Merge a91d9e1084 into e079ffbda6

Authored by: jiru

README.md

@@ -1869,6 +1869,9 @@ The following extractors use this feature:
 #### digitalconcerthall
 * `prefer_combined_hls`: Prefer extracting combined/pre-merged video and audio HLS formats. This will exclude 4K/HEVC video and lossless/FLAC audio formats, which are only available as split video/audio HLS formats
 
+#### sonylivseries
+* `sort_order`: Episode sort order for series extraction - one of `asc` (ascending, oldest first) or `desc` (descending, newest first). Default is `asc`
+
 **Note**: These options may be changed/removed in the future without concern for backward compatibility
 
 <!-- MANPAGE: MOVE "INSTALLATION" SECTION HERE -->

test/test_parsing.py

@@ -0,0 +1,359 @@
+import textwrap
+import unittest
+
+from yt_dlp.compat import compat_HTMLParseError
+from yt_dlp.parsing import (
+    MatchingElementParser,
+    HTMLIgnoreRanges,
+    HTMLTagParser,
+)
+
+extract_attributes = MatchingElementParser.extract_attributes
+get_element_by_attribute = MatchingElementParser.get_element_by_attribute
+get_element_by_class = MatchingElementParser.get_element_by_class
+get_element_html_by_attribute = MatchingElementParser.get_element_html_by_attribute
+get_element_html_by_class = MatchingElementParser.get_element_html_by_class
+get_element_text_and_html_by_tag = MatchingElementParser.get_element_text_and_html_by_tag
+get_elements_by_attribute = MatchingElementParser.get_elements_by_attribute
+get_elements_by_class = MatchingElementParser.get_elements_by_class
+get_elements_html_by_attribute = MatchingElementParser.get_elements_html_by_attribute
+get_elements_html_by_class = MatchingElementParser.get_elements_html_by_class
+get_elements_text_and_html_by_attribute = MatchingElementParser.get_elements_text_and_html_by_attribute
+get_elements_text_and_html_by_tag = MatchingElementParser.get_elements_text_and_html_by_tag
+
+
+class TestParsing(unittest.TestCase):
+    def test_extract_attributes(self):
+        self.assertEqual(extract_attributes('<e x="y">'), {'x': 'y'})
+        self.assertEqual(extract_attributes("<e x='y'>"), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x=y>'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x="a \'b\' c">'), {'x': "a 'b' c"})
+        self.assertEqual(extract_attributes('<e x=\'a "b" c\'>'), {'x': 'a "b" c'})
+        self.assertEqual(extract_attributes('<e x="&#121;">'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x="&#x79;">'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x="&amp;">'), {'x': '&'})  # XML
+        self.assertEqual(extract_attributes('<e x="&quot;">'), {'x': '"'})
+        self.assertEqual(extract_attributes('<e x="&pound;">'), {'x': '£'})  # HTML 3.2
+        self.assertEqual(extract_attributes('<e x="&lambda;">'), {'x': 'λ'})  # HTML 4.0
+        self.assertEqual(extract_attributes('<e x="&foo">'), {'x': '&foo'})
+        self.assertEqual(extract_attributes('<e x="\'">'), {'x': "'"})
+        self.assertEqual(extract_attributes('<e x=\'"\'>'), {'x': '"'})
+        self.assertEqual(extract_attributes('<e x >'), {'x': None})
+        self.assertEqual(extract_attributes('<e x=y a>'), {'x': 'y', 'a': None})
+        self.assertEqual(extract_attributes('<e x= y>'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x=1 y=2 x=3>'), {'y': '2', 'x': '3'})
+        self.assertEqual(extract_attributes('<e \nx=\ny\n>'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e \nx=\n"y"\n>'), {'x': 'y'})
+        self.assertEqual(extract_attributes("<e \nx=\n'y'\n>"), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e \nx="\ny\n">'), {'x': '\ny\n'})
+        self.assertEqual(extract_attributes('<e CAPS=x>'), {'caps': 'x'})  # Names lowercased
+        self.assertEqual(extract_attributes('<e x=1 X=2>'), {'x': '2'})
+        self.assertEqual(extract_attributes('<e X=1 x=2>'), {'x': '2'})
+        self.assertEqual(extract_attributes('<e _:funny-name1=1>'), {'_:funny-name1': '1'})
+        self.assertEqual(extract_attributes('<e x="Fáilte 世界 \U0001f600">'), {'x': 'Fáilte 世界 \U0001f600'})
+        self.assertEqual(extract_attributes('<e x="décompose&#769;">'), {'x': 'décompose\u0301'})
+        # "Narrow" Python builds don't support unicode code points outside BMP.
+        try:
+            chr(0x10000)
+            supports_outside_bmp = True
+        except ValueError:
+            supports_outside_bmp = False
+        if supports_outside_bmp:
+            self.assertEqual(extract_attributes('<e x="Smile &#128512;!">'), {'x': 'Smile \U0001f600!'})
+        # Malformed HTML should not break attributes extraction on older Python
+        self.assertEqual(extract_attributes('<mal"formed/>'), {})
+
+    GET_ELEMENT_BY_CLASS_TEST_STRING = '''
+        <span class="foo bar">nice</span>
+        <div class="foo bar">also nice</div>
+    '''
+
+    def test_get_element_by_class(self):
+        html = self.GET_ELEMENT_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_element_by_class('foo', html), 'nice')
+        self.assertEqual(get_element_by_class('no-such-class', html), None)
+
+    def test_get_element_html_by_class(self):
+        html = self.GET_ELEMENT_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_element_html_by_class('foo', html),
+                         '<span class="foo bar">nice</span>')
+        self.assertEqual(get_element_by_class('no-such-class', html), None)
+
+    GET_ELEMENT_BY_ATTRIBUTE_TEST_STRING = '''
+        <div itemprop="author" itemscope>foo</div>
+    '''
+
+    def test_get_element_by_attribute(self):
+        html = self.GET_ELEMENT_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_element_by_attribute('class', 'foo bar', html), 'nice')
+        self.assertEqual(get_element_by_attribute('class', 'foo', html), None)
+        self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None)
+        self.assertEqual(get_element_by_attribute('class', 'foo bar', html, tag='div'), 'also nice')
+
+        html = self.GET_ELEMENT_BY_ATTRIBUTE_TEST_STRING
+
+        self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo')
+
+    def test_get_element_html_by_attribute(self):
