X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/cb80b21bcb31049702a24f5ca2587d4f5bccc4b1..a475f326a4099c540835dacdda9dbbf92a56130d:/youtube-dl?ds=inline diff --git a/youtube-dl b/youtube-dl index 5afff4e..a4c8f24 100755 --- a/youtube-dl +++ b/youtube-dl @@ -2,7 +2,16 @@ # -*- coding: utf-8 -*- # Author: Ricardo Garcia Gonzalez # Author: Danny Colligan +# Author: Benjamin Johnson +# Author: Vasyl' Vavrychuk +# Author: Witold Baryluk +# Author: Paweł Paprota # License: Public domain code +import cookielib +import ctypes +import datetime +import email.utils +import gzip import htmlentitydefs import httplib import locale @@ -13,15 +22,25 @@ 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: + 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 (X11; Linux x86_64; rv:2.0b10) Gecko/20100101 Firefox/4.0b10', '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': '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', } @@ -33,19 +52,84 @@ def preferredencoding(): 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: - pref = locale.getpreferredencoding() - # Mac OSX systems have this problem sometimes - if pref == '': - return 'UTF-8' - return pref - except: - sys.stderr.write('WARNING: problem obtaining preferred encoding. Falling back to UTF-8.\n') - return 'UTF-8' + 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) + 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) + +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. @@ -68,7 +152,7 @@ class PostProcessingError(Exception): """ pass -class UnavailableFormatError(Exception): +class UnavailableVideoError(Exception): """Unavailable Format exception. This exception will be thrown when a video is requested @@ -91,6 +175,64 @@ 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. @@ -118,33 +260,49 @@ 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. - outtmpl: Template for output names. - ignoreerrors: Do not stop on download errors. - ratelimit: Download speed limit, in bytes/sec. - nooverwrites: Prevent overwriting files. - continuedl: Try to continue downloads if possible. + 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 _ies = [] _pps = [] _download_retcode = None + _num_downloads = None + _screen_file = None def __init__(self, params): """Create a FileDownloader object with the given options.""" self._ies = [] 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".""" @@ -154,7 +312,7 @@ 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: @@ -219,36 +377,42 @@ 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): + + def to_screen(self, message, skip_eol=False, ignore_encoding_errors=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(preferredencoding()), - sys.stdout.flush() - + try: + if not self.params.get('quiet', False): + 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) @@ -279,80 +443,145 @@ 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) - + 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.""" - self.to_stdout(u'\r[download] %s of %s at %s ETA %s' % + if self.params.get('noprogress', False): + return + 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 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.""" - self.to_stdout(u'[download] %s has already been downloaded' % file_name) - + try: + self.to_screen(u'[download] %s has already been downloaded' % file_name) + except (UnicodeEncodeError), err: + 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.""" - self.to_stdout(u'') + if self.params.get('noprogress', False): + self.to_screen(u'[download] Download completed') + else: + 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): - try: - info_dict['url'] = self.verify_url(info_dict['url']) - 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()) + print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace') if self.params.get('forceurl', False): - print info_dict['url'].encode(preferredencoding()) + 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 - - try: - template_dict = dict(info_dict) - template_dict['epoch'] = unicode(long(time.time())) - 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): - self.to_stderr(u'WARNING: file exists: %s; skipping' % filename) + + 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) 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']) + 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): @@ -377,7 +606,7 @@ 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 @@ -389,41 +618,123 @@ class FileDownloader(object): info = pp.run(info) if info is None: break - - def _do_download(self, filename, url): + + def _download_with_rtmpdump(self, filename, url, player_url): + self.report_destination(filename) + tmpfilename = self.temp_name(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'] + [[], ['-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(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(tmpfilename) + if prevsize == cursize and retval == 1: + break + if retval == 0: + self.to_screen(u'\r[rtmpdump] %s bytes' % os.path.getsize(tmpfilename)) + self.try_rename(tmpfilename, filename) + return True + else: + 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 = 'ab' + 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) - # Attempt to resume download with "continuedl" option - if os.path.isfile(filename): - resume_len = os.path.getsize(filename) + # Establish possible resume length + if os.path.isfile(tmpfilename): + resume_len = os.path.getsize(tmpfilename) 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: - data = urllib2.urlopen(request) - except (urllib2.HTTPError, ), err: - if err.code != 416: # 416 is 'Requested range not satisfiable' - raise - data = urllib2.urlopen(basic_request) - content_length = data.info()['Content-Length'] - if content_length is not None and long(content_length) == resume_len: - self.report_file_already_downloaded(filename) - return True - else: - self.report_unable_to_resume() - open_mode = 'wb' + count = 0 + retries = self.params.get('retries', 0) + while count <= retries: + # Establish connection + try: + data = urllib2.urlopen(request) + break + except (urllib2.HTTPError, ), err: + if (err.code < 500 or err.code >= 600) and err.code != 416: + # Unexpected HTTP error + raise + 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: @@ -431,34 +742,45 @@ 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 = 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 - stream.write(data_block) - block_size = self.best_block_size(after - before, data_block_len) + try: + stream.write(data_block) + except (IOError, OSError), err: + 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): @@ -479,6 +801,16 @@ class InfoExtractor(object): title: Literal title. stitle: Simplified title. ext: Video filename extension. + format: Video format. + player_url: SWF Player URL (may be None). + + The following fields are optional. Their primary purpose is to allow + youtube-dl to serve as the backend for a video search function, such + as the one in youtube2mp3. They are only used when their respective + forced printing functions are called: + + thumbnail: Full URL to a video thumbnail image. + 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. @@ -513,7 +845,7 @@ 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 @@ -525,70 +857,60 @@ 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).