X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=inline;f=youtube_dl%2Fextractor%2Fcommon.py;h=48f58aae384c7330648cca771062742fe4337718;hb=cd298882cd6f8ed2571dd372f684ec17e992fd9d;hp=611cf95f1125ec9340a96554df9542d1fcbd62b4;hpb=852f8641e8fb910160b809bb05f7c2533882195d;p=youtube-dl diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index 611cf95f1..48f58aae3 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -12,13 +12,18 @@ import sys import time import xml.etree.ElementTree -from ..utils import ( +from ..compat import ( + compat_cookiejar, + compat_HTTPError, compat_http_client, compat_urllib_error, compat_urllib_parse_urlparse, compat_urlparse, compat_str, - +) +from ..utils import ( + age_restricted, + bug_reports_message, clean_html, compiled_regex_type, ExtractorError, @@ -38,11 +43,15 @@ class InfoExtractor(object): information about the video (or videos) the URL refers to. This information includes the real video URL, the video title, author and others. The information is stored in a dictionary which is then - passed to the FileDownloader. The FileDownloader processes this + passed to the YoutubeDL. The YoutubeDL processes this information possibly downloading the video to the file system, among other possible outcomes. - The dictionaries must include the following fields: + The type field determines the type of the result. + By far the most common value (and the default if _type is missing) is + "video", which indicates a single video. + + For a video, the dictionaries must include the following fields: id: Video identifier. title: Video title, unescaped. @@ -72,6 +81,7 @@ class InfoExtractor(object): * acodec Name of the audio codec in use * asr Audio sampling rate in Hertz * vbr Average video bitrate in KBit/s + * fps Frame rate * vcodec Name of the video codec in use * container Name of the container format * filesize The number of bytes, if known in advance @@ -79,22 +89,39 @@ class InfoExtractor(object): * player_url SWF Player URL (used for rtmpdump). * protocol The protocol that will be used for the actual download, lower-case. - "http", "https", "rtsp", "rtmp", "m3u8" or so. + "http", "https", "rtsp", "rtmp", "rtmpe", + "m3u8", or "m3u8_native". * preference Order number of this format. If this field is present and not None, the formats get sorted by this field, regardless of all other values. -1 for default (order by other properties), -2 or smaller for less than default. + < -1000 to hide the format (if there is + another one which is strictly better) + * language_preference Is this in the correct requested + language? + 10 if it's what the URL is about, + -1 for default (don't know), + -10 otherwise, other values reserved for now. * quality Order number of the video quality of this format, irrespective of the file format. -1 for default (order by other properties), -2 or smaller for less than default. - * http_referer HTTP Referer header value to set. + * source_preference Order number for this video source + (quality takes higher priority) + -1 for default (order by other properties), + -2 or smaller for less than default. * http_method HTTP method to use for the download. * http_headers A dictionary of additional HTTP headers to add to the request. * http_post_data Additional data to send with a POST request. + * stretched_ratio If given and not 1, indicates that the + video's pixels are not square. + width : height ratio as float. + * no_resume The server does not support resuming the + (HTTP or RTMP) download. Boolean. + url: Final video URL. ext: Video filename extension. format: The video format, defaults to ext (used for --get-format) @@ -102,31 +129,53 @@ class InfoExtractor(object): The following fields are optional: + alt_title: A secondary title of the video. display_id An alternative identifier for the video, not necessarily unique, but available before title. Typically, id is something like "4234987", title "Dancing naked mole rats", and display_id "dancing-naked-mole-rats" thumbnails: A list of dictionaries, with the following entries: + * "id" (optional, string) - Thumbnail format ID * "url" + * "preference" (optional, int) - quality of the image * "width" (optional, int) * "height" (optional, int) * "resolution" (optional, string "{width}x{height"}, deprecated) thumbnail: Full URL to a video thumbnail image. - description: One-line video description. + description: Full video description. uploader: Full name of the video uploader. + creator: The main artist who created the video. timestamp: UNIX timestamp of the moment the video became available. upload_date: Video upload date (YYYYMMDD). If not explicitly set, calculated from timestamp. uploader_id: Nickname or id of the video uploader. location: Physical location where the video was filmed. - subtitles: The subtitle file contents as a dictionary in the format - {language: subtitles}. + subtitles: The available subtitles as a dictionary in the format + {language: subformats}. "subformats" is a list sorted from + 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 + automatic_captions: Like 'subtitles', used by the YoutubeIE for + automatically generated captions duration: Length of the video in seconds, as an integer. 