+        html = self.GET_ELEMENT_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_element_html_by_attribute('class', 'foo bar', html),
+                         '<span class="foo bar">nice</span>')
+        self.assertEqual(get_element_html_by_attribute('class', 'foo', html), None)
+        self.assertEqual(get_element_html_by_attribute('class', 'no-such-foo', html), None)
+
+        html = self.GET_ELEMENT_BY_ATTRIBUTE_TEST_STRING
+
+        self.assertEqual(get_element_html_by_attribute('itemprop', 'author', html), html.strip())
+
+    GET_ELEMENTS_BY_CLASS_TEST_STRING = '''
+        <span class="foo bar">nice</span>
+        <span class="foo bar">also nice</span>
+    '''
+    GET_ELEMENTS_BY_CLASS_RES = [
+        '<span class="foo bar">nice</span>',
+        '<span class="foo bar">also nice</span>'
+    ]
+
+    def test_get_elements_by_class(self):
+        html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_elements_by_class('foo', html), ['nice', 'also nice'])
+        self.assertEqual(get_elements_by_class('no-such-class', html), [])
+
+    def test_get_elements_html_by_class(self):
+        html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_elements_html_by_class('foo', html), self.GET_ELEMENTS_BY_CLASS_RES)
+        self.assertEqual(get_elements_html_by_class('no-such-class', html), [])
+
+    def test_get_elements_by_attribute(self):
+        html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_elements_by_attribute('class', 'foo bar', html), ['nice', 'also nice'])
+        self.assertEqual(get_elements_by_attribute('class', 'foo', html), [])
+        self.assertEqual(get_elements_by_attribute('class', 'no-such-foo', html), [])
+
+    def test_get_elements_html_by_attribute(self):
+        html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+        self.assertEqual(get_elements_html_by_attribute('class', 'foo bar', html),
+                         self.GET_ELEMENTS_BY_CLASS_RES)
+        self.assertEqual(get_elements_html_by_attribute('class', 'foo', html), [])
+        self.assertEqual(get_elements_html_by_attribute('class', 'no-such-foo', html), [])
+
+    def test_get_elements_text_and_html_by_attribute(self):
+        html = self.GET_ELEMENTS_BY_CLASS_TEST_STRING
+
+        self.assertEqual(
+            get_elements_text_and_html_by_attribute('class', 'foo bar', html),
+            list(zip(['nice', 'also nice'], self.GET_ELEMENTS_BY_CLASS_RES)))
+        self.assertEqual(get_elements_text_and_html_by_attribute('class', 'foo', html), [])
+        self.assertEqual(get_elements_text_and_html_by_attribute('class', 'no-such-foo', html), [])
+
+        self.assertEqual(get_elements_text_and_html_by_attribute(
+            'class', 'foo', '<a class="foo">nice</a><span class="foo">not nice</span>', tag='a'),
+            [('nice', '<a class="foo">nice</a>')])
+
+    def test_get_element_text_and_html_by_tag(self):
+        get_element_by_tag_test_string = '''
+        random text lorem ipsum</p>
+        <div>
+            this should be returned
+            <span>this should also be returned</span>
+            <div>
+                this should also be returned
+            </div>
+            closing tag above should not trick, so this should also be returned
+        </div>
+        but this text should not be returned
+        '''
+        html = textwrap.indent(textwrap.dedent(get_element_by_tag_test_string), ' ' * 4)
+        get_element_by_tag_res_outerdiv_html = html.strip()[32:276]
+        get_element_by_tag_res_outerdiv_text = get_element_by_tag_res_outerdiv_html[5:-6]
+        get_element_by_tag_res_innerspan_html = html.strip()[78:119]
+        get_element_by_tag_res_innerspan_text = get_element_by_tag_res_innerspan_html[6:-7]
+
+        self.assertEqual(
+            get_element_text_and_html_by_tag('div', html),
+            (get_element_by_tag_res_outerdiv_text, get_element_by_tag_res_outerdiv_html))
+        self.assertEqual(
+            get_element_text_and_html_by_tag('span', html),
+            (get_element_by_tag_res_innerspan_text, get_element_by_tag_res_innerspan_html))
+        self.assertIsNone(get_element_text_and_html_by_tag('article', html))
+
+    def test_get_elements_text_and_html_by_tag(self):
+        class StrictParser(MatchingElementParser):
+            STRICT = True
+
+        test_string = '''
+            <img src="a.png">
+            <img src="b.png" />
+            <span>ignore</span>
+        '''
+        items = get_elements_text_and_html_by_tag('img', test_string)
+        self.assertEqual(items, [('', '<img src="a.png">'), ('', '<img src="b.png" />')])
+
+        self.assertEqual(
+            StrictParser.get_element_text_and_html_by_tag('use', '<use><img></use>'),
+            ('<img>', '<use><img></use>'))
+
+    def test_get_element_text_and_html_by_tag_malformed(self):
+        inner_text = 'inner text'
+        malnested_elements = f'<malnested_a><malnested_b>{inner_text}</malnested_a></malnested_b>'
+        commented_html = '<!--<div>inner comment</div>-->'
+        outerdiv_html = f'<div>{malnested_elements}</div>'
+        html = f'{commented_html}{outerdiv_html}'
+
+        self.assertEqual(
+            get_element_text_and_html_by_tag('div', html), (malnested_elements, outerdiv_html))
+        self.assertEqual(
+            get_element_text_and_html_by_tag('malnested_a', html),
+            (f'<malnested_b>{inner_text}',
+             f'<malnested_a><malnested_b>{inner_text}</malnested_a>'))
+        self.assertEqual(
+            get_element_text_and_html_by_tag('malnested_b', html),
+            (f'{inner_text}</malnested_a>',
+             f'<malnested_b>{inner_text}</malnested_a></malnested_b>'))
+        self.assertEqual(
+            get_element_text_and_html_by_tag('orphan', f'<orphan>{html}'), ('', '<orphan>'))
+        self.assertIsNone(get_element_text_and_html_by_tag('orphan', f'{html}</orphan>'))
+
+        # ignore case on tags
+        ci_html = f'<SpAn>{html}</sPaN>'
+        self.assertEqual(get_element_text_and_html_by_tag('span', ci_html), (html, ci_html))
+
+    def test_strict_html_parsing(self):
+        class StrictTagParser(HTMLTagParser):
+            STRICT = True
+