+)?$' - _LANG_URL = r'http://uk.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' - _LOGIN_URL = 'http://www.youtube.com/signup?next=/&gl=US&hl=en' + _VALID_URL = r'^((?:https?://)?(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/)(?:(?:(?:v|embed)/)|(?:(?:watch(?:_popup)?(?:\.php)?)?(?:\?|#!?)(?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' + _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' - _available_formats = ['22', '35', '18', '5', '17', '13', None] # listed in order of priority for -b flag + # 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', '18': 'mp4', '22': 'mp4', + '37': 'mp4', + '38': 'video', # You actually don't know if this will be MOV, AVI or whatever + '43': 'webm', + '45': 'webm', } @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') + 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_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_screen(u'[youtube] RTMP download detected') + def _real_initialize(self): if self._downloader is None: return @@ -614,7 +936,7 @@ 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() @@ -634,7 +956,7 @@ 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() @@ -644,13 +966,13 @@ 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() @@ -666,99 +988,151 @@ class YoutubeIE(InfoExtractor): return video_id = mobj.group(2) - # Downloader parameters - best_quality = 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_param = self._available_formats[quality_index] - best_quality = 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 info - video_info_url = 'http://www.youtube.com/get_video_info?&video_id=%s&el=detailpage&ps=default&eurl=&gl=US&hl=en' % video_id - request = urllib2.Request(video_info_url, 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: - self.report_video_info_webpage_download(video_id) 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) + 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 - # "t" param - mobj = re.search(r'(?m)&token=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: - # Attempt to see if YouTube has issued an error message - mobj = re.search(r'(?m)&reason=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: - 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)) - self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8')) - return - token = urllib.unquote(mobj.group(1)) - 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) + # Start extracting information + self.report_information_extraction(video_id) - # uploader - mobj = re.search(r'(?m)&author=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: - self._downloader.trouble(u'ERROR: unable to extract uploader nickname') - return - video_uploader = urllib.unquote(mobj.group(1)) + # uploader + if 'author' not in video_info: + self._downloader.trouble(u'ERROR: unable to extract uploader nickname') + return + video_uploader = urllib.unquote_plus(video_info['author'][0]) - # title - mobj = re.search(r'(?m)&title=([^&]+)(?:&|$)', video_info_webpage) - if mobj is None: - self._downloader.trouble(u'ERROR: unable to extract video title') + # 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'] + 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: + 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 'fmt_url_map' in video_info: + url_map = dict(tuple(pair.split('|')) for pair in video_info['fmt_url_map'][0].split(',')) + 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: + 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_title = urllib.unquote(mobj.group(1)) - 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'%') + 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 + + 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 - # simplified title - simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) - simple_title = simple_title.strip(ur'_') + 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'), + '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'), + 'player_url': player_url, }) - - return - - except UnavailableFormatError, err: - if best_quality: - if quality_index == len(self._available_formats) - 1: - # I don't ever expect this to happen - self._downloader.trouble(u'ERROR: no known formats available for video') - return - else: - self.report_unavailable_format(video_id, format_param) - quality_index += 1 - 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): @@ -779,23 +1153,23 @@ 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() @@ -808,14 +1182,14 @@ 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) @@ -831,8 +1205,10 @@ class MetacafeIE(InfoExtractor): self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % mobj2.group(1)) return + # At this point we have a new video + 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) @@ -846,28 +1222,42 @@ 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)(.*) - Video', webpage) if mobj is None: 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)
  • .*?Submitter:.*?(.*?)<', webpage) + mobj = re.search(r'(?ms)By:\s*(.+?)<', webpage) if mobj is None: self._downloader.trouble(u'ERROR: unable to extract uploader nickname') return @@ -879,78 +1269,714 @@ class MetacafeIE(InfoExtractor): 'id': video_id.decode('utf-8'), 'url': video_url.decode('utf-8'), 'uploader': video_uploader.decode('utf-8'), + 'upload_date': u'NA', 'title': video_title, 'stitle': simple_title, 'ext': video_extension.decode('utf-8'), + 'format': u'NA', + 'player_url': None, }) - except UnavailableFormatError: - self._downloader.trouble(u'ERROR: format not available for video') + except UnavailableVideoError: + self._downloader.trouble(u'\nERROR: unable to download video') -class YoutubeSearchIE(InfoExtractor): - """Information Extractor for YouTube search queries.""" - _VALID_QUERY = r'ytsearch(\d+|all)?:[\s\S]+' - _TEMPLATE_URL = 'http://www.youtube.com/results?search_query=%s&page=%s&gl=US&hl=en' - _VIDEO_INDICATOR = r'href="/watch\?v=.+?"' - _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*' - _youtube_ie = None - _max_youtube_results = 1000 +class DailymotionIE(InfoExtractor): + """Information Extractor for Dailymotion""" - def __init__(self, youtube_ie, downloader=None): + _VALID_URL = r'(?i)(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/video/([^_/]+)_([^/]+)' + + def __init__(self, downloader=None): InfoExtractor.