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 + 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 + properties (all but one of text or html optional): + * "author" - human-readable name of the comment author + * "author_id" - user ID of the comment author + * "id" - Comment ID + * "html" - Comment as HTML + * "text" - Plain text of the comment + * "timestamp" - UNIX timestamp of comment + * "parent" - ID of the comment this one is replying to. + 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 should allow to get the same result again. (It will be set @@ -138,6 +187,41 @@ class InfoExtractor(object): Unless mentioned otherwise, the fields should be Unicode strings. + Unless mentioned otherwise, None is equivalent to absence of information. + + + _type "playlist" indicates multiple videos. + 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). + + + _type "multi_video" indicates that there are multiple videos that + form a single show, for examples multiple acts of an opera or TV episode. + It must have an entries key like a playlist and contain all the keys + required for a video at the same time. + + + _type "url" indicates that the video must be extracted from another + location, possibly by a different extractor. Its only required key is: + "url" - the next URL to extract. + The key "ie_key" can be set to the class name (minus the trailing "IE", + e.g. "Youtube") if the extractor class is known in advance. + Additionally, the dictionary may have any properties of the resolved entity + known in advance, for example "title" if the title of the referred video is + known ahead of time. + + + _type "url_transparent" entities have the same specification as "url", but + indicate that the given additional information is more precise than the one + associated with the resolved URL. + This is useful when a site employs a video service that hosts the video and + its technical metadata, but that video service does not embed a useful + title, description etc. + + Subclasses of this one should re-define the _real_initialize() and _real_extract() methods and define a _VALID_URL regexp. Probably, they should also be added to the list of extractors. @@ -187,8 +271,15 @@ class InfoExtractor(object): def extract(self, url): """Extracts URL information and returns it in list of dicts.""" - self.initialize() - return self._real_extract(url) + try: + self.initialize() + return self._real_extract(url) + except ExtractorError: + raise + except compat_http_client.IncompleteRead as e: + raise ExtractorError('A network error has occured.', cause=e, expected=True) + except (KeyError, StopIteration) as e: + raise ExtractorError('An extractor error has occured.', cause=e) def set_downloader(self, downloader): """Sets the downloader for this IE.""" @@ -234,9 +325,8 @@ class InfoExtractor(object): self._downloader.report_warning(errmsg) return False - def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True): + def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True, encoding=None): """ Returns a tuple (page content as string, URL handle) """ - # Strip hashes from the URL (#1038) if isinstance(url_or_request, (compat_str, str)): url_or_request = url_or_request.partition('#')[0] @@ -245,8 +335,11 @@ class InfoExtractor(object): if urlh is False: assert not fatal return False - content_type = urlh.headers.get('Content-Type', '') - webpage_bytes = urlh.read() + content = self._webpage_read_content(urlh, url_or_request, video_id, note, errnote, fatal, encoding=encoding) + return (content, urlh) + + @staticmethod + def _guess_encoding_from_content(content_type, webpage_bytes): m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) if m: encoding = m.group(1) @@ -259,6 +352,16 @@ class InfoExtractor(object): encoding = 'utf-16' else: encoding = 'utf-8' + + return encoding + + def _webpage_read_content(self, urlh, url_or_request, video_id, note=None, errnote=None, fatal=True, prefix=None, encoding=None): + content_type = urlh.headers.get('Content-Type', '') + webpage_bytes = urlh.read() + if prefix is not None: + webpage_bytes = prefix + webpage_bytes + if not encoding: + encoding = self._guess_encoding_from_content(content_type, webpage_bytes) if self._downloader.params.get('dump_intermediate_pages', False): try: url = url_or_request.get_full_url() @@ -279,6 +382,12 @@ class InfoExtractor(object): raw_filename = basen + '.dump' filename = sanitize_filename(raw_filename, restricted=True) self.to_screen('Saving request to ' + filename) + # Working around MAX_PATH limitation on Windows (see + # http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx) + if os.name == 'nt': + absfilepath = os.path.abspath(filename) + if len(absfilepath) > 259: + filename = '\\\\?\\' + absfilepath with open(filename, 'wb') as outf: outf.write(webpage_bytes) @@ -296,12 +405,32 @@ class InfoExtractor(object): if blocked_iframe: msg += ' Visit %s for more details' % blocked_iframe raise ExtractorError(msg, expected=True) + if '
(.*?)