+        parser = StrictTagParser()
+        with self.assertRaisesRegex(compat_HTMLParseError, "stray closing tag 'p'"):
+            parser.taglist('</p>', reset=True)
+        with self.assertRaisesRegex(compat_HTMLParseError, "unclosed tag 'p', 'div'"):
+            parser.taglist('<div><p>', reset=True)
+        with self.assertRaisesRegex(compat_HTMLParseError, "malnested closing tag 'div', expected after '</p>'"):
+            parser.taglist('<div><p></div></p>', reset=True)
+        with self.assertRaisesRegex(compat_HTMLParseError, "malnested closing tag 'div', expected after '</p>'"):
+            parser.taglist('<div><p>/p></div>', reset=True)
+        with self.assertRaisesRegex(compat_HTMLParseError, "malformed closing tag 'p<<'"):
+            parser.taglist('<div><p></p<< </div>', reset=True)
+        with self.assertRaisesRegex(compat_HTMLParseError, "stray closing tag 'img'"):
+            parser.taglist('<img>must be empty</img>', reset=True)
+
+    def test_relaxed_html_parsing(self):
+        Tag = HTMLTagParser.Tag
+        parser = HTMLTagParser()
+
+        self.assertEqual(parser.taglist('</p>', reset=True), [])
+
+        tags = parser.taglist('<div><p>', reset=True)
+        self.assertEqual(tags, [Tag('div'), Tag('p')])
+        self.assertEqual(tags[0].text_and_html(), ('', '<div>'))
+        self.assertEqual(tags[1].text_and_html(), ('', '<p>'))
+
+        tags = parser.taglist('<div><p></div></p>', reset=True)
+        self.assertEqual(tags, [Tag('div'), Tag('p')])
+        self.assertEqual(tags[0].text_and_html(), ('<p>', '<div><p></div>'))
+        self.assertEqual(tags[1].text_and_html(), ('</div>', '<p></div></p>'))
+
+        tags = parser.taglist('<div><p>/p></div>', reset=True)
+        self.assertEqual(tags, [Tag('div'), Tag('p')])
+        self.assertEqual(tags[0].text_and_html(), ('<p>/p>', '<div><p>/p></div>'))
+        self.assertEqual(tags[1].text_and_html(), ('', '<p>'))
+
+        tags = parser.taglist('<div><p>paragraph</p<ignored></div>', reset=True)
+        self.assertEqual(tags, [Tag('div'), Tag('p')])
+        self.assertEqual(tags[0].text_and_html(),
+                         ('<p>paragraph</p<ignored>', '<div><p>paragraph</p<ignored></div>'))
+        self.assertEqual(tags[1].text_and_html(), ('paragraph', '<p>paragraph</p<ignored>'))
+
+        tags = parser.taglist('<img width="300px">must be empty</img>', reset=True)
+        self.assertEqual(tags, [Tag('img')])
+        self.assertEqual(tags[0].text_and_html(), ('', '<img width="300px">'))
+
+    def test_compliant_html_parsing(self):
+        # certain elements don't need to be closed (see HTMLTagParser.VOID_TAGS)
+        Tag = HTMLTagParser.Tag
+        html = '''
+            no error without closing tag: <img>
+            self closing is ok: <img />
+        '''
+        parser = HTMLTagParser()
+        tags = parser.taglist(html, reset=True)
+        self.assertEqual(tags, [Tag('img'), Tag('img')])
+
+        # don't get fooled by '>' in attributes
+        html = '''<img greater_a='1>0' greater_b="1>0">'''
+        tags = parser.taglist(html, reset=True)
+        self.assertEqual(tags[0].text_and_html(), ('', html))
+
+    def test_tag_return_order(self):
+        Tag = HTMLTagParser.Tag
+        html = '''
+        <t0>
+            <t1>
+                <t2>
+                    <t3 /> <t4 />
+                </t2>
+            </t1>
+            <t5>
+                <t6 />
+            </t5>
+        </t0>
+        <t7>
+            <t8 />
+        </t7>
+        '''
+        parser = HTMLTagParser()
+        tags = parser.taglist(html, reset=True)
+        self.assertEqual(
+            str(tags), str([Tag('t0'), Tag('t1'), Tag('t2'), Tag('t3'), Tag('t4'),
+                            Tag('t5'), Tag('t6'), Tag('t7'), Tag('t8')]))
+
+        tags = parser.taglist(html, reset=True, depth_first=True)
+        self.assertEqual(
+            str(tags), str([Tag('t3'), Tag('t4'), Tag('t2'), Tag('t1'), Tag('t6'),
+                            Tag('t5'), Tag('t0'), Tag('t8'), Tag('t7')]))
+
+        # return tags in nested order
+        tags = parser.taglist(html, reset=True, depth_first=None)
+        self.assertEqual(
+            str(tags), str([
+                [Tag('t0'),
+                 [Tag('t1'),
+                  [Tag('t2'), [Tag('t3')], [Tag('t4')]]],
+                 [Tag('t5'), [Tag('t6')]]],
+                [Tag('t7'), [Tag('t8')]]]))
+
+    def test_html_ignored_ranges(self):
+        def mark_comments(_string, char='^', nochar='-'):
+            cmts = HTMLIgnoreRanges(_string)
+            return "".join(char if _idx in cmts else nochar for _idx in range(len(_string)))
+
+        html_string = '''
+        no              comments         in            this              line
+        ---------------------------------------------------------------------
+        <!--                 whole line represents a comment              -->
+        ----^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^---
+        before <!--                      comment                  -->   after
+        -----------^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^-----------
+        this is a leftover comment -->     <!-- a new comment without closing
+        ^^^^^^^^^^^^^^^^^^^^^^^^^^^------------^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+        here   is   <!-- a comment -->   and   <!-- another comment --!>  end
+        ----------------^^^^^^^^^^^----------------^^^^^^^^^^^^^^^^^---------
+        <script> ignore here </script>            <SCRIPT> and here </SCRIPT>
+        --------^^^^^^^^^^^^^-----------------------------^^^^^^^^^^---------
+        '''
+
+        lines = textwrap.dedent(html_string).strip().splitlines()
+        for line, marker in zip(lines[0::2], lines[1::2]):
+            self.assertEqual((line, mark_comments(line)), (line, marker))
+
+        # yet we must be able to match script elements
+        test_string = '''<script type="text/javascript">var foo = 'bar';</script>'''
+        items = get_element_text_and_html_by_tag('script', test_string)
+        self.assertEqual(items, ("var foo = 'bar';", test_string))