__init__(self, downloader) - self._youtube_ie = youtube_ie - + @staticmethod def suitable(url): - return (re.match(YoutubeSearchIE._VALID_QUERY, url) is not None) + return (re.match(DailymotionIE._VALID_URL, url) is not None) - def report_download_page(self, query, pagenum): - """Report attempt to download playlist page with given number.""" - self._downloader.to_stdout(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) + def report_download_webpage(self, video_id): + """Report webpage download.""" + self._downloader.to_screen(u'[dailymotion] %s: Downloading webpage' % video_id) + + def report_extraction(self, video_id): + """Report information extraction.""" + self._downloader.to_screen(u'[dailymotion] %s: Extracting information' % video_id) def _real_initialize(self): - self._youtube_ie.initialize() - - def _real_extract(self, query): - mobj = re.match(self._VALID_QUERY, query) - if mobj is None: - self._downloader.trouble(u'ERROR: invalid search query "%s"' % query) - return + return - prefix, query = query.split(':') - prefix = prefix[8:] - if prefix == '': - self._download_n_results(query, 1) - return - elif prefix == 'all': - self._download_n_results(query, self._max_youtube_results) + def _real_extract(self, url): + # Extract id and simplified title from URL + mobj = re.match(self._VALID_URL, url) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid URL: %s' % url) return - else: - try: - n = long(prefix) - if n <= 0: - self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) - return - elif n > self._max_youtube_results: - self._downloader.to_stderr(u'WARNING: ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n)) - n = self._max_youtube_results - self._download_n_results(query, n) - return - except ValueError: # parsing prefix as integer fails - self._download_n_results(query, 1) - return - def _download_n_results(self, query, n): - """Downloads a specified number of results for a query""" + # At this point we have a new video + self._downloader.increment_downloads() + video_id = mobj.group(1) - video_ids = [] - already_seen = set() - pagenum = 1 + simple_title = mobj.group(2).decode('utf-8') + video_extension = 'flv' - while True: - self.report_download_page(query, pagenum) + # Retrieve video webpage to extract further information + request = urllib2.Request(url) + try: + self.report_download_webpage(video_id) + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable retrieve video webpage: %s' % str(err)) + return + + # Extract URL, uploader and title from webpage + self.report_extraction(video_id) + mobj = re.search(r'(?i)addVariable\(\"video\"\s*,\s*\"([^\"]*)\"\)', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + mediaURL = urllib.unquote(mobj.group(1)) + + # if needed add http://www.dailymotion.com/ if relative URL + + video_url = mediaURL + + # '' + mobj = re.search(r'(?im)Dailymotion\s*[\-:]\s*(.+?)', webpage) + if mobj is None: + 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'(?im)(.+?)', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract uploader nickname') + return + video_uploader = mobj.group(1) + + try: + # Process video information + self._downloader.process_info({ + 'id': video_id.decode('utf-8'), + 'url': video_url.decode('utf-8'), + 'uploader': video_uploader.decode('utf-8'), + 'upload_date': u'NA', + 'title': video_title, + 'stitle': simple_title, + 'ext': video_extension.decode('utf-8'), + 'format': u'NA', + 'player_url': None, + }) + except UnavailableVideoError: + self._downloader.trouble(u'\nERROR: unable to download video') + +class GoogleIE(InfoExtractor): + """Information extractor for video.google.com.""" + + _VALID_URL = r'(?:http://)?video\.google\.(?:com(?:\.au)?|co\.(?:uk|jp|kr|cr)|ca|de|es|fr|it|nl|pl)/videoplay\?docid=([^\&]+).*' + + def __init__(self, downloader=None): + InfoExtractor.__init__(self, downloader) + + @staticmethod + def suitable(url): + return (re.match(GoogleIE._VALID_URL, url) is not None) + + def report_download_webpage(self, video_id): + """Report webpage download.""" + self._downloader.to_screen(u'[video.google] %s: Downloading webpage' % video_id) + + def report_extraction(self, video_id): + """Report information extraction.""" + self._downloader.to_screen(u'[video.google] %s: Extracting information' % video_id) + + def _real_initialize(self): + return + + def _real_extract(self, url): + # Extract id from URL + mobj = re.match(self._VALID_URL, url) + if mobj is None: + self._downloader.trouble(u'ERROR: Invalid URL: %s' % url) + return + + # At this point we have a new video + self._downloader.increment_downloads() + video_id = mobj.group(1) + + video_extension = 'mp4' + + # Retrieve video webpage to extract further information + request = urllib2.Request('http://video.google.com/videoplay?docid=%s&hl=en&oe=utf-8' % video_id) + try: + self.report_download_webpage(video_id) + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + + # Extract URL, uploader, and title from webpage + self.report_extraction(video_id) + mobj = re.search(r"download_url:'([^']+)'", webpage) + if mobj is None: + video_extension = 'flv' + mobj = re.search(r"(?i)videoUrl\\x3d(.+?)\\x26", webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + mediaURL = urllib.unquote(mobj.group(1)) + mediaURL = mediaURL.replace('\\x3d', '\x3d') + mediaURL = mediaURL.replace('\\x26', '\x26') + + video_url = mediaURL + + mobj = re.search(r'(.*)', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract title') + return + video_title = mobj.group(1).decode('utf-8') + video_title = sanitize_title(video_title) + simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) + + # Extract video description + mobj = re.search(r'([^<]*)', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video description') + return + video_description = mobj.group(1).decode('utf-8') + if not video_description: + video_description = 'No description available.' + + # Extract video thumbnail + if self._downloader.params.get('forcethumbnail', False): + request = urllib2.Request('http://video.google.com/videosearch?q=%s+site:video.google.com&hl=en' % abs(int(video_id))) + try: + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video thumbnail') + return + video_thumbnail = mobj.group(1) + else: # we need something to pass to process_info + video_thumbnail = '' + + + try: + # Process video information + self._downloader.process_info({ + 'id': video_id.decode('utf-8'), + 'url': video_url.decode('utf-8'), + 'uploader': u'NA', + 'upload_date': u'NA', + 'title': video_title, + 'stitle': simple_title, + 'ext': video_extension.decode('utf-8'), + 'format': u'NA', + 'player_url': None, + }) + except UnavailableVideoError: + self._downloader.trouble(u'\nERROR: unable to download video') + + +class PhotobucketIE(InfoExtractor): + """Information extractor for photobucket.com.""" + + _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)' + + def __init__(self, downloader=None): + InfoExtractor.__init__(self, downloader) + + @staticmethod + def suitable(url): + return (re.match(PhotobucketIE._VALID_URL, url) is not None) + + def report_download_webpage(self, video_id): + """Report webpage download.""" + self._downloader.to_screen(u'[photobucket] %s: Downloading webpage' % video_id) + + def report_extraction(self, video_id): + """Report information extraction.""" + self._downloader.to_screen(u'[photobucket] %s: Extracting information' % video_id) + + def _real_initialize(self): + return + + def _real_extract(self, url): + # Extract id from URL + mobj = re.match(self._VALID_URL, url) + if mobj is None: + self._downloader.trouble(u'ERROR: Invalid URL: %s' % url) + return + + # At this point we have a new video + self._downloader.increment_downloads() + video_id = mobj.group(1) + + video_extension = 'flv' + + # Retrieve video webpage to extract further information + request = urllib2.Request(url) + try: + self.report_download_webpage(video_id) + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + + # Extract URL, uploader, and title from webpage + self.report_extraction(video_id) + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract media URL') + return + mediaURL = urllib.unquote(mobj.group(1)) + + video_url = mediaURL + + mobj = re.search(r'(.