[itv] remove old extractio method and fix series metadata extraction

closes #23177
closes #26897

youtube_dl/extractor/itv.py

@@ -1,29 +1,21 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import uuid
-import xml.etree.ElementTree as etree
 import json
 import re
 
 from .common import InfoExtractor
 from .brightcove import BrightcoveNewIE
-from ..compat import (
-    compat_str,
-    compat_etree_register_namespace,
-)
 from ..utils import (
     determine_ext,
-    ExtractorError,
     extract_attributes,
-    int_or_none,
+    get_element_by_class,
+    JSON_LD_RE,
     merge_dicts,
     parse_duration,
     smuggle_url,
+    strip_or_none,
     url_or_none,
-    xpath_with_ns,
-    xpath_element,
-    xpath_text,
 )
 
 
@@ -31,14 +23,18 @@ class ITVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/hub/[^/]+/(?P<id>[0-9a-zA-Z]+)'
     _GEO_COUNTRIES = ['GB']
     _TESTS = [{
-        'url': 'http://www.itv.com/hub/mr-bean-animated-series/2a2936a0053',
+        'url': 'https://www.itv.com/hub/liar/2a4547a0012',
         'info_dict': {
-            'id': '2a2936a0053',
-            'ext': 'flv',
-            'title': 'Home Movie',
+            'id': '2a4547a0012',
+            'ext': 'mp4',
+            'title': 'Liar - Series 2 - Episode 6',
+            'description': 'md5:d0f91536569dec79ea184f0a44cca089',
+            'series': 'Liar',
+            'season_number': 2,
+            'episode_number': 6,
         },
         'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
     }, {
@@ -61,139 +57,8 @@ class ITVIE(InfoExtractor):
         params = extract_attributes(self._search_regex(
             r'(?s)(<[^>]+id="video"[^>]*>)', webpage, 'params'))
 