yt_dlp/extractor/_extractors.py

@@ -1139,12 +1139,6 @@ from .microsoftembed import (
     MicrosoftMediusIE,
 )
 from .microsoftstream import MicrosoftStreamIE
-from .mildom import (
-    MildomClipIE,
-    MildomIE,
-    MildomUserVodIE,
-    MildomVodIE,
-)
 from .minds import (
     MindsChannelIE,
     MindsGroupIE,
@@ -1526,8 +1520,8 @@ from .pgatour import PGATourIE
 from .philharmoniedeparis import PhilharmonieDeParisIE
 from .phoenix import PhoenixIE
 from .photobucket import PhotobucketIE
+from .pialive import PiaLiveIE
 from .piapro import PiaproIE
-from .piaulizaportal import PIAULIZAPortalIE
 from .picarto import (
     PicartoIE,
     PicartoVodIE,
@@ -1563,10 +1557,6 @@ from .podbayfm import (
 )
 from .podchaser import PodchaserIE
 from .podomatic import PodomaticIE
-from .pokemon import (
-    PokemonIE,
-    PokemonWatchIE,
-)
 from .pokergo import (
     PokerGoCollectionIE,
     PokerGoIE,
@@ -2260,6 +2250,10 @@ from .ufctv import (
 )
 from .ukcolumn import UkColumnIE
 from .uktvplay import UKTVPlayIE
+from .uliza import (
+    UlizaPlayerIE,
+    UlizaPortalIE,
+)
 from .umg import UMGDeIE
 from .unistra import UnistraIE
 from .unity import UnityIE
@@ -2288,10 +2282,6 @@ from .utreon import UtreonIE
 from .varzesh3 import Varzesh3IE
 from .vbox7 import Vbox7IE
 from .veo import VeoIE
-from .veoh import (
-    VeohIE,
-    VeohUserIE,
-)
 from .vesti import VestiIE
 from .vevo import (
     VevoIE,

yt_dlp/extractor/ctvnews.py

@@ -1,14 +1,27 @@
+import json
 import re
+import urllib.parse
 
 from .common import InfoExtractor
-from ..utils import orderedSet
+from .ninecninemedia import NineCNineMediaIE
+from ..utils import extract_attributes, orderedSet
+from ..utils.traversal import find_element, traverse_obj
 
