system charset: %s' % str(err)) - if self.params['nooverwrites'] and os.path.exists(filename): + if self.params.get('nooverwrites', False) and os.path.exists(filename): self.to_stderr(u'WARNING: file exists: %s; skipping' % filename) return @@ -320,7 +401,7 @@ class FileDownloader(object): return try: - success = self._do_download(filename, info_dict['url']) + success = self._do_download(filename, info_dict['url'].encode('utf-8')) except (OSError, IOError), err: raise UnavailableFormatError except (urllib2.URLError, httplib.HTTPException, socket.error), err: @@ -372,21 +453,53 @@ class FileDownloader(object): if info is None: break + def _download_with_rtmpdump(self, filename, url): + self.report_destination(filename) + + # Check for rtmpdump first + try: + subprocess.call(['rtmpdump', '-h'], stdout=(file(os.path.devnull, 'w')), stderr=subprocess.STDOUT) + except (OSError, IOError): + self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run') + return False + + # Download using rtmpdump. rtmpdump returns exit code 2 when + # the connection was interrumpted and resuming appears to be + # possible. This is part of rtmpdump's normal usage, AFAIK. + basic_args = ['rtmpdump', '-q', '-r', url, '-o', filename] + retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]) + while retval == 2 or retval == 1: + self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename), skip_eol=True) + time.sleep(2.0) # This seems to be needed + retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1]) + if retval == 0: + self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename)) + return True + else: + self.trouble('ERROR: rtmpdump exited with code %d' % retval) + return False + def _do_download(self, filename, url): - stream = None - open_mode = 'ab' + # Attempt to download using rtmpdump + if url.startswith('rtmp'): + return self._download_with_rtmpdump(filename, url) + stream = None + open_mode = 'wb' basic_request = urllib2.Request(url, None, std_headers) request = urllib2.Request(url, None, std_headers) - # Attempt to resume download with "continuedl" option + # Establish possible resume length if os.path.isfile(filename): resume_len = os.path.getsize(filename) else: resume_len = 0 - if self.params['continuedl'] and 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' # Establish connection try: @@ -394,12 +507,16 @@ class FileDownloader(object): except (urllib2.HTTPError, ), err: if err.code != 416: # 416 is 'Requested range not satisfiable' raise + # Unable to resume data = urllib2.urlopen(basic_request) content_length = data.info()['Content-Length'] + if content_length is not None and long(content_length) == resume_len: + # Because the file had already been fully downloaded self.report_file_already_downloaded(filename) return True else: + # Because the server didn't let us self.report_unable_to_resume() open_mode = 'wb' @@ -421,7 +538,7 @@ class FileDownloader(object): # Open file just in time if stream is None: try: - stream = open(filename, open_mode) + (stream, filename) = sanitize_open(filename, open_mode) self.report_destination(filename) except (OSError, IOError), err: self.trouble('ERROR: unable to open for writing: %s' % str(err)) @@ -507,46 +624,24 @@ class InfoExtractor(object): class YoutubeIE(InfoExtractor): """Information extractor for youtube.com.""" - _VALID_URL = r'^((?:http://)?(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:\.php)?)?\?(?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' + _VALID_URL = r'^((?:http://)?(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:\.php)?)?[\?#](?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' _LANG_URL = r'http://uk.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' _LOGIN_URL = 'http://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' - _available_formats = ['22', '35', '18', '5', '17', '13', None] # listed in order of priority for -b flag + _available_formats = ['37', '22', '35', '18', '5', '17', '13', None] # listed in order of priority for -b flag _video_extensions = { '13': '3gp', '17': 'mp4', '18': 'mp4', '22': 'mp4', + '37': 'mp4', } @staticmethod def suitable(url): return (re.match(YoutubeIE._VALID_URL, url) is not None) - @staticmethod - def htmlentity_transform(matchobj): - """Transforms an HTML entity to a Unicode character.""" - entity = matchobj.group(1) - - # Known non-numeric HTML entity - if entity in htmlentitydefs.name2codepoint: - return unichr(htmlentitydefs.name2codepoint[entity]) - - # Unicode character - mobj = re.match(ur'(?u)#(x?