diff options
Diffstat (limited to 'youtube_dl/extractor/common.py')
| -rw-r--r-- | youtube_dl/extractor/common.py | 90 |
1 files changed, 60 insertions, 30 deletions
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 1e7db8a9b..655207447 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -10,20 +10,17 @@ import re import socket import sys import time -import xml.etree.ElementTree from ..compat import ( compat_cookiejar, compat_cookies, compat_getpass, - compat_HTTPError, compat_http_client, compat_urllib_error, compat_urllib_parse, - compat_urllib_parse_urlparse, - compat_urllib_request, compat_urlparse, compat_str, + compat_etree_fromstring, ) from ..utils import ( NO_DEFAULT, @@ -32,16 +29,20 @@ from ..utils import ( clean_html, compiled_regex_type, determine_ext, + error_to_compat_str, ExtractorError, fix_xml_ampersands, float_or_none, int_or_none, RegexNotFoundError, sanitize_filename, + sanitized_Request, unescapeHTML, + unified_strdate, url_basename, xpath_text, xpath_with_ns, + determine_protocol, ) @@ -152,6 +153,7 @@ class InfoExtractor(object): description: Full video description. uploader: Full name of the video uploader. creator: The main artist who created the video. + release_date: The date (YYYYMMDD) when the video was released. timestamp: UNIX timestamp of the moment the video became available. upload_date: Video upload date (YYYYMMDD). If not explicitly set, calculated from timestamp. @@ -163,12 +165,14 @@ class InfoExtractor(object): with the "ext" entry and one of: * "data": The subtitles file contents * "url": A URL pointing to the subtitles file + "ext" will be calculated from URL if missing automatic_captions: Like 'subtitles', used by the YoutubeIE for automatically generated captions - duration: Length of the video in seconds, as an integer. + duration: Length of the video in seconds, as an integer or float. view_count: How many users have watched the video on the platform. like_count: Number of positive ratings of the video dislike_count: Number of negative ratings of the video + repost_count: Number of reposts of the video average_rating: Average rating give by users, the scale used depends on the webpage comment_count: Number of comments on the video comments: A list of comments, each with one or more of the following @@ -307,11 +311,11 @@ class InfoExtractor(object): @classmethod def ie_key(cls): """A string for getting the InfoExtractor with get_info_extractor""" - return cls.__name__[:-2] + return compat_str(cls.__name__[:-2]) @property def IE_NAME(self): - return type(self).__name__[:-2] + return compat_str(type(self).__name__[:-2]) def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): """ Returns the response handle """ @@ -329,7 +333,8 @@ class InfoExtractor(object): return False if errnote is None: errnote = 'Unable to download webpage' - errmsg = '%s: %s' % (errnote, compat_str(err)) + + errmsg = '%s: %s' % (errnote, error_to_compat_str(err)) if fatal: raise ExtractorError(errmsg, sys.exc_info()[2], cause=err) else: @@ -458,7 +463,7 @@ class InfoExtractor(object): return xml_string if transform_source: xml_string = transform_source(xml_string) - return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8')) + return compat_etree_fromstring(xml_string.encode('utf-8')) def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', @@ -619,7 +624,7 @@ class InfoExtractor(object): else: raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) except (IOError, netrc.NetrcParseError) as err: - self._downloader.report_warning('parsing .netrc: %s' % compat_str(err)) + self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err)) return (username, password) @@ -642,8 +647,9 @@ class InfoExtractor(object): # Helper functions for extracting OpenGraph info @staticmethod def _og_regexes(prop): - content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')' - property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop) + content_re = r'content=(?:"([^"]+?)"|\'([^\']+?)\'|\s*([^\s"\'=<>`]+?))' + property_re = (r'(?:name|property)=(?:\'og:%(prop)s\'|"og:%(prop)s"|\s*og:%(prop)s\b)' + % {'prop': re.escape(prop)}) template = r'<meta[^>]+?%s[^>]+?%s' return [ template % (property_re, content_re), @@ -772,14 +778,12 @@ class InfoExtractor(object): preference = f.get('preference') if preference is None: - proto = f.get('protocol') - if proto is None: - proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme - - preference = 0 if proto in ['http', 'https'] else -0.1 + preference = 0 if f.get('ext') in ['f4f', 'f4m']: # Not yet supported preference -= 0.5 + proto_preference = 0 if determine_protocol(f) in ['http', 'https'] else -0.1 + if f.get('vcodec') == 'none': # audio only if self._downloader.params.get('prefer_free_formats'): ORDER = ['aac', 'mp3', 'm4a', 'webm', 'ogg', 'opus'] @@ -810,6 +814,7 @@ class InfoExtractor(object): f.get('vbr') if f.get('vbr') is not None else -1, f.get('height') if f.get('height') is not None else -1, f.get('width') if f.get('width') is not None else -1, + proto_preference, ext_preference, f.get('abr') if