 
 class CTVNewsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:.+?\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
+    _BASE_REGEX = r'https?://(?:[^.]+\.)?ctvnews\.ca/'
+    _VIDEO_ID_RE = r'(?P<id>\d{5,})'
+    _PLAYLIST_ID_RE = r'(?P<id>\d\.\d{5,})'
+    _VALID_URL = [
+        rf'{_BASE_REGEX}video/c{_VIDEO_ID_RE}',
+        rf'{_BASE_REGEX}video(?:-gallery)?/?\?clipId={_VIDEO_ID_RE}',
+        rf'{_BASE_REGEX}video/?\?(?:playlist|bin)Id={_PLAYLIST_ID_RE}',
+        rf'{_BASE_REGEX}(?!video/)[^?#]*?{_PLAYLIST_ID_RE}/?(?:$|[?#])',
+        rf'{_BASE_REGEX}(?!video/)[^?#]+\?binId={_PLAYLIST_ID_RE}',
+    ]
     _TESTS = [{
         'url': 'http://www.ctvnews.ca/video?clipId=901995',
-        'md5': '9b8624ba66351a23e0b6e1391971f9af',
+        'md5': 'b608f466c7fa24b9666c6439d766ab7e',
         'info_dict': {
             'id': '901995',
             'ext': 'flv',
@@ -16,6 +29,33 @@ class CTVNewsIE(InfoExtractor):
             'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
             'timestamp': 1467286284,
             'upload_date': '20160630',
+            'categories': [],
+            'season_number': 0,
+            'season': 'Season 0',
+            'tags': [],
+            'series': 'CTV News National | Archive | Stories 2',
+            'season_id': '57981',
+            'thumbnail': r're:https?://.*\.jpg$',
+            'duration': 764.631,
+        },
+    }, {
+        'url': 'https://barrie.ctvnews.ca/video/c3030933-here_s-what_s-making-news-for-nov--15?binId=1272429',
+        'md5': '8b8c2b33c5c1803e3c26bc74ff8694d5',
+        'info_dict': {
+            'id': '3030933',
+            'ext': 'flv',
+            'title': 'Here’s what’s making news for Nov. 15',
+            'description': 'Here are the top stories we’re working on for CTV News at 11 for Nov. 15',
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2021_2_22_a602e68e-1514-410e-a67a-e1f7cccbacab_png_2000x1125.jpg',
+            'season_id': '58104',
+            'season_number': 0,
+            'tags': [],
+            'season': 'Season 0',
+            'categories': [],
+            'series': 'CTV News Barrie',
+            'upload_date': '20241116',
+            'duration': 42.943,
+            'timestamp': 1731722452,
         },
     }, {
         'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
@@ -31,6 +71,72 @@ class CTVNewsIE(InfoExtractor):
             'id': '1.2876780',
         },
         'playlist_mincount': 100,
+    }, {
+        'url': 'https://www.ctvnews.ca/it-s-been-23-years-since-toronto-called-in-the-army-after-a-major-snowstorm-1.5736957',
+        'info_dict':
+        {
+            'id': '1.5736957',
+        },
+        'playlist_mincount': 6,
+    }, {
+        'url': 'https://www.ctvnews.ca/business/respondents-to-bank-of-canada-questionnaire-largely-oppose-creating-a-digital-loonie-1.6665797',
+        'md5': '24bc4b88cdc17d8c3fc01dfc228ab72c',
+        'info_dict': {
+            'id': '2695026',
+            'ext': 'flv',
+            'season_id': '89852',
+            'series': 'From CTV News Channel',
+            'description': 'md5:796a985a23cacc7e1e2fafefd94afd0a',
+            'season': '2023',
+            'title': 'Bank of Canada asks public about digital currency',
+            'categories': [],
+            'tags': [],
+            'upload_date': '20230526',
+            'season_number': 2023,
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'timestamp': 1685105157,
+            'duration': 253.553,
+        },
+    }, {
+        'url': 'https://stox.ctvnews.ca/video-gallery?clipId=582589',
+        'md5': '135cc592df607d29dddc931f1b756ae2',
+        'info_dict': {
+            'id': '582589',
+            'ext': 'flv',
+            'categories': [],
+            'timestamp': 1427906183,
+            'season_number': 0,
+            'duration': 125.559,
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'series': 'CTV News Stox',
+            'description': 'CTV original footage of the rise and fall of the Berlin Wall.',
+            'title': 'Berlin Wall',
+            'season_id': '63817',
+            'season': 'Season 0',
+            'tags': [],
+            'upload_date': '20150401',
+        },
+    }, {
+        'url': 'https://ottawa.ctvnews.ca/features/regional-contact/regional-contact-archive?binId=1.1164587#3023759',
+        'md5': 'a14c0603557decc6531260791c23cc5e',
+        'info_dict': {
+            'id': '3023759',
+            'ext': 'flv',
+            'season_number': 2024,
+            'timestamp': 1731798000,
+            'season': '2024',
+            'episode': 'Episode 125',
+            'description': 'CTV News Ottawa at Six',
+            'duration': 2712.076,
+            'episode_number': 125,
+            'upload_date': '20241116',
+            'title': 'CTV News Ottawa at Six for Saturday, November 16, 2024',
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'categories': [],
+            'tags': [],
+            'series': 'CTV News Ottawa at Six',
+            'season_id': '92667',
+        },
     }, {
         'url': 'http://www.ctvnews.ca/1.810401',
         'only_matching': True,
@@ -42,29 +148,35 @@ class CTVNewsIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _ninecninemedia_url_result(self, clip_id):
+        return self.url_result(f'9c9media:ctvnews_web:{clip_id}', NineCNineMediaIE, clip_id)
+
     def _real_extract(self, url):
         page_id = self._match_id(url)
 
-        def ninecninemedia_url_result(clip_id):
-            return {
-                '_type': 'url_transparent',
-                'id': clip_id,
-                'url': f'9c9media:ctvnews_web:{clip_id}',
-                'ie_key': 'NineCNineMedia',
-            }
+        if mobj := re.fullmatch(self._VIDEO_ID_RE, urllib.parse.urlparse(url).fragment):
+            page_id = mobj.group('id')
 
