X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcommon.py;h=e9306d806b44cd531178520a47bf0b63cb17b59d;hb=4433bb02457f961dc2a753b3d4350a4a8cae138f;hp=c291bc1dfcb5c29a06495670eeae27814a2dfaab;hpb=79d2077edc9c4ac8c71b672d8eff06ef847842c8;p=youtube-dl

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index c291bc1df..e9306d806 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -44,6 +44,7 @@ from ..utils import (
     compiled_regex_type,
     determine_ext,
     determine_protocol,
+    dict_get,
     error_to_compat_str,
     ExtractorError,
     extract_attributes,
@@ -56,13 +57,17 @@ from ..utils import (
     JSON_LD_RE,
     mimetype2ext,
     orderedSet,
+    parse_bitrate,
     parse_codecs,
     parse_duration,
     parse_iso8601,
     parse_m3u8_attributes,
+    parse_resolution,
     RegexNotFoundError,
     sanitized_Request,
     sanitize_filename,
+    str_or_none,
+    strip_or_none,
     unescapeHTML,
     unified_strdate,
     unified_timestamp,
@@ -113,7 +118,7 @@ class InfoExtractor(object):
                                        unfragmented media)
                                      - URL of the MPD manifest or base URL
                                        representing the media if MPD manifest
-                                       is parsed froma string (in case of
+                                       is parsed from a string (in case of
                                        fragmented media)
                                    for MSS - URL of the ISM manifest.
                     * manifest_url
@@ -215,7 +220,7 @@ class InfoExtractor(object):
                         * "preference" (optional, int) - quality of the image
                         * "width" (optional, int)
                         * "height" (optional, int)
-                        * "resolution" (optional, string "{width}x{height"},
+                        * "resolution" (optional, string "{width}x{height}",
                                         deprecated)
                         * "filesize" (optional, int)
     thumbnail:      Full URL to a video thumbnail image.
@@ -538,11 +543,11 @@ class InfoExtractor(object):
             raise ExtractorError('An extractor error has occurred.', cause=e)
 
     def __maybe_fake_ip_and_retry(self, countries):
-        if (not self._downloader.params.get('geo_bypass_country', None) and
-                self._GEO_BYPASS and
-                self._downloader.params.get('geo_bypass', True) and
-                not self._x_forwarded_for_ip and
-                countries):
+        if (not self._downloader.params.get('geo_bypass_country', None)
+                and self._GEO_BYPASS
+                and self._downloader.params.get('geo_bypass', True)
+                and not self._x_forwarded_for_ip
+                and countries):
             country_code = random.choice(countries)
             self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code)
             if self._x_forwarded_for_ip:
@@ -678,8 +683,8 @@ class InfoExtractor(object):
 
     def __check_blocked(self, content):
         first_block = content[:512]
-        if ('<title>Access to this site is blocked</title>' in content and
-                'Websense' in first_block):
+        if ('<title>Access to this site is blocked</title>' in content
+                and 'Websense' in first_block):
             msg = 'Access to this webpage has been blocked by Websense filtering software in your network.'
             blocked_iframe = self._html_search_regex(
                 r'<iframe src="([^"]+)"', content,
@@ -697,8 +702,8 @@ class InfoExtractor(object):
             if block_msg:
                 msg += ' (Message: "%s")' % block_msg.replace('\n', ' ')
             raise ExtractorError(msg, expected=True)
-        if ('<title>TTK :: Доступ к ресурсу ограничен</title>' in content and
-                'blocklist.rkn.gov.ru' in content):
+        if ('<title>TTK :: Доступ к ресурсу ограничен</title>' in content
+                and 'blocklist.rkn.gov.ru' in content):
             raise ExtractorError(
                 'Access to this webpage has been blocked by decision of the Russian government. '
                 'Visit http://blocklist.rkn.gov.ru/ for a block reason.',
@@ -1177,16 +1182,33 @@ class InfoExtractor(object):
                                       'twitter card player')
 