*) video by (.*) - Photobucket', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract title') + return + video_title = mobj.group(1).decode('utf-8') + video_title = sanitize_title(video_title) + simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) + + video_uploader = mobj.group(2).decode('utf-8') + + try: + # Process video information + self._downloader.process_info({ + 'id': video_id.decode('utf-8'), + 'url': video_url.decode('utf-8'), + 'uploader': video_uploader, + 'upload_date': u'NA', + 'title': video_title, + 'stitle': simple_title, + 'ext': video_extension.decode('utf-8'), + 'format': u'NA', + 'player_url': None, + }) + except UnavailableVideoError: + self._downloader.trouble(u'\nERROR: unable to download video') + + +class YahooIE(InfoExtractor): + """Information extractor for video.yahoo.com.""" + + # _VALID_URL matches all Yahoo! Video URLs + # _VPAGE_URL matches only the extractable '/watch/' URLs + _VALID_URL = r'(?:http://)?(?:[a-z]+\.)?video\.yahoo\.com/(?:watch|network)/([0-9]+)(?:/|\?v=)([0-9]+)(?:[#\?].*)?' + _VPAGE_URL = r'(?:http://)?video\.yahoo\.com/watch/([0-9]+)/([0-9]+)(?:[#\?].*)?' + + def __init__(self, downloader=None): + InfoExtractor.__init__(self, downloader) + + @staticmethod + def suitable(url): + return (re.match(YahooIE._VALID_URL, url) is not None) + + def report_download_webpage(self, video_id): + """Report webpage download.""" + self._downloader.to_screen(u'[video.yahoo] %s: Downloading webpage' % video_id) + + def report_extraction(self, video_id): + """Report information extraction.""" + self._downloader.to_screen(u'[video.yahoo] %s: Extracting information' % video_id) + + def _real_initialize(self): + return + + def _real_extract(self, url, new_video=True): + # Extract ID from URL + mobj = re.match(self._VALID_URL, url) + if mobj is None: + self._downloader.trouble(u'ERROR: Invalid URL: %s' % url) + return + + # At this point we have a new video + self._downloader.increment_downloads() + video_id = mobj.group(2) + video_extension = 'flv' + + # Rewrite valid but non-extractable URLs as + # extractable English language /watch/ URLs + if re.match(self._VPAGE_URL, url) is None: + request = urllib2.Request(url) + try: + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + + mobj = re.search(r'\("id", "([0-9]+)"\);', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: Unable to extract id field') + return + yahoo_id = mobj.group(1) + + mobj = re.search(r'\("vid", "([0-9]+)"\);', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: Unable to extract vid field') + return + yahoo_vid = mobj.group(1) + + url = 'http://video.yahoo.com/watch/%s/%s' % (yahoo_vid, yahoo_id) + return self._real_extract(url, new_video=False) + + # Retrieve video webpage to extract further information + request = urllib2.Request(url) + try: + self.report_download_webpage(video_id) + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + + # Extract uploader and title from webpage + self.report_extraction(video_id) + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video title') + return + video_title = mobj.group(1).decode('utf-8') + simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) + + mobj = re.search(r'

    (.*)

    ', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video uploader') + return + video_uploader = mobj.group(1).decode('utf-8') + + # Extract video thumbnail + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video thumbnail') + return + video_thumbnail = mobj.group(1).decode('utf-8') + + # Extract video description + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video description') + return + video_description = mobj.group(1).decode('utf-8') + if not video_description: video_description = 'No description available.' + + # Extract video height and width + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video height') + return + yv_video_height = mobj.group(1) + + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract video width') + return + yv_video_width = mobj.group(1) + + # Retrieve video playlist to extract media URL + # I'm not completely sure what all these options are, but we + # seem to need most of them, otherwise the server sends a 401. + yv_lg = 'R0xx6idZnW2zlrKP8xxAIR' # not sure what this represents + yv_bitrate = '700' # according to Wikipedia this is hard-coded + request = urllib2.Request('http://cosmos.bcst.yahoo.com/up/yep/process/getPlaylistFOP.php?node_id=' + video_id + + '&tech=flash&mode=playlist&lg=' + yv_lg + '&bitrate=' + yv_bitrate + '&vidH=' + yv_video_height + + '&vidW=' + yv_video_width + '&swf=as3&rd=video.yahoo.com&tk=null&adsupported=v1,v2,&eventid=1301797') + try: + self.report_download_webpage(video_id) + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve video webpage: %s' % str(err)) + return + + # Extract media URL from playlist XML + mobj = re.search(r'(.*)', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract title') + return + video_title = mobj.group(1).decode('utf-8') + video_title = sanitize_title(video_title) + simple_title = re.sub(ur'(?u)([^%s]+)' % simple_title_chars, ur'_', video_title) + + # video uploader is domain name + mobj = re.match(r'(?:https?://)?([^/]*)/.*', url) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract title') + return + video_uploader = mobj.group(1).decode('utf-8') + + try: + # Process video information + self._downloader.process_info({ + 'id': video_id.decode('utf-8'), + 'url': video_url.decode('utf-8'), + 'uploader': video_uploader, + 'upload_date': u'NA', + 'title': video_title, + 'stitle': simple_title, + 'ext': video_extension.decode('utf-8'), + 'format': u'NA', + 'player_url': None, + }) + except UnavailableVideoError, err: + self._downloader.trouble(u'\nERROR: unable to download video') + + +class YoutubeSearchIE(InfoExtractor): + """Information Extractor for YouTube search queries.""" + _VALID_QUERY = r'ytsearch(\d+|all)?:[\s\S]+' + _TEMPLATE_URL = 'http://www.youtube.com/results?search_query=%s&page=%s&gl=US&hl=en' + _VIDEO_INDICATOR = r'href="/watch\?v=.+?"' + _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*' + _youtube_ie = None + _max_youtube_results = 1000 + + def __init__(self, youtube_ie, downloader=None): + InfoExtractor.