-        ns_map = {
-            'soapenv': 'http://schemas.xmlsoap.org/soap/envelope/',
-            'tem': 'http://tempuri.org/',
-            'itv': 'http://schemas.datacontract.org/2004/07/Itv.BB.Mercury.Common.Types',
-            'com': 'http://schemas.itv.com/2009/05/Common',
-        }
-        for ns, full_ns in ns_map.items():
-            compat_etree_register_namespace(ns, full_ns)
-
-        def _add_ns(name):
-            return xpath_with_ns(name, ns_map)
-
-        def _add_sub_element(element, name):
-            return etree.SubElement(element, _add_ns(name))
-
-        production_id = (
-            params.get('data-video-autoplay-id')
-            or '%s#001' % (
-                params.get('data-video-episode-id')
-                or video_id.replace('a', '/')))
-
-        req_env = etree.Element(_add_ns('soapenv:Envelope'))
-        _add_sub_element(req_env, 'soapenv:Header')
-        body = _add_sub_element(req_env, 'soapenv:Body')
-        get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
-        request = _add_sub_element(get_playlist, 'tem:request')
-        _add_sub_element(request, 'itv:ProductionId').text = production_id
-        _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
-        vodcrid = _add_sub_element(request, 'itv:Vodcrid')
-        _add_sub_element(vodcrid, 'com:Id')
-        _add_sub_element(request, 'itv:Partition')
-        user_info = _add_sub_element(get_playlist, 'tem:userInfo')
-        _add_sub_element(user_info, 'itv:Broadcaster').text = 'Itv'
-        _add_sub_element(user_info, 'itv:DM')
-        _add_sub_element(user_info, 'itv:RevenueScienceValue')
-        _add_sub_element(user_info, 'itv:SessionId')
-        _add_sub_element(user_info, 'itv:SsoToken')
-        _add_sub_element(user_info, 'itv:UserToken')
-        site_info = _add_sub_element(get_playlist, 'tem:siteInfo')
-        _add_sub_element(site_info, 'itv:AdvertisingRestriction').text = 'None'
-        _add_sub_element(site_info, 'itv:AdvertisingSite').text = 'ITV'
-        _add_sub_element(site_info, 'itv:AdvertisingType').text = 'Any'
-        _add_sub_element(site_info, 'itv:Area').text = 'ITVPLAYER.VIDEO'
-        _add_sub_element(site_info, 'itv:Category')
-        _add_sub_element(site_info, 'itv:Platform').text = 'DotCom'
-        _add_sub_element(site_info, 'itv:Site').text = 'ItvCom'
-        device_info = _add_sub_element(get_playlist, 'tem:deviceInfo')
-        _add_sub_element(device_info, 'itv:ScreenSize').text = 'Big'
-        player_info = _add_sub_element(get_playlist, 'tem:playerInfo')
-        _add_sub_element(player_info, 'itv:Version').text = '2'
-
-        headers = self.geo_verification_headers()
-        headers.update({
-            'Content-Type': 'text/xml; charset=utf-8',
-            'SOAPAction': 'http://tempuri.org/PlaylistService/GetPlaylist',
-        })
-
-        info = self._search_json_ld(webpage, video_id, default={})
-        formats = []
-        subtitles = {}
-
-        def extract_subtitle(sub_url):
-            ext = determine_ext(sub_url, 'ttml')
-            subtitles.setdefault('en', []).append({
-                'url': sub_url,
-                'ext': 'ttml' if ext == 'xml' else ext,
-            })
-
-        resp_env = self._download_xml(
-            params['data-playlist-url'], video_id,
-            headers=headers, data=etree.tostring(req_env), fatal=False)
-        if resp_env:
-            playlist = xpath_element(resp_env, './/Playlist')
-            if playlist is None:
-                fault_code = xpath_text(resp_env, './/faultcode')
-                fault_string = xpath_text(resp_env, './/faultstring')
-                if fault_code == 'InvalidGeoRegion':
-                    self.raise_geo_restricted(
-                        msg=fault_string, countries=self._GEO_COUNTRIES)
-                elif fault_code not in (
-                        'InvalidEntity', 'InvalidVodcrid', 'ContentUnavailable'):
-                    raise ExtractorError(
-                        '%s said: %s' % (self.IE_NAME, fault_string), expected=True)
-                info.update({
-                    'title': self._og_search_title(webpage),
-                    'episode_title': params.get('data-video-episode'),
-                    'series': params.get('data-video-title'),
-                })
-            else:
-                title = xpath_text(playlist, 'EpisodeTitle', default=None)
-                info.update({
-                    'title': title,
-                    'episode_title': title,
-                    'episode_number': int_or_none(xpath_text(playlist, 'EpisodeNumber')),
-                    'series': xpath_text(playlist, 'ProgrammeTitle'),
-                    'duration': parse_duration(xpath_text(playlist, 'Duration')),
-                })
-                video_element = xpath_element(playlist, 'VideoEntries/Video', fatal=True)
-                media_files = xpath_element(video_element, 'MediaFiles', fatal=True)
-                rtmp_url = media_files.attrib['base']
-
-                for media_file in media_files.findall('MediaFile'):
-                    play_path = xpath_text(media_file, 'URL')
-                    if not play_path:
-                        continue
-                    tbr = int_or_none(media_file.get('bitrate'), 1000)
-                    f = {
-                        'format_id': 'rtmp' + ('-%d' % tbr if tbr else ''),
-                        'play_path': play_path,
-                        # Providing this swfVfy allows to avoid truncated downloads
-                        'player_url': 'http://www.itv.com/mercury/Mercury_VideoPlayer.swf',
-                        'page_url': url,
-                        'tbr': tbr,
-                        'ext': 'flv',
-                    }
-                    app = self._search_regex(
-                        'rtmpe?://[^/]+/(.+)$', rtmp_url, 'app', default=None)
-                    if app:
-                        f.update({
-                            'url': rtmp_url.split('?', 1)[0],
-                            'app': app,
-                        })
-                    else:
-                        f['url'] = rtmp_url
-                    formats.append(f)
-
-                for caption_url in video_element.findall('ClosedCaptioningURIs/URL'):
-                    if caption_url.text:
-                        extract_subtitle(caption_url.text)
-
-        ios_playlist_url = params.get('data-video-playlist') or params.get('data-video-id')
-        hmac = params.get('data-video-hmac')
-        if ios_playlist_url and hmac and re.match(r'https?://', ios_playlist_url):
+        ios_playlist_url = params.get('data-video-playlist') or params['data-video-id']
+        hmac = params['data-video-hmac']
         headers = self.geo_verification_headers()
         headers.update({
             'Accept': 'application/vnd.itv.vod.playlist.v2+json',
@@ -227,11 +92,12 @@ class ITVIE(InfoExtractor):
                     },
                     'platformTag': 'dotcom'
                 }
-                }).encode(), headers=headers, fatal=False)
-            if ios_playlist:
-                video_data = ios_playlist.get('Playlist', {}).get('Video', {})
+            }).encode(), headers=headers)
+        video_data = ios_playlist['Playlist']['Video']
         ios_base_url = video_data.get('Base')
-                for media_file in video_data.get('MediaFiles', []):
+
+        formats = []
+        for media_file in (video_data.get('MediaFiles') or []):
             href = media_file.get('Href')
             if not href:
                 continue
@@ -246,35 +112,42 @@ class ITVIE(InfoExtractor):
                 formats.append({
                     'url': href,
                 })
-                subs = video_data.get('Subtitles')
-                if isinstance(subs, list):
+        self._sort_formats(formats)
+
+        subtitles = {}
+        subs = video_data.get('Subtitles') or []
         for sub in subs:
             if not isinstance(sub, dict):
                 continue
             href = url_or_none(sub.get('Href'))
-                        if href:
-                            extract_subtitle(href)
-                if not info.get('duration'):
-                    info['duration'] = parse_duration(video_data.get('Duration'))
-
-        self._sort_formats(formats)
-
-        info.update({
-            'id': video_id,
-            'formats': formats,
-            'subtitles': subtitles,
+            if not href:
+                continue
+            subtitles.setdefault('en', []).append({
+                'url': href,
+                'ext': determine_ext(href, 'vtt'),
             })
 
