diff options
Diffstat (limited to 'youtube_dl')
78 files changed, 3791 insertions, 1432 deletions
diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 00af78e06..cad6b026e 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -21,24 +21,24 @@ import subprocess import socket import sys import time +import tokenize import traceback if os.name == 'nt': import ctypes from .compat import ( - compat_basestring, compat_cookiejar, compat_expanduser, compat_get_terminal_size, compat_http_client, compat_kwargs, compat_str, + compat_tokenize_tokenize, compat_urllib_error, compat_urllib_request, ) from .utils import ( - escape_url, ContentTooShortError, date_from_str, DateRange, @@ -49,7 +49,6 @@ from .utils import ( ExtractorError, format_bytes, formatSeconds, - HEADRequest, locked_file, make_HTTPS_handler, MaxDownloadsReached, @@ -853,8 +852,8 @@ class YoutubeDL(object): else: raise Exception('Invalid result type: %s' % result_type) - def _apply_format_filter(self, format_spec, available_formats): - " Returns a tuple of the remaining format_spec and filtered formats " + def _build_format_filter(self, filter_spec): + " Returns a function to filter the formats according to the filter_spec " OPERATORS = { '<': operator.lt, @@ -864,13 +863,13 @@ class YoutubeDL(object): '=': operator.eq, '!=': operator.ne, } - operator_rex = re.compile(r'''(?x)\s*\[ + operator_rex = re.compile(r'''(?x)\s* (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps) \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) - \]$ + $ ''' % '|'.join(map(re.escape, OPERATORS.keys()))) - m = operator_rex.search(format_spec) + m = operator_rex.search(filter_spec) if m: try: comparison_value = int(m.group('value')) @@ -881,7 +880,7 @@ class YoutubeDL(object): if comparison_value is None: raise ValueError( 'Invalid value %r in format specification %r' % ( - m.group('value'), format_spec)) + m.group('value'), filter_spec)) op = OPERATORS[m.group('op')] if not m: @@ -889,85 +888,283 @@ class YoutubeDL(object): '=': operator.eq, '!=': operator.ne, } - str_operator_rex = re.compile(r'''(?x)\s*\[ + str_operator_rex = re.compile(r'''(?x) \s*(?P<key>ext|acodec|vcodec|container|protocol) \s*(?P<op>%s)(?P<none_inclusive>\s*\?)? \s*(?P<value>[a-zA-Z0-9_-]+) - \s*\]$ + \s*$ ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) - m = str_operator_rex.search(format_spec) + m = str_operator_rex.search(filter_spec) if m: comparison_value = m.group('value') op = STR_OPERATORS[m.group('op')] if not m: - raise ValueError('Invalid format specification %r' % format_spec) + raise ValueError('Invalid filter specification %r' % filter_spec) def _filter(f): actual_value = f.get(m.group('key')) if actual_value is None: return m.group('none_inclusive') return op(actual_value, comparison_value) - new_formats = [f for f in available_formats if _filter(f)] + return _filter + + def build_format_selector(self, format_spec): + def syntax_error(note, start): + message = ( + 'Invalid format specification: ' + '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) + return SyntaxError(message) + + PICKFIRST = 'PICKFIRST' + MERGE = 'MERGE' + SINGLE = 'SINGLE' + GROUP = 'GROUP' + FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) + + def _parse_filter(tokens): + filter_parts = [] + for type, string, start, _, _ in tokens: + if type == tokenize.OP and string == ']': + return ''.join(filter_parts) + else: + filter_parts.append(string) + + def _remove_unused_ops(tokens): + # Remove operators that we don't use and join them with the sourrounding strings + # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' + ALLOWED_OPS = ('/', '+', ',', '(', ')') + last_string, last_start, last_end, last_line = None, None, None, None + for type, string, start, end, line in tokens: + if type == tokenize.OP and string == '[': + if last_string: + yield tokenize.NAME, last_string, last_start, last_end, last_line + last_string = None + yield type, string, start, end, line + # everything inside brackets will be handled by _parse_filter + for type, string, start, end, line in tokens: + yield type, string, start, end, line + if type == tokenize.OP and string == ']': + break + elif type == tokenize.OP and string in ALLOWED_OPS: + if last_string: + yield tokenize.NAME, last_string, last_start, last_end, last_line + last_string = None + yield type, string, start, end, line + elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: + if not last_string: + last_string = string + last_start = start + last_end = end + else: + last_string += string + if last_string: + yield tokenize.NAME, last_string, last_start, last_end, last_line + + def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): + selectors = [] + current_selector = None + for type, string, start, _, _ in tokens: + # ENCODING is only defined in python 3.x + if type == getattr(tokenize, 'ENCODING', None): + continue + elif type in [tokenize.NAME, tokenize.NUMBER]: + current_selector = FormatSelector(SINGLE, string, []) + elif type == tokenize.OP: + if string == ')': + if not inside_group: + # ')' will be handled by the parentheses group + tokens.restore_last_token() + break + elif inside_merge and string in ['/', ',']: + tokens.restore_last_token() + break + elif inside_choice and string == ',': + tokens.restore_last_token() + break + elif string == ',': + if not current_selector: + raise syntax_error('"," must follow a format selector', start) + selectors.append(current_selector) + current_selector = None + elif string == '/': + if not current_selector: + raise syntax_error('"/" must follow a format selector', start) + first_choice = current_selector + second_choice = _parse_format_selection(tokens, inside_choice=True) + current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) + elif string == '[': + if not current_selector: + current_selector = FormatSelector(SINGLE, 'best', []) + format_filter = _parse_filter(tokens) + current_selector.filters.append(format_filter) + elif string == '(': + if current_selector: + raise syntax_error('Unexpected "("', start) + group = _parse_format_selection(tokens, inside_group=True) + current_selector = FormatSelector(GROUP, group, []) + elif string == '+': + video_selector = current_selector + audio_selector = _parse_format_selection(tokens, inside_merge=True) + if not video_selector or not audio_selector: + raise syntax_error('"+" must be between two format selectors', start) + current_selector = FormatSelector(MERGE, (video_selector, audio_selector), []) + else: + raise syntax_error('Operator not recognized: "{0}"'.format(string), start) + elif type == tokenize.ENDMARKER: + break + if current_selector: + selectors.append(current_selector) + return selectors + + def _build_selector_function(selector): + if isinstance(selector, list): + fs = [_build_selector_function(s) for s in selector] + + def selector_function(formats): + for f in fs: + for format in f(formats): + yield format + return selector_function + elif selector.type == GROUP: + selector_function = _build_selector_function(selector.selector) + elif selector.type == PICKFIRST: + fs = [_build_selector_function(s) for s in selector.selector] + + def selector_function(formats): + for f in fs: + picked_formats = list(f(formats)) + if picked_formats: + return picked_formats + return [] + elif selector.type == SINGLE: + format_spec = selector.selector + + def selector_function(formats): + formats = list(formats) + if not formats: + return + if format_spec == 'all': + for f in formats: + yield f + elif format_spec in ['best', 'worst', None]: + format_idx = 0 if format_spec == 'worst' else -1 + audiovideo_formats = [ + f for f in formats + if f.get('vcodec') != 'none' and f.get('acodec') != 'none'] + if audiovideo_formats: + yield audiovideo_formats[format_idx] + # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format + elif (all(f.get('acodec') != 'none' for f in formats) or + all(f.get('vcodec') != 'none' for f in formats)): + yield formats[format_idx] + elif format_spec == 'bestaudio': + audio_formats = [ + f for f in formats + if f.get('vcodec') == 'none'] + if audio_formats: + yield audio_formats[-1] + elif format_spec == 'worstaudio': + audio_formats = [ + f for f in formats + if f.get('vcodec') == 'none'] + if audio_formats: + yield audio_formats[0] + elif format_spec == 'bestvideo': + video_formats = [ + f for f in formats + if f.get('acodec') == 'none'] + if video_formats: + yield video_formats[-1] + elif format_spec == 'worstvideo': + video_formats = [ + f for f in formats + if f.get('acodec') == 'none'] + if video_formats: + yield video_formats[0] + else: + extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] + if format_spec in extensions: + filter_f = lambda f: f['ext'] == format_spec + else: + filter_f = lambda f: f['format_id'] == format_spec + matches = list(filter(filter_f, formats)) + if matches: + yield matches[-1] + elif selector.type == MERGE: + def _merge(formats_info): + format_1, format_2 = [f['format_id'] for f in formats_info] + # The first format must contain the video and the + # second the audio + if formats_info[0].get('vcodec') == 'none': + self.report_error('The first format must ' + 'contain the video, try using ' + '"-f %s+%s"' % (format_2, format_1)) + return + output_ext = ( + formats_info[0]['ext'] + if self.params.get('merge_output_format') is None + else self.params['merge_output_format']) + return { + 'requested_formats': formats_info, + 'format': '%s+%s' % (formats_info[0].get('format'), + formats_info[1].get('format')), + 'format_id': '%s+%s' % (formats_info[0].get('format_id'), + formats_info[1].get('format_id')), + 'width': formats_info[0].get('width'), + 'height': formats_info[0].get('height'), + 'resolution': formats_info[0].get('resolution'), + 'fps': formats_info[0].get('fps'), + 'vcodec': formats_info[0].get('vcodec'), + 'vbr': formats_info[0].get('vbr'), + 'stretched_ratio': formats_info[0].get('stretched_ratio'), + 'acodec': formats_info[1].get('acodec'), + 'abr': formats_info[1].get('abr'), + 'ext': output_ext, + } + video_selector, audio_selector = map(_build_selector_function, selector.selector) - new_format_spec = format_spec[:-len(m.group(0))] - if not new_format_spec: - new_format_spec = 'best' + def selector_function(formats): + formats = list(formats) + for pair in itertools.product(video_selector(formats), audio_selector(formats)): + yield _merge(pair) - return (new_format_spec, new_formats) + filters = [self._build_format_filter(f) for f in selector.filters] - def select_format(self, format_spec, available_formats): - while format_spec.endswith(']'): - format_spec, available_formats = self._apply_format_filter( - format_spec, available_formats) - if not available_formats: - return None + def final_selector(formats): + for _filter in filters: + formats = list(filter(_filter, formats)) + return selector_function(formats) + return final_selector - if format_spec in ['best', 'worst', None]: - format_idx = 0 if format_spec == 'worst' else -1 - audiovideo_formats = [ - f for f in available_formats - if f.get('vcodec') != 'none' and f.get('acodec') != 'none'] - if audiovideo_formats: - return audiovideo_formats[format_idx] - # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format - elif (all(f.get('acodec') != 'none' for f in available_formats) or - all(f.get('vcodec') != 'none' for f in available_formats)): - return available_formats[format_idx] - elif format_spec == 'bestaudio': - audio_formats = [ - f for f in available_formats - if f.get('vcodec') == 'none'] - if audio_formats: - return audio_formats[-1] - elif format_spec == 'worstaudio': - audio_formats = [ - f for f in available_formats - if f.get('vcodec') == 'none'] - if audio_formats: - return audio_formats[0] - elif format_spec == 'bestvideo': - video_formats = [ - f for f in available_formats - if f.get('acodec') == 'none'] - if video_formats: - return video_formats[-1] - elif format_spec == 'worstvideo': - video_formats = [ - f for f in available_formats - if f.get('acodec') == 'none'] - if video_formats: - return video_formats[0] - else: - extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] - if format_spec in extensions: - filter_f = lambda f: f['ext'] == format_spec - else: - filter_f = lambda f: f['format_id'] == format_spec - matches = list(filter(filter_f, available_formats)) - if matches: - return matches[-1] - return None + stream = io.BytesIO(format_spec.encode('utf-8')) + try: + tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) + except tokenize.TokenError: + raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) + + class TokenIterator(object): + def __init__(self, tokens): + self.tokens = tokens + self.counter = 0 + + def __iter__(self): + return self + + def __next__(self): + if self.counter >= len(self.tokens): + raise StopIteration() + value = self.tokens[self.counter] + self.counter += 1 + return value + + next = __next__ + + def restore_last_token(self): + self.counter -= 1 + + parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) + return _build_selector_function(parsed_selector) def _calc_headers(self, info_dict): res = std_headers.copy() @@ -1104,62 +1301,15 @@ class YoutubeDL(object): if req_format is None: req_format_list = [] if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and - info_dict['extractor'] in ['youtube', 'ted']): + info_dict['extractor'] in ['youtube', 'ted'] and + not info_dict.get('is_live')): merger = FFmpegMergerPP(self) if merger.available and merger.can_merge(): req_format_list.append('bestvideo+bestaudio') req_format_list.append('best') req_format = '/'.join(req_format_list) - formats_to_download = [] - if req_format == 'all': - formats_to_download = formats - else: - for rfstr in req_format.split(','): - # We can accept formats requested in the format: 34/5/best, we pick - # the first that is available, starting from left - req_formats = rfstr.split('/') - for rf in req_formats: - if re.match(r'.+?\+.+?', rf) is not None: - # Two formats have been requested like '137+139' - format_1, format_2 = rf.split('+') - formats_info = (self.select_format(format_1, formats), - self.select_format(format_2, formats)) - if all(formats_info): - # The first format must contain the video and the - # second the audio - if formats_info[0].get('vcodec') == 'none': - self.report_error('The first format must ' - 'contain the video, try using ' - '"-f %s+%s"' % (format_2, format_1)) - return - output_ext = ( - formats_info[0]['ext'] - if self.params.get('merge_output_format') is None - else self.params['merge_output_format']) - selected_format = { - 'requested_formats': formats_info, - 'format': '%s+%s' % (formats_info[0].get('format'), - formats_info[1].get('format')), - 'format_id': '%s+%s' % (formats_info[0].get('format_id'), - formats_info[1].get('format_id')), - 'width': formats_info[0].get('width'), - 'height': formats_info[0].get('height'), - 'resolution': formats_info[0].get('resolution'), - 'fps': formats_info[0].get('fps'), - 'vcodec': formats_info[0].get('vcodec'), - 'vbr': formats_info[0].get('vbr'), - 'stretched_ratio': formats_info[0].get('stretched_ratio'), - 'acodec': formats_info[1].get('acodec'), - 'abr': formats_info[1].get('abr'), - 'ext': output_ext, - } - else: - selected_format = None - else: - selected_format = self.select_format(rf, formats) - if selected_format is not None: - formats_to_download.append(selected_format) - break + format_selector = self.build_format_selector(req_format) + formats_to_download = list(format_selector(formats)) if not formats_to_download: raise ExtractorError('requested format not available', expected=True) @@ -1707,27 +1857,6 @@ class YoutubeDL(object): def urlopen(self, req): """ Start an HTTP download """ - - # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not - # always respected by websites, some tend to give out URLs with non percent-encoded - # non-ASCII characters (see telemb.py, ard.py [#3412]) - # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) - # To work around aforementioned issue we will replace request's original URL with - # percent-encoded one - req_is_string = isinstance(req, compat_basestring) - url = req if req_is_string else req.get_full_url() - url_escaped = escape_url(url) - - # Substitute URL if any change after escaping - if url != url_escaped: - if req_is_string: - req = url_escaped - else: - req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request - req = req_type( - url_escaped, data=req.data, headers=req.headers, - origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) - return self._opener.open(req, timeout=self._socket_timeout) def print_debug_header(self): diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py index 0c57c7aeb..ace5bd716 100644 --- a/youtube_dl/compat.py +++ b/youtube_dl/compat.py @@ -43,6 +43,11 @@ except ImportError: # Python 2 import cookielib as compat_cookiejar try: + import http.cookies as compat_cookies +except ImportError: # Python 2 + import Cookie as compat_cookies + +try: import html.entities as compat_html_entities except ImportError: # Python 2 import htmlentitydefs as compat_html_entities @@ -431,11 +436,17 @@ except TypeError: # Python 2.6 yield n n += step +if sys.version_info >= (3, 0): + from tokenize import tokenize as compat_tokenize_tokenize +else: + from tokenize import generate_tokens as compat_tokenize_tokenize + __all__ = [ 'compat_HTTPError', 'compat_basestring', 'compat_chr', 'compat_cookiejar', + 'compat_cookies', 'compat_expanduser', 'compat_get_terminal_size', 'compat_getenv', @@ -451,6 +462,7 @@ __all__ = [ 'compat_socket_create_connection', 'compat_str', 'compat_subprocess_get_DEVNULL', + 'compat_tokenize_tokenize', 'compat_urllib_error', 'compat_urllib_parse', 'compat_urllib_parse_unquote', diff --git a/youtube_dl/downloader/__init__.py b/youtube_dl/downloader/__init__.py index f110830c4..dccc59212 100644 --- a/youtube_dl/downloader/__init__.py +++ b/youtube_dl/downloader/__init__.py @@ -8,6 +8,7 @@ from .hls import NativeHlsFD from .http import HttpFD from .rtsp import RtspFD from .rtmp import RtmpFD +from .dash import DashSegmentsFD from ..utils import ( determine_protocol, @@ -20,6 +21,7 @@ PROTOCOL_MAP = { 'mms': RtspFD, 'rtsp': RtspFD, 'f4m': F4mFD, + 'http_dash_segments': DashSegmentsFD, } diff --git a/youtube_dl/downloader/dash.py b/youtube_dl/downloader/dash.py new file mode 100644 index 000000000..8b6fa2753 --- /dev/null +++ b/youtube_dl/downloader/dash.py @@ -0,0 +1,66 @@ +from __future__ import unicode_literals + +import re + +from .common import FileDownloader +from ..compat import compat_urllib_request + + +class DashSegmentsFD(FileDownloader): + """ + Download segments in a DASH manifest + """ + def real_download(self, filename, info_dict): + self.report_destination(filename) + tmpfilename = self.temp_name(filename) + base_url = info_dict['url'] + segment_urls = info_dict['segment_urls'] + + is_test = self.params.get('test', False) + remaining_bytes = self._TEST_FILE_SIZE if is_test else None + byte_counter = 0 + + def append_url_to_file(outf, target_url, target_name, remaining_bytes=None): + self.to_screen('[DashSegments] %s: Downloading %s' % (info_dict['id'], target_name)) + req = compat_urllib_request.Request(target_url) + if remaining_bytes is not None: + req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1)) + + data = self.ydl.urlopen(req).read() + + if remaining_bytes is not None: + data = data[:remaining_bytes] + + outf.write(data) + return len(data) + + def combine_url(base_url, target_url): + if re.match(r'^https?://', target_url): + return target_url + return '%s%s%s' % (base_url, '' if base_url.endswith('/') else '/', target_url) + + with open(tmpfilename, 'wb') as outf: + append_url_to_file( + outf, combine_url(base_url, info_dict['initialization_url']), + 'initialization segment') + for i, segment_url in enumerate(segment_urls): + segment_len = append_url_to_file( + outf, combine_url(base_url, segment_url), + 'segment %d / %d' % (i + 1, len(segment_urls)), + remaining_bytes) + byte_counter += segment_len + if remaining_bytes is not None: + remaining_bytes -= segment_len + if remaining_bytes <= 0: + break + + self.try_rename(tmpfilename, filename) + + self._hook_progress({ + 'downloaded_bytes': byte_counter, + 'total_bytes': byte_counter, + 'filename': filename, + 'status': 'finished', + }) + + return True diff --git a/youtube_dl/downloader/external.py b/youtube_dl/downloader/external.py index 1d5cc9904..6c310346c 100644 --- a/youtube_dl/downloader/external.py +++ b/youtube_dl/downloader/external.py @@ -45,11 +45,13 @@ class ExternalFD(FileDownloader): def supports(cls, info_dict): return info_dict['protocol'] in ('http', 'https', 'ftp', 'ftps') - def _source_address(self, command_option): - source_address = self.params.get('source_address') - if source_address is None: + def _option(self, command_option, param): + param = self.params.get(param) + if param is None: return [] - return [command_option, source_address] + if isinstance(param, bool): + return [command_option] + return [command_option, param] def _configuration_args(self, default=[]): ex_args = self.params.get('external_downloader_args') @@ -77,7 +79,17 @@ class CurlFD(ExternalFD): cmd = [self.exe, '--location', '-o', tmpfilename] for key, val in info_dict['http_headers'].items(): cmd += ['--header', '%s: %s' % (key, val)] - cmd += self._source_address('--interface') + cmd += self._option('--interface', 'source_address') + cmd += self._configuration_args() + cmd += ['--', info_dict['url']] + return cmd + + +class AxelFD(ExternalFD): + def _make_cmd(self, tmpfilename, info_dict): + cmd = [self.exe, '-o', tmpfilename] + for key, val in info_dict['http_headers'].items(): + cmd += ['-H', '%s: %s' % (key, val)] cmd += self._configuration_args() cmd += ['--', info_dict['url']] return cmd @@ -88,7 +100,9 @@ class WgetFD(ExternalFD): cmd = [self.exe, '-O', tmpfilename, '-nv', '--no-cookies'] for key, val in info_dict['http_headers'].items(): cmd += ['--header', '%s: %s' % (key, val)] - cmd += self._source_address('--bind-address') + cmd += self._option('--bind-address', 'source_address') + cmd += self._option('--proxy', 'proxy') + cmd += self._option('--no-check-certificate', 'nocheckcertificate') cmd += self._configuration_args() cmd += ['--', info_dict['url']] return cmd @@ -105,7 +119,8 @@ class Aria2cFD(ExternalFD): cmd += ['--out', os.path.basename(tmpfilename)] for key, val in info_dict['http_headers'].items(): cmd += ['--header', '%s: %s' % (key, val)] - cmd += self._source_address('--interface') + cmd += self._option('--interface', 'source_address') + cmd += self._option('--all-proxy', 'proxy') cmd += ['--', info_dict['url']] return cmd diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py index b1a858c45..275564b59 100644 --- a/youtube_dl/downloader/f4m.py +++ b/youtube_dl/downloader/f4m.py @@ -7,8 +7,7 @@ import os import time import xml.etree.ElementTree as etree -from .common import FileDownloader -from .http import HttpFD +from .fragment import FragmentFD from ..compat import ( compat_urlparse, compat_urllib_error, @@ -16,8 +15,6 @@ from ..compat import ( from ..utils import ( struct_pack, struct_unpack, - encodeFilename, - sanitize_open, xpath_text, ) @@ -226,16 +223,13 @@ def _add_ns(prop): return '{http://ns.adobe.com/f4m/1.0}%s' % prop -class HttpQuietDownloader(HttpFD): - def to_screen(self, *args, **kargs): - pass - - -class F4mFD(FileDownloader): +class F4mFD(FragmentFD): """ A downloader for f4m manifests or AdobeHDS. """ + FD_NAME = 'f4m' + def _get_unencrypted_media(self, doc): media = doc.findall(_add_ns('media')) if not media: @@ -288,7 +282,7 @@ class F4mFD(FileDownloader): def real_download(self, filename, info_dict): man_url = info_dict['url'] requested_bitrate = info_dict.get('tbr') - self.to_screen('[download] Downloading f4m manifest') + self.to_screen('[%s] Downloading f4m manifest' % self.FD_NAME) manifest = self.ydl.urlopen(man_url).read() doc = etree.fromstring(manifest) @@ -320,67 +314,20 @@ class F4mFD(FileDownloader): # For some akamai manifests we'll need to add a query to the fragment url akamai_pv = xpath_text(doc, _add_ns('pv-2.0')) - self.report_destination(filename) - http_dl = HttpQuietDownloader( - self.ydl, - { - 'continuedl': True, - 'quiet': True, - 'noprogress': True, - 'ratelimit': self.params.get('ratelimit', None), - 'test': self.params.get('test', False), - } - ) - tmpfilename = self.temp_name(filename) - (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb') + ctx = { + 'filename': filename, + 'total_frags': total_frags, + } + + self._prepare_frag_download(ctx) + + dest_stream = ctx['dest_stream'] write_flv_header(dest_stream) if not live: write_metadata_tag(dest_stream, metadata) - # This dict stores the download progress, it's updated by the progress - # hook - state = { - 'status': 'downloading', - 'downloaded_bytes': 0, - 'frag_index': 0, - 'frag_count': total_frags, - 'filename': filename, - 'tmpfilename': tmpfilename, - } - start = time.time() - - def frag_progress_hook(s): - if s['status'] not in ('downloading', 'finished'): - return - - frag_total_bytes = s.get('total_bytes', 0) - if s['status'] == 'finished': - state['downloaded_bytes'] += frag_total_bytes - state['frag_index'] += 1 - - estimated_size = ( - (state['downloaded_bytes'] + frag_total_bytes) / - (state['frag_index'] + 1) * total_frags) - time_now = time.time() - state['total_bytes_estimate'] = estimated_size - state['elapsed'] = time_now - start - - if s['status'] == 'finished': - progress = self.calc_percent(state['frag_index'], total_frags) - else: - frag_downloaded_bytes = s['downloaded_bytes'] - frag_progress = self.calc_percent(frag_downloaded_bytes, - frag_total_bytes) - progress = self.calc_percent(state['frag_index'], total_frags) - progress += frag_progress / float(total_frags) - - state['eta'] = self.calc_eta( - start, time_now, estimated_size, state['downloaded_bytes'] + frag_downloaded_bytes) - state['speed'] = s.get('speed') - self._hook_progress(state) - - http_dl.add_progress_hook(frag_progress_hook) + self._start_frag_download(ctx) frags_filenames = [] while fragments_list: @@ -391,9 +338,9 @@ class F4mFD(FileDownloader): url += '?' + akamai_pv.strip(';') if info_dict.get('extra_param_to_segment_url'): url += info_dict.get('extra_param_to_segment_url') - frag_filename = '%s-%s' % (tmpfilename, name) + frag_filename = '%s-%s' % (ctx['tmpfilename'], name) try: - success = http_dl.download(frag_filename, {'url': url}) + success = ctx['dl'].download(frag_filename, {'url': url}) if not success: return False with open(frag_filename, 'rb') as down: @@ -425,20 +372,9 @@ class F4mFD(FileDownloader): msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1)) self.report_warning(msg) - dest_stream.close() + self._finish_frag_download(ctx) - elapsed = time.time() - start - self.try_rename(tmpfilename, filename) for frag_file in frags_filenames: os.remove(frag_file) - fsize = os.path.getsize(encodeFilename(filename)) - self._hook_progress({ - 'downloaded_bytes': fsize, - 'total_bytes': fsize, - 'filename': filename, - 'status': 'finished', - 'elapsed': elapsed, - }) - return True diff --git a/youtube_dl/downloader/fragment.py b/youtube_dl/downloader/fragment.py new file mode 100644 index 000000000..5a64b29ee --- /dev/null +++ b/youtube_dl/downloader/fragment.py @@ -0,0 +1,111 @@ +from __future__ import division, unicode_literals + +import os +import time + +from .common import FileDownloader +from .http import HttpFD +from ..utils import ( + encodeFilename, + sanitize_open, +) + + +class HttpQuietDownloader(HttpFD): + def to_screen(self, *args, **kargs): + pass + + +class FragmentFD(FileDownloader): + """ + A base file downloader class for fragmented media (e.g. f4m/m3u8 manifests). + """ + + def _prepare_and_start_frag_download(self, ctx): + self._prepare_frag_download(ctx) + self._start_frag_download(ctx) + + def _prepare_frag_download(self, ctx): + self.to_screen('[%s] Total fragments: %d' % (self.FD_NAME, ctx['total_frags'])) + self.report_destination(ctx['filename']) + dl = HttpQuietDownloader( + self.ydl, + { + 'continuedl': True, + 'quiet': True, + 'noprogress': True, + 'ratelimit': self.params.get('ratelimit', None), + 'retries': self.params.get('retries', 0), + 'test': self.params.get('test', False), + } + ) + tmpfilename = self.temp_name(ctx['filename']) + dest_stream, tmpfilename = sanitize_open(tmpfilename, 'wb') + ctx.update({ + 'dl': dl, + 'dest_stream': dest_stream, + 'tmpfilename': tmpfilename, + }) + + def _start_frag_download(self, ctx): + total_frags = ctx['total_frags'] + # This dict stores the download progress, it's updated by the progress + # hook + state = { + 'status': 'downloading', + 'downloaded_bytes': 0, + 'frag_index': 0, + 'frag_count': total_frags, + 'filename': ctx['filename'], + 'tmpfilename': ctx['tmpfilename'], + } + start = time.time() + ctx['started'] = start + + def frag_progress_hook(s): + if s['status'] not in ('downloading', 'finished'): + return + + frag_total_bytes = s.get('total_bytes', 0) + if s['status'] == 'finished': + state['downloaded_bytes'] += frag_total_bytes + state['frag_index'] += 1 + + estimated_size = ( + (state['downloaded_bytes'] + frag_total_bytes) / + (state['frag_index'] + 1) * total_frags) + time_now = time.time() + state['total_bytes_estimate'] = estimated_size + state['elapsed'] = time_now - start + + if s['status'] == 'finished': + progress = self.calc_percent(state['frag_index'], total_frags) + else: + frag_downloaded_bytes = s['downloaded_bytes'] + frag_progress = self.calc_percent(frag_downloaded_bytes, + frag_total_bytes) + progress = self.calc_percent(state['frag_index'], total_frags) + progress += frag_progress / float(total_frags) + + state['eta'] = self.calc_eta( + start, time_now, estimated_size, state['downloaded_bytes'] + frag_downloaded_bytes) + state['speed'] = s.get('speed') + self._hook_progress(state) + + ctx['dl'].add_progress_hook(frag_progress_hook) + + return start + + def _finish_frag_download(self, ctx): + ctx['dest_stream'].close() + elapsed = time.time() - ctx['started'] + self.try_rename(ctx['tmpfilename'], ctx['filename']) + fsize = os.path.getsize(encodeFilename(ctx['filename'])) + + self._hook_progress({ + 'downloaded_bytes': fsize, + 'total_bytes': fsize, + 'filename': ctx['filename'], + 'status': 'finished', + 'elapsed': elapsed, + }) diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py index 8be4f4249..2b6c3370f 100644 --- a/youtube_dl/downloader/hls.py +++ b/youtube_dl/downloader/hls.py @@ -4,12 +4,11 @@ import os import re import subprocess -from ..postprocessor.ffmpeg import FFmpegPostProcessor from .common import FileDownloader -from ..compat import ( - compat_urlparse, - compat_urllib_request, -) +from .fragment import FragmentFD + +from ..compat import compat_urlparse +from ..postprocessor.ffmpeg import FFmpegPostProcessor from ..utils import ( encodeArgument, encodeFilename, @@ -33,6 +32,8 @@ class HlsFD(FileDownloader): for opt in (ffpp.executable, '-y', '-i', url, '-f', 'mp4', '-c', 'copy', '-bsf:a', 'aac_adtstoasc')] args.append(encodeFilename(tmpfilename, True)) + self._debug_cmd(args) + retval = subprocess.call(args) if retval == 0: fsize = os.path.getsize(encodeFilename(tmpfilename)) @@ -51,54 +52,50 @@ class HlsFD(FileDownloader): return False -class NativeHlsFD(FileDownloader): +class NativeHlsFD(FragmentFD): """ A more limited implementation that does not require ffmpeg """ + FD_NAME = 'hlsnative' + def real_download(self, filename, info_dict): - url = info_dict['url'] - self.report_destination(filename) - tmpfilename = self.temp_name(filename) + man_url = info_dict['url'] + self.to_screen('[%s] Downloading m3u8 manifest' % self.FD_NAME) + manifest = self.ydl.urlopen(man_url).read() - self.to_screen( - '[hlsnative] %s: Downloading m3u8 manifest' % info_dict['id']) - data = self.ydl.urlopen(url).read() - s = data.decode('utf-8', 'ignore') - segment_urls = [] + s = manifest.decode('utf-8', 'ignore') + fragment_urls = [] for line in s.splitlines(): line = line.strip() if line and not line.startswith('#'): segment_url = ( line if re.match(r'^https?://', line) - else compat_urlparse.urljoin(url, line)) - segment_urls.append(segment_url) - - is_test = self.params.get('test', False) - remaining_bytes = self._TEST_FILE_SIZE if is_test else None - byte_counter = 0 - with open(tmpfilename, 'wb') as outf: - for i, segurl in enumerate(segment_urls): - self.to_screen( - '[hlsnative] %s: Downloading segment %d / %d' % - (info_dict['id'], i + 1, len(segment_urls))) - seg_req = compat_urllib_request.Request(segurl) - if remaining_bytes is not None: - seg_req.add_header('Range', 'bytes=0-%d' % (remaining_bytes - 1)) - - segment = self.ydl.urlopen(seg_req).read() - if remaining_bytes is not None: - segment = segment[:remaining_bytes] - remaining_bytes -= len(segment) - outf.write(segment) - byte_counter += len(segment) - if remaining_bytes is not None and remaining_bytes <= 0: + else compat_urlparse.urljoin(man_url, line)) + fragment_urls.append(segment_url) + # We only download the first fragment during the test + if self.params.get('test', False): break - self._hook_progress({ - 'downloaded_bytes': byte_counter, - 'total_bytes': byte_counter, + ctx = { 'filename': filename, - 'status': 'finished', - }) - self.try_rename(tmpfilename, filename) + 'total_frags': len(fragment_urls), + } + + self._prepare_and_start_frag_download(ctx) + + frags_filenames = [] + for i, frag_url in enumerate(fragment_urls): + frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i) + success = ctx['dl'].download(frag_filename, {'url': frag_url}) + if not success: + return False + with open(frag_filename, 'rb') as down: + ctx['dest_stream'].write(down.read()) + frags_filenames.append(frag_filename) + + self._finish_frag_download(ctx) + + for frag_file in frags_filenames: + os.remove(frag_file) + return True diff --git a/youtube_dl/downloader/http.py b/youtube_dl/downloader/http.py index b7f144af9..a29f5cf31 100644 --- a/youtube_dl/downloader/http.py +++ b/youtube_dl/downloader/http.py @@ -4,6 +4,7 @@ import errno import os import socket import time +import re from .common import FileDownloader from ..compat import ( @@ -57,6 +58,24 @@ class HttpFD(FileDownloader): # Establish connection try: data = self.ydl.urlopen(request) + # When trying to resume, Content-Range HTTP header of response has to be checked + # to match the value of requested Range HTTP header. This is due to a webservers + # that don't support resuming and serve a whole file with no Content-Range + # set in response despite of requested Range (see + # https://github.com/rg3/youtube-dl/issues/6057#issuecomment-126129799) + if resume_len > 0: + content_range = data.headers.get('Content-Range') + if content_range: + content_range_m = re.search(r'bytes (\d+)-', content_range) + # Content-Range is present and matches requested Range, resume is possible + if content_range_m and resume_len == int(content_range_m.group(1)): + break + # Content-Range is either not present or invalid. Assuming remote webserver is + # trying to send the whole file, resume is not possible, so wiping the local file + # and performing entire redownload + self.report_unable_to_resume() + resume_len = 0 + open_mode = 'wb' break except (compat_urllib_error.HTTPError, ) as err: if (err.code < 500 or err.code >= 600) and err.code != 416: diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 6c1d4f08a..7459a1944 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -19,9 +19,14 @@ from .anysex import AnySexIE from .aol import AolIE from .allocine import AllocineIE from .aparat import AparatIE +from .appleconnect import AppleConnectIE from .appletrailers import AppleTrailersIE from .archiveorg import ArchiveOrgIE -from .ard import ARDIE, ARDMediathekIE +from .ard import ( + ARDIE, + ARDMediathekIE, + SportschauIE, +) from .arte import ( ArteTvIE, ArteTVPlus7IE, @@ -38,7 +43,10 @@ from .azubu import AzubuIE from .baidu import BaiduVideoIE from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE -from .bbccouk import BBCCoUkIE +from .bbc import ( + BBCCoUkIE, + BBCIE, +) from .beeg import BeegIE from .behindkink import BehindKinkIE from .beatportpro import BeatportProIE @@ -110,6 +118,7 @@ from .dailymotion import ( ) from .daum import DaumIE from .dbtv import DBTVIE +from .dcn import DCNIE from .dctp import DctpTvIE from .deezer import DeezerPlaylistIE from .dfb import DFBIE @@ -149,6 +158,7 @@ from .eporner import EpornerIE from .eroprofile import EroProfileIE from .escapist import EscapistIE from .espn import ESPNIE +from .esri import EsriVideoIE from .everyonesmixtape import EveryonesMixtapeIE from .exfm import ExfmIE from .expotv import ExpoTVIE @@ -233,11 +243,16 @@ from .imdb import ( ) from .imgur import ImgurIE from .ina import InaIE +from .indavideo import ( + IndavideoIE, + IndavideoEmbedIE, +) from .infoq import InfoQIE from .instagram import InstagramIE, InstagramUserIE from .internetvideoarchive import InternetVideoArchiveIE from .iprima import IPrimaIE from .iqiyi import IqiyiIE +from .ir90tv import Ir90TvIE from .ivi import ( IviIE, IviCompilationIE @@ -270,6 +285,7 @@ from .kuwo import ( ) from .la7 import LA7IE from .laola1tv import Laola1TvIE +from .lecture2go import Lecture2GoIE from .letv import ( LetvIE, LetvTvIE, @@ -421,6 +437,10 @@ from .orf import ( from .parliamentliveuk import ParliamentLiveUKIE from .patreon import PatreonIE from .pbs import PBSIE +from .periscope import ( + PeriscopeIE, + QuickscopeIE, +) from .philharmoniedeparis import PhilharmonieDeParisIE from .phoenix import PhoenixIE from .photobucket import PhotobucketIE @@ -476,6 +496,7 @@ from .rtl2 import RTL2IE from .rtp import RTPIE from .rts import RTSIE from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE +from .rtvnh import RTVNHIE from .ruhd import RUHDIE from .rutube import ( RutubeIE, @@ -582,6 +603,7 @@ from .techtalks import TechTalksIE from .ted import TEDIE from .telebruxelles import TeleBruxellesIE from .telecinco import TelecincoIE +from .telegraaf import TelegraafIE from .telemb import TeleMBIE from .teletask import TeleTaskIE from .tenplay import TenPlayIE diff --git a/youtube_dl/extractor/appleconnect.py b/youtube_dl/extractor/appleconnect.py new file mode 100644 index 000000000..ea7a70393 --- /dev/null +++ b/youtube_dl/extractor/appleconnect.py @@ -0,0 +1,50 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + str_to_int, + ExtractorError +) + + +class AppleConnectIE(InfoExtractor): + _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)' + _TEST = { + 'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3', + 'md5': '10d0f2799111df4cb1c924520ca78f98', + 'info_dict': { + 'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3', + 'ext': 'm4v', + 'title': 'Energy', + 'uploader': 'Drake', + 'thumbnail': 'http://is5.mzstatic.com/image/thumb/Video5/v4/78/61/c5/7861c5fa-ad6d-294b-1464-cf7605b911d6/source/1920x1080sr.jpg', + 'upload_date': '20150710', + 'timestamp': 1436545535, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + try: + video_json = self._html_search_regex( + r'class="auc-video-data">(\{.*?\})', webpage, 'json') + except ExtractorError: + raise ExtractorError('This post doesn\'t contain a video', expected=True) + + video_data = self._parse_json(video_json, video_id) + timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp')) + like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count')) + + return { + 'id': video_id, + 'url': video_data['sslSrc'], + 'title': video_data['title'], + 'description': video_data['description'], + 'uploader': video_data['artistName'], + 'thumbnail': video_data['artworkUrl'], + 'timestamp': timestamp, + 'like_count': like_count, + } diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 6a35ea463..6f465789b 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -8,6 +8,7 @@ from .generic import GenericIE from ..utils import ( determine_ext, ExtractorError, + get_element_by_attribute, qualities, int_or_none, parse_duration, @@ -22,19 +23,125 @@ class ARDMediathekIE(InfoExtractor): _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?' _TESTS = [{ - 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', - 'only_matching': True, + 'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114', + 'info_dict': { + 'id': '29582122', + 'ext': 'mp4', + 'title': 'Ich liebe das Leben trotzdem', + 'description': 'md5:45e4c225c72b27993314b31a84a5261c', + 'duration': 4557, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { - 'url': 'http://www.ardmediathek.de/tv/Tatort/Das-Wunder-von-Wolbeck-Video-tgl-ab-20/Das-Erste/Video?documentId=22490580&bcastId=602916', + 'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916', + 'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e', 'info_dict': { - 'id': '22490580', + 'id': '29522730', 'ext': 'mp4', - 'title': 'Das Wunder von Wolbeck (Video tgl. ab 20 Uhr)', - 'description': 'Auf einem restaurierten Hof bei Wolbeck wird der Heilpraktiker Raffael Lembeck eines morgens von seiner Frau Stella tot aufgefunden. Das Opfer war offensichtlich in seiner Praxis zu Fall gekommen und ist dann verblutet, erklärt Prof. Boerne am Tatort.', + 'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)', + 'description': 'md5:196392e79876d0ac94c94e8cdb2875f1', + 'duration': 5252, }, - 'skip': 'Blocked outside of Germany', + }, { + # audio + 'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086', + 'md5': '219d94d8980b4f538c7fcb0865eb7f2c', + 'info_dict': { + 'id': '28488308', + 'ext': 'mp3', + 'title': 'Tod eines Fußballers', + 'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef', + 'duration': 3240, + }, + }, { + 'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht', + 'only_matching': True, }] + def _extract_media_info(self, media_info_url, webpage, video_id): + media_info = self._download_json( + media_info_url, video_id, 'Downloading media JSON') + + formats = self._extract_formats(media_info, video_id) + + if not formats: + if '"fsk"' in webpage: + raise ExtractorError( + 'This video is only available after 20:00', expected=True) + elif media_info.get('_geoblocked'): + raise ExtractorError('This video is not available due to geo restriction', expected=True) + + self._sort_formats(formats) + + duration = int_or_none(media_info.get('_duration')) + thumbnail = media_info.get('_previewImage') + + subtitles = {} + subtitle_url = media_info.get('_subtitleUrl') + if subtitle_url: + subtitles['de'] = [{ + 'ext': 'srt', + 'url': subtitle_url, + }] + + return { + 'id': video_id, + 'duration': duration, + 'thumbnail': thumbnail, + 'formats': formats, + 'subtitles': subtitles, + } + + def _extract_formats(self, media_info, video_id): + type_ = media_info.get('_type') + media_array = media_info.get('_mediaArray', []) + formats = [] + for num, media in enumerate(media_array): + for stream in media.get('_mediaStreamArray', []): + stream_urls = stream.get('_stream') + if not stream_urls: + continue + if not isinstance(stream_urls, list): + stream_urls = [stream_urls] + quality = stream.get('_quality') + server = stream.get('_server') + for stream_url in stream_urls: + ext = determine_ext(stream_url) + if ext == 'f4m': + formats.extend(self._extract_f4m_formats( + stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', + video_id, preference=-1, f4m_id='hds')) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + stream_url, video_id, 'mp4', preference=1, m3u8_id='hls')) + else: + if server and server.startswith('rtmp'): + f = { + 'url': server, + 'play_path': stream_url, + 'format_id': 'a%s-rtmp-%s' % (num, quality), + } + elif stream_url.startswith('http'): + f = { + 'url': stream_url, + 'format_id': 'a%s-%s-%s' % (num, ext, quality) + } + else: + continue + m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url) + if m: + f.update({ + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + if type_ == 'audio': + f['vcodec'] = 'none' + formats.append(f) + return formats + def _real_extract(self, url): # determine video id from url m = re.match(self._VALID_URL, url) @@ -92,46 +199,22 @@ class ARDMediathekIE(InfoExtractor): 'format_id': fid, 'url': furl, }) + self._sort_formats(formats) + info = { + 'formats': formats, + } else: # request JSON file - media_info = self._download_json( - 'http://www.ardmediathek.de/play/media/%s' % video_id, video_id) - # The second element of the _mediaArray contains the standard http urls - streams = media_info['_mediaArray'][1]['_mediaStreamArray'] - if not streams: - if '"fsk"' in webpage: - raise ExtractorError('This video is only available after 20:00') - - formats = [] - for s in streams: - if type(s['_stream']) == list: - for index, url in enumerate(s['_stream'][::-1]): - quality = s['_quality'] + index - formats.append({ - 'quality': quality, - 'url': url, - 'format_id': '%s-%s' % (determine_ext(url), quality) - }) - continue - - format = { - 'quality': s['_quality'], - 'url': s['_stream'], - } - - format['format_id'] = '%s-%s' % ( - determine_ext(format['url']), format['quality']) + info = self._extract_media_info( + 'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id) - formats.append(format) - - self._sort_formats(formats) - - return { + info.update({ 'id': video_id, 'title': title, 'description': description, - 'formats': formats, 'thumbnail': thumbnail, - } + }) + + return info class ARDIE(InfoExtractor): @@ -189,3 +272,41 @@ class ARDIE(InfoExtractor): 'upload_date': upload_date, 'thumbnail': thumbnail, } + + +class SportschauIE(ARDMediathekIE): + IE_NAME = 'Sportschau' + _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html' + _TESTS = [{ + 'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html', + 'info_dict': { + 'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100', + 'ext': 'mp4', + 'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + base_url = mobj.group('baseurl') + + webpage = self._download_webpage(url, video_id) + title = get_element_by_attribute('class', 'headline', webpage) + description = self._html_search_meta('description', webpage, 'description') + + info = self._extract_media_info( + base_url + '-mc_defaultQuality-h.json', webpage, video_id) + + info.update({ + 'title': title, + 'description': description, + }) + + return info diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py new file mode 100644 index 000000000..abc5a44a1 --- /dev/null +++ b/youtube_dl/extractor/bbc.py @@ -0,0 +1,780 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + float_or_none, + int_or_none, + parse_duration, + parse_iso8601, +) +from ..compat import compat_HTTPError + + +class BBCCoUkIE(InfoExtractor): + IE_NAME = 'bbc.co.uk' + IE_DESC = 'BBC iPlayer' + _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})' + + _MEDIASELECTOR_URLS = [ + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s', + ] + + _TESTS = [ + { + 'url': 'http://www.bbc.co.uk/programmes/b039g8p7', + 'info_dict': { + 'id': 'b039d07m', + 'ext': 'flv', + 'title': 'Kaleidoscope, Leonard Cohen', + 'description': 'The Canadian poet and songwriter reflects on his musical career.', + 'duration': 1740, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/', + 'info_dict': { + 'id': 'b00yng1d', + 'ext': 'flv', + 'title': 'The Man in Black: Series 3: The Printed Name', + 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.", + 'duration': 1800, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Episode is no longer available on BBC iPlayer Radio', + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/', + 'info_dict': { + 'id': 'b00yng1d', + 'ext': 'flv', + 'title': 'The Voice UK: Series 3: Blind Auditions 5', + 'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.", + 'duration': 5100, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', + }, + { + 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion', + 'info_dict': { + 'id': 'b03k3pb7', + 'ext': 'flv', + 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction", + 'description': '2. Invasion', + 'duration': 3600, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', + }, { + 'url': 'http://www.bbc.co.uk/programmes/b04v20dw', + 'info_dict': { + 'id': 'b04v209v', + 'ext': 'flv', + 'title': 'Pete Tong, The Essential New Tune Special', + 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!", + 'duration': 10800, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3', + 'note': 'Audio', + 'info_dict': { + 'id': 'p02frcch', + 'ext': 'flv', + 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix', + 'description': 'French house superstar Madeon takes us out of the club and onto the after party.', + 'duration': 3507, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz', + 'note': 'Video', + 'info_dict': { + 'id': 'p025c103', + 'ext': 'flv', + 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)', + 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014', + 'duration': 226, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls', + 'info_dict': { + 'id': 'p02n76xf', + 'ext': 'flv', + 'title': 'Natural World, 2015-2016: 2. Super Powered Owls', + 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d', + 'duration': 3540, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'geolocation', + }, { + 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition', + 'info_dict': { + 'id': 'b05zmgw1', + 'ext': 'flv', + 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.', + 'title': 'Royal Academy Summer Exhibition', + 'duration': 3540, + }, + 'params': { + # rtmp download + 'skip_download': True, + }, + 'skip': 'geolocation', + }, { + 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3', + 'only_matching': True, + }, { + 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo', + 'only_matching': True, + } + ] + + class MediaSelectionError(Exception): + def __init__(self, id): + self.id = id + + def _extract_asx_playlist(self, connection, programme_id): + asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist') + return [ref.get('href') for ref in asx.findall('./Entry/ref')] + + def _extract_connection(self, connection, programme_id): + formats = [] + protocol = connection.get('protocol') + supplier = connection.get('supplier') + if protocol == 'http': + href = connection.get('href') + transfer_format = connection.get('transferFormat') + # ASX playlist + if supplier == 'asx': + for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)): + formats.append({ + 'url': ref, + 'format_id': 'ref%s_%s' % (i, supplier), + }) + # Skip DASH until supported + elif transfer_format == 'dash': + pass + # Direct link + else: + formats.append({ + 'url': href, + 'format_id': supplier, + }) + elif protocol == 'rtmp': + application = connection.get('application', 'ondemand') + auth_string = connection.get('authString') + identifier = connection.get('identifier') + server = connection.get('server') + formats.append({ + 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string), + 'play_path': identifier, + 'app': '%s?%s' % (application, auth_string), + 'page_url': 'http://www.bbc.co.uk', + 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf', + 'rtmp_live': False, + 'ext': 'flv', + 'format_id': supplier, + }) + return formats + + def _extract_items(self, playlist): + return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item') + + def _extract_medias(self, media_selection): + error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error') + if error is not None: + raise BBCCoUkIE.MediaSelectionError(error.get('id')) + return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media') + + def _extract_connections(self, media): + return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection') + + def _extract_video(self, media, programme_id): + formats = [] + vbr = int_or_none(media.get('bitrate')) + vcodec = media.get('encoding') + service = media.get('service') + width = int_or_none(media.get('width')) + height = int_or_none(media.get('height')) + file_size = int_or_none(media.get('media_file_size')) + for connection in self._extract_connections(media): + conn_formats = self._extract_connection(connection, programme_id) + for format in conn_formats: + format.update({ + 'format_id': '%s_%s' % (service, format['format_id']), + 'width': width, + 'height': height, + 'vbr': vbr, + 'vcodec': vcodec, + 'filesize': file_size, + }) + formats.extend(conn_formats) + return formats + + def _extract_audio(self, media, programme_id): + formats = [] + abr = int_or_none(media.get('bitrate')) + acodec = media.get('encoding') + service = media.get('service') + for connection in self._extract_connections(media): + conn_formats = self._extract_connection(connection, programme_id) + for format in conn_formats: + format.update({ + 'format_id': '%s_%s' % (service, format['format_id']), + 'abr': abr, + 'acodec': acodec, + }) + formats.extend(conn_formats) + return formats + + def _get_subtitles(self, media, programme_id): + subtitles = {} + for connection in self._extract_connections(media): + captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions') + lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en') + subtitles[lang] = [ + { + 'url': connection.get('href'), + 'ext': 'ttml', + }, + ] + return subtitles + + def _raise_extractor_error(self, media_selection_error): + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, media_selection_error.id), + expected=True) + + def _download_media_selector(self, programme_id): + last_exception = None + for mediaselector_url in self._MEDIASELECTOR_URLS: + try: + return self._download_media_selector_url( + mediaselector_url % programme_id, programme_id) + except BBCCoUkIE.MediaSelectionError as e: + if e.id == 'notukerror': + last_exception = e + continue + self._raise_extractor_error(e) + self._raise_extractor_error(last_exception) + + def _download_media_selector_url(self, url, programme_id=None): + try: + media_selection = self._download_xml( + url, programme_id, 'Downloading media selection XML') + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: + media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8')) + else: + raise + return self._process_media_selector(media_selection, programme_id) + + def _process_media_selector(self, media_selection, programme_id): + formats = [] + subtitles = None + + for media in self._extract_medias(media_selection): + kind = media.get('kind') + if kind == 'audio': + formats.extend(self._extract_audio(media, programme_id)) + elif kind == 'video': + formats.extend(self._extract_video(media, programme_id)) + elif kind == 'captions': + subtitles = self.extract_subtitles(media, programme_id) + return formats, subtitles + + def _download_playlist(self, playlist_id): + try: + playlist = self._download_json( + 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id, + playlist_id, 'Downloading playlist JSON') + + version = playlist.get('defaultAvailableVersion') + if version: + smp_config = version['smpConfig'] + title = smp_config['title'] + description = smp_config['summary'] + for item in smp_config['items']: + kind = item['kind'] + if kind != 'programme' and kind != 'radioProgramme': + continue + programme_id = item.get('vpid') + duration = int_or_none(item.get('duration')) + formats, subtitles = self._download_media_selector(programme_id) + return programme_id, title, description, duration, formats, subtitles + except ExtractorError as ee: + if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404): + raise + + # fallback to legacy playlist + return self._process_legacy_playlist(playlist_id) + + def _process_legacy_playlist_url(self, url, display_id): + playlist = self._download_legacy_playlist_url(url, display_id) + return self._extract_from_legacy_playlist(playlist, display_id) + + def _process_legacy_playlist(self, playlist_id): + return self._process_legacy_playlist_url( + 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, playlist_id) + + def _download_legacy_playlist_url(self, url, playlist_id=None): + return self._download_xml( + url, playlist_id, 'Downloading legacy playlist XML') + + def _extract_from_legacy_playlist(self, playlist, playlist_id): + no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems') + if no_items is not None: + reason = no_items.get('reason') + if reason == 'preAvailability': + msg = 'Episode %s is not yet available' % playlist_id + elif reason == 'postAvailability': + msg = 'Episode %s is no longer available' % playlist_id + elif reason == 'noMedia': + msg = 'Episode %s is not currently available' % playlist_id + else: + msg = 'Episode %s is not available: %s' % (playlist_id, reason) + raise ExtractorError(msg, expected=True) + + for item in self._extract_items(playlist): + kind = item.get('kind') + if kind != 'programme' and kind != 'radioProgramme': + continue + title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text + description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text + + def get_programme_id(item): + def get_from_attributes(item): + for p in('identifier', 'group'): + value = item.get(p) + if value and re.match(r'^[pb][\da-z]{7}$', value): + return value + get_from_attributes(item) + mediator = item.find('./{http://bbc.co.uk/2008/emp/playlist}mediator') + if mediator is not None: + return get_from_attributes(mediator) + + programme_id = get_programme_id(item) + duration = int_or_none(item.get('duration')) + # TODO: programme_id can be None and media items can be incorporated right inside + # playlist's item (e.g. http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu) + # as f4m and m3u8 + formats, subtitles = self._download_media_selector(programme_id) + + return programme_id, title, description, duration, formats, subtitles + + def _real_extract(self, url): + group_id = self._match_id(url) + + webpage = self._download_webpage(url, group_id, 'Downloading video page') + + programme_id = None + + tviplayer = self._search_regex( + r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById', + webpage, 'player', default=None) + + if tviplayer: + player = self._parse_json(tviplayer, group_id).get('player', {}) + duration = int_or_none(player.get('duration')) + programme_id = player.get('vpid') + + if not programme_id: + programme_id = self._search_regex( + r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None) + + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + title = self._og_search_title(webpage) + description = self._search_regex( + r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>', + webpage, 'description', fatal=False) + else: + programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id) + + self._sort_formats(formats) + + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'duration': duration, + 'formats': formats, + 'subtitles': subtitles, + } + + +class BBCIE(BBCCoUkIE): + IE_NAME = 'bbc' + IE_DESC = 'BBC' + _VALID_URL = r'https?://(?:www\.)