__init__(self, downloader) + self._youtube_ie = youtube_ie + + @staticmethod + def suitable(url): + return (re.match(YoutubeSearchIE._VALID_QUERY, url) is not None) + + def report_download_page(self, query, pagenum): + """Report attempt to download playlist page with given number.""" + query = query.decode(preferredencoding()) + self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) + + def _real_initialize(self): + self._youtube_ie.initialize() + + def _real_extract(self, query): + mobj = re.match(self._VALID_QUERY, query) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid search query "%s"' % query) + return + + prefix, query = query.split(':') + prefix = prefix[8:] + query = query.encode('utf-8') + if prefix == '': + self._download_n_results(query, 1) + return + elif prefix == 'all': + self._download_n_results(query, self._max_youtube_results) + return + else: + try: + n = long(prefix) + if n <= 0: + self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) + return + elif n > self._max_youtube_results: + self._downloader.to_stderr(u'WARNING: ytsearch returns max %i results (you requested %i)' % (self._max_youtube_results, n)) + n = self._max_youtube_results + self._download_n_results(query, n) + return + except ValueError: # parsing prefix as integer fails + self._download_n_results(query, 1) + return + + def _download_n_results(self, query, n): + """Downloads a specified number of results for a query""" + + video_ids = [] + already_seen = set() + pagenum = 1 + + while True: + self.report_download_page(query, pagenum) + result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) + request = urllib2.Request(result_url) + try: + page = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err)) + return + + # Extract video identifiers + for mobj in re.finditer(self._VIDEO_INDICATOR, page): + video_id = page[mobj.span()[0]:mobj.span()[1]].split('=')[2][:-1] + if video_id not in already_seen: + video_ids.append(video_id) + already_seen.add(video_id) + if len(video_ids) == n: + # Specified n videos reached + for id in video_ids: + self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) + return + + if re.search(self._MORE_PAGES_INDICATOR, page) is None: + for id in video_ids: + self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) + return + + pagenum = pagenum + 1 + +class GoogleSearchIE(InfoExtractor): + """Information Extractor for Google Video search queries.""" + _VALID_QUERY = r'gvsearch(\d+|all)?:[\s\S]+' + _TEMPLATE_URL = 'http://video.google.com/videosearch?q=%s+site:video.google.com&start=%s&hl=en' + _VIDEO_INDICATOR = r'videoplay\?docid=([^\&>]+)\&' + _MORE_PAGES_INDICATOR = r'Next' + _google_ie = None + _max_google_results = 1000 + + def __init__(self, google_ie, downloader=None): + InfoExtractor.__init__(self, downloader) + self._google_ie = google_ie + + @staticmethod + def suitable(url): + return (re.match(GoogleSearchIE._VALID_QUERY, url) is not None) + + def report_download_page(self, query, pagenum): + """Report attempt to download playlist page with given number.""" + query = query.decode(preferredencoding()) + self._downloader.to_screen(u'[video.google] query "%s": Downloading page %s' % (query, pagenum)) + + def _real_initialize(self): + self._google_ie.initialize() + + def _real_extract(self, query): + mobj = re.match(self._VALID_QUERY, query) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid search query "%s"' % query) + return + + prefix, query = query.split(':') + prefix = prefix[8:] + query = query.encode('utf-8') + if prefix == '': + self._download_n_results(query, 1) + return + elif prefix == 'all': + self._download_n_results(query, self._max_google_results) + return + else: + try: + n = long(prefix) + if n <= 0: + self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) + return + elif n > self._max_google_results: + self._downloader.to_stderr(u'WARNING: gvsearch returns max %i results (you requested %i)' % (self._max_google_results, n)) + n = self._max_google_results + self._download_n_results(query, n) + return + except ValueError: # parsing prefix as integer fails + self._download_n_results(query, 1) + return + + def _download_n_results(self, query, n): + """Downloads a specified number of results for a query""" + + video_ids = [] + already_seen = set() + pagenum = 1 + + while True: + self.report_download_page(query, pagenum) result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) - request = urllib2.Request(result_url, None, std_headers) + request = urllib2.Request(result_url) try: page = urllib2.urlopen(request).read() except (urllib2.URLError, httplib.HTTPException, socket.error), err: @@ -959,19 +1985,110 @@ class YoutubeSearchIE(InfoExtractor): # Extract video identifiers for mobj in re.finditer(self._VIDEO_INDICATOR, page): - video_id = page[mobj.span()[0]:mobj.span()[1]].split('=')[2][:-1] + video_id = mobj.group(1) if video_id not in already_seen: video_ids.append(video_id) already_seen.add(video_id) if len(video_ids) == n: # Specified n videos reached for id in video_ids: - self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) + self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id) return if re.search(self._MORE_PAGES_INDICATOR, page) is None: for id in video_ids: - self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) + self._google_ie.extract('http://video.google.com/videoplay?docid=%s' % id) + return + + pagenum = pagenum + 1 + +class YahooSearchIE(InfoExtractor): + """Information Extractor for Yahoo! Video search queries.""" + _VALID_QUERY = r'yvsearch(\d+|all)?:[\s\S]+' + _TEMPLATE_URL = 'http://video.yahoo.com/search/?p=%s&o=%s' + _VIDEO_INDICATOR = r'href="http://video\.yahoo\.com/watch/([0-9]+/[0-9]+)"' + _MORE_PAGES_INDICATOR = r'\s*Next' + _yahoo_ie = None + _max_yahoo_results = 1000 + + def __init__(self, yahoo_ie, downloader=None): + InfoExtractor.__init__(self, downloader) + self._yahoo_ie = yahoo_ie + + @staticmethod + def suitable(url): + return (re.match(YahooSearchIE._VALID_QUERY, url) is not None) + + def report_download_page(self, query, pagenum): + """Report attempt to download playlist page with given number.""" + query = query.decode(preferredencoding()) + self._downloader.to_screen(u'[video.yahoo] query "%s": Downloading page %s' % (query, pagenum)) + + def _real_initialize(self): + self._yahoo_ie.initialize() + + def _real_extract(self, query): + mobj = re.match(self._VALID_QUERY, query) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid search query "%s"' % query) + return + + prefix, query = query.split(':') + prefix = prefix[8:] + query = query.encode('utf-8') + if prefix == '': + self._download_n_results(query, 1) + return + elif prefix == 'all': + self._download_n_results(query, self._max_yahoo_results) + return + else: + try: + n = long(prefix) + if n <= 0: + self._downloader.trouble(u'ERROR: invalid download number %s for query "%s"' % (n, query)) + return + elif n > self._max_yahoo_results: + self._downloader.to_stderr(u'WARNING: yvsearch returns max %i results (you requested %i)' % (self._max_yahoo_results, n)) + n = self._max_yahoo_results + self._download_n_results(query, n) + return + except ValueError: # parsing prefix as integer fails + self._download_n_results(query, 1) + return + + def _download_n_results(self, query, n): + """Downloads a specified number of results for a query""" + + video_ids = [] + already_seen = set() + pagenum = 1 + + while True: + self.report_download_page(query, pagenum) + result_url = self._TEMPLATE_URL % (urllib.quote_plus(query), pagenum) + request = urllib2.Request(result_url) + try: + page = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err)) + return + + # Extract video identifiers + for mobj in re.finditer(self._VIDEO_INDICATOR, page): + video_id = mobj.group(1) + if video_id not in already_seen: + video_ids.append(video_id) + already_seen.add(video_id) + if len(video_ids) == n: + # Specified n videos reached + for id in video_ids: + self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id) + return + + if re.search(self._MORE_PAGES_INDICATOR, page) is None: + for id in video_ids: + self._yahoo_ie.extract('http://video.yahoo.com/watch/%s' % id) return pagenum = pagenum + 1 @@ -979,27 +2096,27 @@ class YoutubeSearchIE(InfoExtractor): class YoutubePlaylistIE(InfoExtractor): """Information Extractor for YouTube playlists.""" - _VALID_URL = r'(?:http://)?