     def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
-        json_ld = self._search_regex(
-            JSON_LD_RE, html, 'JSON-LD', group='json_ld', **kwargs)
+        json_ld_list = list(re.finditer(JSON_LD_RE, html))
         default = kwargs.get('default', NO_DEFAULT)
-        if not json_ld:
-            return default if default is not NO_DEFAULT else {}
         # JSON-LD may be malformed and thus `fatal` should be respected.
         # At the same time `default` may be passed that assumes `fatal=False`
         # for _search_regex. Let's simulate the same behavior here as well.
         fatal = kwargs.get('fatal', True) if default == NO_DEFAULT else False
-        return self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type)
+        json_ld = []
+        for mobj in json_ld_list:
+            json_ld_item = self._parse_json(
+                mobj.group('json_ld'), video_id, fatal=fatal)
+            if not json_ld_item:
+                continue
+            if isinstance(json_ld_item, dict):
+                json_ld.append(json_ld_item)
+            elif isinstance(json_ld_item, (list, tuple)):
+                json_ld.extend(json_ld_item)
+        if json_ld:
+            json_ld = self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type)
+        if json_ld:
+            return json_ld
+        if default is not NO_DEFAULT:
+            return default
+        elif fatal:
+            raise RegexNotFoundError('Unable to extract JSON-LD')
+        else:
+            self._downloader.report_warning('unable to extract JSON-LD %s' % bug_reports_message())
+            return {}
 
     def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
         if isinstance(json_ld, compat_str):
@@ -1251,10 +1273,10 @@ class InfoExtractor(object):
             extract_interaction_statistic(e)
 
         for e in json_ld:
-            if isinstance(e.get('@context'), compat_str) and re.match(r'^https?://schema.org/?$', e.get('@context')):
+            if '@context' in e:
                 item_type = e.get('@type')
                 if expected_type is not None and expected_type != item_type:
-                    return info
+                    continue
                 if item_type in ('TVEpisode', 'Episode'):
                     episode_name = unescapeHTML(e.get('name'))
                     info.update({
@@ -1288,11 +1310,17 @@ class InfoExtractor(object):
                     })
                 elif item_type == 'VideoObject':
                     extract_video_object(e)
-                    continue
+                    if expected_type is None:
+                        continue
+                    else:
+                        break
                 video = e.get('video')
                 if isinstance(video, dict) and video.get('@type') == 'VideoObject':
                     extract_video_object(video)
-                break
+                if expected_type is None:
+                    continue
+                else:
+                    break
         return dict((k, v) for k, v in info.items() if v is not None)
 
     @staticmethod
@@ -1419,12 +1447,10 @@ class InfoExtractor(object):
         try:
             self._request_webpage(url, video_id, 'Checking %s URL' % item, headers=headers)
             return True
-        except ExtractorError as e:
-            if isinstance(e.cause, compat_urllib_error.URLError):
-                self.to_screen(
-                    '%s: %s URL is invalid, skipping' % (video_id, item))
-                return False
-            raise
+        except ExtractorError:
+            self.to_screen(
+                '%s: %s URL is invalid, skipping' % (video_id, item))
+            return False
 
     def http_scheme(self):
         """ Either "http:" or "https:", depending on the user's preferences """
@@ -1452,14 +1478,14 @@ class InfoExtractor(object):
 
     def _extract_f4m_formats(self, manifest_url, video_id, preference=None, f4m_id=None,
                              transform_source=lambda s: fix_xml_ampersands(s).strip(),
-                             fatal=True, m3u8_id=None):
+                             fatal=True, m3u8_id=None, data=None, headers={}, query={}):
         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/ytdl-org/youtube-dl/issues/6215#issuecomment-121704244)
             transform_source=transform_source,
-            fatal=fatal)
+            fatal=fatal, data=data, headers=headers, query=query)
 
         if manifest is False:
             return []
@@ -1583,12 +1609,13 @@ class InfoExtractor(object):
     def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                               entry_protocol='m3u8', preference=None,
                               m3u8_id=None, note=None, errnote=None,
-                              fatal=True, live=False):
+                              fatal=True, live=False, data=None, headers={},
+                              query={}):
         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)
+            fatal=fatal, data=data, headers=headers, query=query)
 