-        webpage_info = self._search_json_ld(webpage, video_id, default={})
-        if not webpage_info.get('title'):
-            webpage_info['title'] = self._html_search_regex(
-                r'(?s)<h\d+[^>]+\bclass=["\'][^>]*episode-title["\'][^>]*>([^<]+)<',
-                webpage, 'title', default=None) or self._og_search_title(
-                webpage, default=None) or self._html_search_meta(
-                'twitter:title', webpage, 'title',
-                default=None) or webpage_info['episode']
+        info = self._search_json_ld(webpage, video_id, default={})
+        if not info:
+            json_ld = self._parse_json(self._search_regex(
+                JSON_LD_RE, webpage, 'JSON-LD', '{}',
+                group='json_ld'), video_id, fatal=False)
+            if json_ld and json_ld.get('@type') == 'BreadcrumbList':
+                for ile in (json_ld.get('itemListElement:') or []):
+                    item = ile.get('item:') or {}
+                    if item.get('@type') == 'TVEpisode':
+                        item['@context'] = 'http://schema.org'
+                        info = self._json_ld(item, video_id, fatal=False) or {}
+                        break
 
-        return merge_dicts(info, webpage_info)
+        return merge_dicts({
+            'id': video_id,
+            'title': self._html_search_meta(['og:title', 'twitter:title'], webpage),
+            'formats': formats,
+            'subtitles': subtitles,
+            'duration': parse_duration(video_data.get('Duration')),
+            'description': strip_or_none(get_element_by_class('episode-info__synopsis', webpage)),
+        }, info)
 
 
 class ITVBTCCIE(InfoExtractor):