?bbc\.(?:com|co\.uk)/(?:[^/]+/)+(?P<id>[^/#?]+)' + + _MEDIASELECTOR_URLS = [ + # Provides more formats, namely direct mp4 links, but fails on some videos with + # notukerror for non UK (?) users (e.g. + # http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) + 'http://open.live.bbc.co.uk/mediaselector/4/mtis/stream/%s', + # Provides fewer formats, but works everywhere for everybody (hopefully) + 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/journalism-pc/vpid/%s', + ] + + _TESTS = [{ + # article with multiple videos embedded with data-media-meta containing + # playlist.sxml, externalId and no direct video links + 'url': 'http://www.bbc.com/news/world-europe-32668511', + 'info_dict': { + 'id': 'world-europe-32668511', + 'title': 'Russia stages massive WW2 parade despite Western boycott', + 'description': 'md5:00ff61976f6081841f759a08bf78cc9c', + }, + 'playlist_count': 2, + }, { + # article with multiple videos embedded with data-media-meta (more videos) + 'url': 'http://www.bbc.com/news/business-28299555', + 'info_dict': { + 'id': 'business-28299555', + 'title': 'Farnborough Airshow: Video highlights', + 'description': 'BBC reports and video highlights at the Farnborough Airshow.', + }, + 'playlist_count': 9, + 'skip': 'Save time', + }, { + # article with multiple videos embedded with `new SMP()` + 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460', + 'info_dict': { + 'id': '3662a707-0af9-3149-963f-47bea720b460', + 'title': 'BBC Blogs - Adam Curtis - BUGGER', + }, + 'playlist_count': 18, + }, { + # single video embedded with mediaAssetPage.init() + 'url': 'http://www.bbc.com/news/world-europe-32041533', + 'info_dict': { + 'id': 'p02mprgb', + 'ext': 'mp4', + 'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV', + 'duration': 47, + 'timestamp': 1427219242, + 'upload_date': '20150324', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # article with single video embedded with data-media-meta containing + # direct video links (for now these are extracted) and playlist.xml (with + # media items as f4m and m3u8 - currently unsupported) + 'url': 'http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu', + 'info_dict': { + 'id': '150615_telabyad_kentin_cogu', + 'ext': 'mp4', + 'title': "YPG: Tel Abyad'ın tamamı kontrolümüzde", + 'duration': 47, + 'timestamp': 1434397334, + 'upload_date': '20150615', + }, + 'params': { + 'skip_download': True, + } + }, { + # single video embedded with mediaAssetPage.init() (regional section) + 'url': 'http://www.bbc.com/mundo/video_fotos/2015/06/150619_video_honduras_militares_hospitales_corrupcion_aw', + 'info_dict': { + 'id': '150619_video_honduras_militares_hospitales_corrupcion_aw', + 'ext': 'mp4', + 'title': 'Honduras militariza sus hospitales por nuevo escándalo de corrupción', + 'duration': 87, + 'timestamp': 1434713142, + 'upload_date': '20150619', + }, + 'params': { + 'skip_download': True, + } + }, { + # single video from video playlist embedded with vxp-playlist-data JSON + 'url': 'http://www.bbc.com/news/video_and_audio/must_see/33376376', + 'info_dict': { + 'id': 'p02w6qjc', + 'ext': 'mp4', + 'title': '''Judge Mindy Glazer: "I'm sorry to see you here... I always wondered what happened to you"''', + 'duration': 56, + }, + 'params': { + 'skip_download': True, + } + }, { + # single video story with digitalData + 'url': 'http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret', + 'info_dict': { + 'id': 'p02q6gc4', + 'ext': 'flv', + 'title': 'Sri Lanka’s spicy secret', + 'description': 'As a new train line to Jaffna opens up the country’s north, travellers can experience a truly distinct slice of Tamil culture.', + 'timestamp': 1437674293, + 'upload_date': '20150723', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # single video story without digitalData + 'url': 'http://www.bbc.com/autos/story/20130513-hyundais-rock-star', + 'info_dict': { + 'id': 'p018zqqg', + 'ext': 'mp4', + 'title': 'Hyundai Santa Fe Sport: Rock star', + 'description': 'md5:b042a26142c4154a6e472933cf20793d', + 'timestamp': 1368473503, + 'upload_date': '20130513', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # single video with playlist.sxml URL + 'url': 'http://www.bbc.com/sport/0/football/33653409', + 'info_dict': { + 'id': 'p02xycnp', + 'ext': 'mp4', + 'title': 'Transfers: Cristiano Ronaldo to Man Utd, Arsenal to spend?', + 'description': 'md5:398fca0e2e701c609d726e034fa1fc89', + 'duration': 140, + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }, { + # single video with playlist URL from weather section + 'url': 'http://www.bbc.com/weather/features/33601775', + 'only_matching': True, + }, { + # custom redirection to www.bbc.com + 'url': 'http://www.bbc.co.uk/news/science-environment-33661876', + 'only_matching': True, + }] + + @classmethod + def suitable(cls, url): + return False if BBCCoUkIE.suitable(url) else super(BBCIE, cls).suitable(url) + + def _extract_from_media_meta(self, media_meta, video_id): + # Direct links to media in media metadata (e.g. + # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu) + # TODO: there are also f4m and m3u8 streams incorporated in playlist.sxml + source_files = media_meta.get('sourceFiles') + if source_files: + return [{ + 'url': f['url'], + 'format_id': format_id, + 'ext': f.get('encoding'), + 'tbr': float_or_none(f.get('bitrate'), 1000), + 'filesize': int_or_none(f.get('filesize')), + } for format_id, f in source_files.items() if f.get('url')], [] + + programme_id = media_meta.get('externalId') + if programme_id: + return self._download_media_selector(programme_id) + + # Process playlist.sxml as legacy playlist + href = media_meta.get('href') + if href: + playlist = self._download_legacy_playlist_url(href) + _, _, _, _, formats, subtitles = self._extract_from_legacy_playlist(playlist, video_id) + return formats, subtitles + + return [], [] + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + timestamp = parse_iso8601(self._search_regex( + [r'"datePublished":\s*"([^"]+)', + r'<meta[^>]+property="article:published_time"[^>]+content="([^"]+)"', + r'itemprop="datePublished"[^>]+datetime="([^"]+)"'], + webpage, 'date', default=None)) + + # single video with playlist.sxml URL (e.g. http://www.bbc.com/sport/0/football/3365340ng) + playlist = self._search_regex( + r'<param[^>]+name="playlist"[^>]+value="([^"]+)"', + webpage, 'playlist', default=None) + if playlist: + programme_id, title, description, duration, formats, subtitles = \ + self._process_legacy_playlist_url(playlist, playlist_id) + self._sort_formats(formats) + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + } + + # single video story (e.g. http://www.bbc.com/travel/story/20150625-sri-lankas-spicy-secret) + programme_id = self._search_regex( + [r'data-video-player-vpid="([\da-z]{8})"', + r'<param[^>]+name="externalIdentifier"[^>]+value="([\da-z]{8})"'], + webpage, 'vpid', default=None) + if programme_id: + formats, subtitles = self._download_media_selector(programme_id) + self._sort_formats(formats) + # digitalData may be missing (e.g. http://www.bbc.com/autos/story/20130513-hyundais-rock-star) + digital_data = self._parse_json( + self._search_regex( + r'var\s+digitalData\s*=\s*({.+?});?\n', webpage, 'digital data', default='{}'), + programme_id, fatal=False) + page_info = digital_data.get('page', {}).get('pageInfo', {}) + title = page_info.get('pageName') or self._og_search_title(webpage) + description = page_info.get('description') or self._og_search_description(webpage) + timestamp = parse_iso8601(page_info.get('publicationDate')) or timestamp + return { + 'id': programme_id, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + } + + playlist_title = self._html_search_regex( + r'<title>(.*?)(?:\s*-\s*BBC [^ ]+)?</title>', webpage, 'playlist title') + playlist_description = self._og_search_description(webpage, default=None) + + def extract_all(pattern): + return list(filter(None, map( + lambda s: self._parse_json(s, playlist_id, fatal=False), + re.findall(pattern, webpage)))) + + # Multiple video article (e.g. + # http://www.bbc.co.uk/blogs/adamcurtis/entries/3662a707-0af9-3149-963f-47bea720b460) + EMBED_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:[^/]+/)+[\da-z]{8}(?:\b[^"]+)?' + entries = [] + for match in extract_all(r'new\s+SMP\(({.+?})\)'): + embed_url = match.get('playerSettings', {}).get('externalEmbedUrl') + if embed_url and re.match(EMBED_URL, embed_url): + entries.append(embed_url) + entries.extend(re.findall( + r'setPlaylist\("(%s)"\)' % EMBED_URL, webpage)) + if entries: + return self.playlist_result( + [self.url_result(entry, 'BBCCoUk') for entry in entries], + playlist_id, playlist_title, playlist_description) + + # Multiple video article (e.g. http://www.bbc.com/news/world-europe-32668511) + medias = extract_all(r"data-media-meta='({[^']+})'") + + if not medias: + # Single video article (e.g. http://www.bbc.com/news/video_and_audio/international) + media_asset = self._search_regex( + r'mediaAssetPage\.init\(\s*({.+?}), "/', + webpage, 'media asset', default=None) + if media_asset: + media_asset_page = self._parse_json(media_asset, playlist_id, fatal=False) + medias = [] + for video in media_asset_page.get('videos', {}).values(): + medias.extend(video.values()) + + if not medias: + # Multiple video playlist with single `now playing` entry (e.g. + # http://www.bbc.com/news/video_and_audio/must_see/33767813) + vxp_playlist = self._parse_json( + self._search_regex( + r'<script[^>]+class="vxp-playlist-data"[^>]+type="application/json"[^>]*>([^<]+)</script>', + webpage, 'playlist data'), + playlist_id) + playlist_medias = [] + for item in vxp_playlist: + media = item.get('media') + if not media: + continue + playlist_medias.append(media) + # Download single video if found media with asset id matching the video id from URL + if item.get('advert', {}).get('assetId') == playlist_id: + medias = [media] + break + # Fallback to the whole playlist + if not medias: + medias = playlist_medias + + entries = [] + for num, media_meta in enumerate(medias, start=1): + formats, subtitles = self._extract_from_media_meta(media_meta, playlist_id) + if not formats: + continue + self._sort_formats(formats) + + video_id = media_meta.get('externalId') + if not video_id: + video_id = playlist_id if len(medias) == 1 else '%s-%s' % (playlist_id, num) + + title = media_meta.get('caption') + if not title: + title = playlist_title if len(medias) == 1 else '%s - Video %s' % (playlist_title, num) + + duration = int_or_none(media_meta.get('durationInSeconds')) or parse_duration(media_meta.get('duration')) + + images = [] + for image in media_meta.get('images', {}).values(): + images.extend(image.values()) + if 'image' in media_meta: + images.append(media_meta['image']) + + thumbnails = [{ + 'url': image.get('href'), + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } for image in images] + + entries.append({ + 'id': video_id, + 'title': title, + 'thumbnails': thumbnails, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + 'subtitles': subtitles, + }) + + return self.playlist_result(entries, playlist_id, playlist_title, playlist_description) diff --git a/youtube_dl/extractor/bbccouk.py b/youtube_dl/extractor/bbccouk.py deleted file mode 100644 index 5825d2867..000000000 --- a/youtube_dl/extractor/bbccouk.py +++ /dev/null @@ -1,379 +0,0 @@ -from __future__ import unicode_literals - -import xml.etree.ElementTree - -from .common import InfoExtractor -from ..utils import ( - ExtractorError, - int_or_none, -) -from ..compat import compat_HTTPError - - -class BBCCoUkIE(InfoExtractor): - IE_NAME = 'bbc.co.uk' - IE_DESC = 'BBC iPlayer' - _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})' - - _TESTS = [ - { - 'url': 'http://www.bbc.co.uk/programmes/b039g8p7', - 'info_dict': { - 'id': 'b039d07m', - 'ext': 'flv', - 'title': 'Kaleidoscope, Leonard Cohen', - 'description': 'The Canadian poet and songwriter reflects on his musical career.', - 'duration': 1740, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, - { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/', - 'info_dict': { - 'id': 'b00yng1d', - 'ext': 'flv', - 'title': 'The Man in Black: Series 3: The Printed Name', - 'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.", - 'duration': 1800, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'Episode is no longer available on BBC iPlayer Radio', - }, - { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/', - 'info_dict': { - 'id': 'b00yng1d', - 'ext': 'flv', - 'title': 'The Voice UK: Series 3: Blind Auditions 5', - 'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.", - 'duration': 5100, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', - }, - { - 'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion', - 'info_dict': { - 'id': 'b03k3pb7', - 'ext': 'flv', - 'title': "Tomorrow's Worlds: The Unearthly History of Science Fiction", - 'description': '2. Invasion', - 'duration': 3600, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only', - }, { - 'url': 'http://www.bbc.co.uk/programmes/b04v20dw', - 'info_dict': { - 'id': 'b04v209v', - 'ext': 'flv', - 'title': 'Pete Tong, The Essential New Tune Special', - 'description': "Pete has a very special mix - all of 2014's Essential New Tunes!", - 'duration': 10800, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://www.bbc.co.uk/music/clips/p02frcc3', - 'note': 'Audio', - 'info_dict': { - 'id': 'p02frcch', - 'ext': 'flv', - 'title': 'Pete Tong, Past, Present and Future Special, Madeon - After Hours mix', - 'description': 'French house superstar Madeon takes us out of the club and onto the after party.', - 'duration': 3507, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://www.bbc.co.uk/music/clips/p025c0zz', - 'note': 'Video', - 'info_dict': { - 'id': 'p025c103', - 'ext': 'flv', - 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)', - 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014', - 'duration': 226, - }, - 'params': { - # rtmp download - 'skip_download': True, - } - }, { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b054fn09/ad/natural-world-20152016-2-super-powered-owls', - 'info_dict': { - 'id': 'p02n76xf', - 'ext': 'flv', - 'title': 'Natural World, 2015-2016: 2. Super Powered Owls', - 'description': 'md5:e4db5c937d0e95a7c6b5e654d429183d', - 'duration': 3540, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'geolocation', - }, { - 'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition', - 'info_dict': { - 'id': 'b05zmgw1', - 'ext': 'flv', - 'description': 'Kirsty Wark and Morgan Quaintance visit the Royal Academy as it prepares for its annual artistic extravaganza, meeting people who have come together to make the show unique.', - 'title': 'Royal Academy Summer Exhibition', - 'duration': 3540, - }, - 'params': { - # rtmp download - 'skip_download': True, - }, - 'skip': 'geolocation', - }, { - 'url': 'http://www.bbc.co.uk/iplayer/playlist/p01dvks4', - 'only_matching': True, - }, { - 'url': 'http://www.bbc.co.uk/music/clips#p02frcc3', - 'only_matching': True, - }, { - 'url': 'http://www.bbc.co.uk/iplayer/cbeebies/episode/b0480276/bing-14-atchoo', - 'only_matching': True, - } - ] - - def _extract_asx_playlist(self, connection, programme_id): - asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist') - return [ref.get('href') for ref in asx.findall('./Entry/ref')] - - def _extract_connection(self, connection, programme_id): - formats = [] - protocol = connection.get('protocol') - supplier = connection.get('supplier') - if protocol == 'http': - href = connection.get('href') - # ASX playlist - if supplier == 'asx': - for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)): - formats.append({ - 'url': ref, - 'format_id': 'ref%s_%s' % (i, supplier), - }) - # Direct link - else: - formats.append({ - 'url': href, - 'format_id': supplier, - }) - elif protocol == 'rtmp': - application = connection.get('application', 'ondemand') - auth_string = connection.get('authString') - identifier = connection.get('identifier') - server = connection.get('server') - formats.append({ - 'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string), - 'play_path': identifier, - 'app': '%s?%s' % (application, auth_string), - 'page_url': 'http://www.bbc.co.uk', - 'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf', - 'rtmp_live': False, - 'ext': 'flv', - 'format_id': supplier, - }) - return formats - - def _extract_items(self, playlist): - return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item') - - def _extract_medias(self, media_selection): - error = media_selection.find('./{http://bbc.co.uk/2008/mp/mediaselection}error') - if error is not None: - raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, error.get('id')), expected=True) - return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media') - - def _extract_connections(self, media): - return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection') - - def _extract_video(self, media, programme_id): - formats = [] - vbr = int(media.get('bitrate')) - vcodec = media.get('encoding') - service = media.get('service') - width = int(media.get('width')) - height = int(media.get('height')) - file_size = int(media.get('media_file_size')) - for connection in self._extract_connections(media): - conn_formats = self._extract_connection(connection, programme_id) - for format in conn_formats: - format.update({ - 'format_id': '%s_%s' % (service, format['format_id']), - 'width': width, - 'height': height, - 'vbr': vbr, - 'vcodec': vcodec, - 'filesize': file_size, - }) - formats.extend(conn_formats) - return formats - - def _extract_audio(self, media, programme_id): - formats = [] - abr = int(media.get('bitrate')) - acodec = media.get('encoding') - service = media.get('service') - for connection in self._extract_connections(media): - conn_formats = self._extract_connection(connection, programme_id) - for format in conn_formats: - format.update({ - 'format_id': '%s_%s' % (service, format['format_id']), - 'abr': abr, - 'acodec': acodec, - }) - formats.extend(conn_formats) - return formats - - def _get_subtitles(self, media, programme_id): - subtitles = {} - for connection in self._extract_connections(media): - captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions') - lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en') - subtitles[lang] = [ - { - 'url': connection.get('href'), - 'ext': 'ttml', - }, - ] - return subtitles - - def _download_media_selector(self, programme_id): - try: - media_selection = self._download_xml( - 'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s' % programme_id, - programme_id, 'Downloading media selection XML') - except ExtractorError as ee: - if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403: - media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8')) - else: - raise - - formats = [] - subtitles = None - - for media in self._extract_medias(media_selection): - kind = media.get('kind') - if kind == 'audio': - formats.extend(self._extract_audio(media, programme_id)) - elif kind == 'video': - formats.extend(self._extract_video(media, programme_id)) - elif kind == 'captions': - subtitles = self.extract_subtitles(media, programme_id) - - return formats, subtitles - - def _download_playlist(self, playlist_id): - try: - playlist = self._download_json( - 'http://www.bbc.co.uk/programmes/%s/playlist.json' % playlist_id, - playlist_id, 'Downloading playlist JSON') - - version = playlist.get('defaultAvailableVersion') - if version: - smp_config = version['smpConfig'] - title = smp_config['title'] - description = smp_config['summary'] - for item in smp_config['items']: - kind = item['kind'] - if kind != 'programme' and kind != 'radioProgramme': - continue - programme_id = item.get('vpid') - duration = int(item.get('duration')) - formats, subtitles = self._download_media_selector(programme_id) - return programme_id, title, description, duration, formats, subtitles - except ExtractorError as ee: - if not (isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404): - raise - - # fallback to legacy playlist - playlist = self._download_xml( - 'http://www.bbc.co.uk/iplayer/playlist/%s' % playlist_id, - playlist_id, 'Downloading legacy playlist XML') - - no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems') - if no_items is not None: - reason = no_items.get('reason') - if reason == 'preAvailability': - msg = 'Episode %s is not yet available' % playlist_id - elif reason == 'postAvailability': - msg = 'Episode %s is no longer available' % playlist_id - elif reason == 'noMedia': - msg = 'Episode %s is not currently available' % playlist_id - else: - msg = 'Episode %s is not available: %s' % (playlist_id, reason) - raise ExtractorError(msg, expected=True) - - for item in self._extract_items(playlist): - kind = item.get('kind') - if kind != 'programme' and kind != 'radioProgramme': - continue - title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text - description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text - programme_id = item.get('identifier') - duration = int(item.get('duration')) - formats, subtitles = self._download_media_selector(programme_id) - - return programme_id, title, description, duration, formats, subtitles - - def _real_extract(self, url): - group_id = self._match_id(url) - - webpage = self._download_webpage(url, group_id, 'Downloading video page') - - programme_id = None - - tviplayer = self._search_regex( - r'mediator\.bind\(({.+?})\s*,\s*document\.getElementById', - webpage, 'player', default=None) - - if tviplayer: - player = self._parse_json(tviplayer, group_id).get('player', {}) - duration = int_or_none(player.get('duration')) - programme_id = player.get('vpid') - - if not programme_id: - programme_id = self._search_regex( - r'"vpid"\s*:\s*"([\da-z]{8})"', webpage, 'vpid', fatal=False, default=None) - - if programme_id: - formats, subtitles = self._download_media_selector(programme_id) - title = self._og_search_title(webpage) - description = self._search_regex( - r'<p class="[^"]*medium-description[^"]*">([^<]+)</p>', - webpage, 'description', fatal=False) - else: - programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id) - - self._sort_formats(formats) - - return { - 'id': programme_id, - 'title': title, - 'description': description, - 'thumbnail': self._og_search_thumbnail(webpage, default=None), - 'duration': duration, - 'formats': formats, - 'subtitles': subtitles, - } diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py index 809287d14..aa08051b1 100644 --- a/youtube_dl/extractor/breakcom.py +++ b/youtube_dl/extractor/breakcom.py @@ -18,6 +18,7 @@ class BreakIE(InfoExtractor): 'id': '2468056', 'ext': 'mp4', 'title': 'When Girls Act Like D-Bags', + 'age_limit': 13, } }, { 'url': 'http://www.break.com/video/ugc/baby-flex-2773063', diff --git a/youtube_dl/extractor/canalplus.py b/youtube_dl/extractor/canalplus.py index 699b4f7d0..57e0cda2c 100644 --- a/youtube_dl/extractor/canalplus.py +++ b/youtube_dl/extractor/canalplus.py @@ -106,15 +106,11 @@ class CanalplusIE(InfoExtractor): continue format_id = fmt.tag if format_id == 'HLS': - hls_formats = self._extract_m3u8_formats(format_url, video_id, 'flv') - for fmt in hls_formats: - fmt['preference'] = preference(format_id) - formats.extend(hls_formats) + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', preference=preference(format_id))) elif format_id == 'HDS': - hds_formats = self._extract_f4m_formats(format_url + '?hdcore=2.11.3', video_id) - for fmt in hds_formats: - fmt['preference'] = preference(format_id) - formats.extend(hds_formats) + formats.extend(self._extract_f4m_formats( + format_url + '?hdcore=2.11.3', video_id, preference=preference(format_id))) else: formats.append({ 'url': format_url, diff --git a/youtube_dl/extractor/clipfish.py b/youtube_dl/extractor/clipfish.py index a5c3cb7c6..7af903571 100644 --- a/youtube_dl/extractor/clipfish.py +++ b/youtube_dl/extractor/clipfish.py @@ -1,53 +1,68 @@ from __future__ import unicode_literals import re -import time -import xml.etree.ElementTree from .common import InfoExtractor from ..utils import ( - ExtractorError, - parse_duration, + determine_ext, + int_or_none, + js_to_json, + parse_iso8601, + remove_end, ) class ClipfishIE(InfoExtractor): - IE_NAME = 'clipfish' - - _VALID_URL = r'^https?://(?:www\.)?clipfish\.de/.*?/video/(?P<id>[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?clipfish\.de/(?:[^/]+/)+video/(?P<id>[0-9]+)' _TEST = { 'url': 'http://www.clipfish.de/special/game-trailer/video/3966754/fifa-14-e3-2013-trailer/', - 'md5': '2521cd644e862936cf2e698206e47385', + 'md5': '79bc922f3e8a9097b3d68a93780fd475', 'info_dict': { 'id': '3966754', 'ext': 'mp4', 'title': 'FIFA 14 - E3 2013 Trailer', + 'timestamp': 1370938118, + 'upload_date': '20130611', 'duration': 82, - }, - 'skip': 'Blocked in the US' + } } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1) - - info_url = ('http://www.clipfish.de/devxml/videoinfo/%s?ts=%d' % - (video_id, int(time.time()))) - doc = self._download_xml( - info_url, video_id, note='Downloading info page') - title = doc.find('title').text - video_url = doc.find('filename').text - if video_url is None: - xml_bytes = xml.etree.ElementTree.tostring(doc) - raise ExtractorError('Cannot find video URL in document %r' % - xml_bytes) - thumbnail = doc.find('imageurl').text - duration = parse_duration(doc.find('duration').text) + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + video_info = self._parse_json( + js_to_json(self._html_search_regex( + '(?s)videoObject\s*=\s*({.+?});', webpage, 'video object')), + video_id) + + formats = [] + for video_url in re.findall(r'var\s+videourl\s*=\s*"([^"]+)"', webpage): + ext = determine_ext(video_url) + if ext == 'm3u8': + formats.append({ + 'url': video_url.replace('de.hls.fra.clipfish.de', 'hls.fra.clipfish.de'), + 'ext': 'mp4', + 'format_id': 'hls', + }) + else: + formats.append({ + 'url': video_url, + 'format_id': ext, + }) + self._sort_formats(formats) + + title = remove_end(self._og_search_title(webpage), ' - Video') + thumbnail = self._og_search_thumbnail(webpage) + duration = int_or_none(video_info.get('length')) + timestamp = parse_iso8601(self._html_search_meta('uploadDate', webpage, 'upload date')) return { 'id': video_id, 'title': title, - 'url': video_url, + 'formats': formats, 'thumbnail': thumbnail, 'duration': duration, + 'timestamp': timestamp, } diff --git a/youtube_dl/extractor/comcarcoff.py b/youtube_dl/extractor/comcarcoff.py index 9c25b2223..81f3d7697 100644 --- a/youtube_dl/extractor/comcarcoff.py +++ b/youtube_dl/extractor/comcarcoff.py @@ -36,7 +36,7 @@ class ComCarCoffIE(InfoExtractor): webpage, 'full data json')) video_id = full_data['activeVideo']['video'] - video_data = full_data['videos'][video_id] + video_data = full_data.get('videos', {}).get(video_id) or full_data['singleshots'][video_id] thumbnails = [{ 'url': video_data['images']['thumb'], }, { diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index b9014fc23..16ae4b98f 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -14,10 +14,13 @@ import xml.etree.ElementTree from ..compat import ( compat_cookiejar, + compat_cookies, compat_HTTPError, compat_http_client, compat_urllib_error, + compat_urllib_parse, compat_urllib_parse_urlparse, + compat_urllib_request, compat_urlparse, compat_str, ) @@ -35,6 +38,9 @@ from ..utils import ( RegexNotFoundError, sanitize_filename, unescapeHTML, + url_basename, + xpath_text, + xpath_with_ns, ) @@ -65,7 +71,7 @@ class InfoExtractor(object): Potential fields: * url Mandatory. The URL of the video file - * ext Will be calculated from url if missing + * ext Will be calculated from URL if missing * format A human-readable description of the format ("mp4 container with h264/opus"). Calculated from the format_id, width, height. @@ -155,7 +161,7 @@ class InfoExtractor(object): lower to higher preference, each element is a dictionary with the "ext" entry and one of: * "data": The subtitles file contents - * "url": A url pointing to the subtitles file + * "url": A URL pointing to the subtitles file automatic_captions: Like 'subtitles', used by the YoutubeIE for automatically generated captions duration: Length of the video in seconds, as an integer. @@ -176,13 +182,18 @@ class InfoExtractor(object): Set to "root" to indicate that this is a comment to the original video. age_limit: Age restriction for the video, as an integer (years) - webpage_url: The url to the video webpage, if given to youtube-dl it + webpage_url: The URL to the video webpage, if given to youtube-dl it should allow to get the same result again. (It will be set by YoutubeDL if it's missing) categories: A list of categories that the video falls in, for example ["Sports", "Berlin"] + tags: A list of tags assigned to the video, e.g. ["sweden", "pop music"] is_live: True, False, or None (=unknown). Whether this video is a live stream that goes on instead of a fixed-length video. + start_time: Time in seconds where the reproduction should start, as + specified in the URL. + end_time: Time in seconds where the reproduction should end, as + specified in the URL. Unless mentioned otherwise, the fields should be Unicode strings. @@ -193,8 +204,8 @@ class InfoExtractor(object): There must be a key "entries", which is a list, an iterable, or a PagedList object, each element of which is a valid dictionary by this specification. - Additionally, playlists can have "title" and "id" attributes with the same - semantics as videos (see above). + Additionally, playlists can have "title", "description" and "id" attributes + with the same semantics as videos (see above). _type "multi_video" indicates that there are multiple videos that @@ -501,7 +512,7 @@ class InfoExtractor(object): # Methods for following #608 @staticmethod def url_result(url, ie=None, video_id=None, video_title=None): - """Returns a url that points to a page that should be processed""" + """Returns a URL that points to a page that should be processed""" # TODO: ie should be the class used for getting the info video_info = {'_type': 'url', 'url': url, @@ -626,6 +637,12 @@ class InfoExtractor(object): template % (content_re, property_re), ] + @staticmethod + def _meta_regex(prop): + return r'''(?isx)<meta + (?