         if res is False:
             return []
@@ -1705,8 +1732,8 @@ class InfoExtractor(object):
                 continue
             else:
                 tbr = float_or_none(
-                    last_stream_inf.get('AVERAGE-BANDWIDTH') or
-                    last_stream_inf.get('BANDWIDTH'), scale=1000)
+                    last_stream_inf.get('AVERAGE-BANDWIDTH')
+                    or last_stream_inf.get('BANDWIDTH'), scale=1000)
                 format_id = []
                 if m3u8_id:
                     format_id.append(m3u8_id)
@@ -1762,6 +1789,19 @@ class InfoExtractor(object):
                         # the same GROUP-ID
                         f['acodec'] = 'none'
                 formats.append(f)
+
+                # for DailyMotion
+                progressive_uri = last_stream_inf.get('PROGRESSIVE-URI')
+                if progressive_uri:
+                    http_f = f.copy()
+                    del http_f['manifest_url']
+                    http_f.update({
+                        'format_id': f['format_id'].replace('hls-', 'http-'),
+                        'protocol': 'http',
+                        'url': progressive_uri,
+                    })
+                    formats.append(http_f)
+
                 last_stream_inf = {}
         return formats
 
@@ -2006,15 +2046,17 @@ class InfoExtractor(object):
             })
         return entries
 
-    def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}):
+    def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}, data=None, headers={}, query={}):
         res = self._download_xml_handle(
             mpd_url, video_id,
             note=note or 'Downloading MPD manifest',
             errnote=errnote or 'Failed to download MPD manifest',
-            fatal=fatal)
+            fatal=fatal, data=data, headers=headers, query=query)
         if res is False:
             return []
         mpd_doc, urlh = res
+        if mpd_doc is None:
+            return []
         mpd_base_url = base_url(urlh.geturl())
 
         return self._parse_mpd_formats(
@@ -2312,15 +2354,17 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
-    def _extract_ism_formats(self, ism_url, video_id, ism_id=None, note=None, errnote=None, fatal=True):
+    def _extract_ism_formats(self, ism_url, video_id, ism_id=None, note=None, errnote=None, fatal=True, data=None, headers={}, query={}):
         res = self._download_xml_handle(
             ism_url, video_id,
             note=note or 'Downloading ISM manifest',
             errnote=errnote or 'Failed to download ISM manifest',
-            fatal=fatal)
+            fatal=fatal, data=data, headers=headers, query=query)
         if res is False:
             return []
         ism_doc, urlh = res
+        if ism_doc is None:
+            return []
 
         return self._parse_ism_formats(ism_doc, urlh.geturl(), ism_id)
 
@@ -2474,25 +2518,50 @@ class InfoExtractor(object):
                 'subtitles': {},
             }
             media_attributes = extract_attributes(media_tag)
-            src = media_attributes.get('src')
+            src = strip_or_none(media_attributes.get('src'))
             if src:
                 _, formats = _media_formats(src, media_type)
                 media_info['formats'].extend(formats)
             media_info['thumbnail'] = absolute_url(media_attributes.get('poster'))
             if media_content:
                 for source_tag in re.findall(r'<source[^>]+>', media_content):
-                    source_attributes = extract_attributes(source_tag)
-                    src = source_attributes.get('src')
+                    s_attr = extract_attributes(source_tag)
+                    # data-video-src and data-src are non standard but seen
+                    # several times in the wild
+                    src = strip_or_none(dict_get(s_attr, ('src', 'data-video-src', 'data-src')))
                     if not src:
                         continue
-                    f = parse_content_type(source_attributes.get('type'))
+                    f = parse_content_type(s_attr.get('type'))
                     is_plain_url, formats = _media_formats(src, media_type, f)
                     if is_plain_url:
-                        # res attribute is not standard but seen several times
-                        # in the wild
+                        # width, height, res, label and title attributes are
+                        # all not standard but seen several times in the wild
+                        labels = [
+                            s_attr.get(lbl)
+                            for lbl in ('label', 'title')
+                            if str_or_none(s_attr.get(lbl))
+                        ]
+                        width = int_or_none(s_attr.get('width'))
+                        height = (int_or_none(s_attr.get('height'))
+                                  or int_or_none(s_attr.get('res')))
+                        if not width or not height:
+                            for lbl in labels:
+                                resolution = parse_resolution(lbl)
+                                if not resolution:
+                                    continue
+                                width = width or resolution.get('width')
+                                height = height or resolution.get('height')
+                        for lbl in labels:
+                            tbr = parse_bitrate(lbl)
+                            if tbr:
+                                break
+                        else:
+                            tbr = None
                         f.update({
-                            'height': int_or_none(source_attributes.get('res')),
-                            'format_id': source_attributes.get('label'),
+                            'width': width,
+                            'height': height,
+                            'tbr': tbr,
+                            'format_id': s_attr.get('label') or s_attr.get('title'),
                         })
                         f.update(formats[0])
                         media_info['formats'].append(f)
@@ -2502,7 +2571,7 @@ class InfoExtractor(object):
                     track_attributes = extract_attributes(track_tag)
                     kind = track_attributes.get('kind')
                     if not kind or kind in ('subtitles', 'captions'):
-                        src = track_attributes.get('src')
+                        src = strip_or_none(track_attributes.get('src'))
                         if not src:
                             continue
                         lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label')
@@ -2659,7 +2728,7 @@ class InfoExtractor(object):
             entry = {
                 'id': this_video_id,
                 'title': unescapeHTML(video_data['title'] if require_title else video_data.get('title')),
-                'description': video_data.get('description'),
+                'description': clean_html(video_data.get('description')),
                 'thumbnail': urljoin(base_url, self._proto_relative_url(video_data.get('image'))),
                 'timestamp': int_or_none(video_data.get('pubdate')),
                 'duration': float_or_none(jwplayer_data.get('duration') or video_data.get('duration')),
@@ -2786,6 +2855,33 @@ class InfoExtractor(object):
         self._downloader.cookiejar.add_cookie_header(req)
         return compat_cookies.SimpleCookie(req.get_header('Cookie'))
 