-        if page_id.isdigit():
-            return ninecninemedia_url_result(page_id)
-        else:
-            webpage = self._download_webpage(f'http://www.ctvnews.ca/{page_id}', page_id, query={
-                'ot': 'example.AjaxPageLayout.ot',
-                'maxItemsPerPage': 1000000,
-            })
-            entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
-                re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
-            if not entries:
-                webpage = self._download_webpage(url, page_id)
-                if 'getAuthStates("' in webpage:
-                    entries = [ninecninemedia_url_result(clip_id) for clip_id in
-                               self._search_regex(r'getAuthStates\("([\d+,]+)"', webpage, 'clip ids').split(',')]
-            return self.playlist_result(entries, page_id)
+        if re.fullmatch(self._VIDEO_ID_RE, page_id):
+            return self._ninecninemedia_url_result(page_id)
+
+        webpage = self._download_webpage(f'https://www.ctvnews.ca/{page_id}', page_id, query={
+            'ot': 'example.AjaxPageLayout.ot',
+            'maxItemsPerPage': 1000000,
+        })
+        entries = [self._ninecninemedia_url_result(clip_id)
+                   for clip_id in orderedSet(re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
+        if not entries:
+            webpage = self._download_webpage(url, page_id)
+            if 'getAuthStates("' in webpage:
+                entries = [self._ninecninemedia_url_result(clip_id) for clip_id in
+                           self._search_regex(r'getAuthStates\("([\d+,]+)"', webpage, 'clip ids').split(',')]
+            else:
+                entries = [
+                    self._ninecninemedia_url_result(clip_id) for clip_id in
+                    traverse_obj(webpage, (
+                        {find_element(tag='jasper-player-container', html=True)},
+                        {extract_attributes}, 'axis-ids', {json.loads}, ..., 'axisId'))
+                ]
+
+        return self.playlist_result(entries, page_id)

yt_dlp/extractor/litv.py

@@ -1,30 +1,32 @@
 import json
+import uuid
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
+    join_nonempty,
     smuggle_url,
     traverse_obj,
     try_call,
     unsmuggle_url,
+    urljoin,
 )
 
 
 class LiTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
-
-    _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?content_id=%s'
-
+    _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:[^/?#]+/watch/|vod/[^/?#]+/content\.do\?content_id=)(?P<id>[\w-]+)'
+    _URL_TEMPLATE = 'https://www.litv.tv/%s/watch/%s'
+    _GEO_COUNTRIES = ['TW']
     _TESTS = [{
-        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'url': 'https://www.litv.tv/drama/watch/VOD00041610',
         'info_dict': {
             'id': 'VOD00041606',
             'title': '花千骨',
         },
         'playlist_count': 51,  # 50 episodes + 1 trailer
     }, {
-        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'url': 'https://www.litv.tv/drama/watch/VOD00041610',
         'md5': 'b90ff1e9f1d8f5cfcd0a44c3e2b34c7a',
         'info_dict': {
             'id': 'VOD00041610',
@@ -32,16 +34,15 @@ class LiTVIE(InfoExtractor):
             'title': '花千骨第1集',
             'thumbnail': r're:https?://.*\.jpg$',
             'description': '《花千骨》陸劇線上看。十六年前，平靜的村莊內，一名女嬰隨異相出生，途徑此地的蜀山掌門清虛道長算出此女命運非同一般，她體內散發的異香易招惹妖魔。一念慈悲下，他在村莊周邊設下結界阻擋妖魔入侵，讓其年滿十六後去蜀山，並賜名花千骨。',
-            'categories': ['奇幻', '愛情', '中國', '仙俠'],
+            'categories': ['奇幻', '愛情', '仙俠', '古裝'],
             'episode': 'Episode 1',
             'episode_number': 1,
         },
         'params': {
             'noplaylist': True,
         },
-        'skip': 'Georestricted to Taiwan',
     }, {
-        'url': 'https://www.litv.tv/promo/miyuezhuan/?content_id=VOD00044841&',
+        'url': 'https://www.litv.tv/drama/watch/VOD00044841',
         'md5': '88322ea132f848d6e3e18b32a832b918',
         'info_dict': {
             'id': 'VOD00044841',
@@ -55,94 +56,62 @@ class LiTVIE(InfoExtractor):
     def _extract_playlist(self, playlist_data, content_type):
         all_episodes = [
             self.url_result(smuggle_url(
-                self._URL_TEMPLATE % (content_type, episode['contentId']),
+                self._URL_TEMPLATE % (content_type, episode['content_id']),
                 {'force_noplaylist': True}))  # To prevent infinite recursion
-            for episode in traverse_obj(playlist_data, ('seasons', ..., 'episode', lambda _, v: v['contentId']))]
+            for episode in traverse_obj(playlist_data, ('seasons', ..., 'episodes', lambda _, v: v['content_id']))]
 
-        return self.playlist_result(all_episodes, playlist_data['contentId'], playlist_data.get('title'))
+        return self.playlist_result(all_episodes, playlist_data['content_id'], playlist_data.get('title'))
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
-
         video_id = self._match_id(url)
-
         webpage = self._download_webpage(url, video_id)
+        vod_data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']
 
-        if self._search_regex(
-                r'(?i)<meta\s[^>]*http-equiv="refresh"\s[^>]*content="[0-9]+;\s*url=https://www\.litv\.tv/"',
-                webpage, 'meta refresh redirect', default=False, group=0):
-            raise ExtractorError('No such content found', expected=True)
+        program_info = traverse_obj(vod_data, ('programInformation', {dict})) or {}
+        playlist_data = traverse_obj(vod_data, ('seriesTree'))
+        if playlist_data and self._yes_playlist(program_info.get('series_id'), video_id, smuggled_data):
+            return self._extract_playlist(playlist_data, program_info.get('content_type'))
 