=[^>]+(?:itemprop|name|property|id|http-equiv)=(["\']?)%s\1) + [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(prop) + def _og_search_property(self, prop, html, name=None, **kargs): if name is None: name = 'OpenGraph %s' % prop @@ -635,7 +652,7 @@ class InfoExtractor(object): return unescapeHTML(escaped) def _og_search_thumbnail(self, html, **kargs): - return self._og_search_property('image', html, 'thumbnail url', fatal=False, **kargs) + return self._og_search_property('image', html, 'thumbnail URL', fatal=False, **kargs) def _og_search_description(self, html, **kargs): return self._og_search_property('description', html, fatal=False, **kargs) @@ -656,9 +673,7 @@ class InfoExtractor(object): if display_name is None: display_name = name return self._html_search_regex( - r'''(?isx)<meta - (?=[^>]+(?:itemprop|name|property)=(["\']?)%s\1) - [^>]+?content=(["\'])(?P<content>.*?)\2''' % re.escape(name), + self._meta_regex(name), html, display_name, fatal=fatal, group='content', **kwargs) def _dc_search_uploader(self, html): @@ -967,69 +982,210 @@ class InfoExtractor(object): self._sort_formats(formats) return formats - # TODO: improve extraction - def _extract_smil_formats(self, smil_url, video_id, fatal=True): - smil = self._download_xml( - smil_url, video_id, 'Downloading SMIL file', - 'Unable to download SMIL file', fatal=fatal) + @staticmethod + def _xpath_ns(path, namespace=None): + if not namespace: + return path + out = [] + for c in path.split('/'): + if not c or c == '.': + out.append(c) + else: + out.append('{%s}%s' % (namespace, c)) + return '/'.join(out) + + def _extract_smil_formats(self, smil_url, video_id, fatal=True, f4m_params=None): + smil = self._download_smil(smil_url, video_id, fatal=fatal) + if smil is False: assert not fatal return [] - base = smil.find('./head/meta').get('base') + namespace = self._parse_smil_namespace(smil) + + return self._parse_smil_formats( + smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) + + def _extract_smil_info(self, smil_url, video_id, fatal=True, f4m_params=None): + smil = self._download_smil(smil_url, video_id, fatal=fatal) + if smil is False: + return {} + return self._parse_smil(smil, smil_url, video_id, f4m_params=f4m_params) + + def _download_smil(self, smil_url, video_id, fatal=True): + return self._download_xml( + smil_url, video_id, 'Downloading SMIL file', + 'Unable to download SMIL file', fatal=fatal) + + def _parse_smil(self, smil, smil_url, video_id, f4m_params=None): + namespace = self._parse_smil_namespace(smil) + + formats = self._parse_smil_formats( + smil, smil_url, video_id, namespace=namespace, f4m_params=f4m_params) + subtitles = self._parse_smil_subtitles(smil, namespace=namespace) + + video_id = os.path.splitext(url_basename(smil_url))[0] + title = None + description = None + for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): + name = meta.attrib.get('name') + content = meta.attrib.get('content') + if not name or not content: + continue + if not title and name == 'title': + title = content + elif not description and name in ('description', 'abstract'): + description = content + + return { + 'id': video_id, + 'title': title or video_id, + 'description': description, + 'formats': formats, + 'subtitles': subtitles, + } + + def _parse_smil_namespace(self, smil): + return self._search_regex( + r'(?i)^{([^}]+)?}smil$', smil.tag, 'namespace', default=None) + + def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None): + base = smil_url + for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): + b = meta.get('base') or meta.get('httpBase') + if b: + base = b + break formats = [] rtmp_count = 0 - if smil.findall('./body/seq/video'): - video = smil.findall('./body/seq/video')[0] - fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count) - formats.extend(fmts) - else: - for video in smil.findall('./body/switch/video'): - fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count) - formats.extend(fmts) + http_count = 0 + + videos = smil.findall(self._xpath_ns('.//video', namespace)) + for video in videos: + src = video.get('src') + if not src: + continue + + bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) + filesize = int_or_none(video.get('size') or video.get('fileSize')) + width = int_or_none(video.get('width')) + height = int_or_none(video.get('height')) + proto = video.get('proto') + ext = video.get('ext') + src_ext = determine_ext(src) + streamer = video.get('streamer') or base + + if proto == 'rtmp' or streamer.startswith('rtmp'): + rtmp_count += 1 + formats.append({ + 'url': streamer, + 'play_path': src, + 'ext': 'flv', + 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), + 'tbr': bitrate, + 'filesize': filesize, + 'width': width, + 'height': height, + }) + continue + + src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src) + + if proto == 'm3u8' or src_ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + src_url, video_id, ext or 'mp4', m3u8_id='hls')) + continue + + if src_ext == 'f4m': + f4m_url = src_url + if not f4m_params: + f4m_params = { + 'hdcore': '3.2.0', + 'plugin': 'flowplayer-3.2.0.1', + } + f4m_url += '&' if '?' in f4m_url else '?' + f4m_url += compat_urllib_parse.urlencode(f4m_params) + formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds')) + continue + + if src_url.startswith('http'): + http_count += 1 + formats.append({ + 'url': src_url, + 'ext': ext or src_ext or 'flv', + 'format_id': 'http-%d' % (bitrate or http_count), + 'tbr': bitrate, + 'filesize': filesize, + 'width': width, + 'height': height, + }) + continue self._sort_formats(formats) return formats - def _parse_smil_video(self, video, video_id, base, rtmp_count): - src = video.get('src') - if not src: - return [], rtmp_count - bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) - width = int_or_none(video.get('width')) - height = int_or_none(video.get('height')) - proto = video.get('proto') - if not proto: - if base: - if base.startswith('rtmp'): - proto = 'rtmp' - elif base.startswith('http'): - proto = 'http' - ext = video.get('ext') - if proto == 'm3u8': - return self._extract_m3u8_formats(src, video_id, ext), rtmp_count - elif proto == 'rtmp': - rtmp_count += 1 - streamer = video.get('streamer') or base - return ([{ - 'url': streamer, - 'play_path': src, - 'ext': 'flv', - 'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate), - 'tbr': bitrate, - 'width': width, - 'height': height, - }], rtmp_count) - elif proto.startswith('http'): - return ([{ - 'url': base + src, - 'ext': ext or 'flv', - 'tbr': bitrate, - 'width': width, - 'height': height, - }], rtmp_count) + def _parse_smil_subtitles(self, smil, namespace=None): + subtitles = {} + for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))): + src = textstream.get('src') + if not src: + continue + ext = textstream.get('ext') or determine_ext(src) + if not ext: + type_ = textstream.get('type') + if type_ == 'text/srt': + ext = 'srt' + lang = textstream.get('systemLanguage') or textstream.get('systemLanguageName') + subtitles.setdefault(lang, []).append({ + 'url': src, + 'ext': ext, + }) + return subtitles + + def _extract_xspf_playlist(self, playlist_url, playlist_id, fatal=True): + xspf = self._download_xml( + playlist_url, playlist_id, 'Downloading xpsf playlist', + 'Unable to download xspf manifest', fatal=fatal) + if xspf is False: + return [] + return self._parse_xspf(xspf, playlist_id) + + def _parse_xspf(self, playlist, playlist_id): + NS_MAP = { + 'xspf': 'http://xspf.org/ns/0/', + 's1': 'http://static.streamone.nl/player/ns/0', + } + + entries = [] + for track in playlist.findall(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)): + title = xpath_text( + track, xpath_with_ns('./xspf:title', NS_MAP), 'title', default=playlist_id) + description = xpath_text( + track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') + thumbnail = xpath_text( + track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') + duration = float_or_none( + xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), 1000) + + formats = [{ + 'url': location.text, + 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), + 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), + 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), + } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] + self._sort_formats(formats) + + entries.append({ + 'id': playlist_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + }) + return entries def _live_title(self, name): """ Generate the title for a live video """ @@ -1065,6 +1221,12 @@ class InfoExtractor(object): None, '/', True, False, expire_time, '', None, None, None) self._downloader.cookiejar.set_cookie(cookie) + def _get_cookies(self, url): + """ Return a compat_cookies.SimpleCookie with the cookies for the url """ + req = compat_urllib_request.Request(url) + self._downloader.cookiejar.add_cookie_header(req) + return compat_cookies.SimpleCookie(req.get_header('Cookie')) + def get_testcases(self, include_onlymatching=False): t = getattr(self, '_TEST', None) if t: @@ -1116,7 +1278,7 @@ class InfoExtractor(object): class SearchInfoExtractor(InfoExtractor): """ Base class for paged search queries extractors. - They accept urls in the format _SEARCH_KEY(|all|[0-9]):{query} + They accept URLs in the format _SEARCH_KEY(|all|[0-9]):{query} Instances should define _SEARCH_KEY and _MAX_RESULTS. """ diff --git a/youtube_dl/extractor/dailymotion.py b/youtube_dl/extractor/dailymotion.py index 1a41c0db1..2d90b2224 100644 --- a/youtube_dl/extractor/dailymotion.py +++ b/youtube_dl/extractor/dailymotion.py @@ -13,8 +13,9 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + determine_ext, int_or_none, - orderedSet, + parse_iso8601, str_to_int, unescapeHTML, ) @@ -28,10 +29,16 @@ class DailymotionBaseInfoExtractor(InfoExtractor): request.add_header('Cookie', 'family_filter=off; ff=off') return request + def _download_webpage_handle_no_ff(self, url, *args, **kwargs): + request = self._build_request(url) + return self._download_webpage_handle(request, *args, **kwargs) + + def _download_webpage_no_ff(self, url, *args, **kwargs): + request = self._build_request(url) + return self._download_webpage(request, *args, **kwargs) -class DailymotionIE(DailymotionBaseInfoExtractor): - """Information Extractor for Dailymotion""" +class DailymotionIE(DailymotionBaseInfoExtractor): _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)' IE_NAME = 'dailymotion' @@ -50,10 +57,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'info_dict': { 'id': 'x2iuewm', 'ext': 'mp4', - 'uploader': 'IGN', 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News', - 'upload_date': '20150306', + 'description': 'Several come bundled with the Steam Controller.', + 'thumbnail': 're:^https?:.*\.(?:jpg|png)$', 'duration': 74, + 'timestamp': 1425657362, + 'upload_date': '20150306', + 'uploader': 'IGN', + 'uploader_id': 'xijv66', + 'age_limit': 0, + 'view_count': int, + 'comment_count': int, } }, # Vevo video @@ -87,38 +101,106 @@ class DailymotionIE(DailymotionBaseInfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - url = 'https://www.dailymotion.com/video/%s' % video_id - # Retrieve video webpage to extract further information - request = self._build_request(url) - webpage = self._download_webpage(request, video_id) + webpage = self._download_webpage_no_ff( + 'https://www.dailymotion.com/video/%s' % video_id, video_id) + + age_limit = self._rta_search(webpage) - # Extract URL, uploader and title from webpage - self.report_extraction(video_id) + description = self._og_search_description(webpage) or self._html_search_meta( + 'description', webpage, 'description') - # It may just embed a vevo video: - m_vevo = re.search( + view_count = str_to_int(self._search_regex( + [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"', + r'video_views_count[^>]+>\s+([\d\.,]+)'], + webpage, 'view count', fatal=False)) + comment_count = int_or_none(self._search_regex( + r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"', + webpage, 'comment count', fatal=False)) + + player_v5 = self._search_regex( + r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);', + webpage, 'player v5', default=None) + if player_v5: + player = self._parse_json(player_v5, video_id) + metadata = player['metadata'] + formats = [] + for quality, media_list in metadata['qualities'].items(): + for media in media_list: + media_url = media.get('url') + if not media_url: + continue + type_ = media.get('type') + if type_ == 'application/vnd.lumberjack.manifest': + continue + if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + media_url, video_id, 'mp4', m3u8_id='hls')) + else: + f = { + 'url': media_url, + 'format_id': quality, + } + m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url) + if m: + f.update({ + 'width': int(m.group('width')), + 'height': int(m.group('height')), + }) + formats.append(f) + self._sort_formats(formats) + + title = metadata['title'] + duration = int_or_none(metadata.get('duration')) + timestamp = int_or_none(metadata.get('created_time')) + thumbnail = metadata.get('poster_url') + uploader = metadata.get('owner', {}).get('screenname') + uploader_id = metadata.get('owner', {}).get('id') + + subtitles = {} + for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items(): + subtitles[subtitle_lang] = [{ + 'ext': determine_ext(subtitle_url), + 'url': subtitle_url, + } for subtitle_url in subtitle.get('urls', [])] + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'age_limit': age_limit, + 'view_count': view_count, + 'comment_count': comment_count, + 'formats': formats, + 'subtitles': subtitles, + } + + # vevo embed + vevo_id = self._search_regex( r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)', - webpage) - if m_vevo is not None: - vevo_id = m_vevo.group('id') - self.to_screen('Vevo video detected: %s' % vevo_id) - return self.url_result('vevo:%s' % vevo_id, ie='Vevo') + webpage, 'vevo embed', default=None) + if vevo_id: + return self.url_result('vevo:%s' % vevo_id, 'Vevo') - age_limit = self._rta_search(webpage) + # fallback old player + embed_page = self._download_webpage_no_ff( + 'https://www.dailymotion.com/embed/video/%s' % video_id, + video_id, 'Downloading embed page') + + timestamp = parse_iso8601(self._html_search_meta( + 'video:release_date', webpage, 'upload date')) + + info = self._parse_json( + self._search_regex( + r'var info = ({.*?}),$', embed_page, + 'video info', flags=re.MULTILINE), + video_id) - video_upload_date = None - mobj = re.search(r'<meta property="video:release_date" content="([0-9]{4})-([0-9]{2})-([0-9]{2}).+?"/>', webpage) - if mobj is not None: - video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3) - - embed_url = 'https://www.dailymotion.com/embed/video/%s' % video_id - embed_request = self._build_request(embed_url) - embed_page = self._download_webpage( - embed_request, video_id, 'Downloading embed page') - info = self._search_regex(r'var info = ({.*?}),$', embed_page, - 'video info', flags=re.MULTILINE) - info = json.loads(info) if info.get('error') is not None: msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title'] raise ExtractorError(msg, expected=True) @@ -139,16 +221,11 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'width': width, 'height': height, }) - if not formats: - raise ExtractorError('Unable to extract video URL') + self._sort_formats(formats) # subtitles video_subtitles = self.extract_subtitles(video_id, webpage) - view_count = str_to_int(self._search_regex( - r'video_views_count[^>]+>\s+([\d\.,]+)', - webpage, 'view count', fatal=False)) - title = self._og_search_title(webpage, default=None) if title is None: title = self._html_search_regex( @@ -159,8 +236,9 @@ class DailymotionIE(DailymotionBaseInfoExtractor): 'id': video_id, 'formats': formats, 'uploader': info['owner.screenname'], - 'upload_date': video_upload_date, + 'timestamp': timestamp, 'title': title, + 'description': description, 'subtitles': video_subtitles, 'thumbnail': info['thumbnail_url'], 'age_limit': age_limit, @@ -199,18 +277,26 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): }] def _extract_entries(self, id): - video_ids = [] + video_ids = set() + processed_urls = set() for pagenum in itertools.count(1): - request = self._build_request(self._PAGE_TEMPLATE % (id, pagenum)) - webpage = self._download_webpage(request, - id, 'Downloading page %s' % pagenum) + page_url = self._PAGE_TEMPLATE % (id, pagenum) + webpage, urlh = self._download_webpage_handle_no_ff( + page_url, id, 'Downloading page %s' % pagenum) + if urlh.geturl() in processed_urls: + self.report_warning('Stopped at duplicated page %s, which is the same as %s' % ( + page_url, urlh.geturl()), id) + break - video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage)) + processed_urls.add(urlh.geturl()) + + for video_id in re.findall(r'data-xid="(.+?)"', webpage): + if video_id not in video_ids: + yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') + video_ids.add(video_id) if re.search(self._MORE_PAGES_INDICATOR, webpage) is None: break - return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion') - for video_id in orderedSet(video_ids)] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -227,7 +313,7 @@ class DailymotionPlaylistIE(DailymotionBaseInfoExtractor): class DailymotionUserIE(DailymotionPlaylistIE): IE_NAME = 'dailymotion:user' - _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?:(?:old/)?user/)?(?P<user>[^/]+)$' + _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)' _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s' _TESTS = [{ 'url': 'https://www.dailymotion.com/user/nqtv', @@ -236,6 +322,17 @@ class DailymotionUserIE(DailymotionPlaylistIE): 'title': 'Rémi Gaillard', }, 'playlist_mincount': 100, + }, { + 'url': 'http://www.dailymotion.com/user/UnderProject', + 'info_dict': { + 'id': 'UnderProject', + 'title': 'UnderProject', + }, + 'playlist_mincount': 1800, + 'expected_warnings': [ + 'Stopped at duplicated page', + ], + 'skip': 'Takes too long time', }] def _real_extract(self, url): @@ -286,8 +383,7 @@ class DailymotionCloudIE(DailymotionBaseInfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - request = self._build_request(url) - webpage = self._download_webpage(request, video_id) + webpage = self._download_webpage_no_ff(url, video_id) title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title') diff --git a/youtube_dl/extractor/dcn.py b/youtube_dl/extractor/dcn.py new file mode 100644 index 000000000..82261e25c --- /dev/null +++ b/youtube_dl/extractor/dcn.py @@ -0,0 +1,84 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse, + compat_urllib_request, +) +from ..utils import ( + int_or_none, + parse_iso8601, +) + + +class DCNIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?(?:video/.+|show/\d+/.+?)/(?P<id>\d+)' + _TEST = { + 'url': 'http://www.dcndigital.ae/#/show/199074/%D8%B1%D8%AD%D9%84%D8%A9-%D8%A7%D9%84%D8%B9%D9%85%D8%B1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/17375/6887', + 'info_dict': + { + 'id': '17375', + 'ext': 'mp4', + 'title': 'رحلة العمر : الحلقة 1', + 'description': 'md5:0156e935d870acb8ef0a66d24070c6d6', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 2041, + 'timestamp': 1227504126, + 'upload_date': '20081124', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + request = compat_urllib_request.Request( + 'http://admin.mangomolo.com/analytics/index.php/plus/video?id=%s' % video_id, + headers={'Origin': 'http://www.dcndigital.ae'}) + + video = self._download_json(request, video_id) + title = video.get('title_en') or video['title_ar'] + + webpage = self._download_webpage( + 'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' + + compat_urllib_parse.urlencode({ + 'id': video['id'], + 'user_id': video['user_id'], + 'signature': video['signature'], + 'countries': 'Q0M=', + 'filter': 'DENY', + }), video_id) + + m3u8_url = self._html_search_regex(r'file:\s*"([^"]+)', webpage, 'm3u8 url') + formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls') + + rtsp_url = self._search_regex( + r'<a[^>]+href="(rtsp://[^"]+)"', webpage, 'rtsp url', fatal=False) + if rtsp_url: + formats.append({ + 'url': rtsp_url, + 'format_id': 'rtsp', + }) + + self._sort_formats(formats) + + img = video.get('img') + thumbnail = 'http://admin.mangomolo.com/analytics/%s' % img if img else None + duration = int_or_none(video.get('duration')) + description = video.get('description_en') or video.get('description_ar') + timestamp = parse_iso8601(video.get('create_time') or video.get('update_time'), ' ') + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'timestamp': timestamp, + 'formats': formats, + } diff --git a/youtube_dl/extractor/dhm.py b/youtube_dl/extractor/dhm.py index 3ed1f1663..44e0c5d4d 100644 --- a/youtube_dl/extractor/dhm.py +++ b/youtube_dl/extractor/dhm.py @@ -1,10 +1,7 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - xpath_text, - parse_duration, -) +from ..utils import parse_duration class DHMIE(InfoExtractor): @@ -34,24 +31,14 @@ class DHMIE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) + playlist_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage(url, playlist_id) playlist_url = self._search_regex( r"file\s*:\s*'([^']+)'", webpage, 'playlist url') - playlist = self._download_xml(playlist_url, video_id) - - track = playlist.find( - './{http://xspf.org/ns/0/}trackList/{http://xspf.org/ns/0/}track') - - video_url = xpath_text( - track, './{http://xspf.org/ns/0/}location', - 'video url', fatal=True) - thumbnail = xpath_text( - track, './{http://xspf.org/ns/0/}image', - 'thumbnail') + entries = self._extract_xspf_playlist(playlist_url, playlist_id) title = self._search_regex( [r'dc:title="([^"]+)"', r'<title> »([^<]+)</title>'], @@ -63,11 +50,10 @@ class DHMIE(InfoExtractor): r'<em>Length\s*</em>\s*:\s*</strong>([^<]+)', webpage, 'duration', default=None)) - return { - 'id': video_id, - 'url': video_url, + entries[0].update({ 'title': title, 'description': description, 'duration': duration, - 'thumbnail': thumbnail, - } + }) + + return self.playlist_result(entries, playlist_id) diff --git a/youtube_dl/extractor/esri.py b/youtube_dl/extractor/esri.py new file mode 100644 index 000000000..bf5d2019f --- /dev/null +++ b/youtube_dl/extractor/esri.py @@ -0,0 +1,74 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import compat_urlparse +from ..utils import ( + int_or_none, + parse_filesize, + unified_strdate, +) + + +class EsriVideoIE(InfoExtractor): + _VALID_URL = r'https?://video\.esri\.com/watch/(?P<id>[0-9]+)' + _TEST = { + 'url': 'https://video.esri.com/watch/1124/arcgis-online-_dash_-developing-applications', + 'md5': 'd4aaf1408b221f1b38227a9bbaeb95bc', + 'info_dict': { + 'id': '1124', + 'ext': 'mp4', + 'title': 'ArcGIS Online - Developing Applications', + 'description': 'Jeremy Bartley demonstrates how to develop applications with ArcGIS Online.', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 185, + 'upload_date': '20120419', + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + formats = [] + for width, height, content in re.findall( + r'(?s)<li><strong>(\d+)x(\d+):</strong>(.+?)</li>', webpage): + for video_url, ext, filesize in re.findall( + r'<a[^>]+href="([^"]+)">([^<]+) \(([^<]+)\)</a>', content): + formats.append({ + 'url': compat_urlparse.urljoin(url, video_url), + 'ext': ext.lower(), + 'format_id': '%s-%s' % (ext.lower(), height), + 'width': int(width), + 'height': int(height), + 'filesize_approx': parse_filesize(filesize), + }) + self._sort_formats(formats) + + title = self._html_search_meta('title', webpage, 'title') + description = self._html_search_meta( + 'description', webpage, 'description', fatal=False) + + thumbnail = self._html_search_meta('thumbnail', webpage, 'thumbnail', fatal=False) + if thumbnail: + thumbnail = re.sub(r'_[st]\.jpg$', '_x.jpg', thumbnail) + + duration = int_or_none(self._search_regex( + [r'var\s+videoSeconds\s*=\s*(\d+)', r"'duration'\s*:\s*(\d+)"], + webpage, 'duration', fatal=False)) + + upload_date = unified_strdate(self._html_search_meta( + 'last-modified', webpage, 'upload date', fatal=None)) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'duration': duration, + 'upload_date': upload_date, + 'formats': formats + } diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index e17bb9aea..178a7ca4c 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -17,6 +17,8 @@ from ..utils import ( int_or_none, limit_length, urlencode_postdata, + get_element_by_id, + clean_html, ) @@ -42,6 +44,7 @@ class FacebookIE(InfoExtractor): 'id': '637842556329505', 'ext': 'mp4', 'title': 're:Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam', + 'uploader': 'Tennis on Facebook', } }, { 'note': 'Video without discernible title', @@ -50,6 +53,7 @@ class FacebookIE(InfoExtractor): 'id': '274175099429670', 'ext': 'mp4', 'title': 'Facebook video #274175099429670', + 'uploader': 'Asif Nawab Butt', }, 'expected_warnings': [ 'title' @@ -161,6 +165,7 @@ class FacebookIE(InfoExtractor): video_title = limit_length(video_title, 80) if not video_title: video_title = 'Facebook video #%s' % video_id + uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage)) return { 'id': video_id, @@ -168,4 +173,5 @@ class FacebookIE(InfoExtractor): 'formats': formats, 'duration': int_or_none(video_data.get('video_duration')), 'thumbnail': video_data.get('thumbnail_src'), + 'uploader': uploader, } diff --git a/youtube_dl/extractor/fc2.py b/youtube_dl/extractor/fc2.py index 1ccc1a964..e4f7195a8 100644 --- a/youtube_dl/extractor/fc2.py +++ b/youtube_dl/extractor/fc2.py @@ -86,7 +86,7 @@ class FC2IE(InfoExtractor): info_url = ( "http://video.fc2.com/ginfo.php?mimi={1:s}&href={2:s}&v={0:s}&fversion=WIN%2011%2C6%2C602%2C180&from=2&otag=0&upid={0:s}&tk=null&". - format(video_id, mimi, compat_urllib_request.quote(refer, safe='').replace('.', '%2E'))) + format(video_id, mimi, compat_urllib_request.quote(refer, safe=b'').replace('.', '%2E'))) info_webpage = self._download_webpage( info_url, video_id, note='Downloading info page') diff --git a/youtube_dl/extractor/fourtube.py b/youtube_dl/extractor/fourtube.py index b2284ab01..3bb4f6239 100644 --- a/youtube_dl/extractor/fourtube.py +++ b/youtube_dl/extractor/fourtube.py @@ -32,6 +32,7 @@ class FourTubeIE(InfoExtractor): 'view_count': int, 'like_count': int, 'categories': list, + 'age_limit': 18, } } diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index b2c984bf2..75723c00d 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -6,15 +6,11 @@ import re import json from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( clean_html, ExtractorError, int_or_none, - float_or_none, parse_duration, determine_ext, ) @@ -59,12 +55,12 @@ class FranceTVBaseInfoExtractor(InfoExtractor): # See https://github.com/rg3/youtube-dl/issues/3963 # m3u8 urls work fine continue - video_url_parsed = compat_urllib_parse_urlparse(video_url) f4m_url = self._download_webpage( - 'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url_parsed.path, + 'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url, video_id, 'Downloading f4m manifest token', fatal=False) if f4m_url: - formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id)) + formats.extend(self._extract_f4m_formats( + f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, 1, format_id)) elif ext == 'm3u8': formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4', m3u8_id=format_id)) elif video_url.startswith('rtmp'): @@ -87,7 +83,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor): 'title': info['titre'], 'description': clean_html(info['synopsis']), 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']), - 'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']), + 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']), 'timestamp': int_or_none(info['diffusion']['timestamp']), 'formats': formats, } @@ -160,11 +156,21 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor): class FranceTVIE(FranceTVBaseInfoExtractor): IE_NAME = 'francetv' IE_DESC = 'France 2, 3, 4, 5 and Ô' - _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/ - (?: - emissions/.*?/(videos|emissions)/(?P<id>[^/?]+) - | (emissions?|jt)/(?P<key>[^/?]+) - )''' + _VALID_URL = r'''(?x) + https?:// + (?: + (?:www\.)