', + content, 'block message', default=None) + if block_msg: + msg += ' (Message: "%s")' % block_msg.replace('\n', ' ') + raise ExtractorError(msg, expected=True) - return (content, urlh) + return content - def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True): + def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True, tries=1, timeout=5, encoding=None): """ Returns the data of the page as a string """ - res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal) + success = False + try_count = 0 + while success is False: + try: + res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal, encoding=encoding) + success = True + except compat_http_client.IncompleteRead as e: + try_count += 1 + if try_count >= tries: + raise e + self._sleep(timeout, video_id) if res is False: return res else: @@ -310,10 +439,10 @@ class InfoExtractor(object): def _download_xml(self, url_or_request, video_id, note='Downloading XML', errnote='Unable to download XML', - transform_source=None, fatal=True): + transform_source=None, fatal=True, encoding=None): """Return the xml as an xml.etree.ElementTree.Element""" xml_string = self._download_webpage( - url_or_request, video_id, note, errnote, fatal=fatal) + url_or_request, video_id, note, errnote, fatal=fatal, encoding=encoding) if xml_string is False: return xml_string if transform_source: @@ -324,11 +453,16 @@ class InfoExtractor(object): note='Downloading JSON metadata', errnote='Unable to download JSON metadata', transform_source=None, - fatal=True): + fatal=True, encoding=None): json_string = self._download_webpage( - url_or_request, video_id, note, errnote, fatal=fatal) + url_or_request, video_id, note, errnote, fatal=fatal, + encoding=encoding) if (not fatal) and json_string is False: return None + return self._parse_json( + json_string, video_id, transform_source=transform_source, fatal=fatal) + + def _parse_json(self, json_string, video_id, transform_source=None, fatal=True): if transform_source: json_string = transform_source(json_string) try: @@ -365,19 +499,22 @@ class InfoExtractor(object): """Report attempt to log in.""" self.to_screen('Logging in') - #Methods for following #608 + # Methods for following #608 @staticmethod - def url_result(url, ie=None, video_id=None): + def url_result(url, ie=None, video_id=None, video_title=None): """Returns a url that points to a page that should be processed""" - #TODO: ie should be the class used for getting the info + # TODO: ie should be the class used for getting the info video_info = {'_type': 'url', 'url': url, 'ie_key': ie} if video_id is not None: video_info['id'] = video_id + if video_title is not None: + video_info['title'] = video_title return video_info + @staticmethod - def playlist_result(entries, playlist_id=None, playlist_title=None): + def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None): """Returns a playlist""" video_info = {'_type': 'playlist', 'entries': entries} @@ -385,9 +522,11 @@ class InfoExtractor(object): video_info['id'] = playlist_id if playlist_title: video_info['title'] = playlist_title + if playlist_description: + video_info['description'] = playlist_description return video_info - def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): + def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None): """ Perform a regex search on the given string, using a single or a list of patterns returning the first matching group. @@ -402,28 +541,30 @@ class InfoExtractor(object): if mobj: break - if os.name != 'nt' and sys.stderr.isatty(): + if not self._downloader.params.get('no_color') and os.name != 'nt' and sys.stderr.isatty(): _name = '\033[0;34m%s\033[0m' % name else: _name = name if mobj: - # return the first matching group - return next(g for g in mobj.groups() if g is not None) + if group is None: + # return the first matching group + return next(g for g in mobj.groups() if g is not None) + else: + return mobj.group(group) elif default is not _NO_DEFAULT: return default elif fatal: raise RegexNotFoundError('Unable to extract %s' % _name) else: - self._downloader.report_warning('unable to extract %s; ' - 'please report this issue on http://yt-dl.org/bug' % _name) + self._downloader.report_warning('unable to extract %s' % _name + bug_reports_message()) return None - def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0): + def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0, group=None): """ Like _search_regex, but strips HTML tags and unescapes entities. """ - res = self._search_regex(pattern, string, name, default, fatal, flags) + res = self._search_regex(pattern, string, name, default, fatal, flags, group) if res: return clean_html(res).strip() else: @@ -431,7 +572,7 @@ class InfoExtractor(object): def _get_login_info(self): """ - Get the the login info as (username, password) + Get the login info as (username, password) It will look in the netrc file using the _NETRC_MACHINE value If there's no info available, return (None, None) """ @@ -456,7 +597,7 @@ class InfoExtractor(object): 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)) - + return (username, password) def _get_tfa_info(self): @@ -516,10 +657,10 @@ class InfoExtractor(object): if display_name is None: display_name = name return self._html_search_regex( - r'''(?ix)]+(?:itemprop|name|property)=["\']?%s["\']?) - [^>]+content=["\']([^"\']+)["\']''' % re.escape(name), - html, display_name, fatal=fatal, **kwargs) + r'''(?isx)]+(?:itemprop|name|property)=(["\']?)%s\1) + [^>]+?content=(["\'])(?P