\d+)', entity) - if mobj is not None: - numstr = mobj.group(1) - if numstr.startswith(u'x'): - base = 16 - numstr = u'0%s' % numstr - else: - base = 10 - return unichr(long(numstr, base)) - - # Unknown entity in name, return its literal representation - return (u'&%s;' % entity) - def report_lang(self): """Report attempt to set language.""" self._downloader.to_stdout(u'[youtube] Setting language') @@ -567,14 +662,14 @@ class YoutubeIE(InfoExtractor): """Report attempt to extract video information.""" self._downloader.to_stdout(u'[youtube] %s: Extracting video information' % video_id) - def report_video_url(self, video_id, video_real_url): - """Report extracted video URL.""" - self._downloader.to_stdout(u'[youtube] %s: URL: %s' % (video_id, video_real_url)) - def report_unavailable_format(self, video_id, format): """Report extracted video URL.""" self._downloader.to_stdout(u'[youtube] %s: Format %s not available' % (video_id, format)) + def report_rtmp_download(self): + """Indicate the download will use the RTMP protocol.""" + self._downloader.to_stdout(u'[youtube] RTMP download detected') + def _real_initialize(self): if self._downloader is None: return @@ -673,47 +768,47 @@ class YoutubeIE(InfoExtractor): try: self.report_video_info_webpage_download(video_id) video_info_webpage = urllib2.urlopen(request).read() + video_info = parse_qs(video_info_webpage) except (urllib2.URLError, httplib.HTTPException, socket.error), err: self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err)) return self.report_information_extraction(video_id) # "t" param - mobj = re.search(r'(?m)&token=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: + if 'token' not in video_info: # Attempt to see if YouTube has issued an error message - mobj = re.search(r'(?m)&reason=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: + if 'reason' not in video_info: self._downloader.trouble(u'ERROR: unable to extract "t" parameter for unknown reason') stream = open('reportme-ydl-%s.dat' % time.time(), 'wb') stream.write(video_info_webpage) stream.close() else: - reason = urllib.unquote_plus(mobj.group(1)) + reason = urllib.unquote_plus(video_info['reason'][0]) self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8')) return - token = urllib.unquote(mobj.group(1)) + token = urllib.unquote_plus(video_info['token'][0]) video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&eurl=&el=detailpage&ps=default&gl=US&hl=en' % (video_id, token) if format_param is not None: video_real_url = '%s&fmt=%s' % (video_real_url, format_param) - self.report_video_url(video_id, video_real_url) + + # Check possible RTMP download + if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): + self.report_rtmp_download() + video_real_url = video_info['conn'][0] # uploader - mobj = re.search(r'(?m)&author=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: + if 'author' not in video_info: self._downloader.trouble(u'ERROR: unable to extract uploader nickname') return - video_uploader = urllib.unquote(mobj.group(1)) + video_uploader = urllib.unquote_plus(video_info['author'][0]) # title - mobj = re.search(r'(?m)&title=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: + if 'title' not in video_info: self._downloader.trouble(u'ERROR: unable to extract video title') return - video_title = urllib.unquote(mobj.group(1)) + video_title = urllib.unquote_plus(video_info['title'][0]) video_title = video_title.decode('utf-8') - video_title = re.sub(ur'(?u)&(.+?);', self.htmlentity_transform, video_title) - video_title = video_title.replace(os.sep, u'%') + video_title = sanitize_title(video_title) # simplified title simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) @@ -853,8 +948,9 @@ class MetacafeIE(InfoExtractor): self._downloader.trouble(u'ERROR: unable to extract title') return video_title = mobj.group(1).decode('utf-8') + video_title = sanitize_title(video_title) - mobj = re.search(r'(?ms)