(?:\w+\.)?youtube.com/(?:view_play_list|my_playlists)\?.*?p=([^&]+).*' + _VALID_URL = r'(?:http://)?(?:\w+\.)?youtube.com/(?:(?:view_play_list|my_playlists)\?.*?p=|user/.*?/user/|p/)([^&]+).*' _TEMPLATE_URL = 'http://www.youtube.com/view_play_list?p=%s&page=%s&gl=US&hl=en' _VIDEO_INDICATOR = r'/watch\?v=(.+?)&' - _MORE_PAGES_INDICATOR = r'/view_play_list?p=%s&page=%s' + _MORE_PAGES_INDICATOR = r'(?m)>\s*Next\s*' _youtube_ie = None def __init__(self, youtube_ie, downloader=None): InfoExtractor.__init__(self, downloader) self._youtube_ie = youtube_ie - + @staticmethod def suitable(url): return (re.match(YoutubePlaylistIE._VALID_URL, url) is not None) def report_download_page(self, playlist_id, pagenum): """Report attempt to download playlist page with given number.""" - self._downloader.to_stdout(u'[youtube] PL %s: Downloading page #%s' % (playlist_id, pagenum)) + self._downloader.to_screen(u'[youtube] PL %s: Downloading page #%s' % (playlist_id, pagenum)) def _real_initialize(self): self._youtube_ie.initialize() - + def _real_extract(self, url): # Extract playlist id mobj = re.match(self._VALID_URL, url) @@ -1014,7 +2131,7 @@ class YoutubePlaylistIE(InfoExtractor): while True: self.report_download_page(playlist_id, pagenum) - request = urllib2.Request(self._TEMPLATE_URL % (playlist_id, pagenum), None, std_headers) + request = urllib2.Request(self._TEMPLATE_URL % (playlist_id, pagenum)) try: page = urllib2.urlopen(request).read() except (urllib2.URLError, httplib.HTTPException, socket.error), err: @@ -1028,14 +2145,188 @@ class YoutubePlaylistIE(InfoExtractor): ids_in_page.append(mobj.group(1)) video_ids.extend(ids_in_page) - if (self._MORE_PAGES_INDICATOR % (playlist_id.upper(), pagenum + 1)) not in page: + if re.search(self._MORE_PAGES_INDICATOR, page) is None: break pagenum = pagenum + 1 + playliststart = self._downloader.params.get('playliststart', 1) - 1 + playlistend = self._downloader.params.get('playlistend', -1) + video_ids = video_ids[playliststart:playlistend] + for id in video_ids: self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % id) return +class YoutubeUserIE(InfoExtractor): + """Information Extractor for YouTube users.""" + + _VALID_URL = r'(?:(?:(?:http://)?(?:\w+\.)?youtube.com/user/)|ytuser:)([A-Za-z0-9_-]+)' + _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' + _GDATA_PAGE_SIZE = 50 + _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' + _VIDEO_INDICATOR = r'/watch\?v=(.+?)&' + _youtube_ie = None + + def __init__(self, youtube_ie, downloader=None): + InfoExtractor.__init__(self, downloader) + self._youtube_ie = youtube_ie + + @staticmethod + def suitable(url): + return (re.match(YoutubeUserIE._VALID_URL, url) is not None) + + def report_download_page(self, username, start_index): + """Report attempt to download user page.""" + self._downloader.to_screen(u'[youtube] user %s: Downloading video ids from %d to %d' % + (username, start_index, start_index + self._GDATA_PAGE_SIZE)) + + def _real_initialize(self): + self._youtube_ie.initialize() + + def _real_extract(self, url): + # Extract username + mobj = re.match(self._VALID_URL, url) + if mobj is None: + self._downloader.trouble(u'ERROR: invalid url: %s' % url) + return + + username = mobj.group(1) + + # Download video ids using YouTube Data API. Result size per + # query is limited (currently to 50 videos) so we need to query + # page by page until there are no video ids - it means we got + # all of them. + + video_ids = [] + pagenum = 0 + + while True: + start_index = pagenum * self._GDATA_PAGE_SIZE + 1 + self.report_download_page(username, start_index) + + request = urllib2.Request(self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)) + + try: + page = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err)) + return + + # Extract video identifiers + ids_in_page = [] + + for mobj in re.finditer(self._VIDEO_INDICATOR, page): + if mobj.group(1) not in ids_in_page: + ids_in_page.append(mobj.group(1)) + + video_ids.extend(ids_in_page) + + # A little optimization - if current page is not + # "full", ie. does not contain PAGE_SIZE video ids then + # we can assume that this page is the last one - there + # are no more ids on further pages - no need to query + # again. + + if len(ids_in_page) < self._GDATA_PAGE_SIZE: + break + + pagenum += 1 + + all_ids_count = len(video_ids) + playliststart = self._downloader.params.get('playliststart', 1) - 1 + playlistend = self._downloader.params.get('playlistend', -1) + + if playlistend == -1: + video_ids = video_ids[playliststart:] + else: + video_ids = video_ids[playliststart:playlistend] + + self._downloader.to_screen("[youtube] user %s: Collected %d video ids (downloading %d of them)" % + (username, all_ids_count, len(video_ids))) + + for video_id in video_ids: + self._youtube_ie.extract('http://www.youtube.com/watch?v=%s' % video_id) + + +class DepositFilesIE(InfoExtractor): + """Information extractor for depositfiles.com""" + + _VALID_URL = r'(?:http://)?(?:\w+\.)?depositfiles.com/(?:../(?#locale))?files/(.+)' + + def __init__(self, downloader=None): + InfoExtractor.__init__(self, downloader) + + @staticmethod + def suitable(url): + return (re.match(DepositFilesIE._VALID_URL, url) is not None) + + def report_download_webpage(self, file_id): + """Report webpage download.""" + self._downloader.to_screen(u'[DepositFiles] %s: Downloading webpage' % file_id) + + def report_extraction(self, file_id): + """Report information extraction.""" + self._downloader.to_screen(u'[DepositFiles] %s: Extracting information' % file_id) + + def _real_initialize(self): + return + + def _real_extract(self, url): + # At this point we have a new file + self._downloader.increment_downloads() + + file_id = url.split('/')[-1] + # Rebuild url in english locale + url = 'http://depositfiles.com/en/files/' + file_id + + # Retrieve file webpage with 'Free download' button pressed + free_download_indication = { 'gateway_result' : '1' } + request = urllib2.Request(url, urllib.urlencode(free_download_indication)) + try: + self.report_download_webpage(file_id) + webpage = urllib2.urlopen(request).read() + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: Unable to retrieve file webpage: %s' % str(err)) + return + + # Search for the real file URL + mobj = re.search(r'