?france[2345o]\.fr/ + (?: + emissions/[^/]+/(?:videos|diffusions)| + emission/[^/]+| + videos| + jt + ) + /| + embed\.francetv\.fr/\?ue= + ) + (?P<id>[^/?]+) + ''' _TESTS = [ # france2 @@ -221,24 +227,46 @@ class FranceTVIE(FranceTVBaseInfoExtractor): }, # franceo { - 'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013', - 'md5': '52f0bfe202848b15915a2f39aaa8981b', + 'url': 'http://www.franceo.fr/jt/info-soir/18-07-2015', + 'md5': '47d5816d3b24351cdce512ad7ab31da8', 'info_dict': { - 'id': '108634970', + 'id': '125377621', 'ext': 'flv', - 'title': 'Infô Afrique', - 'description': 'md5:ebf346da789428841bee0fd2a935ea55', - 'upload_date': '20140915', - 'timestamp': 1410822000, + 'title': 'Infô soir', + 'description': 'md5:01b8c6915a3d93d8bbbd692651714309', + 'upload_date': '20150718', + 'timestamp': 1437241200, + 'duration': 414, + }, + }, + { + # francetv embed + 'url': 'http://embed.francetv.fr/?ue=8d7d3da1e3047c42ade5a5d7dfd3fc87', + 'info_dict': { + 'id': 'EV_30231', + 'ext': 'flv', + 'title': 'Alcaline, le concert avec Calogero', + 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff', + 'upload_date': '20150226', + 'timestamp': 1424989860, + 'duration': 5400, }, }, + { + 'url': 'http://www.france4.fr/emission/highlander/diffusion-du-17-07-2015-04h05', + 'only_matching': True, + }, + { + 'url': 'http://www.franceo.fr/videos/125377617', + 'only_matching': True, + } ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id')) + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) video_id, catalogue = self._html_search_regex( - r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', + r'href="http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video ID').split('@') return self._extract_video(video_id, catalogue) diff --git a/youtube_dl/extractor/funnyordie.py b/youtube_dl/extractor/funnyordie.py index dd87257c4..f5f13689c 100644 --- a/youtube_dl/extractor/funnyordie.py +++ b/youtube_dl/extractor/funnyordie.py @@ -53,7 +53,7 @@ class FunnyOrDieIE(InfoExtractor): for bitrate in bitrates: for link in links: formats.append({ - 'url': '%s%d.%s' % (link[0], bitrate, link[1]), + 'url': self._proto_relative_url('%s%d.%s' % (link[0], bitrate, link[1])), 'format_id': '%s-%d' % (link[1], bitrate), 'vbr': bitrate, }) diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index 43f916412..a6834db43 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -7,7 +7,10 @@ from ..compat import ( compat_urllib_parse, compat_urllib_request, ) -from ..utils import remove_end +from ..utils import ( + remove_end, + HEADRequest, +) class GDCVaultIE(InfoExtractor): @@ -73,10 +76,20 @@ class GDCVaultIE(InfoExtractor): return video_formats def _parse_flv(self, xml_description): - video_formats = [] + formats = [] akamai_url = xml_description.find('./metadata/akamaiHost').text + audios = xml_description.find('./metadata/audios') + if audios is not None: + for audio in audios: + formats.append({ + 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, + 'play_path': remove_end(audio.get('url'), '.flv'), + 'ext': 'flv', + 'vcodec': 'none', + 'format_id': audio.get('code'), + }) slide_video_path = xml_description.find('./metadata/slideVideo').text - video_formats.append({ + formats.append({ 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, 'play_path': remove_end(slide_video_path, '.flv'), 'ext': 'flv', @@ -86,7 +99,7 @@ class GDCVaultIE(InfoExtractor): 'format_id': 'slides', }) speaker_video_path = xml_description.find('./metadata/speakerVideo').text - video_formats.append({ + formats.append({ 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, 'play_path': remove_end(speaker_video_path, '.flv'), 'ext': 'flv', @@ -95,7 +108,7 @@ class GDCVaultIE(InfoExtractor): 'preference': -1, 'format_id': 'speaker', }) - return video_formats + return formats def _login(self, webpage_url, display_id): (username, password) = self._get_login_info() @@ -133,16 +146,18 @@ class GDCVaultIE(InfoExtractor): r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);', start_page, 'url', default=None) if direct_url: - video_url = 'http://www.gdcvault.com/' + direct_url title = self._html_search_regex( r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>', start_page, 'title') + video_url = 'http://www.gdcvault.com' + direct_url + # resolve the url so that we can detect the correct extension + head = self._request_webpage(HEADRequest(video_url), video_id) + video_url = head.geturl() return { 'id': video_id, 'display_id': display_id, 'url': video_url, - 'ext': 'flv', 'title': title, } @@ -168,8 +183,8 @@ class GDCVaultIE(InfoExtractor): # Fallback to the older format xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename') - xml_decription_url = xml_root + 'xml/' + xml_name - xml_description = self._download_xml(xml_decription_url, display_id) + xml_description_url = xml_root + 'xml/' + xml_name + xml_description = self._download_xml(xml_description_url, display_id) video_title = xml_description.find('./metadata/title').text video_formats = self._parse_mp4(xml_description) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index a62287e50..376feecae 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -130,6 +130,89 @@ class GenericIE(InfoExtractor): 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624', } }, + # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng + { + 'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml', + 'info_dict': { + 'id': 'smil', + 'ext': 'mp4', + 'title': 'Automatics, robotics and biocybernetics', + 'description': 'md5:815fc1deb6b3a2bff99de2d5325be482', + 'formats': 'mincount:16', + 'subtitles': 'mincount:1', + }, + 'params': { + 'force_generic_extractor': True, + 'skip_download': True, + }, + }, + # SMIL from http://www1.wdr.de/mediathek/video/livestream/index.html + { + 'url': 'http://metafilegenerator.de/WDR/WDR_FS/hds/hds.smil', + 'info_dict': { + 'id': 'hds', + 'ext': 'flv', + 'title': 'hds', + 'formats': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from https://www.restudy.dk/video/play/id/1637 + { + 'url': 'https://www.restudy.dk/awsmedia/SmilDirectory/video_1637.xml', + 'info_dict': { + 'id': 'video_1637', + 'ext': 'flv', + 'title': 'video_1637', + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm + { + 'url': 'http://services.media.howstuffworks.com/videos/450221/smil-service.smil', + 'info_dict': { + 'id': 'smil-service', + 'ext': 'flv', + 'title': 'smil-service', + 'formats': 'mincount:1', + }, + 'params': { + 'skip_download': True, + }, + }, + # SMIL from http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370 + { + 'url': 'http://api.new.livestream.com/accounts/1570303/events/1585861/videos/4719370.smil', + 'info_dict': { + 'id': '4719370', + 'ext': 'mp4', + 'title': '571de1fd-47bc-48db-abf9-238872a58d1f', + 'formats': 'mincount:3', + }, + 'params': { + 'skip_download': True, + }, + }, + # XSPF playlist from http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html + { + 'url': 'http://www.telegraaf.nl/xml/playlist/2015/8/7/mZlp2ctYIUEB.xspf', + 'info_dict': { + 'id': 'mZlp2ctYIUEB', + 'ext': 'mp4', + 'title': 'Tikibad ontruimd wegens brand', + 'description': 'md5:05ca046ff47b931f9b04855015e163a4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 33, + }, + 'params': { + 'skip_download': True, + }, + }, # google redirect { 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE', @@ -236,6 +319,19 @@ class GenericIE(InfoExtractor): }, 'add_ie': ['Ooyala'], }, + { + # ooyala video embedded with http://player.ooyala.com/iframe.js + 'url': 'http://www.macrumors.com/2015/07/24/steve-jobs-the-man-in-the-machine-first-trailer/', + 'info_dict': { + 'id': 'p0MGJndjoG5SOKqO_hZJuZFPB-Tr5VgB', + 'ext': 'mp4', + 'title': '"Steve Jobs: Man in the Machine" trailer', + 'description': 'The first trailer for the Alex Gibney documentary "Steve Jobs: Man in the Machine."', + }, + 'params': { + 'skip_download': True, + }, + }, # multiple ooyala embeds on SBN network websites { 'url': 'http://www.sbnation.com/college-football-recruiting/2015/2/3/7970291/national-signing-day-rationalizations-itll-be-ok-itll-be-ok', @@ -276,14 +372,6 @@ class GenericIE(InfoExtractor): 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.', }, }, - # BBC iPlayer embeds - { - 'url': 'http://www.bbc.co.uk/blogs/adamcurtis/posts/BUGGER', - 'info_dict': { - 'title': 'BBC - Blogs - Adam Curtis - BUGGER', - }, - 'playlist_mincount': 18, - }, # RUTV embed { 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html', @@ -407,6 +495,26 @@ class GenericIE(InfoExtractor): 'skip_download': 'Requires rtmpdump' } }, + # francetv embed + { + 'url': 'http://www.tsprod.com/replay-du-concert-alcaline-de-calogero', + 'info_dict': { + 'id': 'EV_30231', + 'ext': 'mp4', + 'title': 'Alcaline, le concert avec Calogero', + 'description': 'md5:61f08036dcc8f47e9cfc33aed08ffaff', + 'upload_date': '20150226', + 'timestamp': 1424989860, + 'duration': 5400, + }, + 'params': { + # m3u8 downloads + 'skip_download': True, + }, + 'expected_warnings': [ + 'Forbidden' + ] + }, # Condé Nast embed { 'url': 'http://www.wired.com/2014/04/honda-asimo/', @@ -1098,11 +1206,15 @@ class GenericIE(InfoExtractor): self.report_extraction(video_id) - # Is it an RSS feed? + # Is it an RSS feed, a SMIL file or a XSPF playlist? try: doc = parse_xml(webpage) if doc.tag == 'rss': return self._extract_rss(url, video_id, doc) + elif re.match(r'^(?:{[^}]+})?smil$', doc.tag): + return self._parse_smil(doc, url, video_id) + elif doc.tag == '{http://xspf.org/ns/0/}playlist': + return self.playlist_result(self._parse_xspf(doc, video_id), video_id) except compat_xml_parse_error: pass @@ -1176,6 +1288,12 @@ class GenericIE(InfoExtractor): if vimeo_url is not None: return self.url_result(vimeo_url) + vid_me_embed_url = self._search_regex( + r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]', + webpage, 'vid.me embed', default=None) + if vid_me_embed_url is not None: + return self.url_result(vid_me_embed_url, 'Vidme') + # Look for embedded YouTube player matches = re.findall(r'''(?x) (?: @@ -1302,7 +1420,7 @@ class GenericIE(InfoExtractor): return self.url_result(mobj.group('url')) # Look for Ooyala videos - mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or + mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage) or re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage) or re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage)) @@ -1431,6 +1549,13 @@ class GenericIE(InfoExtractor): if mobj is not None: return self.url_result(mobj.group('url'), 'ArteTVEmbed') + # Look for embedded francetv player + mobj = re.search( + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?://)?embed\.francetv\.fr/\?ue=.+?)\1', + webpage) + if mobj is not None: + return self.url_result(mobj.group('url')) + # Look for embedded smotri.com player smotri_url = SmotriIE._extract_url(webpage) if smotri_url: @@ -1630,7 +1755,7 @@ class GenericIE(InfoExtractor): if not found: # Broaden the findall a little bit: JWPlayer JS loader found = filter_video(re.findall( - r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)) + r'[^A-Za-z0-9]?(?:file|video_url)["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)) if not found: # Flow player found = filter_video(re.findall(r'''(?xs) @@ -1669,7 +1794,7 @@ class GenericIE(InfoExtractor): if refresh_header: found = re.search(REDIRECT_REGEX, refresh_header) if found: - new_url = compat_urlparse.urljoin(url, found.group(1)) + new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1))) self.report_following_redirect(new_url) return { '_type': 'url', @@ -1691,7 +1816,8 @@ class GenericIE(InfoExtractor): # here's a fun little line of code for you: video_id = os.path.splitext(video_id)[0] - if determine_ext(video_url) == 'smil': + ext = determine_ext(video_url) + if ext == 'smil': entries.append({ 'id': video_id, 'formats': self._extract_smil_formats(video_url, video_id), @@ -1699,6 +1825,8 @@ class GenericIE(InfoExtractor): 'title': video_title, 'age_limit': age_limit, }) + elif ext == 'xspf': + return self.playlist_result(self._extract_xspf_playlist(video_url, video_id), video_id) else: entries.append({ 'id': video_id, diff --git a/youtube_dl/extractor/indavideo.py b/youtube_dl/extractor/indavideo.py new file mode 100644 index 000000000..12fb5e8e1 --- /dev/null +++ b/youtube_dl/extractor/indavideo.py @@ -0,0 +1,142 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_age_limit, + parse_iso8601, +) + + +class IndavideoEmbedIE(InfoExtractor): + _VALID_URL = r'https?://(?:(?:embed\.)?indavideo\.hu/player/video/|assets\.indavideo\.hu/swf/player\.swf\?.*\b(?:v(?:ID|id))=)(?P<id>[\da-f]+)' + _TESTS = [{ + 'url': 'http://indavideo.hu/player/video/1bdc3c6d80/', + 'md5': 'f79b009c66194acacd40712a6778acfa', + 'info_dict': { + 'id': '1837039', + 'ext': 'mp4', + 'title': 'Cicatánc', + 'description': '', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'cukiajanlo', + 'uploader_id': '83729', + 'timestamp': 1439193826, + 'upload_date': '20150810', + 'duration': 72, + 'age_limit': 0, + 'tags': ['tánc', 'cica', 'cuki', 'cukiajanlo', 'newsroom'], + }, + }, { + 'url': 'http://embed.indavideo.hu/player/video/1bdc3c6d80?autostart=1&hide=1', + 'only_matching': True, + }, { + 'url': 'http://assets.indavideo.hu/swf/player.swf?v=fe25e500&vID=1bdc3c6d80&autostart=1&hide=1&i=1', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + 'http://amfphp.indavideo.hu/SYm0json.php/player.playerHandler.getVideoData/%s' % video_id, + video_id)['data'] + + title = video['title'] + + video_urls = video.get('video_files', []) + video_file = video.get('video_file') + if video: + video_urls.append(video_file) + video_urls = list(set(video_urls)) + + video_prefix = video_urls[0].rsplit('/', 1)[0] + + for flv_file in video.get('flv_files', []): + flv_url = '%s/%s' % (video_prefix, flv_file) + if flv_url not in video_urls: + video_urls.append(flv_url) + + formats = [{ + 'url': video_url, + 'height': self._search_regex(r'\.(\d{3,4})\.mp4$', video_url, 'height', default=None), + } for video_url in video_urls] + self._sort_formats(formats) + + timestamp = video.get('date') + if timestamp: + # upload date is in CEST + timestamp = parse_iso8601(timestamp + ' +0200', ' ') + + thumbnails = [{ + 'url': self._proto_relative_url(thumbnail) + } for thumbnail in video.get('thumbnails', [])] + + tags = [tag['title'] for tag in video.get('tags', [])] + + return { + 'id': video.get('id') or video_id, + 'title': title, + 'description': video.get('description'), + 'thumbnails': thumbnails, + 'uploader': video.get('user_name'), + 'uploader_id': video.get('user_id'), + 'timestamp': timestamp, + 'duration': int_or_none(video.get('length')), + 'age_limit': parse_age_limit(video.get('age_limit')), + 'tags': tags, + 'formats': formats, + } + + +class IndavideoIE(InfoExtractor): + _VALID_URL = r'https?://(?:.+?\.)?indavideo\.hu/video/(?P<id>[^/#?]+)' + _TESTS = [{ + 'url': 'http://indavideo.hu/video/Vicces_cica_1', + 'md5': '8c82244ba85d2a2310275b318eb51eac', + 'info_dict': { + 'id': '1335611', + 'display_id': 'Vicces_cica_1', + 'ext': 'mp4', + 'title': 'Vicces cica', + 'description': 'Játszik a tablettel. :D', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'Jet_Pack', + 'uploader_id': '491217', + 'timestamp': 1390821212, + 'upload_date': '20140127', + 'duration': 7, + 'age_limit': 0, + 'tags': ['vicces', 'macska', 'cica', 'ügyes', 'nevetés', 'játszik', 'Cukiság', 'Jet_Pack'], + }, + }, { + 'url': 'http://index.indavideo.hu/video/2015_0728_beregszasz', + 'only_matching': True, + }, { + 'url': 'http://auto.indavideo.hu/video/Sajat_utanfutoban_a_kis_tacsko', + 'only_matching': True, + }, { + 'url': 'http://erotika.indavideo.hu/video/Amator_tini_punci', + 'only_matching': True, + }, { + 'url': 'http://film.indavideo.hu/video/f_hrom_nagymamm_volt', + 'only_matching': True, + }, { + 'url': 'http://palyazat.indavideo.hu/video/Embertelen_dal_Dodgem_egyuttes', + 'only_matching': True, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + embed_url = self._search_regex( + r'<link[^>]+rel="video_src"[^>]+href="(.+?)"', webpage, 'embed url') + + return { + '_type': 'url_transparent', + 'ie_key': 'IndavideoEmbed', + 'url': embed_url, + 'display_id': display_id, + } diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index 0f6707d7c..393e67e35 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -3,19 +3,13 @@ from __future__ import unicode_literals import hashlib import math -import os.path import random -import re import time import uuid -import zlib from .common import InfoExtractor from ..compat import compat_urllib_parse -from ..utils import ( - ExtractorError, - url_basename, -) +from ..utils import ExtractorError class IqiyiIE(InfoExtractor): @@ -39,62 +33,57 @@ class IqiyiIE(InfoExtractor): 'title': '名侦探柯南第752集', }, 'playlist': [{ - 'md5': '7e49376fecaffa115d951634917fe105', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '41b75ba13bb7ac0e411131f92bc4f6ca', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '0cee1dd0a3d46a83e71e2badeae2aab0', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '4f8ad72373b0c491b582e7c196b0b1f9', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': 'd89ad028bcfad282918e8098e811711d', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '9cb1e5c95da25dff0660c32ae50903b7', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '155116e0ff1867bbc9b98df294faabc9', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }, { - 'md5': '53f5db77622ae14fa493ed2a278a082b', 'info_dict': { 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8', 'ext': 'f4v', 'title': '名侦探柯南第752集', }, }], + 'params': { + 'skip_download': True, + }, }] _FORMATS_MAP = [ @@ -212,20 +201,7 @@ class IqiyiIE(InfoExtractor): return raw_data def get_enc_key(self, swf_url, video_id): - filename, _ = os.path.splitext(url_basename(swf_url)) - enc_key_json = self._downloader.cache.load('iqiyi-enc-key', filename) - if enc_key_json is not None: - return enc_key_json[0] - - req = self._request_webpage( - swf_url, video_id, note='download swf content') - cn = req.read() - cn = zlib.decompress(cn[8:]) - pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv') - enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8') - - self._downloader.cache.store('iqiyi-enc-key', filename, [enc_key]) - + enc_key = '3601ba290e4f4662848c710e2122007e' # last update at 2015-08-10 for Zombie return enc_key def _real_extract(self, url): diff --git a/youtube_dl/extractor/ir90tv.py b/youtube_dl/extractor/ir90tv.py new file mode 100644 index 000000000..214bcd5b5 --- /dev/null +++ b/youtube_dl/extractor/ir90tv.py @@ -0,0 +1,42 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import remove_start + + +class Ir90TvIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?90tv\.ir/video/(?P<id>[0-9]+)/.*' + _TESTS = [{ + 'url': 'http://90tv.ir/video/95719/%D8%B4%D8%A7%DB%8C%D8%B9%D8%A7%D8%AA-%D9%86%D9%82%D9%84-%D9%88-%D8%A7%D9%86%D8%AA%D9%82%D8%A7%D9%84%D8%A7%D8%AA-%D9%85%D9%87%D9%85-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7-940218', + 'md5': '411dbd94891381960cb9e13daa47a869', + 'info_dict': { + 'id': '95719', + 'ext': 'mp4', + 'title': 'شایعات نقل و انتقالات مهم فوتبال اروپا 94/02/18', + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'url': 'http://www.90tv.ir/video/95719/%D8%B4%D8%A7%DB%8C%D8%B9%D8%A7%D8%AA-%D9%86%D9%82%D9%84-%D9%88-%D8%A7%D9%86%D8%AA%D9%82%D8%A7%D9%84%D8%A7%D8%AA-%D9%85%D9%87%D9%85-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7-940218', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + title = remove_start(self._html_search_regex( + r'<title>([^<]+)</title>', webpage, 'title'), '90tv.ir :: ') + + video_url = self._search_regex( + r'<source[^>]+src="([^"]+)"', webpage, 'video url') + + thumbnail = self._search_regex(r'poster="([^"]+)"', webpage, 'thumbnail url', fatal=False) + + return { + 'url': video_url, + 'id': video_id, + 'title': title, + 'video_url': video_url, + 'thumbnail': thumbnail, + } diff --git a/youtube_dl/extractor/lecture2go.py b/youtube_dl/extractor/lecture2go.py new file mode 100644 index 000000000..40a3d2346 --- /dev/null +++ b/youtube_dl/extractor/lecture2go.py @@ -0,0 +1,62 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + parse_duration, + int_or_none, +) + + +class Lecture2GoIE(InfoExtractor): + _VALID_URL = r'https?://lecture2go\.uni-hamburg\.de/veranstaltungen/-/v/(?P<id>\d+)' + _TEST = { + 'url': 'https://lecture2go.uni-hamburg.de/veranstaltungen/-/v/17473', + 'md5': 'ac02b570883020d208d405d5a3fd2f7f', + 'info_dict': { + 'id': '17473', + 'ext': 'flv', + 'title': '2 - Endliche Automaten und reguläre Sprachen', + 'creator': 'Frank Heitmann', + 'duration': 5220, + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + title = self._html_search_regex(r'<em[^>]+class="title">(.+)</em>', webpage, 'title') + + formats = [] + for url in set(re.findall(r'"src","([^"]+)"', webpage)): + ext = determine_ext(url) + if ext == 'f4m': + formats.extend(self._extract_f4m_formats(url, video_id)) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats(url, video_id)) + else: + formats.append({ + 'url': url, + }) + + self._sort_formats(formats) + + creator = self._html_search_regex( + r'<div[^>]+id="description">([^<]+)</div>', webpage, 'creator', fatal=False) + duration = parse_duration(self._html_search_regex( + r'Duration:\s*</em>\s*<em[^>]*>([^<]+)</em>', webpage, 'duration', fatal=False)) + view_count = int_or_none(self._html_search_regex( + r'Views:\s*</em>\s*<em[^>]+>(\d+)</em>', webpage, 'view count', fatal=False)) + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + 'creator': creator, + 'duration': duration, + 'view_count': view_count, + } diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index ba2ae8085..a28abb0f0 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -15,6 +15,7 @@ from ..utils import ( determine_ext, ExtractorError, parse_iso8601, + int_or_none, ) @@ -134,7 +135,7 @@ class LetvIE(InfoExtractor): } if format_id[-1:] == 'p': - url_info_dict['height'] = format_id[:-1] + url_info_dict['height'] = int_or_none(format_id[:-1]) urls.append(url_info_dict) diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index a00f6e5e5..5b9157ed4 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -11,13 +11,13 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + clean_html, int_or_none, ) class LyndaBaseIE(InfoExtractor): _LOGIN_URL = 'https://www.lynda.com/login/login.aspx' - _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true' _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.' _NETRC_MACHINE = 'lynda' @@ -41,7 +41,7 @@ class LyndaBaseIE(InfoExtractor): request, None, 'Logging in as %s' % username) # Not (yet) logged in - m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page) + m = re.search(r'loginResultJson\s*=\s*\'(?P<json>[^\']+)\';', login_page) if m is not None: response = m.group('json') response_json = json.loads(response) @@ -70,7 +70,16 @@ class LyndaBaseIE(InfoExtractor): request, None, 'Confirming log in and log out from another device') - if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None: + if all(not re.search(p, login_page) for p in ('isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')): + if 'login error' in login_page: + mobj = re.search( + r'(?s)<h1[^>]+class="topmost">(?P<title>[^<]+)</h1>\s*<div>(?P<description>.+?)</div>', + login_page) + if mobj: + raise ExtractorError( + 'lynda returned error: %s - %s' + % (mobj.group('title'), clean_html(mobj.group('description'))), + expected=True) raise ExtractorError('Unable to log in') diff --git a/youtube_dl/extractor/mdr.py b/youtube_dl/extractor/mdr.py index 5fdd19027..fc7499958 100644 --- a/youtube_dl/extractor/mdr.py +++ b/youtube_dl/extractor/mdr.py @@ -29,7 +29,7 @@ class MDRIE(InfoExtractor): doc = self._download_xml(domain + xmlurl, video_id) formats = [] for a in doc.findall('./assets/asset'): - url_el = a.find('.//progressiveDownloadUrl') + url_el = a.find('./progressiveDownloadUrl') if url_el is None: continue abr = int(a.find('bitrateAudio').text) // 1000 diff --git a/youtube_dl/extractor/moniker.py b/youtube_dl/extractor/moniker.py index 88dcd4f73..69e4bcd1a 100644 --- a/youtube_dl/extractor/moniker.py +++ b/youtube_dl/extractor/moniker.py @@ -9,7 +9,10 @@ from ..compat import ( compat_urllib_parse, compat_urllib_request, ) -from ..utils import ExtractorError +from ..utils import ( + ExtractorError, + remove_start, +) class MonikerIE(InfoExtractor): @@ -25,6 +28,14 @@ class MonikerIE(InfoExtractor): 'title': 'youtube-dl test video', }, }, { + 'url': 'http://allmyvideos.net/embed-jih3nce3x6wn', + 'md5': '710883dee1bfc370ecf9fa6a89307c88', + 'info_dict': { + 'id': 'jih3nce3x6wn', + 'ext': 'mp4', + 'title': 'youtube-dl test video', + }, + }, { 'url': 'http://vidspot.net/l2ngsmhs8ci5', 'md5': '710883dee1bfc370ecf9fa6a89307c88', 'info_dict': { @@ -38,7 +49,10 @@ class MonikerIE(InfoExtractor): }] def _real_extract(self, url): - video_id = self._match_id(url) + orig_video_id = self._match_id(url) + video_id = remove_start(orig_video_id, 'embed-') + url = url.replace(orig_video_id, video_id) + assert re.match(self._VALID_URL, url) is not None orig_webpage = self._download_webpage(url, video_id) if '>File Not Found<' in orig_webpage: diff --git a/youtube_dl/extractor/nationalgeographic.py b/youtube_dl/extractor/nationalgeographic.py index f793b72f5..6fc9e7b05 100644 --- a/youtube_dl/extractor/nationalgeographic.py +++ b/youtube_dl/extractor/nationalgeographic.py @@ -8,18 +8,30 @@ from ..utils import ( class NationalGeographicIE(InfoExtractor): - _VALID_URL = r'http://video\.nationalgeographic\.com/video/.*?' - - _TEST = { - 'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo', - 'info_dict': { - 'id': '4DmDACA6Qtk_', - 'ext': 'flv', - 'title': 'Mating Crabs Busted by Sharks', - 'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3', + _VALID_URL = r'http://video\.nationalgeographic\.com/.*?' + + _TESTS = [ + { + 'url': 'http://video.nationalgeographic.com/video/news/150210-news-crab-mating-vin?source=featuredvideo', + 'info_dict': { + 'id': '4DmDACA6Qtk_', + 'ext': 'flv', + 'title': 'Mating Crabs Busted by Sharks', + 'description': 'md5:16f25aeffdeba55aaa8ec37e093ad8b3', + }, + 'add_ie': ['ThePlatform'], }, - 'add_ie': ['ThePlatform'], - } + { + 'url': 'http://video.nationalgeographic.com/wild/when-sharks-attack/the-real-jaws', + 'info_dict': { + 'id': '_JeBD_D7PlS5', + 'ext': 'flv', + 'title': 'The Real Jaws', + 'description': 'md5:8d3e09d9d53a85cd397b4b21b2c77be6', + }, + 'add_ie': ['ThePlatform'], + }, + ] def _real_extract(self, url): name = url_basename(url) @@ -37,5 +49,6 @@ class NationalGeographicIE(InfoExtractor): return self.url_result(smuggle_url( 'http://link.theplatform.com/s/ngs/%s?format=SMIL&formats=MPEG4&manifest=f4m' % theplatform_id, - # For some reason, the normal links don't work and we must force the use of f4m + # For some reason, the normal links don't work and we must force + # the use of f4m {'force_smil_url': True})) diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index dc2091be0..ccdbfb6c9 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -124,7 +124,7 @@ class NBCSportsIE(InfoExtractor): class NBCNewsIE(InfoExtractor): _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/ (?:video/.+?/(?P<id>\d+)| - (?:feature|nightly-news)/[^/]+/(?P<title>.+)) + (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+)) ''' _TESTS = [ @@ -169,6 +169,10 @@ class NBCNewsIE(InfoExtractor): 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5', }, }, + { + 'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952', + 'only_matching': True, + }, ] def _real_extract(self, url): diff --git a/youtube_dl/extractor/nowtv.py b/youtube_dl/extractor/nowtv.py index 0b5ff4760..66c627bec 100644 --- a/youtube_dl/extractor/nowtv.py +++ b/youtube_dl/extractor/nowtv.py @@ -1,12 +1,11 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import compat_str from ..utils import ( ExtractorError, + determine_ext, int_or_none, parse_iso8601, parse_duration, @@ -15,7 +14,7 @@ from ..utils import ( class NowTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?nowtv\.de/(?P<station>rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<id>.+?)/player' + _VALID_URL = r'https?://(?:www\.)?nowtv\.(?:de|at|ch)/(?:rtl|rtl2|rtlnitro|superrtl|ntv|vox)/(?P<id>.+?)/(?:player|preview)' _TESTS = [{ # rtl @@ -23,7 +22,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '203519', 'display_id': 'bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Die neuen Bauern und eine Hochzeit', 'description': 'md5:e234e1ed6d63cf06be5c070442612e7e', 'thumbnail': 're:^https?://.*\.jpg$', @@ -32,7 +31,7 @@ class NowTVIE(InfoExtractor): 'duration': 2786, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -41,7 +40,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '203481', 'display_id': 'berlin-tag-nacht/berlin-tag-nacht-folge-934', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Berlin - Tag & Nacht (Folge 934)', 'description': 'md5:c85e88c2e36c552dfe63433bc9506dd0', 'thumbnail': 're:^https?://.*\.jpg$', @@ -50,7 +49,7 @@ class NowTVIE(InfoExtractor): 'duration': 2641, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -59,7 +58,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '165780', 'display_id': 'alarm-fuer-cobra-11-die-autobahnpolizei/hals-und-beinbruch-2014-08-23-21-10-00', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Hals- und Beinbruch', 'description': 'md5:b50d248efffe244e6f56737f0911ca57', 'thumbnail': 're:^https?://.