[tvopengr] Add extractors (#2297)

Authored by: zmousm
2022-01-19 22:43:02 +02:00 · 2022-01-19 22:43:02 +02:00 · 1a20d29552
commit 1a20d29552
parent f7085283e1
3 changed files with 169 additions and 0 deletions
--- a/yt_dlp/extractor/extractors.py
+++ b/yt_dlp/extractor/extractors.py
@ -1679,6 +1679,10 @@ from .tvnow import (
    TVNowAnnualIE,
    TVNowShowIE,
 )
+from .tvopengr import (
+    TVOpenGrWatchIE,
+    TVOpenGrEmbedIE,
+)
 from .tvp import (
    TVPEmbedIE,
    TVPIE,
--- a/yt_dlp/extractor/generic.py
+++ b/yt_dlp/extractor/generic.py
@ -139,6 +139,7 @@ from .arcpublishing import ArcPublishingIE
 from .medialaan import MedialaanIE
 from .simplecast import SimplecastIE
 from .wimtv import WimTVIE
+from .tvopengr import TVOpenGrEmbedIE
 from .tvp import TVPEmbedIE
 from .blogger import BloggerIE
 from .mainstreaming import MainStreamingIE
@ -2227,6 +2228,22 @@ class GenericIE(InfoExtractor):
                'skip_download': True,
            },
        },
+        {
+            # tvopengr:embed
+            'url': 'https://www.ethnos.gr/World/article/190604/hparosiaxekinoynoisynomiliessthgeneyhmethskiatoypolemoypanoapothnoykrania',
+            'md5': 'eb0c3995d0a6f18f6538c8e057865d7d',
+            'info_dict': {
+                'id': '101119',
+                'ext': 'mp4',
+                'display_id': 'oikarpoitondiapragmateyseonhparosias',
+                'title': 'md5:b979f4d640c568617d6547035528a149',
+                'description': 'md5:e54fc1977c7159b01cc11cd7d9d85550',
+                'timestamp': 1641772800,
+                'upload_date': '20220110',
+                'thumbnail': 'https://opentv-static.siliconweb.com/imgHandler/1920/70bc39fa-895b-4918-a364-c39d2135fc6d.jpg',
+
+            }
+        },
        {
            # blogger embed
            'url': 'https://blog.tomeuvizoso.net/2019/01/a-panfrost-milestone.html',
@ -3671,6 +3688,11 @@ class GenericIE(InfoExtractor):
            return self.playlist_from_matches(
                rumble_urls, video_id, video_title, ie=RumbleEmbedIE.ie_key())

+        # Look for (tvopen|ethnos).gr embeds
+        tvopengr_urls = list(TVOpenGrEmbedIE._extract_urls(webpage))
+        if tvopengr_urls:
+            return self.playlist_from_matches(tvopengr_urls, video_id, video_title, ie=TVOpenGrEmbedIE.ie_key())
+
        tvp_urls = TVPEmbedIE._extract_urls(webpage)
        if tvp_urls:
            return self.playlist_from_matches(tvp_urls, video_id, video_title, ie=TVPEmbedIE.ie_key())
--- a/yt_dlp/extractor/tvopengr.py
+++ b/yt_dlp/extractor/tvopengr.py
@ -0,0 +1,143 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    get_elements_text_and_html_by_attribute,
+    merge_dicts,
+    unescapeHTML,
+)
+
+
+class TVOpenGrBaseIE(InfoExtractor):
+    def _return_canonical_url(self, url, video_id):
+        webpage = self._download_webpage(url, video_id)
+        canonical_url = self._og_search_url(webpage)
+        title = self._og_search_title(webpage)
+        return self.url_result(canonical_url, ie=TVOpenGrWatchIE.ie_key(), video_id=video_id, video_title=title)
+
+
+class TVOpenGrWatchIE(TVOpenGrBaseIE):
+    IE_NAME = 'tvopengr:watch'
+    IE_DESC = 'tvopen.gr (and ethnos.gr) videos'
+    _VALID_URL = r'https?://(?P<netloc>(?:www\.)?(?:tvopen|ethnos)\.gr)/watch/(?P<id>\d+)/(?P<slug>[^/]+)'
+    _API_ENDPOINT = 'https://www.tvopen.gr/templates/data/player'
+
+    _TESTS = [{
+        'url': 'https://www.ethnos.gr/watch/101009/nikoskaprabelosdenexoymekanenanasthenhsemethmethmetallaxhomikron',
+        'md5': '8728570e3a72e0f8d9475ba94859fdc1',
+        'info_dict': {
+            'id': '101009',
+            'title': 'md5:51f68773dcb6c70498cd326f45fefdf0',
+            'display_id': 'nikoskaprabelosdenexoymekanenanasthenhsemethmethmetallaxhomikron',
+            'description': 'md5:78fff49f18fb3effe41b070e5c7685d6',
+            'thumbnail': 'https://opentv-static.siliconweb.com/imgHandler/1920/d573ba71-ec5f-43c6-b4cb-d181f327d3a8.jpg',
+            'ext': 'mp4',
+            'upload_date': '20220109',
+            'timestamp': 1641686400,
+        },
+    }, {
+        'url': 'https://www.tvopen.gr/watch/100979/se28099agapaomenalla7cepeisodio267cmhthrargiapashskakias',
+        'md5': '38f98a1be0c577db4ea2d1b1c0770c48',
+        'info_dict': {
+            'id': '100979',