    (Attention.*?)', webpage, re.DOTALL) + if (mobj is not None) and (mobj.group(1) is not None): + restriction_message = re.sub('\s+', ' ', mobj.group(1)).strip() + self._downloader.trouble(u'ERROR: %s' % restriction_message) + else: + self._downloader.trouble(u'ERROR: unable to extract download URL from: %s' % url) + return + + file_url = mobj.group(1) + file_extension = os.path.splitext(file_url)[1][1:] + + # Search for file title + mobj = re.search(r'', webpage) + if mobj is None: + self._downloader.trouble(u'ERROR: unable to extract title') + return + file_title = mobj.group(1).decode('utf-8') + + try: + # Process file information + self._downloader.process_info({ + 'id': file_id.decode('utf-8'), + 'url': file_url.decode('utf-8'), + 'uploader': u'NA', + 'upload_date': u'NA', + 'title': file_title, + 'stitle': file_title, + 'ext': file_extension.decode('utf-8'), + 'format': u'NA', + 'player_url': None, + }) + except UnavailableVideoError, err: + self._downloader.trouble(u'ERROR: unable to download file') + class PostProcessor(object): """Post Processor class. @@ -1061,7 +2352,7 @@ class PostProcessor(object): def set_downloader(self, downloader): """Sets the downloader for this PP.""" self._downloader = downloader - + def run(self, information): """Run the PostProcessor. @@ -1081,7 +2372,7 @@ class PostProcessor(object): it was called from. """ return information # by default, do nothing - + ### MAIN PROGRAM ### if __name__ == '__main__': try: @@ -1089,15 +2380,32 @@ if __name__ == '__main__': import getpass import optparse - # General configuration - urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler())) - urllib2.install_opener(urllib2.build_opener(urllib2.HTTPCookieProcessor())) - socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words) + # Function to update the program file with the latest version from the repository. + def update_self(downloader, filename): + # Note: downloader only used for options + if not os.access(filename, os.W_OK): + sys.exit('ERROR: no write permissions on %s' % filename) + + downloader.to_screen('Updating to latest stable version...') + try: + latest_url = 'http://github.com/rg3/youtube-dl/raw/master/LATEST_VERSION' + latest_version = urllib.urlopen(latest_url).read().strip() + prog_url = 'http://github.com/rg3/youtube-dl/raw/%s/youtube-dl' % latest_version + newcontent = urllib.urlopen(prog_url).read() + except (IOError, OSError), err: + sys.exit('ERROR: unable to download latest version') + try: + stream = open(filename, 'w') + stream.write(newcontent) + stream.close() + except (IOError, OSError), err: + sys.exit('ERROR: unable to overwrite current version') + downloader.to_screen('Updated to version %s' % latest_version) # Parse command line parser = optparse.OptionParser( usage='Usage: %prog [options] url...', - version='2009.09.13', + version='2011.01.30', conflict_handler='resolve', ) @@ -1105,29 +2413,38 @@ if __name__ == '__main__': action='help', help='print this help text and exit') parser.add_option('-v', '--version', action='version', help='print program version and exit') + parser.add_option('-U', '--update', + action='store_true', dest='update_self', help='update this program to latest stable version') parser.add_option('-i', '--ignore-errors', action='store_true', dest='ignoreerrors', help='continue on download errors', default=False) parser.add_option('-r', '--rate-limit', - dest='ratelimit', metavar='L', help='download rate limit (e.g. 50k or 44.6m)') + dest='ratelimit', metavar='LIMIT', help='download rate limit (e.g. 50k or 44.6m)') + parser.add_option('-R', '--retries', + dest='retries', metavar='RETRIES', help='number of retries (default is 10)', default=10) + parser.add_option('--playlist-start', + dest='playliststart', metavar='NUMBER', help='playlist video to start at (default is 1)', default=1) + parser.add_option('--playlist-end', + dest='playlistend', metavar='NUMBER', help='playlist video to end at (default is last)', default=-1) + parser.add_option('--dump-user-agent', + action='store_true', dest='dump_user_agent', + help='display the current browser identification', default=False) authentication = optparse.OptionGroup(parser, 'Authentication Options') authentication.add_option('-u', '--username', - dest='username', metavar='UN', help='account username') + dest='username', metavar='USERNAME', help='account username') authentication.add_option('-p', '--password', - dest='password', metavar='PW', help='account password') + dest='password', metavar='PASSWORD', help='account password') authentication.add_option('-n', '--netrc', action='store_true', dest='usenetrc', help='use .netrc authentication data', default=False) parser.add_option_group(authentication) video_format = optparse.OptionGroup(parser, 'Video Format Options') video_format.add_option('-f', '--format', - action='store', dest='format', metavar='FMT', help='video format code') - video_format.add_option('-b', '--best-quality', - action='store_const', dest='format', help='download the best quality video possible', const='0') - video_format.add_option('-m', '--mobile-version', - action='store_const', dest='format', help='alias for -f 17', const='17') - video_format.add_option('-d', '--high-def', - action='store_const', dest='format', help='alias for -f 22', const='22') + action='store', dest='format', metavar='FORMAT', help='video format code') + video_format.add_option('--all-formats', + action='store_const', dest='format', help='download all available video formats', const='-1') + video_format.add_option('--max-quality', + action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download') parser.add_option_group(video_format) verbosity = optparse.OptionGroup(parser, 'Verbosity / Simulation Options') @@ -1139,6 +2456,20 @@ if __name__ == '__main__': action='store_true', dest='geturl', help='simulate, quiet but print URL', default=False) verbosity.add_option('-e', '--get-title', action='store_true', dest='gettitle', help='simulate, quiet but print title', default=False) + verbosity.add_option('--get-thumbnail', + action='store_true', dest='getthumbnail', + help='simulate, quiet but print thumbnail URL', default=False) + verbosity.add_option('--get-description', + action='store_true', dest='getdescription', + help='simulate, quiet but print video description', default=False) + verbosity.add_option('--get-filename', + action='store_true', dest='getfilename', + help='simulate, quiet but print output filename', default=False) + verbosity.add_option('--no-progress', + action='store_true', dest='noprogress', help='do not print progress bar', default=False) + verbosity.add_option('--console-title', + action='store_true', dest='consoletitle', + help='display progress in console titlebar', default=False) parser.add_option_group(verbosity) filesystem = optparse.OptionGroup(parser, 'Filesystem Options') @@ -1146,38 +2477,71 @@ if __name__ == '__main__': action='store_true', dest='usetitle', help='use title in file name', default=False) filesystem.add_option('-l', '--literal', action='store_true', dest='useliteral', help='use literal title in file name', default=False) + filesystem.add_option('-A', '--auto-number', + action='store_true', dest='autonumber', + help='number downloaded files starting from 00000', default=False) filesystem.add_option('-o', '--output', - dest='outtmpl', metavar='TPL', help='output filename template') + dest='outtmpl', metavar='TEMPLATE', help='output