-        program_info = self._parse_json(self._search_regex(
-            r'var\s+programInfo\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
-            video_id)
+        asset_id = traverse_obj(program_info, ('assets', 0, 'asset_id', {str}))
+        if asset_id:  # This is a VOD
+            media_type = 'vod'
+        else:  # This is a live stream
+            asset_id = program_info['content_id']
+            media_type = program_info['content_type']
+        puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
+        if puid:
+            endpoint = 'get-urls'
+        else:
+            puid = str(uuid.uuid4())
+            endpoint = 'get-urls-no-auth'
+        video_data = self._download_json(
+            f'https://www.litv.tv/api/{endpoint}', video_id,
+            data=json.dumps({'AssetId': asset_id, 'MediaType': media_type, 'puid': puid}).encode(),
+            headers={'Content-Type': 'application/json'})
 
-        # In browsers `getProgramInfo` request is always issued. Usually this
-        # endpoint gives the same result as the data embedded in the webpage.
-        # If, for some reason, there are no embedded data, we do an extra request.
-        if 'assetId' not in program_info:
-            program_info = self._download_json(
-                'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
-                query={'contentId': video_id},
-                headers={'Accept': 'application/json'})
-
-        series_id = program_info['seriesId']
-        if self._yes_playlist(series_id, video_id, smuggled_data):
-            playlist_data = self._download_json(
-                'https://www.litv.tv/vod/ajax/getSeriesTree', video_id,
-                query={'seriesId': series_id}, headers={'Accept': 'application/json'})
-            return self._extract_playlist(playlist_data, program_info['contentType'])
-
-        video_data = self._parse_json(self._search_regex(
-            r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
-            webpage, 'video data', default='{}'), video_id)
-        if not video_data:
-            payload = {'assetId': program_info['assetId']}
-            puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
-            if puid:
-                payload.update({
-                    'type': 'auth',
-                    'puid': puid,
-                })
-                endpoint = 'getUrl'
-            else:
-                payload.update({
-                    'watchDevices': program_info['watchDevices'],
-                    'contentType': program_info['contentType'],
-                })
-                endpoint = 'getMainUrlNoAuth'
-            video_data = self._download_json(
-                f'https://www.litv.tv/vod/ajax/{endpoint}', video_id,
-                data=json.dumps(payload).encode(),
-                headers={'Content-Type': 'application/json'})
-
-        if not video_data.get('fullpath'):
-            error_msg = video_data.get('errorMessage')
-            if error_msg == 'vod.error.outsideregionerror':
+        if error := traverse_obj(video_data, ('error', {dict})):
+            error_msg = traverse_obj(error, ('message', {str}))
+            if error_msg and 'OutsideRegionError' in error_msg:
                 self.raise_geo_restricted('This video is available in Taiwan only')
-            if error_msg:
+            elif error_msg:
                 raise ExtractorError(f'{self.IE_NAME} said: {error_msg}', expected=True)
-            raise ExtractorError(f'Unexpected result from {self.IE_NAME}')
+            raise ExtractorError(f'Unexpected error from {self.IE_NAME}')
 
         formats = self._extract_m3u8_formats(
-            video_data['fullpath'], video_id, ext='mp4',
-            entry_protocol='m3u8_native', m3u8_id='hls')
+            video_data['result']['AssetURLs'][0], video_id, ext='mp4', m3u8_id='hls')
         for a_format in formats:
             # LiTV HLS segments doesn't like compressions
             a_format.setdefault('http_headers', {})['Accept-Encoding'] = 'identity'
 
-        title = program_info['title'] + program_info.get('secondaryMark', '')
-        description = program_info.get('description')
-        thumbnail = program_info.get('imageFile')
-        categories = [item['name'] for item in program_info.get('category', [])]
-        episode = int_or_none(program_info.get('episode'))
-
         return {
             'id': video_id,
             'formats': formats,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
-            'episode_number': episode,
+            'title': join_nonempty('title', 'secondary_mark', delim='', from_dict=program_info),
+            **traverse_obj(program_info, {
+                'description': ('description', {str}),
+                'thumbnail': ('picture', {urljoin('https://p-cdnstatic.svc.litv.tv/')}),
+                'categories': ('genres', ..., 'name', {str}),
+                'episode_number': ('episode', {int_or_none}),
+            }),
         }