f.get('abr') is not None else -1, audio_ext_preference, @@ -837,7 +842,7 @@ class InfoExtractor(object): self._request_webpage(url, video_id, 'Checking %s URL' % item) return True except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError): + if isinstance(e.cause, compat_urllib_error.URLError): self.to_screen( '%s: %s URL is invalid, skipping' % (video_id, item)) return False @@ -868,13 +873,18 @@ class InfoExtractor(object): time.sleep(timeout) def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None, - transform_source=lambda s: fix_xml_ampersands(s).strip()): + transform_source=lambda s: fix_xml_ampersands(s).strip(), + fatal=True): manifest = self._download_xml( manifest_url, video_id, 'Downloading f4m manifest', 'Unable to download f4m manifest', # Some manifests may be malformed, e.g. prosiebensat1 generated manifests # (see https://github.com/rg3/youtube-dl/issues/6215#issuecomment-121704244) - transform_source=transform_source) + transform_source=transform_source, + fatal=fatal) + + if manifest is False: + return [] formats = [] manifest_version = '1.0' @@ -882,6 +892,11 @@ class InfoExtractor(object): if not media_nodes: manifest_version = '2.0' media_nodes = manifest.findall('{http://ns.adobe.com/f4m/2.0}media') + base_url = xpath_text( + manifest, ['{http://ns.adobe.com/f4m/1.0}baseURL', '{http://ns.adobe.com/f4m/2.0}baseURL'], + 'base URL', default=None) + if base_url: + base_url = base_url.strip() for i, media_el in enumerate(media_nodes): if manifest_version == '2.0': media_url = media_el.attrib.get('href') or media_el.attrib.get('url') @@ -889,13 +904,14 @@ class InfoExtractor(object): continue manifest_url = ( media_url if media_url.startswith('http://') or media_url.startswith('https://') - else ('/'.join(manifest_url.split('/')[:-1]) + '/' + media_url)) + else ((base_url or '/'.join(manifest_url.split('/')[:-1])) + '/' + media_url)) # If media_url is itself a f4m manifest do the recursive extraction # since bitrates in parent manifest (this one) and media_url manifest # may differ leading to inability to resolve the format by requested # bitrate in f4m downloader if determine_ext(manifest_url) == 'f4m': - formats.extend(self._extract_f4m_formats(manifest_url, video_id, preference, f4m_id)) + formats.extend(self._extract_f4m_formats( + manifest_url, video_id, preference, f4m_id, fatal=fatal)) continue tbr = int_or_none(media_el.attrib.get('bitrate')) formats.append({ @@ -931,13 +947,15 @@ class InfoExtractor(object): if re.match(r'^https?://', u) else compat_urlparse.urljoin(m3u8_url, u)) - m3u8_doc = self._download_webpage( + res = self._download_webpage_handle( m3u8_url, video_id, note=note or 'Downloading m3u8 information', errnote=errnote or 'Failed to download m3u8 information', fatal=fatal) - if m3u8_doc is False: - return m3u8_doc + if res is False: + return [] + m3u8_doc, urlh = res + m3u8_url = urlh.geturl() last_info = None last_media = None kv_rex = re.compile( @@ -1043,6 +1061,7 @@ class InfoExtractor(object): video_id = os.path.splitext(url_basename(smil_url))[0] title = None description = None + upload_date = None for meta in smil.findall(self._xpath_ns('./head/meta', namespace)): name = meta.attrib.get('name') content = meta.attrib.get('content') @@ -1052,11 +1071,22 @@ class InfoExtractor(object): title = content elif not description and name in ('description', 'abstract'): description = content + elif not upload_date and name == 'date': + upload_date = unified_strdate(content) + + thumbnails = [{ + 'id': image.get('type'), + 'url': image.get('src'), + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } for image in smil.findall(self._xpath_ns('.//image', namespace)) if image.get('src')] return { 'id': video_id, 'title': title or video_id, 'description': description, + 'upload_date': upload_date, + 'thumbnails': thumbnails, 'formats': formats, 'subtitles': subtitles, } @@ -1083,7 +1113,7 @@ class InfoExtractor(object): if not src: continue - bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) + bitrate = float_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')) @@ -1116,7 +1146,7 @@ class InfoExtractor(object): if proto == 'm3u8' or src_ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - src_url, video_id, ext or 'mp4', m3u8_id='hls')) + src_url, video_id, ext or 'mp4', m3u8_id='hls', fatal=False)) continue if src_ext == 'f4m': @@ -1128,10 +1158,10 @@ class InfoExtractor(object): } 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')) + formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False)) continue - if src_url.startswith('http'): + if src_url.startswith('http') and self._is_valid_url(src, video_id): http_count += 1 formats.append({ 'url': src_url, @@ -1250,7 +1280,7 @@ class InfoExtractor(object): def _get_cookies(self, url): """ Return a compat_cookies.SimpleCookie with the cookies for the url """ - req = compat_urllib_request.Request(url) + req = sanitized_Request(url) self._downloader.cookiejar.add_cookie_header(req) return compat_cookies.SimpleCookie(req.get_header('Cookie')) |