*\.jpg$', @@ -68,7 +67,7 @@ class NowTVIE(InfoExtractor): 'duration': 2742, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -77,7 +76,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '99205', 'display_id': 'medicopter-117/angst', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Angst!', 'description': 'md5:30cbc4c0b73ec98bcd73c9f2a8c17c4e', 'thumbnail': 're:^https?://.*\.jpg$', @@ -86,7 +85,7 @@ class NowTVIE(InfoExtractor): 'duration': 3025, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -95,7 +94,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '203521', 'display_id': 'ratgeber-geld/thema-ua-der-erste-blick-die-apple-watch', - 'ext': 'mp4', + 'ext': 'flv', 'title': 'Thema u.a.: Der erste Blick: Die Apple Watch', 'description': 'md5:4312b6c9d839ffe7d8caf03865a531af', 'thumbnail': 're:^https?://.*\.jpg$', @@ -104,7 +103,7 @@ class NowTVIE(InfoExtractor): 'duration': 1083, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, }, { @@ -113,7 +112,7 @@ class NowTVIE(InfoExtractor): 'info_dict': { 'id': '128953', 'display_id': 'der-hundeprofi/buero-fall-chihuahua-joel', - 'ext': 'mp4', + 'ext': 'flv', 'title': "Büro-Fall / Chihuahua 'Joel'", 'description': 'md5:e62cb6bf7c3cc669179d4f1eb279ad8d', 'thumbnail': 're:^https?://.*\.jpg$', @@ -122,15 +121,19 @@ class NowTVIE(InfoExtractor): 'duration': 3092, }, 'params': { - # m3u8 download + # rtmp download 'skip_download': True, }, + }, { + 'url': 'http://www.nowtv.de/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit/preview', + 'only_matching': True, + }, { + 'url': 'http://www.nowtv.at/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit/preview?return=/rtl/bauer-sucht-frau/die-neuen-bauern-und-eine-hochzeit', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('id') - station = mobj.group('station') + display_id = self._match_id(url) info = self._download_json( 'https://api.nowtv.de/v3/movies/%s?fields=id,title,free,geoblocked,articleLong,articleShort,broadcastStartDate,seoUrl,duration,format,files' % display_id, @@ -148,29 +151,19 @@ class NowTVIE(InfoExtractor): raise ExtractorError( 'Video %s is not available for free' % video_id, expected=True) - f = info.get('format', {}) - station = f.get('station') or station - - STATIONS = { - 'rtl': 'rtlnow', - 'rtl2': 'rtl2now', - 'vox': 'voxnow', - 'nitro': 'rtlnitronow', - 'ntv': 'n-tvnow', - 'superrtl': 'superrtlnow' - } - formats = [] for item in files['items']: - item_path = remove_start(item['path'], '/') - tbr = int_or_none(item['bitrate']) - m3u8_url = 'http://hls.fra.%s.de/hls-vod-enc/%s.m3u8' % (STATIONS[station], item_path) - m3u8_url = m3u8_url.replace('now/', 'now/videos/') + if determine_ext(item['path']) != 'f4v': + continue + app, play_path = remove_start(item['path'], '/').split('/', 1) formats.append({ - 'url': m3u8_url, - 'format_id': '%s-%sk' % (item['id'], tbr), - 'ext': 'mp4', - 'tbr': tbr, + 'url': 'rtmpe://fms.rtl.de', + 'app': app, + 'play_path': 'mp4:%s' % play_path, + 'ext': 'flv', + 'page_url': url, + 'player_url': 'http://rtl-now.rtl.de/includes/nc_player.swf', + 'tbr': int_or_none(item.get('bitrate')), }) self._sort_formats(formats) @@ -178,6 +171,8 @@ class NowTVIE(InfoExtractor): description = info.get('articleLong') or info.get('articleShort') timestamp = parse_iso8601(info.get('broadcastStartDate'), ' ') duration = parse_duration(info.get('duration')) + + f = info.get('format', {}) thumbnail = f.get('defaultImage169Format') or f.get('defaultImage169Logo') return { diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index 0c2d02c10..eb12fb810 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -407,6 +407,7 @@ class NPORadioFragmentIE(InfoExtractor): class VPROIE(NPOIE): + IE_NAME = 'vpro' _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html' _TESTS = [ diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py index 215ffe87b..003d27de7 100644 --- a/youtube_dl/extractor/odnoklassniki.py +++ b/youtube_dl/extractor/odnoklassniki.py @@ -16,15 +16,17 @@ class OdnoklassnikiIE(InfoExtractor): _TESTS = [{ # metadata in JSON 'url': 'http://ok.ru/video/20079905452', - 'md5': '8e24ad2da6f387948e7a7d44eb8668fe', + 'md5': '6ba728d85d60aa2e6dd37c9e70fdc6bc', 'info_dict': { 'id': '20079905452', 'ext': 'mp4', 'title': 'Культура меняет нас (прекрасный ролик!))', 'duration': 100, + 'upload_date': '20141207', 'uploader_id': '330537914540', 'uploader': 'Виталий Добровольский', 'like_count': int, + 'age_limit': 0, }, }, { # metadataUrl @@ -35,9 +37,11 @@ class OdnoklassnikiIE(InfoExtractor): 'ext': 'mp4', 'title': 'Девушка без комплексов ...', 'duration': 191, + 'upload_date': '20150518', 'uploader_id': '534380003155', - 'uploader': 'Андрей Мещанинов', + 'uploader': '☭ Андрей Мещанинов ☭', 'like_count': int, + 'age_limit': 0, }, }, { 'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452', diff --git a/youtube_dl/extractor/pbs.py b/youtube_dl/extractor/pbs.py index fec5d65ad..683c81de3 100644 --- a/youtube_dl/extractor/pbs.py +++ b/youtube_dl/extractor/pbs.py @@ -32,7 +32,7 @@ class PBSIE(InfoExtractor): 'info_dict': { 'id': '2365006249', 'ext': 'mp4', - 'title': 'A More Perfect Union', + 'title': 'Constitution USA with Peter Sagal - A More Perfect Union', 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a', 'duration': 3190, }, @@ -46,7 +46,7 @@ class PBSIE(InfoExtractor): 'info_dict': { 'id': '2365297690', 'ext': 'mp4', - 'title': 'Losing Iraq', + 'title': 'FRONTLINE - Losing Iraq', 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e', 'duration': 5050, }, @@ -60,7 +60,7 @@ class PBSIE(InfoExtractor): 'info_dict': { 'id': '2201174722', 'ext': 'mp4', - 'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist', + 'title': 'PBS NewsHour - Cyber Schools Gain Popularity, but Quality Questions Persist', 'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28', 'duration': 801, }, @@ -72,7 +72,7 @@ class PBSIE(InfoExtractor): 'id': '2365297708', 'ext': 'mp4', 'description': 'md5:68d87ef760660eb564455eb30ca464fe', - 'title': 'Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full', + 'title': 'Great Performances - Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full', 'duration': 6559, 'thumbnail': 're:^https?://.*\.jpg$', }, @@ -88,10 +88,11 @@ class PBSIE(InfoExtractor): 'display_id': 'killer-typhoon', 'ext': 'mp4', 'description': 'md5:c741d14e979fc53228c575894094f157', - 'title': 'Killer Typhoon', + 'title': 'NOVA - Killer Typhoon', 'duration': 3172, 'thumbnail': 're:^https?://.*\.jpg$', 'upload_date': '20140122', + 'age_limit': 10, }, 'params': { 'skip_download': True, # requires ffmpeg @@ -110,7 +111,7 @@ class PBSIE(InfoExtractor): 'id': '2280706814', 'display_id': 'player', 'ext': 'mp4', - 'title': 'Death and the Civil War', + 'title': 'American Experience - Death and the Civil War', 'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.', 'duration': 6705, 'thumbnail': 're:^https?://.*\.jpg$', @@ -118,6 +119,21 @@ class PBSIE(InfoExtractor): 'params': { 'skip_download': True, # requires ffmpeg }, + }, + { + 'url': 'http://video.pbs.org/video/2365367186/', + 'info_dict': { + 'id': '2365367186', + 'display_id': '2365367186', + 'ext': 'mp4', + 'title': 'To Catch A Comet - Full Episode', + 'description': 'On November 12, 2014, billions of kilometers from Earth, spacecraft orbiter Rosetta and lander Philae did what no other had dared to attempt \u2014 land on the volatile surface of a comet as it zooms around the sun at 67,000 km/hr. The European Space Agency hopes this mission can help peer into our past and unlock secrets of our origins.', + 'duration': 3342, + 'thumbnail': 're:^https?://.*\.jpg$', + }, + 'params': { + 'skip_download': True, # requires ffmpeg + }, } ] @@ -232,6 +248,12 @@ class PBSIE(InfoExtractor): 'url': closed_captions_url, }] + # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc) + # Try turning it to 'program - title' naming scheme if possible + alt_title = info.get('program', {}).get('title') + if alt_title: + info['title'] = alt_title + ' - ' + re.sub(r'^' + alt_title + '[\s\-:]+', '', info['title']) + return { 'id': video_id, 'display_id': display_id, diff --git a/youtube_dl/extractor/periscope.py b/youtube_dl/extractor/periscope.py new file mode 100644 index 000000000..8ad936758 --- /dev/null +++ b/youtube_dl/extractor/periscope.py @@ -0,0 +1,99 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import ( + compat_urllib_parse, + compat_urllib_request, +) +from ..utils import parse_iso8601 + + +class PeriscopeIE(InfoExtractor): + IE_DESC = 'Periscope' + _VALID_URL = r'https?://(?:www\.)?periscope\.tv/w/(?P<id>[^/?#]+)' + _TEST = { + 'url': 'https://www.periscope.tv/w/aJUQnjY3MjA3ODF8NTYxMDIyMDl2zCg2pECBgwTqRpQuQD352EMPTKQjT4uqlM3cgWFA-g==', + 'md5': '65b57957972e503fcbbaeed8f4fa04ca', + 'info_dict': { + 'id': '56102209', + 'ext': 'mp4', + 'title': 'Bec Boop - 🚠✈️🇬🇧 Fly above #London in Emirates Air Line cable car at night 🇬🇧✈️🚠 #BoopScope 🎀💗', + 'timestamp': 1438978559, + 'upload_date': '20150807', + 'uploader': 'Bec Boop', + 'uploader_id': '1465763', + }, + 'skip': 'Expires in 24 hours', + } + + def _call_api(self, method, token): + return self._download_json( + 'https://api.periscope.tv/api/v2/%s?token=%s' % (method, token), token) + + def _real_extract(self, url): + token = self._match_id(url) + + broadcast_data = self._call_api('getBroadcastPublic', token) + broadcast = broadcast_data['broadcast'] + status = broadcast['status'] + + uploader = broadcast.get('user_display_name') or broadcast_data.get('user', {}).get('display_name') + uploader_id = broadcast.get('user_id') or broadcast_data.get('user', {}).get('id') + + title = '%s - %s' % (uploader, status) if uploader else status + state = broadcast.get('state').lower() + if state == 'running': + title = self._live_title(title) + timestamp = parse_iso8601(broadcast.get('created_at')) + + thumbnails = [{ + 'url': broadcast[image], + } for image in ('image_url', 'image_url_small') if broadcast.get(image)] + + stream = self._call_api('getAccessPublic', token) + + formats = [] + for format_id in ('replay', 'rtmp', 'hls', 'https_hls'): + video_url = stream.get(format_id + '_url') + if not video_url: + continue + f = { + 'url': video_url, + 'ext': 'flv' if format_id == 'rtmp' else 'mp4', + } + if format_id != 'rtmp': + f['protocol'] = 'm3u8_native' if state == 'ended' else 'm3u8' + formats.append(f) + self._sort_formats(formats) + + return { + 'id': broadcast.get('id') or token, + 'title': title, + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'thumbnails': thumbnails, + 'formats': formats, + } + + +class QuickscopeIE(InfoExtractor): + IE_DESC = 'Quick Scope' + _VALID_URL = r'https?://watchonperiscope\.com/broadcast/(?P<id>\d+)' + _TEST = { + 'url': 'https://watchonperiscope.com/broadcast/56180087', + 'only_matching': True, + } + + def _real_extract(self, url): + broadcast_id = self._match_id(url) + request = compat_urllib_request.Request( + 'https://watchonperiscope.com/api/accessChannel', compat_urllib_parse.urlencode({ + 'broadcast_id': broadcast_id, + 'entry_ticket': '', + 'from_push': 'false', + 'uses_sessions': 'true', + }).encode('utf-8')) + return self.url_result( + self._download_json(request, broadcast_id)['share_url'], 'Periscope') diff --git a/youtube_dl/extractor/porn91.py b/youtube_dl/extractor/porn91.py index 72d1b2718..3e15533e9 100644 --- a/youtube_dl/extractor/porn91.py +++ b/youtube_dl/extractor/porn91.py @@ -22,6 +22,7 @@ class Porn91IE(InfoExtractor): 'title': '18岁大一漂亮学妹，水嫩性感，再爽一次！', 'ext': 'mp4', 'duration': 431, + 'age_limit': 18, } } @@ -68,4 +69,5 @@ class Porn91IE(InfoExtractor): 'url': video_url, 'duration': duration, 'comment_count': comment_count, + 'age_limit': self._rta_search(webpage), } diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 0b7886840..7b0cdc41a 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -81,7 +81,7 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment') - video_urls = list(map(compat_urllib_parse_unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage))) + video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"player_quality_[0-9]{3}p\s*=\s*'([^']+)'", webpage))) if webpage.find('"encrypted":true') != -1: password = compat_urllib_parse_unquote_plus( self._search_regex(r'"video_title":"([^"]+)', webpage, 'password')) @@ -94,7 +94,7 @@ class PornHubIE(InfoExtractor): format = path.split('/')[5].split('_')[:2] format = "-".join(format) - m = re.match(r'^(?P<height>[0-9]+)P-(?P<tbr>[0-9]+)K$', format) + m = re.match(r'^(?P<height>[0-9]+)[pP]-(?P<tbr>[0-9]+)[kK]$', format) if m is None: height = None tbr = None diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py index fec008ce7..effcf1db3 100644 --- a/youtube_dl/extractor/prosiebensat1.py +++ b/youtube_dl/extractor/prosiebensat1.py @@ -9,7 +9,9 @@ from ..compat import ( compat_urllib_parse, ) from ..utils import ( + ExtractorError, determine_ext, + float_or_none, int_or_none, unified_strdate, ) @@ -224,10 +226,13 @@ class ProSiebenSat1IE(InfoExtractor): 'ids': clip_id, }) - videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON') + video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0] - duration = float(videos[0]['duration']) - source_ids = [source['id'] for source in videos[0]['sources']] + if video.get('is_protected') is True: + raise ExtractorError('This video is DRM protected.', expected=True) + + duration = float_or_none(video.get('duration')) + source_ids = [source['id'] for source in video['sources']] source_ids_str = ','.join(map(str, source_ids)) g = '01!8d8F_)r9]4s[qeuXfP%' diff --git a/youtube_dl/extractor/rtlnl.py b/youtube_dl/extractor/rtlnl.py index a4d3d73ff..543d94417 100644 --- a/youtube_dl/extractor/rtlnl.py +++ b/youtube_dl/extractor/rtlnl.py @@ -44,6 +44,21 @@ class RtlNlIE(InfoExtractor): 'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.', } }, { + # empty synopsis and missing episodes (see https://github.com/rg3/youtube-dl/issues/6275) + 'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a/autoplay=false', + 'info_dict': { + 'id': 'f536aac0-1dc3-4314-920e-3bd1c5b3811a', + 'ext': 'mp4', + 'title': 'RTL Nieuws - Meer beelden van overval juwelier', + 'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$', + 'timestamp': 1437233400, + 'upload_date': '20150718', + 'duration': 30.474, + }, + 'params': { + 'skip_download': True, + }, + }, { # encrypted m3u8 streams, georestricted 'url': 'http://www.rtlxl.nl/#!/afl-2-257632/52a74543-c504-4cde-8aa8-ec66fe8d68a7', 'only_matching': True, @@ -59,22 +74,29 @@ class RtlNlIE(InfoExtractor): uuid) material = info['material'][0] - progname = info['abstracts'][0]['name'] - subtitle = material['title'] or info['episodes'][0]['name'] - description = material.get('synopsis') or info['episodes'][0]['synopsis'] + title = info['abstracts'][0]['name'] + subtitle = material.get('title') + if subtitle: + title += ' - %s' % subtitle + description = material.get('synopsis') meta = info.get('meta', {}) - # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118) - # NB: nowadays, recent ffmpeg and avconv can handle these encrypted streams, so - # this adaptive -> flash workaround is not required in general, but it also - # allows bypassing georestriction therefore is retained for now. - videopath = material['videopath'].replace('/adaptive/', '/flash/') + # m3u8 streams are encrypted and may not be handled properly by older ffmpeg/avconv. + # To workaround this previously adaptive -> flash trick was used to obtain + # unencrypted m3u8 streams (see https://github.com/rg3/youtube-dl/issues/4118) + # and bypass georestrictions as well. + # Currently, unencrypted m3u8 playlists are (intentionally?) invalid and therefore + # unusable albeit can be fixed by simple string replacement (see + # https://github.com/rg3/youtube-dl/pull/6337) + # Since recent ffmpeg and avconv handle encrypted streams just fine encrypted + # streams are used now. + videopath = material['videopath'] m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4') - video_urlpart = videopath.split('/flash/')[1][:-5] + video_urlpart = videopath.split('/adaptive/')[1][:-5] PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4' formats.extend([ @@ -107,7 +129,7 @@ class RtlNlIE(InfoExtractor): return { 'id': uuid, - 'title': '%s - %s' % (progname, subtitle), + 'title': title, 'formats': formats, 'timestamp': material['original_date'], 'description': description, diff --git a/youtube_dl/extractor/rts.py b/youtube_dl/extractor/rts.py index 9fbe239d8..12639f08b 100644 --- a/youtube_dl/extractor/rts.py +++ b/youtube_dl/extractor/rts.py @@ -19,7 +19,16 @@ from ..utils import ( class RTSIE(InfoExtractor): IE_DESC = 'RTS.ch' - _VALID_URL = r'https?://(?:www\.)?rts\.ch/(?:(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html|play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+))' + _VALID_URL = r'''(?x) + (?: + rts:(?P<rts_id>\d+)| + https?:// + (?:www\.)?rts\.ch/ + (?: + (?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html| + play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+) + ) + )''' _TESTS = [ { @@ -123,6 +132,15 @@ class RTSIE(InfoExtractor): }, }, { + # article with videos on rhs + 'url': 'http://www.rts.ch/sport/hockey/6693917-hockey-davos-decroche-son-31e-titre-de-champion-de-suisse.html', + 'info_dict': { + 'id': '6693917', + 'title': 'Hockey: Davos décroche son 31e titre de champion de Suisse', + }, + 'playlist_mincount': 5, + }, + { 'url': 'http://www.rts.ch/play/tv/le-19h30/video/le-chantier-du-nouveau-parlement-vaudois-a-permis-une-trouvaille-historique?id=6348280', 'only_matching': True, } @@ -130,7 +148,7 @@ class RTSIE(InfoExtractor): def _real_extract(self, url): m = re.match(self._VALID_URL, url) - video_id = m.group('id') or m.group('id_new') + video_id = m.group('rts_id') or m.group('id') or m.group('id_new') display_id = m.group('display_id') or m.group('display_id_new') def download_json(internal_id): @@ -143,6 +161,15 @@ class RTSIE(InfoExtractor): # video_id extracted out of URL is not always a real id if 'video' not in all_info and 'audio' not in all_info: page = self._download_webpage(url, display_id) + + # article with videos on rhs + videos = re.findall( + r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:rts:video:(\d+)"', + page) + if videos: + entries = [self.url_result('rts:%s' % video_urn, 'RTS') for video_urn in videos] + return self.playlist_result(entries, video_id, self._og_search_title(page)) + internal_id = self._html_search_regex( r'<(?:video|audio) data-id="([0-9]+)"', page, 'internal video id') diff --git a/youtube_dl/extractor/rtvnh.py b/youtube_dl/extractor/rtvnh.py new file mode 100644 index 000000000..7c9d4b0cd --- /dev/null +++ b/youtube_dl/extractor/rtvnh.py @@ -0,0 +1,47 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ExtractorError + + +class RTVNHIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?rtvnh\.nl/video/(?P<id>[0-9]+)' + _TEST = { + 'url': 'http://www.rtvnh.nl/video/131946', + 'md5': '6e1d0ab079e2a00b6161442d3ceacfc1', + 'info_dict': { + 'id': '131946', + 'ext': 'mp4', + 'title': 'Grote zoektocht in zee bij Zandvoort naar vermiste vrouw', + 'thumbnail': 're:^https?:.*\.jpg$' + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + + meta = self._parse_json(self._download_webpage( + 'http://www.rtvnh.nl/video/json?m=' + video_id, video_id), video_id) + + status = meta.get('status') + if status != 200: + raise ExtractorError( + '%s returned error code %d' % (self.IE_NAME, status), expected=True) + + formats = self._extract_smil_formats( + 'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id, fatal=False) + + for item in meta['source']['fb']: + if item.get('type') == 'hls': + formats.extend(self._extract_m3u8_formats( + item['file'], video_id, ext='mp4', entry_protocol='m3u8_native')) + elif item.get('type') == '': + formats.append({'url': item['file']}) + + return { + 'id': video_id, + 'title': meta['title'].strip(), + 'thumbnail': meta.get('image'), + 'formats': formats + } diff --git a/youtube_dl/extractor/rutube.py b/youtube_dl/extractor/rutube.py index 5b1c3577a..d94dc7399 100644 --- a/youtube_dl/extractor/rutube.py +++ b/youtube_dl/extractor/rutube.py @@ -30,6 +30,7 @@ class RutubeIE(InfoExtractor): 'uploader': 'NTDRussian', 'uploader_id': '29790', 'upload_date': '20131016', + 'age_limit': 0, }, 'params': { # It requires ffmpeg (m3u8 download) diff --git a/youtube_dl/extractor/screenwavemedia.py b/youtube_dl/extractor/screenwavemedia.py index d1ab66b32..3bc84989e 100644 --- a/youtube_dl/extractor/screenwavemedia.py +++ b/youtube_dl/extractor/screenwavemedia.py @@ -1,12 +1,11 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( int_or_none, unified_strdate, + js_to_json, ) @@ -22,59 +21,48 @@ class ScreenwaveMediaIE(InfoExtractor): video_id = self._match_id(url) playerdata = self._download_webpage( - 'http://player.screenwavemedia.com/play/player.php?id=%s' % video_id, + 'http://player.screenwavemedia.com/player.php?id=%s' % video_id, video_id, 'Downloading player webpage') vidtitle = self._search_regex( r'\'vidtitle\'\s*:\s*"([^"]+)"', playerdata, 'vidtitle').replace('\\/', '/') - vidurl = self._search_regex( - r'\'vidurl\'\s*:\s*"([^"]+)"', playerdata, 'vidurl').replace('\\/', '/') - - videolist_url = None - - mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata) - if mobj: - videoserver = mobj.group('videoserver') - mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata) - vidid = mobj.group('vidid') if mobj else video_id - videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid) - else: - mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata) - if mobj: - videolist_url = mobj.group('smil') - - if videolist_url: - videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML') - formats = [] - baseurl = vidurl[:vidurl.rfind('/') + 1] - for video in videolist.findall('.//video'): - src = video.get('src') - if not src: - continue - file_ = src.partition(':')[-1] - width = int_or_none(video.get('width')) - height = int_or_none(video.get('height')) - bitrate = int_or_none(video.get('system-bitrate'), scale=1000) - format = { - 'url': baseurl + file_, - 'format_id': src.rpartition('.')[0].rpartition('_')[-1], - } - if width or height: - format.update({ - 'tbr': bitrate, - 'width': width, - 'height': height, - }) - else: - format.update({ - 'abr': bitrate, - 'vcodec': 'none', - }) - formats.append(format) - else: - formats = [{ - 'url': vidurl, - }] + + playerconfig = self._download_webpage( + 'http://player.screenwavemedia.com/player.js', + video_id, 'Downloading playerconfig webpage') + + videoserver = self._search_regex(r"\[ipaddress\]\s*=>\s*([\d\.]+)", playerdata, 'videoserver') + + sources = self._parse_json( + js_to_json( + self._search_regex( + r"sources\s*:\s*(\[[^\]]+?\])", playerconfig, + 'sources', + ).replace( + "' + thisObj.options.videoserver + '", + videoserver + ).replace( + "' + playerVidId + '", + video_id + ) + ), + video_id + ) + + formats = [] + for source in sources: + if source['type'] == 'hls': + formats.extend(self._extract_m3u8_formats(source['file'], video_id)) + else: + format_label = source.get('label') + height = int_or_none(self._search_regex( + r'^(\d+)[pP]', format_label, 'height', default=None)) + formats.append({ + 'url': source['file'], + 'format': format_label, + 'ext': source.get('type'), + 'height': height, + }) self._sort_formats(formats) return { diff --git a/youtube_dl/extractor/sexykarma.py b/youtube_dl/extractor/sexykarma.py index 6446d26dc..e33483674 100644 --- a/youtube_dl/extractor/sexykarma.py +++ b/youtube_dl/extractor/sexykarma.py @@ -29,6 +29,7 @@ class SexyKarmaIE(InfoExtractor): 'view_count': int, 'comment_count': int, 'categories': list, + 'age_limit': 18, } }, { 'url': 'http://www.sexykarma.com/gonewild/video/pot-pixie-tribute-8Id6EZPbuHf.html', diff --git a/youtube_dl/extractor/snagfilms.py b/youtube_dl/extractor/snagfilms.py index cf495f310..6977afb27 100644 --- a/youtube_dl/extractor/snagfilms.py +++ b/youtube_dl/extractor/snagfilms.py @@ -24,6 +24,15 @@ class SnagFilmsEmbedIE(InfoExtractor): 'title': '#whilewewatch', } }, { + # invalid labels, 360p is better that 480p + 'url': 'http://www.snagfilms.com/embed/player?filmId=17ca0950-a74a-11e0-a92a-0026bb61d036', + 'md5': '882fca19b9eb27ef865efeeaed376a48', + 'info_dict': { + 'id': '17ca0950-a74a-11e0-a92a-0026bb61d036', + 'ext': 'mp4', + 'title': 'Life in Limbo', + } + }, { 'url': 'http://www.snagfilms.com/embed/player?filmId=0000014c-de2f-d5d6-abcf-ffef58af0017', 'only_matching': True, }] @@ -52,14 +61,15 @@ class SnagFilmsEmbedIE(InfoExtractor): if not file_: continue type_ = source.get('type') - format_id = source.get('label') ext = determine_ext(file_) - if any(_ == 'm3u8' for _ in (type_, ext)): + format_id = source.get('label') or ext + if all(v == 'm3u8' for v in (type_, ext)): formats.extend(self._extract_m3u8_formats( file_, video_id, 'mp4', m3u8_id='hls')) else: bitrate = int_or_none(self._search_regex( - r'(\d+)kbps', file_, 'bitrate', default=None)) + [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext], + file_, 'bitrate', default=None)) height = int_or_none(self._search_regex( r'^(\d+)[pP]$', format_id, 'height', default=None)) formats.append({ diff --git a/youtube_dl/extractor/soundcloud.py b/youtube_dl/extractor/soundcloud.py index 118ca4832..6ce86cbcd 100644 --- a/youtube_dl/extractor/soundcloud.py +++ b/youtube_dl/extractor/soundcloud.py @@ -29,7 +29,7 @@ class SoundcloudIE(InfoExtractor): _VALID_URL = r'''(?x)^(?:https?://)? (?:(?:(?:www\.|m\.)?soundcloud\.com/ (?P<uploader>[\w\d-]+)/ - (?!sets/|(?:likes|tracks)/?(?:$|[?#])) + (?!(?:tracks|sets(?:/[^/?#]+)?|reposts|likes|spotlight)/?(?:$|[?#])) (?P<title>[\w\d-]+)/? (?P<token>[^?]+?)?(?:[?].*)?$) |(?:api\.soundcloud\.com/tracks/(?P<track_id>\d+) @@ -282,69 +282,150 @@ class SoundcloudSetIE(SoundcloudIE): msgs = (compat_str(err['error_message']) for err in info['errors']) raise ExtractorError('unable to download video webpage: %s' % ','.join(msgs)) + entries = [self.url_result(track['permalink_url'], 'Soundcloud') for track in info['tracks']] + return { '_type': 'playlist', - 'entries': [self._extract_info_dict(track, secret_token=token) for track in info['tracks']], + 'entries': entries, 'id': '%s' % info['id'], 'title': info['title'], } class SoundcloudUserIE(SoundcloudIE): - _VALID_URL = r'https?://(?:(?:www|m)\.)?soundcloud\.com/(?P<user>[^/]+)/?((?P<rsrc>tracks|likes)/?)?(\?.*)?$' + _VALID_URL = r'''(?x) + https?:// + (?:(?:www|m)\.)?soundcloud\.com/ + (?P<user>[^/]+) + (?:/ + (?P<rsrc>tracks|sets|reposts|likes|spotlight) + )? + /?(?:[?#].*)?$ + ''' IE_NAME = 'soundcloud:user' _TESTS = [{ - 'url': 'https://soundcloud.com/the-concept-band', + 'url': 'https://soundcloud.com/the-akashic-chronicler', 'info_dict': { - 'id': '9615865', - 'title': 'The Royal Concept', + 'id': '114582580', + 'title': 'The Akashic Chronicler (All)', }, - 'playlist_mincount': 12 + 'playlist_mincount': 112, }, { - 'url': 'https://soundcloud.com/the-concept-band/likes', + 'url': 'https://soundcloud.com/the-akashic-chronicler/tracks', 'info_dict': { - 'id': '9615865', - 'title': 'The Royal Concept', + 'id': '114582580', + 'title': 'The Akashic Chronicler (Tracks)', }, - 'playlist_mincount': 1, + 'playlist_mincount': 50, }, { - 'url': 'https://soundcloud.com/the-akashic-chronicler/tracks', - 'only_matching': True, + 'url': 'https://soundcloud.com/the-akashic-chronicler/sets', + 'info_dict': { + 'id': '114582580', + 'title': 'The Akashic Chronicler (Playlists)', + }, + 'playlist_mincount': 3, + }, { + 'url': 'https://soundcloud.com/the-akashic-chronicler/reposts', + 'info_dict': { + 'id': '114582580', + 'title': 'The Akashic Chronicler (Reposts)', + }, + 'playlist_mincount': 9, + }, { + 'url': 'https://soundcloud.com/the-akashic-chronicler/likes', + 'info_dict': { + 'id': '114582580', + 'title': 'The Akashic Chronicler (Likes)', + }, + 'playlist_mincount': 333, + }, { + 'url': 'https://soundcloud.com/grynpyret/spotlight', + 'info_dict': { + 'id': '7098329', + 'title': 'Grynpyret (Spotlight)', + }, + 'playlist_mincount': 1, }] + _API_BASE = 'https://api.soundcloud.com' + _API_V2_BASE = 'https://api-v2.soundcloud.com' + + _BASE_URL_MAP = { + 'all': '%s/profile/soundcloud:users:%%s' % _API_V2_BASE, + 'tracks': '%s/users/%%s/tracks' % _API_BASE, + 'sets': '%s/users/%%s/playlists' % _API_V2_BASE, + 'reposts': '%s/profile/soundcloud:users:%%s/reposts' % _API_V2_BASE, + 'likes': '%s/users/%%s/likes' % _API_V2_BASE, + 'spotlight': '%s/users/%%s/spotlight' % _API_V2_BASE, + } + + _TITLE_MAP = { + 'all': 'All', + 'tracks': 'Tracks', + 'sets': 'Playlists', + 'reposts': 'Reposts', + 'likes': 'Likes', + 'spotlight': 'Spotlight', + } + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) uploader = mobj.group('user') - resource = mobj.group('rsrc') - if resource is None: - resource = 'tracks' - elif resource == 'likes': - resource = 'favorites' url = 'http://soundcloud.com/%s/' % uploader resolv_url = self._resolv_url(url) user = self._download_json( resolv_url, uploader, 'Downloading user info') - base_url = 'http://api.soundcloud.com/users/%s/%s.json?' % (uploader, resource) + + resource = mobj.group('rsrc') or 'all' + base_url = self._BASE_URL_MAP[resource] % user['id'] + + next_href = None entries = [] for i in itertools.count(): - data = compat_urllib_parse.urlencode({ - 'offset': i * 50, - 'limit': 50, - 'client_id': self._CLIENT_ID, - }) - new_entries = self._download_json( - base_url + data, uploader, 'Downloading track page %s' % (i + 1)) - if len(new_entries) == 0: + if not next_href: + data = compat_urllib_parse.urlencode({ + 'offset': i * 50, + 'limit': 50, + 'client_id': self._CLIENT_ID, + 'linked_partitioning': '1', + 'representation': 'speedy', + }) + next_href = base_url + '?' + data + + response = self._download_json( + next_href, uploader, 'Downloading track page %s' % (i + 1)) + + collection = response['collection'] + + if not collection: self.to_screen('%s: End page received' % uploader) break - entries.extend(self.url_result(e['permalink_url'], 'Soundcloud') for e in new_entries) + + def resolve_permalink_url(candidates): + for cand in candidates: + if isinstance(cand, dict): + permalink_url = cand.get('permalink_url') + if permalink_url and permalink_url.startswith('http'): + return permalink_url + + for e in collection: + permalink_url = resolve_permalink_url((e, e.get('track'), e.get('playlist'))) + if permalink_url: + entries.append(self.url_result(permalink_url)) + + if 'next_href' in response: + next_href = response['next_href'] + if not next_href: + break + else: + next_href = None return { '_type': 'playlist', 'id': compat_str(user['id']), - 'title': user['username'], + 'title': '%s (%s)' % (user['username'], self._TITLE_MAP[resource]), 'entries': entries, } @@ -379,9 +460,7 @@ class SoundcloudPlaylistIE(SoundcloudIE): data = self._download_json( base_url + data, playlist_id, 'Downloading playlist') - entries = [ - self._extract_info_dict(t, quiet=True, secret_token=token) - for t in data['tracks']] + entries = [self.url_result(track['permalink_url'], 'Soundcloud') for track in data['tracks']] return { '_type': 'playlist', diff --git a/youtube_dl/extractor/southpark.py b/youtube_dl/extractor/southpark.py index 7fb165a87..87b650468 100644 --- a/youtube_dl/extractor/southpark.py +++ b/youtube_dl/extractor/southpark.py @@ -45,6 +45,14 @@ class SouthParkDeIE(SouthParkIE): 'title': 'The Government Won\'t Respect My Privacy', 'description': 'Cartman explains the benefits of "Shitter" to Stan, Kyle and Craig.', }, + }, { + # non-ASCII characters in initial URL + 'url': 'http://www.southpark.de/alle-episoden/s18e09-hashtag-aufwärmen', + 'playlist_count': 4, + }, { + # non-ASCII characters in redirect URL + 'url': 'http://www.southpark.de/alle-episoden/s18e09', + 'playlist_count': 4, }] diff --git a/youtube_dl/extractor/spiegel.py b/youtube_dl/extractor/spiegel.py index b868241d5..5bd3c0087 100644 --- a/youtube_dl/extractor/spiegel.py +++ b/youtube_dl/extractor/spiegel.py @@ -9,7 +9,7 @@ from .spiegeltv import SpiegeltvIE class SpiegelIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed)?(?:\.html)?(?:#.*)?$' + _VALID_URL = r'https?://(?:www\.)?spiegel\.de/video/[^/]*-(?P<id>[0-9]+)(?:-embed|-iframe)?(?:\.html)?(?:#.*)?$' _TESTS = [{ 'url': 'http://www.spiegel.de/video/vulkan-tungurahua-in-ecuador-ist-wieder-aktiv-video-1259285.html', 'md5': '2c2754212136f35fb4b19767d242f66e', @@ -39,6 +39,9 @@ class SpiegelIE(InfoExtractor): 'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.', 'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"', } + }, { + 'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/tagesschau.py b/youtube_dl/extractor/tagesschau.py index bfe07b024..73e7657d4 100644 --- a/youtube_dl/extractor/tagesschau.py +++ b/youtube_dl/extractor/tagesschau.py @@ -8,17 +8,17 @@ from ..utils import parse_filesize class TagesschauIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:sendung/ts|video/video)(?P<id>-?[0-9]+)\.html' + _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:[^/]+/)*?[^/#?]+?(?P<id>-?[0-9]+)(?:~_[^/#?]+?)?\.html' _TESTS = [{ - 'url': 'http://www.tagesschau.de/multimedia/video/video1399128.html', - 'md5': 'bcdeac2194fb296d599ce7929dfa4009', + 'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html', + 'md5': '917a228bc7df7850783bc47979673a09', 'info_dict': { - 'id': '1399128', + 'id': '102143', 'ext': 'mp4', - 'title': 'Harald Range, Generalbundesanwalt, zu den Ermittlungen', - 'description': 'md5:69da3c61275b426426d711bde96463ab', - 'thumbnail': 're:^http:.*\.jpg$', + 'title': 'Regierungsumbildung in Athen: Neue Minister in Griechenland vereidigt', + 'description': 'md5:171feccd9d9b3dd54d05d501568f6359', + 'thumbnail': 're:^https?:.*\.jpg$', }, }, { 'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html', @@ -28,8 +28,39 @@ class TagesschauIE(InfoExtractor): 'ext': 'mp4', 'description': 'md5:695c01bfd98b7e313c501386327aea59', 'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr', - 'thumbnail': 're:^http:.*\.jpg$', - } + 'thumbnail': 're:^https?:.*\.jpg$', + }, + }, { + 'url': 'http://www.tagesschau.de/multimedia/politikimradio/audio-18407.html', + 'md5': 'aef45de271c4bf0a5db834aa40bf774c', + 'info_dict': { + 'id': '18407', + 'ext': 'mp3', + 'title': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', + 'description': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich', + 'thumbnail': 're:^https?:.*\.jpg$', + }, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/tt-3827.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/nm-3475.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/weltspiegel-3167.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/tsvorzwanzig-959.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/sendung/bab/bab-3299~_bab-sendung-209.html', + 'only_matching': True, + }, { + 'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html', + 'only_matching': True, }] _FORMATS = { @@ -49,19 +80,26 @@ class TagesschauIE(InfoExtractor): playerpage = self._download_webpage( player_url, display_id, 'Downloading player page') - medias = re.findall( - r'"(http://media.+?)", type:"video/(.+?)", quality:"(.+?)"', - playerpage) formats = [] - for url, ext, res in medias: + for media in re.finditer( + r'''(?x) + (?P<q_url>["\'])(?P<url>http://media.+?)(?P=q_url) + ,\s*type:(?P<q_type>["\'])(?P<type>video|audio)/(?P<ext>.+?)(?P=q_type) + (?:,\s*quality:(?P<q_quality>["\'])(?P<quality>.+?)(?P=q_quality))? + ''', playerpage): + url = media.group('url') + type_ = media.group('type') + ext = media.group('ext') + res = media.group('quality') f = { - 'format_id': res + '_' + ext, + 'format_id': '%s_%s' % (res, ext) if res else ext, 'url': url, 'ext': ext, + 'vcodec': 'none' if type_ == 'audio' else None, } f.update(self._FORMATS.get(res, {})) formats.append(f) - thumbnail_fn = re.findall(r'"(/multimedia/.+?\.jpg)"', playerpage)[-1] + thumbnail = self._og_search_thumbnail(playerpage) title = self._og_search_title(webpage).strip() description = self._og_search_description(webpage).strip() else: @@ -99,17 +137,14 @@ class TagesschauIE(InfoExtractor): 'filesize_approx': parse_filesize(m.group('filesize_approx')), }) formats.append(format) - thumbnail_fn = self._search_regex( - r'(?s)<img alt="Sendungsbild".*?src="([^"]+)"', - webpage, 'thumbnail', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) description = self._html_search_regex( r'(?s)<p class="teasertext">(.*?)</p>', - webpage, 'description', fatal=False) + webpage, 'description', default=None) title = self._html_search_regex( r'<span class="headline".*?>(.*?)</span>', webpage, 'title') self._sort_formats(formats) - thumbnail = 'http://www.tagesschau.de' + thumbnail_fn return { 'id': display_id, diff --git a/youtube_dl/extractor/telegraaf.py b/youtube_dl/extractor/telegraaf.py new file mode 100644 index 000000000..6f8333cfc --- /dev/null +++ b/youtube_dl/extractor/telegraaf.py @@ -0,0 +1,35 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import remove_end + + +class TelegraafIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/tv/(?:[^/]+/)+(?P<id>\d+)/[^/]+\.html' + _TEST = { + 'url': 'http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html', + 'md5': '83245a9779bcc4a24454bfd53c65b6dc', + 'info_dict': { + 'id': '24353229', + 'ext': 'mp4', + 'title': 'Tikibad ontruimd wegens brand', + 'description': 'md5:05ca046ff47b931f9b04855015e163a4', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 33, + }, + } + + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + playlist_url = self._search_regex( + r"iframe\.loadPlayer\('([^']+)'", webpage, 'player') + + entries = self._extract_xspf_playlist(playlist_url, playlist_id) + title = remove_end(self._og_search_title(webpage), ' - VIDEO') + description = self._og_search_description(webpage) + + return self.playlist_result(entries, playlist_id, title, description) diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 83d833e30..0643eccaf 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -108,7 +108,11 @@ class ThePlatformIE(InfoExtractor): config_url = config_url.replace('swf/', 'config/') config_url = config_url.replace('onsite/', 'onsite/config/') config = self._download_json(config_url, video_id, 'Downloading config') - smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m' + if 'releaseUrl' in config: + release_url = config['releaseUrl'] + else: + release_url = 'http://link.theplatform.com/s/%s?mbr=true' % path + smil_url = release_url + '&format=SMIL&formats=MPEG4&manifest=f4m' else: smil_url = 'http://link.theplatform.com/s/%s/meta.smil?format=smil&mbr=true' % path diff --git a/youtube_dl/extractor/tudou.py b/youtube_dl/extractor/tudou.py index c89de5ba4..84fe71aef 100644 --- a/youtube_dl/extractor/tudou.py +++ b/youtube_dl/extractor/tudou.py @@ -29,6 +29,8 @@ class TudouIE(InfoExtractor): } }] + _PLAYER_URL = 'http://js.tudouui.com/bin/lingtong/PortalPlayer_177.swf' + def _url_for_id(self, id, quality=None): info_url = "http://v2.tudou.com/f?id=" + str(id) if quality: @@ -54,6 +56,10 @@ class TudouIE(InfoExtractor): thumbnail_url = self._search_regex( r",pic:\s*[\"'](.+?)[\"']", webpage, 'thumbnail URL', fatal=False) + player_url = self._search_regex( + r"playerUrl\s*:\s*['\"](.+?\.swf)[\"']", + webpage, 'player URL', default=self._PLAYER_URL) + segs_json = self._search_regex(r'segs: \'(.*)\'', webpage, 'segments') segments = json.loads(segs_json) # It looks like the keys are the arguments that have to be passed as @@ -76,6 +82,9 @@ class TudouIE(InfoExtractor): 'ext': ext, 'title': title, 'thumbnail': thumbnail_url, + 'http_headers': { + 'Referer': player_url, + }, } result.append(part_info) diff --git a/youtube_dl/extractor/tumblr.py b/youtube_dl/extractor/tumblr.py index 9ead13a91..3d3b635e4 100644 --- a/youtube_dl/extractor/tumblr.py +++ b/youtube_dl/extractor/tumblr.py @@ -4,8 +4,6 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from .pornhub import PornHubIE -from .vimeo import VimeoIE class TumblrIE(InfoExtractor): @@ -60,26 +58,16 @@ class TumblrIE(InfoExtractor): blog = m_url.group('blog_name') url = 'http://%s.tumblr.com/post/%s/' % (blog, video_id) - webpage = self._download_webpage(url, video_id) - - vid_me_embed_url = self._search_regex( - r'src=[\'"](https?://vid\.me/[^\'"]+)[\'"]', - webpage, 'vid.me embed', default=None) - if vid_me_embed_url is not None: - return self.url_result(vid_me_embed_url, 'Vidme') - - pornhub_url = PornHubIE._extract_url(webpage) - if pornhub_url: - return self.url_result(pornhub_url, 'PornHub') - - vimeo_url = VimeoIE._extract_vimeo_url(url, webpage) - if vimeo_url: - return self.url_result(vimeo_url, 'Vimeo') + webpage, urlh = self._download_webpage_handle(url, video_id) iframe_url = self._search_regex( r'src=\'(https?://www\.tumblr\.com/video/[^\']+)\'', - webpage, 'iframe url') - iframe = self._download_webpage(iframe_url, video_id) + webpage, 'iframe url', default=None) + if iframe_url is None: + return self.url_result(urlh.geturl(), 'Generic') + + iframe = self._download_webpage(iframe_url, video_id, + 'Downloading iframe page') video_url = self._search_regex(r'<source src="([^"]+)"', iframe, 'video url') diff --git a/youtube_dl/extractor/tvplay.py b/youtube_dl/extractor/tvplay.py index 79863e781..b4683de54 100644 --- a/youtube_dl/extractor/tvplay.py +++ b/youtube_dl/extractor/tvplay.py @@ -104,6 +104,7 @@ class TVPlayIE(InfoExtractor): 'duration': 1492, 'timestamp': 1330522854, 'upload_date': '20120229', + 'age_limit': 18, }, 'params': { # rtmp download diff --git a/youtube_dl/extractor/tweakers.py b/youtube_dl/extractor/tweakers.py index c80ec15cf..f3198fb85 100644 --- a/youtube_dl/extractor/tweakers.py +++ b/youtube_dl/extractor/tweakers.py @@ -1,19 +1,13 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import ( - xpath_text, - xpath_with_ns, - int_or_none, - float_or_none, -) class TweakersIE(InfoExtractor): _VALID_URL = r'https?://tweakers\.net/video/(?P<id>\d+)' _TEST = { 'url': 'https://tweakers.net/video/9926/new-nintendo-3ds-xl-op-alle-fronten-beter.html', - 'md5': '1b5afa817403bb5baa08359dca31e6df', + 'md5': '3147e4ddad366f97476a93863e4557c8', 'info_dict': { 'id': '9926', 'ext': 'mp4', @@ -25,41 +19,7 @@ class TweakersIE(InfoExtractor): } def _real_extract(self, url): - video_id = self._match_id(url) - - playlist = self._download_xml( - 'https://tweakers.net/video/s1playlist/%s/playlist.xspf' % video_id, - video_id) - - NS_MAP = { - 'xspf': 'http://xspf.org/ns/0/', - 's1': 'http://static.streamone.nl/player/ns/0', - } - - track = playlist.find(xpath_with_ns('./xspf:trackList/xspf:track', NS_MAP)) - - title = xpath_text( - track, xpath_with_ns('./xspf:title', NS_MAP), 'title') - description = xpath_text( - track, xpath_with_ns('./xspf:annotation', NS_MAP), 'description') - thumbnail = xpath_text( - track, xpath_with_ns('./xspf:image', NS_MAP), 'thumbnail') - duration = float_or_none( - xpath_text(track, xpath_with_ns('./xspf:duration', NS_MAP), 'duration'), - 1000) - - formats = [{ - 'url': location.text, - 'format_id': location.get(xpath_with_ns('s1:label', NS_MAP)), - 'width': int_or_none(location.get(xpath_with_ns('s1:width', NS_MAP))), - 'height': int_or_none(location.get(xpath_with_ns('s1:height', NS_MAP))), - } for location in track.findall(xpath_with_ns('./xspf:location', NS_MAP))] - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'duration': duration, - 'formats': formats, - } + playlist_id = self._match_id(url) + entries = self._extract_xspf_playlist( + 'https://tweakers.net/video/s1playlist/%s/playlist.xspf' % playlist_id, playlist_id) + return self.playlist_result(entries, playlist_id) diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index 948c8ce39..a2b6a35aa 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -7,12 +7,15 @@ import random from .common import InfoExtractor from ..compat import ( + compat_parse_qs, compat_str, compat_urllib_parse, + compat_urllib_parse_urlparse, compat_urllib_request, ) from ..utils import ( ExtractorError, + parse_duration, parse_iso8601, ) @@ -185,7 +188,7 @@ class TwitchVodIE(TwitchItemBaseIE): _ITEM_SHORTCUT = 'v' _TEST = { - 'url': 'http://www.twitch.tv/riotgames/v/6528877', + 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s', 'info_dict': { 'id': 'v6528877', 'ext': 'mp4', @@ -197,6 +200,7 @@ class TwitchVodIE(TwitchItemBaseIE): 'uploader': 'Riot Games', 'uploader_id': 'riotgames', 'view_count': int, + 'start_time': 310, }, 'params': { # m3u8 download @@ -216,6 +220,12 @@ class TwitchVodIE(TwitchItemBaseIE): item_id, 'mp4') self._prefer_source(formats) info['formats'] = formats + + parsed_url = compat_urllib_parse_urlparse(url) + query = compat_parse_qs(parsed_url.query) + if 't' in query: + info['start_time'] = parse_duration(query['t'][0]) + return info @@ -310,9 +320,9 @@ class TwitchBookmarksIE(TwitchPlaylistBaseIE): class TwitchStreamIE(TwitchBaseIE): IE_NAME = 'twitch:stream' - _VALID_URL = r'%s/(?P<id>[^/]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE + _VALID_URL = r'%s/(?P<id>[^/#?]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE - _TEST = { + _TESTS = [{ 'url': 'http://www.twitch.tv/shroomztv', 'info_dict': { 'id': '12772022048', @@ -331,7 +341,10 @@ class TwitchStreamIE(TwitchBaseIE): # m3u8 download 'skip_download': True, }, - } + }, { + 'url': 'http://www.twitch.tv/miracle_doto#profile-0', + 'only_matching': True, + }] def _real_extract(self, url): channel_id = self._match_id(url) diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index e2bab52fe..4a0eaf65f 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -77,7 +77,11 @@ class UdemyIE(InfoExtractor): login_popup = self._download_webpage( self._LOGIN_URL, None, 'Downloading login popup') - if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>': + def is_logged(webpage): + return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<']) + + # already logged in + if is_logged(login_popup): return login_form = self._form_hidden_inputs('login-form', login_popup) @@ -95,8 +99,7 @@ class UdemyIE(InfoExtractor): response = self._download_webpage( request, None, 'Logging in as %s' % username) - if all(logout_pattern not in response - for logout_pattern in ['href="https://www.udemy.com/user/logout/', '>Logout<']): + if not is_logged(response): error = self._html_search_regex( r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>', response, 'error message', default=None) diff --git a/youtube_dl/extractor/vice.py b/youtube_dl/extractor/vice.py index 04e2b0ba7..01af7a995 100644 --- a/youtube_dl/extractor/vice.py +++ b/youtube_dl/extractor/vice.py @@ -1,5 +1,4 @@ from __future__ import unicode_literals -import re from .common import InfoExtractor from .ooyala import OoyalaIE @@ -7,25 +6,29 @@ from ..utils import ExtractorError class ViceIE(InfoExtractor): - _VALID_URL = r'http://www\.vice\.com/.*?/(?P<name>.+)' + _VALID_URL = r'https?://(?:.+?\.)?vice\.com/(?:[^/]+/)+(?P<id>.+)' - _TEST = { - 'url': 'http://www.vice.com/Fringes/cowboy-capitalists-part-1', - 'info_dict': { - 'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp', - 'ext': 'mp4', - 'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov', - }, - 'params': { - # Requires ffmpeg (m3u8 manifest) - 'skip_download': True, - }, - } + _TESTS = [ + { + 'url': 'http://www.vice.com/Fringes/cowboy-capitalists-part-1', + 'info_dict': { + 'id': '43cW1mYzpia9IlestBjVpd23Yu3afAfp', + 'ext': 'mp4', + 'title': 'VICE_COWBOYCAPITALISTS_PART01_v1_VICE_WM_1080p.mov', + }, + 'params': { + # Requires ffmpeg (m3u8 manifest) + 'skip_download': True, + }, + }, { + 'url': 'https://news.vice.com/video/experimenting-on-animals-inside-the-monkey-lab', + 'only_matching': True, + } + ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name = mobj.group('name') - webpage = self._download_webpage(url, name) + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) try: embed_code = self._search_regex( r'embedCode=([^&\'"]+)', webpage, diff --git a/youtube_dl/extractor/videolecturesnet.py b/youtube_dl/extractor/videolecturesnet.py index d6a7eb203..ef2da5632 100644 --- a/youtube_dl/extractor/videolecturesnet.py +++ b/youtube_dl/extractor/videolecturesnet.py @@ -12,7 +12,7 @@ from ..utils import ( class VideoLecturesNetIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/' + _VALID_URL = r'http://(?:www\.)?videolectures\.net/(?P<id>[^/#?]+)/*(?:[#?].*)?$' IE_NAME = 'videolectures.net' _TEST = { diff --git a/youtube_dl/extractor/vidme.py b/youtube_dl/extractor/vidme.py index e0b55078b..157bb74fe 100644 --- a/youtube_dl/extractor/vidme.py +++ b/youtube_dl/extractor/vidme.py @@ -22,6 +22,27 @@ class VidmeIE(InfoExtractor): 'timestamp': 1406313244, 'upload_date': '20140725', 'thumbnail': 're:^https?://.*\.jpg', + 'view_count': int, + 'like_count': int, + }, + }, { + # tests uploader field + 'url': 'https://vid.me/4Iib', + 'info_dict': { + 'id': '4Iib', + 'ext': 'mp4', + 'title': 'The Carver', + 'description': 'md5:e9c24870018ae8113be936645b93ba3c', + 'duration': 97.859999999999999, + 'timestamp': 1433203629, + 'upload_date': '20150602', + 'uploader': 'Thomas', + 'thumbnail': 're:^https?://.*\.jpg', + 'view_count': int, + 'like_count': int, + }, + 'params': { + 'skip_download': True, }, }, { # From http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching @@ -40,16 +61,23 @@ class VidmeIE(InfoExtractor): title = self._og_search_title(webpage) description = self._og_search_description(webpage, default='') thumbnail = self._og_search_thumbnail(webpage) - timestamp = int_or_none(self._og_search_property('updated_time', webpage, fatal=False)) - width = int_or_none(self._og_search_property('video:width', webpage, fatal=False)) - height = int_or_none(self._og_search_property('video:height', webpage, fatal=False)) + timestamp = int_or_none(self._og_search_property( + 'updated_time', webpage, fatal=False)) + width = int_or_none(self._og_search_property( + 'video:width', webpage, fatal=False)) + height = int_or_none(self._og_search_property( + 'video:height', webpage, fatal=False)) duration = float_or_none(self._html_search_regex( r'data-duration="([^"]+)"', webpage, 'duration', fatal=False)) view_count = str_to_int(self._html_search_regex( - r'<(?:li|span) class="video_views">\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False)) + r'<(?:li|span) class="video_views">\s*([\d,\.]+)\s*plays?', + webpage, 'view count', fatal=False)) like_count = str_to_int(self._html_search_regex( r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">', webpage, 'like count', fatal=False)) + uploader = self._html_search_regex( + 'class="video_author_username"[^>]*>([^<]+)', + webpage, 'uploader', default=None) return { 'id': video_id, @@ -63,4 +91,5 @@ class VidmeIE(InfoExtractor): 'duration': duration, 'view_count': view_count, 'like_count': like_count, + 'uploader': uploader, } diff --git a/youtube_dl/extractor/viewster.py b/youtube_dl/extractor/viewster.py index 1742e66f4..cda02ba24 100644 --- a/youtube_dl/extractor/viewster.py +++ b/youtube_dl/extractor/viewster.py @@ -1,129 +1,142 @@ +# coding: utf-8 from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_urllib_request +from ..compat import ( + compat_urllib_request, + compat_urllib_parse, + compat_urllib_parse_unquote, +) +from ..utils import ( + determine_ext, + int_or_none, + parse_iso8601, + HEADRequest, +) class ViewsterIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?viewster\.com/movie/(?P<id>\d+-\d+-\d+)' + _VALID_URL = r'http://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)' _TESTS = [{ - # movielink, paymethod=fre - 'url': 'http://www.viewster.com/movie/1293-19341-000/hout-wood/', - 'playlist': [{ - 'md5': '8f9d94b282d80c42b378dffdbb11caf3', - 'info_dict': { - 'id': '1293-19341-000-movie', - 'ext': 'flv', - 'title': "'Hout' (Wood) - Movie", - }, - }], - 'info_dict': { - 'id': '1293-19341-000', - 'title': "'Hout' (Wood)", - 'description': 'md5:925733185a9242ef96f436937683f33b', - } - }, { - # movielink, paymethod=adv + # movie, Type=Movie 'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/', - 'playlist': [{ - 'md5': '77a005453ca7396cbe3d35c9bea30aef', - 'info_dict': { - 'id': '1140-11855-000-movie', - 'ext': 'flv', - 'title': "THE LISTENING PROJECT - Movie", - }, - }], + 'md5': '14d3cfffe66d57b41ae2d9c873416f01', 'info_dict': { 'id': '1140-11855-000', - 'title': "THE LISTENING PROJECT", - 'description': 'md5:714421ae9957e112e672551094bf3b08', - } + 'ext': 'flv', + 'title': 'The listening Project', + 'description': 'md5:bac720244afd1a8ea279864e67baa071', + 'timestamp': 1214870400, + 'upload_date': '20080701', + 'duration': 4680, + }, }, { - # direct links, no movielink - 'url': 'http://www.viewster.com/movie/1198-56411-000/sinister/', - 'playlist': [{ - 'md5': '0307b7eac6bfb21ab0577a71f6eebd8f', - 'info_dict': { - 'id': '1198-56411-000-trailer', - 'ext': 'mp4', - 'title': "Sinister - Trailer", - }, - }, { - 'md5': '80b9ee3ad69fb368f104cb5d9732ae95', - 'info_dict': { - 'id': '1198-56411-000-behind-scenes', - 'ext': 'mp4', - 'title': "Sinister - Behind Scenes", - }, - }, { - 'md5': '3b3ea897ecaa91fca57a8a94ac1b15c5', - 'info_dict': { - 'id': '1198-56411-000-scene-from-movie', - 'ext': 'mp4', - 'title': "Sinister - Scene from movie", - }, - }], + # series episode, Type=Episode + 'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/', + 'md5': 'd5434c80fcfdb61651cc2199a88d6ba3', 'info_dict': { - 'id': '1198-56411-000', - 'title': "Sinister", - 'description': 'md5:014c40b0488848de9683566a42e33372', - } + 'id': '1284-19427-001', + 'ext': 'flv', + 'title': 'The World and a Wall', + 'description': 'md5:24814cf74d3453fdf5bfef9716d073e3', + 'timestamp': 1428192000, + 'upload_date': '20150405', + 'duration': 1500, + }, + }, { + # serie, Type=Serie + 'url': 'http://www.viewster.com/serie/1303-19426-000/', + 'info_dict': { + 'id': '1303-19426-000', + 'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?', + 'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11', + }, + 'playlist_count': 13, + }, { + # unfinished serie, no Type + 'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/', + 'info_dict': { + 'id': '1284-19427-000', + 'title': 'Baby Steps—Season 2', + 'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1', + }, + 'playlist_mincount': 16, }] _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01' - def _real_extract(self, url): - video_id = self._match_id(url) - - request = compat_urllib_request.Request( - 'http://api.live.viewster.com/api/v1/movie/%s' % video_id) + def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True): + request = compat_urllib_request.Request(url) request.add_header('Accept', self._ACCEPT_HEADER) + request.add_header('Auth-token', self._AUTH_TOKEN) + return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal) - movie = self._download_json( - request, video_id, 'Downloading movie metadata JSON') - - title = movie.get('title') or movie['original_title'] - description = movie.get('synopsis') - thumbnail = movie.get('large_artwork') or movie.get('artwork') - - entries = [] - for clip in movie['play_list']: - entry = None - - # movielink api - link_request = clip.get('link_request') - if link_request: - request = compat_urllib_request.Request( - 'http://api.live.viewster.com/api/v1/movielink?movieid=%(movieid)s&action=%(action)s&paymethod=%(paymethod)s&price=%(price)s¤cy=%(currency)s&language=%(language)s&subtitlelanguage=%(subtitlelanguage)s&ischromecast=%(ischromecast)s' - % link_request) - request.add_header('Accept', self._ACCEPT_HEADER) + def _real_extract(self, url): + video_id = self._match_id(url) + # Get 'api_token' cookie + self._request_webpage(HEADRequest(url), video_id) + cookies = self._get_cookies(url) + self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value) - movie_link = self._download_json( - request, video_id, 'Downloading movie link JSON', fatal=False) + info = self._download_json( + 'https://public-api.viewster.com/search/%s' % video_id, + video_id, 'Downloading entry JSON') - if movie_link: - formats = self._extract_f4m_formats( - movie_link['url'] + '&hdcore=3.2.0&plugin=flowplayer-3.2.0.1', video_id) - self._sort_formats(formats) - entry = { - 'formats': formats, - } + entry_id = info.get('Id') or info['id'] - # direct link - clip_url = clip.get('clip_data', {}).get('url') - if clip_url: - entry = { - 'url': clip_url, - 'ext': 'mp4', - } + # unfinished serie has no Type + if info.get('Type') in ['Serie', None]: + episodes = self._download_json( + 'https://public-api.viewster.com/series/%s/episodes' % entry_id, + video_id, 'Downloading series JSON') + entries = [ + self.url_result( + 'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster') + for episode in episodes] + title = (info.get('Title') or info['Synopsis']['Title']).strip() + description = info.get('Synopsis', {}).get('Detailed') + return self.playlist_result(entries, video_id, title, description) - if entry: - entry.update({ - 'id': '%s-%s' % (video_id, clip['canonical_title']), - 'title': '%s - %s' % (title, clip['title']), + formats = [] + for media_type in ('application/f4m+xml', 'application/x-mpegURL'): + media = self._download_json( + 'https://public-api.viewster.com/movies/%s/video?mediaType=%s' + % (entry_id, compat_urllib_parse.quote(media_type)), + video_id, 'Downloading %s JSON' % media_type, fatal=False) + if not media: + continue + video_url = media.get('Uri') + if not video_url: + continue + ext = determine_ext(video_url) + if ext == 'f4m': + video_url += '&' if '?' in video_url else '?' + video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1' + formats.extend(self._extract_f4m_formats( + video_url, video_id, f4m_id='hds')) + elif ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', m3u8_id='hls', + fatal=False # m3u8 sometimes fail + )) + else: + formats.append({ + 'url': video_url, }) - entries.append(entry) + self._sort_formats(formats) - playlist = self.playlist_result(entries, video_id, title, description) - playlist['thumbnail'] = thumbnail - return playlist + synopsis = info.get('Synopsis', {}) + # Prefer title outside synopsis since it's less messy + title = (info.get('Title') or synopsis['Title']).strip() + description = synopsis.get('Detailed') or info.get('Synopsis', {}).get('Short') + duration = int_or_none(info.get('Duration')) + timestamp = parse_iso8601(info.get('ReleaseDate')) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'timestamp': timestamp, + 'duration': duration, + 'formats': formats, + } diff --git a/youtube_dl/extractor/viki.py b/youtube_dl/extractor/viki.py index 51cdc6b65..ddbd395c8 100644 --- a/youtube_dl/extractor/viki.py +++ b/youtube_dl/extractor/viki.py @@ -88,6 +88,14 @@ class VikiBaseIE(InfoExtractor): if not self._token: self.report_warning('Unable to get session token, login has probably failed') + @staticmethod + def dict_selection(dict_obj, preferred_key): + if preferred_key in dict_obj: + return dict_obj.get(preferred_key) + + filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()])) + return filtered_dict[0] if filtered_dict else None + class VikiIE(VikiBaseIE): IE_NAME = 'viki' @@ -173,6 +181,19 @@ class VikiIE(VikiBaseIE): }, { 'url': 'http://www.viki.com/player/44699v', 'only_matching': True, + }, { + # non-English description + 'url': 'http://www.viki.com/videos/158036v-love-in-magic', + 'md5': '1713ae35df5a521b31f6dc40730e7c9c', + 'info_dict': { + 'id': '158036v', + 'ext': 'mp4', + 'uploader': 'I Planet Entertainment', + 'upload_date': '20111122', + 'timestamp': 1321985454, + 'description': 'md5:44b1e46619df3a072294645c770cef36', + 'title': 'Love In Magic', + }, }] def _real_extract(self, url): @@ -181,19 +202,14 @@ class VikiIE(VikiBaseIE): video = self._call_api( 'videos/%s.json' % video_id, video_id, 'Downloading video JSON') - title = None - titles = video.get('titles') - if titles: - title = titles.get('en') or titles[titles.keys()[0]] + title = self.dict_selection(video.get('titles', {}), 'en') if not title: title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id - container_titles = video.get('container', {}).get('titles') - if container_titles: - container_title = container_titles.get('en') or container_titles[container_titles.keys()[0]] - title = '%s - %s' % (container_title, title) + container_titles = video.get('container', {}).get('titles', {}) + container_title = self.dict_selection(container_titles, 'en') + title = '%s - %s' % (container_title, title) - descriptions = video.get('descriptions') - description = descriptions.get('en') or descriptions[titles.keys()[0]] if descriptions else None + description = self.dict_selection(video.get('descriptions', {}), 'en') duration = int_or_none(video.get('duration')) timestamp = parse_iso8601(video.get('created_at')) @@ -242,8 +258,8 @@ class VikiIE(VikiBaseIE): formats = [] for format_id, stream_dict in streams.items(): - height = self._search_regex( - r'^(\d+)[pP]$', format_id, 'height', default=None) + height = int_or_none(self._search_regex( + r'^(\d+)[pP]$', format_id, 'height', default=None)) for protocol, format_dict in stream_dict.items(): if format_id == 'm3u8': formats = self._extract_m3u8_formats( @@ -299,11 +315,9 @@ class VikiChannelIE(VikiBaseIE): 'containers/%s.json' % channel_id, channel_id, 'Downloading channel JSON') - titles = channel['titles'] - title = titles.get('en') or titles[titles.keys()[0]] + title = self.dict_selection(channel['titles'], 'en') - descriptions = channel['descriptions'] - description = descriptions.get('en') or descriptions[descriptions.keys()[0]] + description = self.dict_selection(channel['descriptions'], 'en') entries = [] for video_type in ('episodes', 'clips', 'movies'): diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 10d6745af..50df79ca1 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -29,6 +29,7 @@ from ..utils import ( class VimeoBaseInfoExtractor(InfoExtractor): _NETRC_MACHINE = 'vimeo' _LOGIN_REQUIRED = False + _LOGIN_URL = 'https://vimeo.com/log_in' def _login(self): (username, password) = self._get_login_info() @@ -37,21 +38,25 @@ class VimeoBaseInfoExtractor(InfoExtractor): raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) return self.report_login() - login_url = 'https://vimeo.com/log_in' - webpage = self._download_webpage(login_url, None, False) - token = self._search_regex(r'xsrft":"(.