+            'title': 'md5:e021f3001e16088ee40fa79b20df305b',
+            'display_id': 'se28099agapaomenalla7cepeisodio267cmhthrargiapashskakias',
+            'description': 'md5:ba17db53954134eb8d625d199e2919fb',
+            'thumbnail': 'https://opentv-static.siliconweb.com/imgHandler/1920/9bb71cf1-21da-43a9-9d65-367950fde4e3.jpg',
+            'ext': 'mp4',
+            'upload_date': '20220108',
+            'timestamp': 1641600000,
+        },
+    }]
+
+    def _extract_formats_and_subs(self, response, video_id):
+        formats, subs = [], {}
+        for format_id, format_url in response.items():
+            if format_id not in ('stream', 'httpstream', 'mpegdash'):
+                continue
+            ext = determine_ext(format_url)
+            if ext == 'm3u8':
+                formats_, subs_ = self._extract_m3u8_formats_and_subtitles(
+                    format_url, video_id, 'mp4', m3u8_id=format_id,
+                    fatal=False)
+            elif ext == 'mpd':
+                formats_, subs_ = self._extract_mpd_formats_and_subtitles(
+                    format_url, video_id, 'mp4', fatal=False)
+            else:
+                formats.append({
+                    'url': format_url,
+                    'format_id': format_id,
+                })
+                continue
+            formats.extend(formats_)
+            self._merge_subtitles(subs_, target=subs)
+        self._sort_formats(formats)
+        return formats, subs
+
+    @staticmethod
+    def _scale_thumbnails_to_max_width(formats, thumbnails, url_width_re):
+        _keys = ('width', 'height')
+        max_dimensions = max(
+            [tuple(format.get(k) or 0 for k in _keys) for format in formats],
+            default=(0, 0))
+        if not max_dimensions[0]:
+            return thumbnails
+        return [
+            merge_dicts(
+                {'url': re.sub(url_width_re, str(max_dimensions[0]), thumbnail['url'])},
+                dict(zip(_keys, max_dimensions)), thumbnail)
+            for thumbnail in thumbnails
+        ]
+
+    def _real_extract(self, url):
+        netloc, video_id, display_id = self._match_valid_url(url).group('netloc', 'id', 'slug')
+        if netloc.find('tvopen.gr') == -1:
+            return self._return_canonical_url(url, video_id)
+        webpage = self._download_webpage(url, video_id)
+        info = self._search_json_ld(webpage, video_id, expected_type='VideoObject')
+        info['formats'], info['subtitles'] = self._extract_formats_and_subs(
+            self._download_json(self._API_ENDPOINT, video_id, query={'cid': video_id}),
+            video_id)
+        info['thumbnails'] = self._scale_thumbnails_to_max_width(
+            info['formats'], info['thumbnails'], r'(?<=/imgHandler/)\d+')
+        description, _html = next(get_elements_text_and_html_by_attribute('class', 'description', webpage))
+        if description and _html.startswith('<span '):
+            info['description'] = description
+        info['id'] = video_id
+        info['display_id'] = display_id
+        return info
+
+
+class TVOpenGrEmbedIE(TVOpenGrBaseIE):
+    IE_NAME = 'tvopengr:embed'
+    IE_DESC = 'tvopen.gr embedded videos'
+    _VALID_URL = r'(?:https?:)?//(?:www\.|cdn\.|)(?:tvopen|ethnos).gr/embed/(?P<id>\d+)'
+    _EMBED_RE = re.compile(rf'''<iframe[^>]+?src=(?P<_q1>["'])(?P<url>{_VALID_URL})(?P=_q1)''')
+
+    _TESTS = [{
+        'url': 'https://cdn.ethnos.gr/embed/100963',
+        'md5': '2da147881f45571d81662d94d086628b',
+        'info_dict': {
+            'id': '100963',
+            'display_id': 'koronoiosapotoysdieythyntestonsxoleionselftestgiaosoysdenbrhkan',
+            'title': 'md5:2c71876fadf0cda6043da0da5fca2936',
+            'description': 'md5:17482b4432e5ed30eccd93b05d6ea509',
+            'thumbnail': 'https://opentv-static.siliconweb.com/imgHandler/1920/5804e07f-799a-4247-a696-33842c94ca37.jpg',
+            'ext': 'mp4',
+            'upload_date': '20220108',
+            'timestamp': 1641600000,
+        },
+    }]
+
+    @classmethod
+    def _extract_urls(cls, webpage):
+        for mobj in cls._EMBED_RE.finditer(webpage):
+            yield unescapeHTML(mobj.group('url'))
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        return self._return_canonical_url(url, video_id)