+    def _apply_first_set_cookie_header(self, url_handle, cookie):
+        """
+        Apply first Set-Cookie header instead of the last. Experimental.
+
+        Some sites (e.g. [1-3]) may serve two cookies under the same name
+        in Set-Cookie header and expect the first (old) one to be set rather
+        than second (new). However, as of RFC6265 the newer one cookie
+        should be set into cookie store what actually happens.
+        We will workaround this issue by resetting the cookie to
+        the first one manually.
+        1. https://new.vk.com/
+        2. https://github.com/ytdl-org/youtube-dl/issues/9841#issuecomment-227871201
+        3. https://learning.oreilly.com/
+        """
+        for header, cookies in url_handle.headers.items():
+            if header.lower() != 'set-cookie':
+                continue
+            if sys.version_info[0] >= 3:
+                cookies = cookies.encode('iso-8859-1')
+            cookies = cookies.decode('utf-8')
+            cookie_value = re.search(
+                r'%s=(.+?);.*?\b[Dd]omain=(.+?)(?:[,;]|$)' % cookie, cookies)
+            if cookie_value:
+                value, domain = cookie_value.groups()
+                self._set_cookie(domain, cookie, value)
+                break
+
     def get_testcases(self, include_onlymatching=False):
         t = getattr(self, '_TEST', None)
         if t:
@@ -2816,8 +2912,8 @@ class InfoExtractor(object):
         return not any_restricted
 
     def extract_subtitles(self, *args, **kwargs):
-        if (self._downloader.params.get('writesubtitles', False) or
-                self._downloader.params.get('listsubtitles')):
+        if (self._downloader.params.get('writesubtitles', False)
+                or self._downloader.params.get('listsubtitles')):
             return self._get_subtitles(*args, **kwargs)
         return {}
 
@@ -2842,8 +2938,8 @@ class InfoExtractor(object):
         return ret
 
     def extract_automatic_captions(self, *args, **kwargs):
-        if (self._downloader.params.get('writeautomaticsub', False) or
-                self._downloader.params.get('listsubtitles')):
+        if (self._downloader.params.get('writeautomaticsub', False)
+                or self._downloader.params.get('listsubtitles')):
             return self._get_automatic_captions(*args, **kwargs)
         return {}
 
@@ -2851,9 +2947,9 @@ class InfoExtractor(object):
         raise NotImplementedError('This method must be implemented by subclasses')
 
     def mark_watched(self, *args, **kwargs):
-        if (self._downloader.params.get('mark_watched', False) and
-                (self._get_login_info()[0] is not None or
-                    self._downloader.params.get('cookiefile') is not None)):
+        if (self._downloader.params.get('mark_watched', False)
+                and (self._get_login_info()[0] is not None
+                     or self._downloader.params.get('cookiefile') is not None)):
             self._mark_watched(*args, **kwargs)
 
     def _mark_watched(self, *args, **kwargs):