*?)"', webpage, 'login token') + webpage = self._download_webpage(self._LOGIN_URL, None, False) + token = self._extract_xsrft(webpage) data = urlencode_postdata({ + 'action': 'login', 'email': username, 'password': password, - 'action': 'login', 'service': 'vimeo', 'token': token, }) - login_request = compat_urllib_request.Request(login_url, data) + login_request = compat_urllib_request.Request(self._LOGIN_URL, data) login_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - login_request.add_header('Cookie', 'xsrft=%s' % token) + login_request.add_header('Referer', self._LOGIN_URL) self._download_webpage(login_request, None, False, 'Wrong login info') + def _extract_xsrft(self, webpage): + return self._search_regex( + r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)', + webpage, 'login token', group='xsrft') + class VimeoIE(VimeoBaseInfoExtractor): """Information extractor for vimeo.com.""" @@ -193,7 +198,7 @@ class VimeoIE(VimeoBaseInfoExtractor): password = self._downloader.params.get('videopassword', None) if password is None: raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) - token = self._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token') + token = self._extract_xsrft(webpage) data = urlencode_postdata({ 'password': password, 'token': token, @@ -203,7 +208,7 @@ class VimeoIE(VimeoBaseInfoExtractor): url = url.replace('http://', 'https://') password_request = compat_urllib_request.Request(url + '/password', data) password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Cookie', 'xsrft=%s' % token) + password_request.add_header('Referer', url) return self._download_webpage( password_request, video_id, 'Verifying the password', 'Wrong password') @@ -422,10 +427,11 @@ class VimeoIE(VimeoBaseInfoExtractor): } -class VimeoChannelIE(InfoExtractor): +class VimeoChannelIE(VimeoBaseInfoExtractor): IE_NAME = 'vimeo:channel' _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])' _MORE_PAGES_INDICATOR = r'<a.+?rel="next"' + _TITLE = None _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"' _TESTS = [{ 'url': 'https://vimeo.com/channels/tributes', @@ -440,7 +446,7 @@ class VimeoChannelIE(InfoExtractor): return '%s/videos/page:%d/' % (base_url, pagenum) def _extract_list_title(self, webpage): - return self._html_search_regex(self._TITLE_RE, webpage, 'list title') + return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title') def _login_list_password(self, page_url, list_id, webpage): login_form = self._search_regex( @@ -453,7 +459,7 @@ class VimeoChannelIE(InfoExtractor): if password is None: raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True) fields = self._hidden_inputs(login_form) - token = self._search_regex(r'xsrft[\s=:"\']+([^"\']+)', webpage, 'login token') + token = self._extract_xsrft(webpage) fields['token'] = token fields['password'] = password post = urlencode_postdata(fields) @@ -499,7 +505,7 @@ class VimeoChannelIE(InfoExtractor): class VimeoUserIE(VimeoChannelIE): IE_NAME = 'vimeo:user' - _VALID_URL = r'https://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)' + _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)' _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>' _TESTS = [{ 'url': 'https://vimeo.com/nkistudio/videos', @@ -603,14 +609,14 @@ class VimeoReviewIE(InfoExtractor): return self.url_result(player_url, 'Vimeo', video_id) -class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): +class VimeoWatchLaterIE(VimeoChannelIE): IE_NAME = 'vimeo:watchlater' IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)' - _VALID_URL = r'https://vimeo\.com/home/watchlater|:vimeowatchlater' + _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater' + _TITLE = 'Watch Later' _LOGIN_REQUIRED = True - _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<' _TESTS = [{ - 'url': 'https://vimeo.com/home/watchlater', + 'url': 'https://vimeo.com/watchlater', 'only_matching': True, }] @@ -626,7 +632,7 @@ class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE): return request def _real_extract(self, url): - return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater') + return self._extract_videos('watchlater', 'https://vimeo.com/watchlater') class VimeoLikesIE(InfoExtractor): diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index b4ad513a0..97315750f 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -4,7 +4,6 @@ import re from .common import InfoExtractor from ..utils import ( - ExtractorError, unified_strdate, str_to_int, int_or_none, @@ -22,7 +21,7 @@ class XHamsterIE(InfoExtractor): 'ext': 'mp4', 'title': 'FemaleAgent Shy beauty takes the bait', 'upload_date': '20121014', - 'uploader_id': 'Ruseful2011', + 'uploader': 'Ruseful2011', 'duration': 893, 'age_limit': 18, } @@ -34,7 +33,7 @@ class XHamsterIE(InfoExtractor): 'ext': 'mp4', 'title': 'Britney Spears Sexy Booty', 'upload_date': '20130914', - 'uploader_id': 'jojo747400', + 'uploader': 'jojo747400', 'duration': 200, 'age_limit': 18, } @@ -46,12 +45,12 @@ class XHamsterIE(InfoExtractor): ] def _real_extract(self, url): - def extract_video_url(webpage): - mp4 = re.search(r'<video\s+.*?file="([^"]+)".*?>', webpage) - if mp4 is None: - raise ExtractorError('Unable to extract media URL') - else: - return mp4.group(1) + def extract_video_url(webpage, name): + return self._search_regex( + [r'''file\s*:\s*(?P<q>["'])(?P<mp4>.+?)(?P=q)''', + r'''<a\s+href=(?P<q>["'])(?P<mp4>.+?)(?P=q)\s+class=["']mp4Thumb''', + r'''<video[^>]+file=(?P<q>["'])(?P<mp4>.+?)(?P=q)[^>]*>'''], + webpage, name, group='mp4') def is_hd(webpage): return '<div class=\'icon iconHD\'' in webpage @@ -75,10 +74,14 @@ class XHamsterIE(InfoExtractor): if upload_date: upload_date = unified_strdate(upload_date) - uploader_id = self._html_search_regex(r'<a href=\'/user/[^>]+>(?P<uploader_id>[^<]+)', - webpage, 'uploader id', default='anonymous') + uploader = self._html_search_regex( + r"<a href='[^']+xhamster\.com/user/[^>]+>(?P<uploader>[^<]+)", + webpage, 'uploader', default='anonymous') - thumbnail = self._html_search_regex(r'<video\s+.*?poster="([^"]+)".*?>', webpage, 'thumbnail', fatal=False) + thumbnail = self._search_regex( + [r'''thumb\s*:\s*(?P<q>["'])(?P<thumbnail>.+?)(?P=q)''', + r'''<video[^>]+poster=(?P<q>["'])(?P<thumbnail>.+?)(?P=q)[^>]*>'''], + webpage, 'thumbnail', fatal=False, group='thumbnail') duration = parse_duration(self._html_search_regex(r'<span>Runtime:</span> (\d+:\d+)</div>', webpage, 'duration', fatal=False)) @@ -97,7 +100,9 @@ class XHamsterIE(InfoExtractor): hd = is_hd(webpage) - video_url = extract_video_url(webpage) + format_id = 'hd' if hd else 'sd' + + video_url = extract_video_url(webpage, format_id) formats = [{ 'url': video_url, 'format_id': 'hd' if hd else 'sd', @@ -108,7 +113,7 @@ class XHamsterIE(InfoExtractor): mrss_url = self._search_regex(r'<link rel="canonical" href="([^"]+)', webpage, 'mrss_url') webpage = self._download_webpage(mrss_url + '?hd', video_id, note='Downloading HD webpage') if is_hd(webpage): - video_url = extract_video_url(webpage) + video_url = extract_video_url(webpage, 'hd') formats.append({ 'url': video_url, 'format_id': 'hd', @@ -122,7 +127,7 @@ class XHamsterIE(InfoExtractor): 'title': title, 'description': description, 'upload_date': upload_date, - 'uploader_id': uploader_id, + 'uploader': uploader, 'thumbnail': thumbnail, 'duration': duration, 'view_count': view_count, diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index e7f5c7861..facd837ad 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -19,6 +19,7 @@ from ..compat import ( compat_urllib_parse, compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus, + compat_urllib_parse_urlparse, compat_urllib_request, compat_urlparse, compat_str, @@ -31,9 +32,12 @@ from ..utils import ( get_element_by_id, int_or_none, orderedSet, + parse_duration, + smuggle_url, str_to_int, unescapeHTML, unified_strdate, + unsmuggle_url, uppercase_escape, ISO3166Utils, ) @@ -209,7 +213,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): |(?: # or the v= param in all its forms (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! - (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) + (?:.*?&)?? # any other preceding param (like /?s=tuff&v=xxxx) v= ) )) @@ -279,13 +283,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, # Dash webm - '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, - '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'}, + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'}, '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, @@ -295,11 +299,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, - '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, - '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, - '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, - '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'}, - '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}, # Dash webm audio '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, @@ -317,7 +321,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): IE_NAME = 'youtube' _TESTS = [ { - 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc', + 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9', 'info_dict': { 'id': 'BaW_jenozKc', 'ext': 'mp4', @@ -327,8 +331,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20121002', 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', 'categories': ['Science & Technology'], + 'tags': ['youtube-dl'], 'like_count': int, 'dislike_count': int, + 'start_time': 1, + 'end_time': 9, } }, { @@ -339,7 +346,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'ext': 'mp4', 'upload_date': '20120506', 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]', - 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f', + 'description': 'md5:782e8651347686cba06e58f71ab51773', + 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli', + 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop', + 'iconic ep', 'iconic', 'love', 'it'], 'uploader': 'Icona Pop', 'uploader_id': 'IconaPop', } @@ -355,6 +365,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'description': 'md5:64249768eec3bc4276236606ea996373', 'uploader': 'justintimberlakeVEVO', 'uploader_id': 'justintimberlakeVEVO', + 'age_limit': 18, } }, { @@ -371,6 +382,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } }, { + 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY', + 'note': 'Use the first video ID in the URL', + 'info_dict': { + 'id': 'BaW_jenozKc', + 'ext': 'mp4', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', + 'uploader': 'Philipp Hagemeister', + 'uploader_id': 'phihag', + 'upload_date': '20121002', + 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', + 'categories': ['Science & Technology'], + 'tags': ['youtube-dl'], + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'skip_download': True, + }, + }, + { 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I', 'note': '256k DASH audio (format 141) via DASH manifest', 'info_dict': { @@ -411,7 +442,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'id': 'nfWlot6h_JM', 'ext': 'm4a', 'title': 'Taylor Swift - Shake It Off', - 'description': 'md5:2acfda1b285bdd478ccec22f9918199d', + 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3', 'uploader': 'TaylorSwiftVEVO', 'uploader_id': 'TaylorSwiftVEVO', 'upload_date': '20140818', @@ -445,6 +476,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'The Witcher', 'uploader_id': 'WitcherGame', 'upload_date': '20140605', + 'age_limit': 18, }, }, # Age-gate video with encrypted signature @@ -458,6 +490,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'LloydVEVO', 'uploader_id': 'LloydVEVO', 'upload_date': '20110629', + 'age_limit': 18, }, }, # video_info is None (https://github.com/rg3/youtube-dl/issues/4421) @@ -482,7 +515,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'info_dict': { 'id': 'lqQg6PlCWgI', 'ext': 'mp4', - 'upload_date': '20120731', + 'upload_date': '20120724', 'uploader_id': 'olympic', 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', 'uploader': 'Olympics', @@ -511,7 +544,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'url': 'qEJwOuvDf7I', 'info_dict': { 'id': 'qEJwOuvDf7I', - 'ext': 'mp4', + 'ext': 'webm', 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге', 'description': '', 'upload_date': '20150404', @@ -535,6 +568,77 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'dorappi2000', 'formats': 'mincount:33', }, + }, + # DASH manifest with segment_list + { + 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8', + 'md5': '8ce563a1d667b599d21064e982ab9e31', + 'info_dict': { + 'id': 'CsmdDsKjzN8', + 'ext': 'mp4', + 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510 + 'uploader': 'Airtek', + 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.', + 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ', + 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '135', # bestvideo + } + }, + { + # Multifeed videos (multiple cameras), URL is for Main Camera + 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs', + 'info_dict': { + 'id': 'jqWvoWXjCVs', + 'title': 'teamPGP: Rocket League Noob Stream', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + }, + 'playlist': [{ + 'info_dict': { + 'id': 'jqWvoWXjCVs', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }, { + 'info_dict': { + 'id': '6h8e8xoXJzg', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }, { + 'info_dict': { + 'id': 'PUOgX5z9xZw', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (grizzle)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }, { + 'info_dict': { + 'id': 'teuwxikvS5k', + 'ext': 'mp4', + 'title': 'teamPGP: Rocket League Noob Stream (zim)', + 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'upload_date': '20150721', + 'uploader': 'Beer Games Beer', + 'uploader_id': 'beergamesbeer', + }, + }], + 'params': { + 'skip_download': True, + }, } ] @@ -826,6 +930,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # TODO implement WebVTT downloading pass elif mime_type.startswith('audio/') or mime_type.startswith('video/'): + segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList') format_id = r.attrib['id'] video_url = url_el.text filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) @@ -839,6 +944,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'filesize': filesize, 'fps': int_or_none(r.attrib.get('frameRate')), } + if segment_list is not None: + f.update({ + 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'], + 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')], + 'protocol': 'http_dash_segments', + }) try: existing_format = next( fo for fo in formats @@ -860,10 +971,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return formats def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + proto = ( 'http' if self._downloader.params.get('prefer_insecure', False) else 'https') + start_time = None + end_time = None + parsed_url = compat_urllib_parse_urlparse(url) + for component in [parsed_url.fragment, parsed_url.query]: + query = compat_parse_qs(component) + if start_time is None and 't' in query: + start_time = parse_duration(query['t'][0]) + if start_time is None and 'start' in query: + start_time = parse_duration(query['start'][0]) + if end_time is None and 'end' in query: + end_time = parse_duration(query['end'][0]) + # Extract original video URL from URL with redirection, like age verification, using next_url parameter mobj = re.search(self._NEXT_URL_RE, url) if mobj: @@ -890,6 +1015,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Get video info embed_webpage = None + is_live = None if re.search(r'player-age-gate-content">', video_webpage) is not None: age_gate = True # We simulate the access to the video from www.youtube.com/v/{video_id} @@ -922,6 +1048,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Convert to the same format returned by compat_parse_qs video_info = dict((k, [v]) for k, v in args.items()) add_dash_mpd(video_info) + if args.get('livestream') == '1' or args.get('live_playback') == 1: + is_live = True if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True): # We also try looking in get_video_info since it may contain different dashmpd # URL that points to a DASH manifest with possibly different itag set (some itags @@ -939,7 +1067,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): video_id, note=False, errnote='unable to download video info webpage') get_video_info = compat_parse_qs(video_info_webpage) - add_dash_mpd(get_video_info) + if get_video_info.get('use_cipher_signature') != ['True']: + add_dash_mpd(get_video_info) if not video_info: video_info = get_video_info if 'token' in get_video_info: @@ -948,7 +1077,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if 'reason' in video_info: if 'The uploader has not made this video available in your country.' in video_info['reason']: regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None) - if regions_allowed is not None: + if regions_allowed: raise ExtractorError('YouTube said: This video is available in %s only' % ( ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))), expected=True) @@ -960,6 +1089,55 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '"token" parameter not in video info for unknown reason', video_id=video_id) + # title + if 'title' in video_info: + video_title = video_info['title'][0] + else: + self._downloader.report_warning('Unable to extract video title') + video_title = '_' + + # description + video_description = get_element_by_id("eow-description", video_webpage) + if video_description: + video_description = re.sub(r'''(?x) + <a\s+ + (?:[a-zA-Z-]+="[^"]+"\s+)*? + title="([^"]+)"\s+ + (?:[a-zA-Z-]+="[^"]+"\s+)*? + class="yt-uix-redirect-link"\s*> + [^<]+ + </a> + ''', r'\1', video_description) + video_description = clean_html(video_description) + else: + fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) + if fd_mobj: + video_description = unescapeHTML(fd_mobj.group(1)) + else: + video_description = '' + + if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False): + if not self._downloader.params.get('noplaylist'): + entries = [] + feed_ids = [] + multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0]) + for feed in multifeed_metadata_list.split(','): + feed_data = compat_parse_qs(feed) + entries.append({ + '_type': 'url_transparent', + 'ie_key': 'Youtube', + 'url': smuggle_url( + '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]), + {'force_singlefeed': True}), + 'title': '%s (%s)' % (video_title, feed_data['title'][0]), + }) + feed_ids.append(feed_data['id'][0]) + self.to_screen( + 'Downloading multifeed video (%s) - add --no-playlist to just download video %s' + % (', '.join(feed_ids), video_id)) + return self.playlist_result(entries, video_id, video_title, video_description) + self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + if 'view_count' in video_info: view_count = int(video_info['view_count'][0]) else: @@ -985,13 +1163,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): else: self._downloader.report_warning('unable to extract uploader nickname') - # title - if 'title' in video_info: - video_title = video_info['title'][0] - else: - self._downloader.report_warning('Unable to extract video title') - video_title = '_' - # thumbnail image # We try first to get a high quality image: m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', @@ -1027,25 +1198,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): else: video_categories = None - # description - video_description = get_element_by_id("eow-description", video_webpage) - if video_description: - video_description = re.sub(r'''(?x) - <a\s+ - (?:[a-zA-Z-]+="[^"]+"\s+)*? - title="([^"]+)"\s+ - (?:[a-zA-Z-]+="[^"]+"\s+)*? - class="yt-uix-redirect-link"\s*> - [^<]+ - </a> - ''', r'\1', video_description) - video_description = clean_html(video_description) - else: - fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) - if fd_mobj: - video_description = unescapeHTML(fd_mobj.group(1)) - else: - video_description = '' + video_tags = [ + unescapeHTML(m.group('content')) + for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)] def _extract_count(count_name): return str_to_int(self._search_regex( @@ -1215,6 +1370,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'thumbnail': video_thumbnail, 'description': video_description, 'categories': video_categories, + 'tags': video_tags, 'subtitles': video_subtitles, 'automatic_captions': automatic_captions, 'duration': video_duration, @@ -1226,6 +1382,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'dislike_count': dislike_count, 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]), 'formats': formats, + 'is_live': is_live, + 'start_time': start_time, + 'end_time': end_time, } diff --git a/youtube_dl/options.py b/youtube_dl/options.py index 85365d769..9016e3498 100644 --- a/youtube_dl/options.py +++ b/youtube_dl/options.py @@ -219,7 +219,7 @@ def parseOpts(overrideArguments=None): selection.add_option( '--playlist-items', dest='playlist_items', metavar='ITEM_SPEC', default=None, - help='Playlist video items to download. Specify indices of the videos in the playlist seperated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.') + help='Playlist video items to download. Specify indices of the videos in the playlist separated by commas like: "--playlist-items 1,2,5,8" if you want to download videos indexed 1, 2, 5, 8 in the playlist. You can specify range: "--playlist-items 1-3,7,10-13", it will download the videos at index 1, 2, 3, 7, 10, 11, 12 and 13.') selection.add_option( '--match-title', dest='matchtitle', metavar='REGEX', diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 942f76d24..e265c7574 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -139,21 +139,24 @@ def write_json_file(obj, fn): if sys.version_info >= (2, 7): - def find_xpath_attr(node, xpath, key, val): + def find_xpath_attr(node, xpath, key, val=None): """ Find the xpath xpath[@key=val] """ assert re.match(r'^[a-zA-Z-]+$', key) - assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val) - expr = xpath + "[@%s='%s']" % (key, val) + if val: + assert re.match(r'^[a-zA-Z0-9@\s:._-]*$', val) + expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val)) return node.find(expr) else: - def find_xpath_attr(node, xpath, key, val): + def find_xpath_attr(node, xpath, key, val=None): # Here comes the crazy part: In 2.6, if the xpath is a unicode, # .//node does not match if a node is a direct child of . ! if isinstance(xpath, compat_str): xpath = xpath.encode('ascii') for f in node.findall(xpath): - if f.attrib.get(key) == val: + if key not in f.attrib: + continue + if val is None or f.attrib.get(key) == val: return f return None @@ -576,11 +579,9 @@ class ContentTooShortError(Exception): download is too small for what the server announced first, indicating the connection was probably interrupted. """ - # Both in bytes - downloaded = None - expected = None def __init__(self, downloaded, expected): + # Both in bytes self.downloaded = downloaded self.expected = expected @@ -650,6 +651,26 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): return ret def http_request(self, req): + # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not + # always respected by websites, some tend to give out URLs with non percent-encoded + # non-ASCII characters (see telemb.py, ard.py [#3412]) + # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) + # To work around aforementioned issue we will replace request's original URL with + # percent-encoded one + # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) + # the code of this workaround has been moved here from YoutubeDL.urlopen() + url = req.get_full_url() + url_escaped = escape_url(url) + + # Substitute URL if any change after escaping + if url != url_escaped: + req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request + new_req = req_type( + url_escaped, data=req.data, headers=req.headers, + origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) + new_req.timeout = req.timeout + req = new_req + for h, v in std_headers.items(): # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 # The dict keys are capitalized because of this bug by urllib @@ -694,6 +715,17 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): gz = io.BytesIO(self.deflate(resp.read())) resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) resp.msg = old_resp.msg + # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 + if 300 <= resp.code < 400: + location = resp.headers.get('Location') + if location: + # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 + if sys.version_info >= (3, 0): + location = location.encode('iso-8859-1').decode('utf-8') + location_escaped = escape_url(location) + if location != location_escaped: + del resp.headers['Location'] + resp.headers['Location'] = location_escaped return resp https_request = http_request @@ -1309,10 +1341,10 @@ def parse_duration(s): m = re.match( r'''(?ix)(?:P?T)? (?: - (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*| + (?P<only_mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*| (?P<only_hours>[0-9.]+)\s*(?:hours?)| - \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*| + \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?\.?|minutes?)\s*| (?: (?: (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)? diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 3ad7a2bc0..6462d4477 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2015.07.18' +__version__ = '2015.08.09' |