filename template') filesystem.add_option('-a', '--batch-file', - dest='batchfile', metavar='F', help='file containing URLs to download') + dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)') filesystem.add_option('-w', '--no-overwrites', action='store_true', dest='nooverwrites', help='do not overwrite files', default=False) filesystem.add_option('-c', '--continue', action='store_true', dest='continue_dl', help='resume partially downloaded files', default=False) + filesystem.add_option('--cookies', + dest='cookiefile', metavar='FILE', help='file to dump cookie jar to') + filesystem.add_option('--no-part', + action='store_true', dest='nopart', help='do not use .part files', default=False) + filesystem.add_option('--no-mtime', + action='store_false', dest='updatetime', + help='do not use the Last-modified header to set the file modification time', default=True) parser.add_option_group(filesystem) (opts, args) = parser.parse_args() + # Open appropriate CookieJar + if opts.cookiefile is None: + jar = cookielib.CookieJar() + else: + try: + jar = cookielib.MozillaCookieJar(opts.cookiefile) + if os.path.isfile(opts.cookiefile) and os.access(opts.cookiefile, os.R_OK): + jar.load() + except (IOError, OSError), err: + sys.exit(u'ERROR: unable to open cookie file') + + # Dump user agent + if opts.dump_user_agent: + print std_headers['User-Agent'] + sys.exit(0) + + # General configuration + cookie_processor = urllib2.HTTPCookieProcessor(jar) + urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler(), cookie_processor, YoutubeDLHandler())) + socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words) + # Batch file verification batchurls = [] if opts.batchfile is not None: try: - batchurls = open(opts.batchfile, 'r').readlines() + if opts.batchfile == '-': + batchfd = sys.stdin + else: + batchfd = open(opts.batchfile, 'r') + batchurls = batchfd.readlines() batchurls = [x.strip() for x in batchurls] - batchurls = [x for x in batchurls if len(x) > 0] + batchurls = [x for x in batchurls if len(x) > 0 and not re.search(r'^[#/;]', x)] except IOError: sys.exit(u'ERROR: batch file could not be read') all_urls = batchurls + args # Conflicting, missing and erroneous options - if len(all_urls) < 1: - parser.error(u'you must provide at least one URL') if opts.usenetrc and (opts.username is not None or opts.password is not None): parser.error(u'using .netrc conflicts with giving username/password') if opts.password is not None and opts.username is None: parser.error(u'account username missing') - if opts.outtmpl is not None and (opts.useliteral or opts.usetitle): - parser.error(u'using output template conflicts with using title or literal title') + if opts.outtmpl is not None and (opts.useliteral or opts.usetitle or opts.autonumber): + parser.error(u'using output template conflicts with using title, literal title or auto number') if opts.usetitle and opts.useliteral: parser.error(u'using title conflicts with using literal title') if opts.username is not None and opts.password is None: @@ -1187,37 +2551,112 @@ if __name__ == '__main__': if numeric_limit is None: parser.error(u'invalid rate limit specified') opts.ratelimit = numeric_limit + if opts.retries is not None: + try: + opts.retries = long(opts.retries) + except (TypeError, ValueError), err: + parser.error(u'invalid retry count specified') + try: + opts.playliststart = long(opts.playliststart) + if opts.playliststart <= 0: + raise ValueError + except (TypeError, ValueError), err: + parser.error(u'invalid playlist start number specified') + try: + opts.playlistend = long(opts.playlistend) + if opts.playlistend != -1 and (opts.playlistend <= 0 or opts.playlistend < opts.playliststart): + raise ValueError + except (TypeError, ValueError), err: + parser.error(u'invalid playlist end number specified') # Information extractors youtube_ie = YoutubeIE() metacafe_ie = MetacafeIE(youtube_ie) + dailymotion_ie = DailymotionIE() youtube_pl_ie = YoutubePlaylistIE(youtube_ie) + youtube_user_ie = YoutubeUserIE(youtube_ie) youtube_search_ie = YoutubeSearchIE(youtube_ie) + google_ie = GoogleIE() + google_search_ie = GoogleSearchIE(google_ie) + photobucket_ie = PhotobucketIE() + yahoo_ie = YahooIE() + yahoo_search_ie = YahooSearchIE(yahoo_ie) + deposit_files_ie = DepositFilesIE() + generic_ie = GenericIE() # File downloader fd = FileDownloader({ 'usenetrc': opts.usenetrc, 'username': opts.username, 'password': opts.password, - 'quiet': (opts.quiet or opts.geturl or opts.gettitle), + 'quiet': (opts.quiet or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription or opts.getfilename), 'forceurl': opts.geturl, 'forcetitle': opts.gettitle, - 'simulate': (opts.simulate or opts.geturl or opts.gettitle), + 'forcethumbnail': opts.getthumbnail, + 'forcedescription': opts.getdescription, + 'forcefilename': opts.getfilename, + 'simulate': (opts.simulate or opts.geturl or opts.gettitle or opts.getthumbnail or opts.getdescription or opts.getfilename), 'format': opts.format, + 'format_limit': opts.format_limit, 'outtmpl': ((opts.outtmpl is not None and opts.outtmpl.decode(preferredencoding())) + or (opts.format == '-1' and opts.usetitle and u'%(stitle)s-%(id)s-%(format)s.%(ext)s') + or (opts.format == '-1' and opts.useliteral and u'%(title)s-%(id)s-%(format)s.%(ext)s') + or (opts.format == '-1' and u'%(id)s-%(format)s.%(ext)s') + or (opts.usetitle and opts.autonumber and u'%(autonumber)s-%(stitle)s-%(id)s.%(ext)s') + or (opts.useliteral and opts.autonumber and u'%(autonumber)s-%(title)s-%(id)s.%(ext)s') or (opts.usetitle and u'%(stitle)s-%(id)s.%(ext)s') or (opts.useliteral and u'%(title)s-%(id)s.%(ext)s') + or (opts.autonumber and u'%(autonumber)s-%(id)s.%(ext)s') or u'%(id)s.%(ext)s'), 'ignoreerrors': opts.ignoreerrors, 'ratelimit': opts.ratelimit, 'nooverwrites': opts.nooverwrites, + 'retries': opts.retries, 'continuedl': opts.continue_dl, + 'noprogress': opts.noprogress, + 'playliststart': opts.playliststart, + 'playlistend': opts.playlistend, + 'logtostderr': opts.outtmpl == '-', + 'consoletitle': opts.consoletitle, + 'nopart': opts.nopart, + 'updatetime': opts.updatetime, }) fd.add_info_extractor(youtube_search_ie) fd.add_info_extractor(youtube_pl_ie) + fd.add_info_extractor(youtube_user_ie) fd.add_info_extractor(metacafe_ie) + fd.add_info_extractor(dailymotion_ie) fd.add_info_extractor(youtube_ie) + fd.add_info_extractor(google_ie) + fd.add_info_extractor(google_search_ie) + fd.add_info_extractor(photobucket_ie) + fd.add_info_extractor(yahoo_ie) + fd.add_info_extractor(yahoo_search_ie) + fd.add_info_extractor(deposit_files_ie) + + # This must come last since it's the + # fallback if none of the others work + fd.add_info_extractor(generic_ie) + + # Update version + if opts.update_self: + update_self(fd, sys.argv[0]) + + # Maybe do nothing + if len(all_urls) < 1: + if not opts.update_self: + parser.error(u'you must provide at least one URL') + else: + sys.exit() retcode = fd.download(all_urls) + + # Dump cookie jar if requested + if opts.cookiefile is not None: + try: + jar.save() + except (IOError, OSError), err: + sys.exit(u'ERROR: unable to save cookie jar') + sys.exit(retcode) except DownloadError: