X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/147e1f3ceb674aa51d1837ca812f8176381d5d7b..6cd98fe80fb13ffc76b9e9bd8c01485644d1da8e:/youtube-dl?ds=sidebyside diff --git a/youtube-dl b/youtube-dl index e89c915..e8b19c8 100755 --- a/youtube-dl +++ b/youtube-dl @@ -3,7 +3,16 @@ # Author: Ricardo Garcia Gonzalez # Author: Danny Colligan # Author: Benjamin Johnson +# Author: Vasyl' Vavrychuk +# Author: Witold Baryluk +# Author: PaweÅ Paprota +# Author: Gergely Imreh # License: Public domain code +import cookielib +import ctypes +import datetime +import email.utils +import gzip import htmlentitydefs import httplib import locale @@ -14,11 +23,13 @@ import os.path import re import socket import string +import StringIO import subprocess import sys import time import urllib import urllib2 +import zlib # parse_qs was moved from the cgi module to the urlparse module recently. try: @@ -27,9 +38,10 @@ except ImportError: from cgi import parse_qs std_headers = { - 'User-Agent': 'Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.6) Gecko/20100627 Firefox/3.6.6', + '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', } @@ -53,7 +65,7 @@ def preferredencoding(): 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. """ @@ -94,6 +106,9 @@ def sanitize_open(filename, open_mode): """ try: if filename == u'-': + if sys.platform == 'win32': + import msvcrt + msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) return (sys.stdout, filename) stream = open(filename, open_mode) return (stream, filename) @@ -105,10 +120,17 @@ 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 class DownloadError(Exception): """Download Error exception. - + This exception may be thrown by FileDownloader objects if they are not configured to continue on errors. They will contain the appropriate error message. @@ -131,7 +153,7 @@ class PostProcessingError(Exception): """ pass -class UnavailableFormatError(Exception): +class UnavailableVideoError(Exception): """Unavailable Format exception. This exception will be thrown when a video is requested @@ -154,6 +176,64 @@ class ContentTooShortError(Exception): self.downloaded = downloaded self.expected = expected +class YoutubeDLHandler(urllib2.HTTPHandler): + """Handler for HTTP requests and responses. + + This class, when installed with an OpenerDirector, automatically adds + the standard headers to every HTTP request and handles gzipped and + deflated responses from web servers. If compression is to be avoided in + 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/ + + Andrew Rowls, the author of that code, agreed to release it to the + public domain. + """ + + @staticmethod + def deflate(data): + try: + 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'): + return urllib2.addinfourl(stream, headers, url, code) + 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: + del req.headers[h] + req.add_header(h, std_headers[h]) + if 'Youtubedl-no-compression' in req.headers: + if 'Accept-encoding' in req.headers: + del req.headers['Accept-encoding'] + del req.headers['Youtubedl-no-compression'] + return req + + def http_response(self, req, resp): + old_resp = resp + # gzip + if resp.headers.get('Content-encoding', '') == 'gzip': + gz = gzip.GzipFile(fileobj=StringIO.StringIO(resp.read()), mode='r') + resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) + resp.msg = old_resp.msg + # deflate + if resp.headers.get('Content-encoding', '') == 'deflate': + gz = StringIO.StringIO(self.deflate(resp.read())) + resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) + resp.msg = old_resp.msg + return resp + class FileDownloader(object): """File Downloader class. @@ -181,22 +261,31 @@ class FileDownloader(object): Available options: - username: Username for authentication purposes. - password: Password for authentication purposes. - usenetrc: Use netrc for authentication instead. - quiet: Do not print messages to stdout. - forceurl: Force printing final URL. - forcetitle: Force printing title. - simulate: Do not download the video files. - format: Video format code. - format_limit: Highest quality format to try. - outtmpl: Template for output names. - ignoreerrors: Do not stop on download errors. - ratelimit: Download speed limit, in bytes/sec. - nooverwrites: Prevent overwriting files. - retries: Number of times to retry for HTTP error 503 - continuedl: Try to continue downloads if possible. - noprogress: Do not print the progress bar. + username: Username for authentication purposes. + password: Password for authentication purposes. + usenetrc: Use netrc for authentication instead. + quiet: Do not print messages to stdout. + forceurl: Force printing final URL. + forcetitle: Force printing title. + forcethumbnail: Force printing thumbnail URL. + forcedescription: Force printing description. + forcefilename: Force printing final filename. + simulate: Do not download the video files. + format: Video format code. + format_limit: Highest quality format to try. + outtmpl: Template for output names. + ignoreerrors: Do not stop on download errors. + ratelimit: Download speed limit, in bytes/sec. + nooverwrites: Prevent overwriting files. + retries: Number of times to retry for HTTP error 5xx + continuedl: Try to continue downloads if possible. + noprogress: Do not print the progress bar. + playliststart: Playlist item to start at. + playlistend: Playlist item to end at. + 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. """ params = None @@ -204,6 +293,7 @@ class FileDownloader(object): _pps = [] _download_retcode = None _num_downloads = None + _screen_file = None def __init__(self, params): """Create a FileDownloader object with the given options.""" @@ -211,8 +301,9 @@ class FileDownloader(object): self._pps = [] self._download_retcode = 0 self._num_downloads = 0 + 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".""" @@ -222,7 +313,7 @@ class FileDownloader(object): for dir in aggregate: if not os.path.exists(dir): os.mkdir(dir) - + @staticmethod def format_bytes(bytes): if bytes is None: @@ -287,40 +378,42 @@ class FileDownloader(object): multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower()) return long(round(number * multiplier)) - @staticmethod - def verify_url(url): - """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.""" self._ies.append(ie) ie.set_downloader(self) - + def add_post_processor(self, pp): """Add a PostProcessor object to the end of the chain.""" self._pps.append(pp) pp.set_downloader(self) - - def to_stdout(self, message, skip_eol=False, ignore_encoding_errors=False): + + def to_screen(self, message, skip_eol=False, ignore_encoding_errors=False): """Print message to stdout if not in quiet mode.""" try: if not self.params.get('quiet', False): - print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(preferredencoding()), - sys.stdout.flush() + terminator = [u'\n', u''][skip_eol] + print >>self._screen_file, (u'%s%s' % (message, terminator)).encode(preferredencoding()), + self._screen_file.flush() except (UnicodeEncodeError), err: if not ignore_encoding_errors: raise - + def to_stderr(self, message): """Print message to stderr.""" print >>sys.stderr, message.encode(preferredencoding()) - + + def to_cons_title(self, message): + """Set console/terminal window title to message.""" + if not self.params.get('consoletitle', False): + return + if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): + # c_wchar_p() might not be necessary if `message` is + # already of type unicode() + ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) + elif 'TERM' in os.environ: + sys.stderr.write('\033]0;%s\007' % message.encode(preferredencoding())) + def fixed_template(self): """Checks if the output template is fixed.""" return (re.search(ur'(?u)%\(.+?\)s', self.params['outtmpl']) is None) @@ -351,58 +444,103 @@ class FileDownloader(object): if speed > rate_limit: time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit) + def temp_name(self, filename): + """Returns a temporary filename for the given filename.""" + if self.params.get('nopart', False) or filename == u'-' or \ + (os.path.exists(filename) and not os.path.isfile(filename)): + return filename + return filename + u'.part' + + def undo_temp_name(self, filename): + if filename.endswith(u'.part'): + return filename[:-len(u'.part')] + return filename + + def try_rename(self, old_filename, new_filename): + try: + if old_filename == new_filename: + return + 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: + return + if not os.path.isfile(filename): + return + timestr = last_modified_hdr + if timestr is None: + return + filetime = timeconvert(timestr) + if filetime is None: + return + try: + os.utime(filename,(time.time(), filetime)) + except: + pass + def report_destination(self, filename): """Report destination filename.""" - self.to_stdout(u'[download] Destination: %s' % filename, ignore_encoding_errors=True) - + self.to_screen(u'[download] Destination: %s' % filename, ignore_encoding_errors=True) + def report_progress(self, percent_str, data_len_str, speed_str, eta_str): """Report download progress.""" if self.params.get('noprogress', False): return - self.to_stdout(u'\r[download] %s of %s at %s ETA %s' % + self.to_screen(u'\r[download] %s of %s at %s ETA %s' % (percent_str, data_len_str, speed_str, eta_str), skip_eol=True) + self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' % + (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip())) def report_resuming_byte(self, resume_len): - """Report attemtp to resume at given byte.""" - self.to_stdout(u'[download] Resuming download at byte %s' % resume_len) - + """Report attempt to resume at given byte.""" + self.to_screen(u'[download] Resuming download at byte %s' % resume_len) + def report_retry(self, count, retries): - """Report retry in case of HTTP error 503""" - self.to_stdout(u'[download] Got HTTP error 503. Retrying (attempt %d of %d)...' % (count, retries)) - + """Report retry in case of HTTP error 5xx""" + self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries)) + def report_file_already_downloaded(self, file_name): """Report file has already been fully downloaded.""" try: - self.to_stdout(u'[download] %s has already been downloaded' % file_name) + self.to_screen(u'[download] %s has already been downloaded' % file_name) except (UnicodeEncodeError), err: - self.to_stdout(u'[download] The file has already been downloaded') - + self.to_screen(u'[download] The file has already been downloaded') + def report_unable_to_resume(self): """Report it was impossible to resume download.""" - self.to_stdout(u'[download] Unable to resume') - + self.to_screen(u'[download] Unable to resume') + def report_finish(self): """Report download finished.""" if self.params.get('noprogress', False): - self.to_stdout(u'[download] Download completed') + self.to_screen(u'[download] Download completed') else: - self.to_stdout(u'') - + self.to_screen(u'') + def increment_downloads(self): """Increment the ordinal that assigns a number to each file.""" self._num_downloads += 1 + def prepare_filename(self, info_dict): + """Generate the output filename.""" + try: + template_dict = dict(info_dict) + template_dict['epoch'] = unicode(long(time.time())) + template_dict['autonumber'] = unicode('%05d' % self._num_downloads) + filename = self.params['outtmpl'] % template_dict + return filename + except (ValueError, KeyError), err: + self.trouble(u'ERROR: invalid system charset or erroneous output template') + return None + def process_info(self, info_dict): """Process a single dictionary returned by an InfoExtractor.""" + filename = self.prepare_filename(info_dict) # Do nothing else if in simulate mode if self.params.get('simulate', False): - # Verify URL if it's an HTTP one - if info_dict['url'].startswith('http'): - try: - 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(preferredencoding(), 'xmlcharrefreplace') @@ -412,42 +550,39 @@ class FileDownloader(object): 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 - - try: - template_dict = dict(info_dict) - template_dict['epoch'] = unicode(long(time.time())) - template_dict['ord'] = unicode('%05d' % self._num_downloads) - filename = self.params['outtmpl'] % template_dict - except (ValueError, KeyError), err: - self.trouble('ERROR: invalid output template or system charset: %s' % str(err)) + + if filename is None: + return if self.params.get('nooverwrites', False) and os.path.exists(filename): - self.to_stderr(u'WARNING: file exists: %s; skipping' % filename) + self.to_stderr(u'WARNING: file exists and will be skipped') return try: self.pmkdir(filename) except (OSError, IOError), err: - self.trouble('ERROR: unable to create directories: %s' % str(err)) + self.trouble(u'ERROR: unable to create directories: %s' % str(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 UnavailableFormatError + raise UnavailableVideoError except (urllib2.URLError, httplib.HTTPException, socket.error), err: - self.trouble('ERROR: unable to download video data: %s' % str(err)) + self.trouble(u'ERROR: unable to download video data: %s' % str(err)) return except (ContentTooShortError, ), err: - self.trouble('ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) + 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('ERROR: postprocessing: %s' % str(err)) + self.trouble(u'ERROR: postprocessing: %s' % str(err)) return def download(self, url_list): @@ -472,7 +607,7 @@ class FileDownloader(object): break if not suitable_found: - self.trouble('ERROR: no suitable InfoExtractor: %s' % url) + self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url) return self._download_retcode @@ -484,9 +619,10 @@ class FileDownloader(object): info = pp.run(info) if info is None: break - + def _download_with_rtmpdump(self, filename, url, player_url): self.report_destination(filename) + tmpfilename = self.temp_name(filename) # Check for rtmpdump first try: @@ -498,36 +634,46 @@ class FileDownloader(object): # 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'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', filename] + basic_args = ['rtmpdump', '-q'] + [[], ['-W', player_url]][player_url is not None] + ['-r', url, '-o', tmpfilename] retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]) while retval == 2 or retval == 1: - prevsize = os.path.getsize(filename) - self.to_stdout(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True) + prevsize = os.path.getsize(tmpfilename) + self.to_screen(u'\r[rtmpdump] %s bytes' % prevsize, skip_eol=True) time.sleep(5.0) # This seems to be needed retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1]) - cursize = os.path.getsize(filename) + cursize = os.path.getsize(tmpfilename) if prevsize == cursize and retval == 1: break if retval == 0: - self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename)) + self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(tmpfilename)) + self.try_rename(tmpfilename, filename) return True else: - self.trouble('\nERROR: rtmpdump exited with code %d' % retval) + self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval) return False def _do_download(self, filename, url, player_url): + # 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) + return True + # Attempt to download using rtmpdump if url.startswith('rtmp'): return self._download_with_rtmpdump(filename, url, player_url) + tmpfilename = self.temp_name(filename) stream = None open_mode = 'wb' - basic_request = urllib2.Request(url, None, std_headers) - request = urllib2.Request(url, None, std_headers) + + # Do not include the Accept-Encoding header + headers = {'Youtubedl-no-compression': 'True'} + basic_request = urllib2.Request(url, None, headers) + request = urllib2.Request(url, None, headers) # Establish possible resume length - if os.path.isfile(filename): - resume_len = os.path.getsize(filename) + if os.path.isfile(tmpfilename): + resume_len = os.path.getsize(tmpfilename) else: resume_len = 0 @@ -539,36 +685,57 @@ class FileDownloader(object): count = 0 retries = self.params.get('retries', 0) - while True: + while count <= retries: # Establish connection try: data = urllib2.urlopen(request) break except (urllib2.HTTPError, ), err: - if err.code == 503: - # Retry in case of HTTP error 503 - count += 1 - if count <= retries: - self.report_retry(count, retries) - continue - if err.code != 416: # 416 is 'Requested range not satisfiable' + if (err.code < 500 or err.code >= 600) and err.code != 416: + # Unexpected HTTP error 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' + elif err.code == 416: + # Unable to resume (requested range not satisfiable) + try: + # Open the connection again without the range header + data = urllib2.urlopen(basic_request) + content_length = data.info()['Content-Length'] + except (urllib2.HTTPError, ), err: + if err.code < 500 or err.code >= 600: + raise + else: + # Examine the reported length + if (content_length is not None and + (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, + # changing the file size slightly and causing problems for some users. So + # I decided to implement a suggested change and consider the file + # completely downloaded if the file size differs less than 100 bytes from + # the one in the hard drive. + self.report_file_already_downloaded(filename) + self.try_rename(tmpfilename, filename) + return True + else: + # The length does not match, we start the download over + self.report_unable_to_resume() + open_mode = 'wb' + break + # Retry + count += 1 + if count <= retries: + self.report_retry(count, retries) + + if count > retries: + self.trouble(u'ERROR: giving up after %s retries' % retries) + return False data_len = data.info().get('Content-length', None) + if data_len is not None: + data_len = long(data_len) + resume_len data_len_str = self.format_bytes(data_len) - byte_counter = 0 + byte_counter = 0 + resume_len block_size = 1024 start = time.time() while True: @@ -576,37 +743,45 @@ class FileDownloader(object): before = time.time() data_block = data.read(block_size) after = time.time() - data_block_len = len(data_block) - if data_block_len == 0: + if len(data_block) == 0: break - byte_counter += data_block_len + byte_counter += len(data_block) # Open file just in time if stream is None: try: - (stream, filename) = sanitize_open(filename, open_mode) + (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode) + filename = self.undo_temp_name(tmpfilename) self.report_destination(filename) except (OSError, IOError), err: - self.trouble('ERROR: unable to open for writing: %s' % str(err)) + self.trouble(u'ERROR: unable to open for writing: %s' % str(err)) return False try: stream.write(data_block) except (IOError, OSError), err: - self.trouble('\nERROR: unable to write data: %s' % str(err)) - block_size = self.best_block_size(after - before, data_block_len) + self.trouble(u'\nERROR: unable to write data: %s' % str(err)) + return False + 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, byte_counter) - speed_str = self.calc_speed(start, time.time(), byte_counter) + 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) # Apply rate limit - self.slow_down(start, byte_counter) + self.slow_down(start, byte_counter - resume_len) + stream.close() self.report_finish() - if data_len is not None and str(byte_counter) != data_len: + if data_len is not None and byte_counter != data_len: raise ContentTooShortError(byte_counter, long(data_len)) + self.try_rename(tmpfilename, filename) + + # Update file modification time + if self.params.get('updatetime', True): + self.try_utime(filename, data.info().get('last-modified', None)) + return True class InfoExtractor(object): @@ -671,7 +846,7 @@ class InfoExtractor(object): def set_downloader(self, downloader): """Sets the downloader for this IE.""" self._downloader = downloader - + def _real_initialize(self): """Real initialization process. Redefine in subclasses.""" pass @@ -683,13 +858,13 @@ class InfoExtractor(object): class YoutubeIE(InfoExtractor): """Information extractor for youtube.com.""" - _VALID_URL = r'^((?:http://)?(?:youtu\.be/|(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:_popup)?(?:\.php)?)?[\?#](?:.+&)?v=))))?([0-9A-Za-z_-]+)(?(1).+)?$' - _LANG_URL = r'http://uk.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' - _LOGIN_URL = 'http://www.youtube.com/signup?next=/&gl=US&hl=en' + _VALID_URL = r'^((?:https?://)?(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/)(?:(?:(?:v|embed|e)/)|(?:(?:watch(?:_popup)?(?:\.php)?)?(?:\?|#!?)(?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' + _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 priority for the -b option - _available_formats = ['38', '37', '22', '45', '35', '34', '43', '18', '6', '5', '17', '13', None] + # Listed in order of quality + _available_formats = ['38', '37', '22', '45', '35', '34', '43', '18', '6', '5', '17', '13'] _video_extensions = { '13': '3gp', '17': 'mp4', @@ -707,36 +882,36 @@ class YoutubeIE(InfoExtractor): def report_lang(self): """Report attempt to set language.""" - self._downloader.to_stdout(u'[youtube] Setting language') + self._downloader.to_screen(u'[youtube] Setting language') def report_login(self): """Report attempt to log in.""" - self._downloader.to_stdout(u'[youtube] Logging in') - + self._downloader.to_screen(u'[youtube] Logging in') + def report_age_confirmation(self): """Report attempt to confirm age.""" - self._downloader.to_stdout(u'[youtube] Confirming age') - + self._downloader.to_screen(u'[youtube] Confirming age') + def report_video_webpage_download(self, video_id): """Report attempt to download video webpage.""" - self._downloader.to_stdout(u'[youtube] %s: Downloading video webpage' % video_id) - + self._downloader.to_screen(u'[youtube] %s: Downloading video webpage' % video_id) + def report_video_info_webpage_download(self, video_id): """Report attempt to download video info webpage.""" - self._downloader.to_stdout(u'[youtube] %s: Downloading video info webpage' % video_id) - + self._downloader.to_screen(u'[youtube] %s: Downloading video info webpage' % video_id) + def report_information_extraction(self, video_id): """Report attempt to extract video information.""" - self._downloader.to_stdout(u'[youtube] %s: Extracting video information' % video_id) - + self._downloader.to_screen(u'[youtube] %s: Extracting video information' % video_id) + 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)) - + self._downloader.to_screen(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') - + self._downloader.to_screen(u'[youtube] RTMP download detected') + def _real_initialize(self): if self._downloader is None: return @@ -762,7 +937,7 @@ class YoutubeIE(InfoExtractor): return # Set language - request = urllib2.Request(self._LANG_URL, None, std_headers) + request = urllib2.Request(self._LANG_URL) try: self.report_lang() urllib2.urlopen(request).read() @@ -782,7 +957,7 @@ class YoutubeIE(InfoExtractor): 'username': username, 'password': password, } - request = urllib2.Request(self._LOGIN_URL, urllib.urlencode(login_form), std_headers) + request = urllib2.Request(self._LOGIN_URL, urllib.urlencode(login_form)) try: self.report_login() login_results = urllib2.urlopen(request).read() @@ -792,13 +967,13 @@ class YoutubeIE(InfoExtractor): except (urllib2.URLError, httplib.HTTPException, socket.error), err: self._downloader.to_stderr(u'WARNING: unable to log in: %s' % str(err)) return - + # Confirm age age_form = { 'next_url': '/', 'action_confirm': 'Confirm', } - request = urllib2.Request(self._AGE_URL, urllib.urlencode(age_form), std_headers) + request = urllib2.Request(self._AGE_URL, urllib.urlencode(age_form)) try: self.report_age_confirmation() age_results = urllib2.urlopen(request).read() @@ -812,130 +987,145 @@ class YoutubeIE(InfoExtractor): if mobj is None: self._downloader.trouble(u'ERROR: invalid URL: %s' % url) return - - # At this point we have a new video - if self._downloader is not None: - self._downloader.increment_downloads() video_id = mobj.group(2) - # Downloader parameters - best_quality = False - all_formats = False - format_param = None - quality_index = 0 - if self._downloader is not None: - params = self._downloader.params - format_param = params.get('format', None) - if format_param == '0': - format_limit = params.get('format_limit', None) - if format_limit is not None: - try: - # Start at a different format if the user has limited the maximum quality - quality_index = self._available_formats.index(format_limit) - except ValueError: - pass - format_param = self._available_formats[quality_index] - best_quality = True - elif format_param == '-1': - format_param = self._available_formats[quality_index] - all_formats = True - - while True: - # Extension - video_extension = self._video_extensions.get(format_param, 'flv') + # 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) + try: + video_webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err)) + return - # Get video webpage - self.report_video_webpage_download(video_id) - request = urllib2.Request('http://www.youtube.com/watch?v=%s&gl=US&hl=en' % video_id, None, std_headers) + # Attempt to extract SWF player URL + mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) + if mobj is not None: + player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) + else: + player_url = None + + # Get video info + 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)) + request = urllib2.Request(video_info_url) try: - video_webpage = urllib2.urlopen(request).read() + video_info_webpage = urllib2.urlopen(request).read() + video_info = parse_qs(video_info_webpage) + if 'token' in video_info: + break except (urllib2.URLError, httplib.HTTPException, socket.error), err: - self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err)) + self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err)) return + if 'token' not in video_info: + if 'reason' in video_info: + self._downloader.trouble(u'ERROR: YouTube said: %s' % video_info['reason'][0].decode('utf-8')) + else: + self._downloader.trouble(u'ERROR: "token" parameter not in video info for unknown reason') + return + + # Start extracting information + self.report_information_extraction(video_id) - # Attempt to extract SWF player URL - mobj = re.search(r'swfConfig.*"(http://.*?watch.*?-.*?\.swf)"', video_webpage) + # uploader + if 'author' not in video_info: + self._downloader.trouble(u'ERROR: unable to extract uploader nickname') + return + video_uploader = urllib.unquote_plus(video_info['author'][0]) + + # title + if 'title' not in video_info: + self._downloader.trouble(u'ERROR: unable to extract video title') + return + video_title = urllib.unquote_plus(video_info['title'][0]) + video_title = video_title.decode('utf-8') + 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'_') + + # thumbnail image + if 'thumbnail_url' not in video_info: + self._downloader.trouble(u'WARNING: unable to extract video thumbnail') + video_thumbnail = '' + else: # don't panic if we can't find it + video_thumbnail = urllib.unquote_plus(video_info['thumbnail_url'][0]) + + # upload date + upload_date = u'NA' + 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', '%b %d %Y'] + for expression in format_expressions: + try: + upload_date = datetime.datetime.strptime(upload_date, expression).strftime('%Y%m%d') + except: + 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: - player_url = mobj.group(1) + video_description = mobj.group(1) + + # token + video_token = urllib.unquote_plus(video_info['token'][0]) + + # Decide which formats to download + req_format = self._downloader.params.get('format', None) + + if '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 = [dict(pairStr.split('=') for pairStr in uds.split('&')) for uds in url_data_strs] + url_map = dict((ud['itag'], urllib.unquote(ud['url'])) 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):] else: - player_url = None - - # Get video info - 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)) - request = urllib2.Request(video_info_url, None, std_headers) - try: - video_info_webpage = urllib2.urlopen(request).read() - video_info = parse_qs(video_info_webpage) - if 'token' in video_info: - break - 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 - if 'token' not in video_info: - # Attempt to see if YouTube has issued an error message - 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(video_info['reason'][0]) - self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8')) - return - 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) - - # 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 - if 'author' not in video_info: - self._downloader.trouble(u'ERROR: unable to extract uploader nickname') + format_list = self._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 - video_uploader = urllib.unquote_plus(video_info['author'][0]) + if req_format is None: + video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality + elif req_format == '-1': + video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats + else: + # Specific format + if req_format not in url_map: + self._downloader.trouble(u'ERROR: requested format not available') + return + video_url_list = [(req_format, url_map[req_format])] # Specific format - # title - if 'title' not in video_info: - self._downloader.trouble(u'ERROR: unable to extract video title') - return - video_title = urllib.unquote_plus(video_info['title'][0]) - video_title = video_title.decode('utf-8') - 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'_') - - # thumbnail image - if 'thumbnail_url' not in video_info: - self._downloader.trouble(u'WARNING: unable to extract video thumbnail') - video_thumbnail = '' - else: # don't panic if we can't find it - video_thumbnail = urllib.unquote_plus(video_info['thumbnail_url'][0]) - - # 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) + 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') + return + for format_param, video_real_url in video_url_list: + # At this point we have a new video + self._downloader.increment_downloads() + + # 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({ 'id': video_id.decode('utf-8'), 'url': video_real_url.decode('utf-8'), 'uploader': video_uploader.decode('utf-8'), + 'upload_date': upload_date, 'title': video_title, 'stitle': simple_title, 'ext': video_extension.decode('utf-8'), @@ -944,32 +1134,8 @@ class YoutubeIE(InfoExtractor): 'description': video_description.decode('utf-8'), 'player_url': player_url, }) - - if all_formats: - quality_index += 1 - if quality_index == len(self._available_formats): - # None left to get - return - else: - format_param = self._available_formats[quality_index] - continue - return - - except UnavailableFormatError, err: - if best_quality or all_formats: - quality_index += 1 - if quality_index == len(self._available_formats): - # I don't ever expect this to happen - if not all_formats: - self._downloader.trouble(u'ERROR: no known formats available for video') - return - else: - self.report_unavailable_format(video_id, format_param) - format_param = self._available_formats[quality_index] - continue - else: - self._downloader.trouble('ERROR: format not available for video') - return + except UnavailableVideoError, err: + self._downloader.trouble(u'\nERROR: unable to download video') class MetacafeIE(InfoExtractor): @@ -990,23 +1156,23 @@ class MetacafeIE(InfoExtractor): def report_disclaimer(self): """Report disclaimer retrieval.""" - self._downloader.to_stdout(u'[metacafe] Retrieving disclaimer') + self._downloader.to_screen(u'[metacafe] Retrieving disclaimer') def report_age_confirmation(self): """Report attempt to confirm age.""" - self._downloader.to_stdout(u'[metacafe] Confirming age') - + self._downloader.to_screen(u'[metacafe] Confirming age') + def report_download_webpage(self, video_id): """Report webpage download.""" - self._downloader.to_stdout(u'[metacafe] %s: Downloading webpage' % video_id) - + self._downloader.to_screen(u'[metacafe] %s: Downloading webpage' % video_id) + def report_extraction(self, video_id): """Report information extraction.""" - self._downloader.to_stdout(u'[metacafe] %s: Extracting information' % video_id) + self._downloader.to_screen(u'[metacafe] %s: Extracting information' % video_id) def _real_initialize(self): # Retrieve disclaimer - request = urllib2.Request(self._DISCLAIMER, None, std_headers) + request = urllib2.Request(self._DISCLAIMER) try: self.report_disclaimer() disclaimer = urllib2.urlopen(request).read() @@ -1019,14 +1185,14 @@ class MetacafeIE(InfoExtractor): 'filters': '0', 'submit': "Continue - I'm over 18", } - request = urllib2.Request(self._FILTER_POST, urllib.urlencode(disclaimer_form), std_headers) + request = urllib2.Request(self._FILTER_POST, urllib.urlencode(disclaimer_form)) try: self.report_age_confirmation() disclaimer = urllib2.urlopen(request).read() except (urllib2.URLError, httplib.HTTPException, socket.error), err: self._downloader.trouble(u'ERROR: unable to confirm age: %s' % str(err)) return - + def _real_extract(self, url): # Extract id and simplified title from URL mobj = re.match(self._VALID_URL, url) @@ -1043,11 +1209,9 @@ class MetacafeIE(InfoExtractor): return # At this point we have a new video - if self._downloader is not None: - self._downloader.increment_downloads() + self._downloader.increment_downloads() simple_title = mobj.group(2).decode('utf-8') - video_extension = 'flv' # Retrieve video webpage to extract further information request = urllib2.Request('http://www.metacafe.com/watch/%s/' % video_id) @@ -1061,20 +1225,33 @@ class MetacafeIE(InfoExtractor): # Extract URL, uploader and title from webpage self.report_extraction(video_id) mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage) - if mobj is None: - self._downloader.trouble(u'ERROR: unable to extract media URL') - return - mediaURL = urllib.unquote(mobj.group(1)) - - #mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage) - #if mobj is None: - # self._downloader.trouble(u'ERROR: unable to extract gdaKey') - # return - #gdaKey = mobj.group(1) - # - #video_url = '%s?__gda__=%s' % (mediaURL, gdaKey) + if mobj is not None: + mediaURL = urllib.unquote(mobj.group(1)) + video_extension = mediaURL[-3:] - video_url = mediaURL + # Extract gdaKey if available + mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage) + if mobj is None: + video_url = mediaURL + else: + gdaKey = mobj.group(1) + video_url = '%s?__gda__=%s' % (mediaURL, gdaKey) + else: + mobj = re.search(r' name="flashvars" value="(.*?)"', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + vardict = parse_qs(mobj.group(1)) + if 'mediaData' not in vardict: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + mobj = re.search(r'"mediaURL":"(http.*?)","key":"(.*?)"', vardict['mediaData'][0]) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + mediaURL = mobj.group(1).replace('\\/', '/') + video_extension = mediaURL[-3:] + video_url = '%s?__gda__=%s' % (mediaURL, mobj.group(2)) mobj = re.search(r'(?im)