yt_dlp/extractor/mildom.py

@@ -1,291 +0,0 @@
-import functools
-import json
-import uuid
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    OnDemandPagedList,
-    determine_ext,
-    dict_get,
-    float_or_none,
-    traverse_obj,
-)
-
-
-class MildomBaseIE(InfoExtractor):
-    _GUEST_ID = None
-
-    def _call_api(self, url, video_id, query=None, note='Downloading JSON metadata', body=None):
-        if not self._GUEST_ID:
-            self._GUEST_ID = f'pc-gp-{uuid.uuid4()}'
-
-        content = self._download_json(
-            url, video_id, note=note, data=json.dumps(body).encode() if body else None,
-            headers={'Content-Type': 'application/json'} if body else {},
-            query={
-                '__guest_id': self._GUEST_ID,
-                '__platform': 'web',
-                **(query or {}),
-            })
-
-        if content['code'] != 0:
-            raise ExtractorError(
-                f'Mildom says: {content["message"]} (code {content["code"]})',
-                expected=True)
-        return content['body']
-
-
-class MildomIE(MildomBaseIE):
-    IE_NAME = 'mildom'
-    IE_DESC = 'Record ongoing live by specific user in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/(?P<id>\d+)'
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(f'https://www.mildom.com/{video_id}', video_id)
-
-        enterstudio = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/live/enterstudio', video_id,
-            note='Downloading live metadata', query={'user_id': video_id})
-        result_video_id = enterstudio.get('log_id', video_id)
-
-        servers = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/live/liveserver', result_video_id,
-            note='Downloading live server list', query={
-                'user_id': video_id,
-                'live_server_type': 'hls',
-            })
-
-        playback_token = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/live/token', result_video_id,
-            note='Obtaining live playback token', body={'host_id': video_id, 'type': 'hls'})
-        playback_token = traverse_obj(playback_token, ('data', ..., 'token'), get_all=False)
-        if not playback_token:
-            raise ExtractorError('Failed to obtain live playback token')
-
-        formats = self._extract_m3u8_formats(
-            f'{servers["stream_server"]}/{video_id}_master.m3u8?{playback_token}',
-            result_video_id, 'mp4', headers={
-                'Referer': 'https://www.mildom.com/',
-                'Origin': 'https://www.mildom.com',
-            })
-
-        for fmt in formats:
-            fmt.setdefault('http_headers', {})['Referer'] = 'https://www.mildom.com/'
-
-        return {
-            'id': result_video_id,
-            'title': self._html_search_meta('twitter:description', webpage, default=None) or traverse_obj(enterstudio, 'anchor_intro'),
-            'description': traverse_obj(enterstudio, 'intro', 'live_intro', expected_type=str),
-            'timestamp': float_or_none(enterstudio.get('live_start_ms'), scale=1000),
-            'uploader': self._html_search_meta('twitter:title', webpage, default=None) or traverse_obj(enterstudio, 'loginname'),
-            'uploader_id': video_id,
-            'formats': formats,
-            'is_live': True,
-        }
-
-
-class MildomVodIE(MildomBaseIE):
-    IE_NAME = 'mildom:vod'
-    IE_DESC = 'VOD in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/playback/(?P<user_id>\d+)/(?P<id>(?P=user_id)-[a-zA-Z0-9]+-?[0-9]*)'
-    _TESTS = [{
-        'url': 'https://www.mildom.com/playback/10882672/10882672-1597662269',
-        'info_dict': {
-            'id': '10882672-1597662269',
-            'ext': 'mp4',
-            'title': '始めてのミルダム配信じゃぃ！',
-            'thumbnail': r're:^https?://.*\.(png|jpg)$',
-            'upload_date': '20200817',
-            'duration': 4138.37,
-            'description': 'ゲームをしたくて！',
-            'timestamp': 1597662269.0,
-            'uploader_id': '10882672',
-            'uploader': 'kson組長(けいそん)',
-        },
-    }, {
-        'url': 'https://www.mildom.com/playback/10882672/10882672-1597758589870-477',
-        'info_dict': {
-            'id': '10882672-1597758589870-477',
-            'ext': 'mp4',
-            'title': '【kson】感染メイズ！麻酔銃で無双する',
-            'thumbnail': r're:^https?://.*\.(png|jpg)$',
-            'timestamp': 1597759093.0,
-            'uploader': 'kson組長(けいそん)',
-            'duration': 4302.58,
-            'uploader_id': '10882672',
-            'description': 'このステージ絶対乗り越えたい',
-            'upload_date': '20200818',
-        },
-    }, {
-        'url': 'https://www.mildom.com/playback/10882672/10882672-buha9td2lrn97fk2jme0',
-        'info_dict': {
-            'id': '10882672-buha9td2lrn97fk2jme0',
-            'ext': 'mp4',
-            'title': '【kson組長】CART RACER!!!',
-            'thumbnail': r're:^https?://.*\.(png|jpg)$',
-            'uploader_id': '10882672',
-            'uploader': 'kson組長(けいそん)',
-            'upload_date': '20201104',
-            'timestamp': 1604494797.0,
-            'duration': 4657.25,
-            'description': 'WTF',
-        },
-    }]
-
-    def _real_extract(self, url):
-        user_id, video_id = self._match_valid_url(url).group('user_id', 'id')
-        webpage = self._download_webpage(f'https://www.mildom.com/playback/{user_id}/{video_id}', video_id)
-
-        autoplay = self._call_api(
-            'https://cloudac.mildom.com/nonolive/videocontent/playback/getPlaybackDetail', video_id,
-            note='Downloading playback metadata', query={
-                'v_id': video_id,
-            })['playback']
-
-        formats = [{
-            'url': autoplay['audio_url'],
-            'format_id': 'audio',
-            'protocol': 'm3u8_native',
-            'vcodec': 'none',
-            'acodec': 'aac',
-            'ext': 'm4a',
-        }]
-        for fmt in autoplay['video_link']:
-            formats.append({
-                'format_id': 'video-{}'.format(fmt['name']),
-                'url': fmt['url'],
-                'protocol': 'm3u8_native',
-                'width': fmt['level'] * autoplay['video_width'] // autoplay['video_height'],
-                'height': fmt['level'],
-                'vcodec': 'h264',
-                'acodec': 'aac',
-                'ext': 'mp4',
-            })
-
-        return {
-            'id': video_id,
-            'title': self._html_search_meta(('og:description', 'description'), webpage, default=None) or autoplay.get('title'),
-            'description': traverse_obj(autoplay, 'video_intro'),
-            'timestamp': float_or_none(autoplay.get('publish_time'), scale=1000),
-            'duration': float_or_none(autoplay.get('video_length'), scale=1000),
-            'thumbnail': dict_get(autoplay, ('upload_pic', 'video_pic')),
-            'uploader': traverse_obj(autoplay, ('author_info', 'login_name')),
-            'uploader_id': user_id,
-            'formats': formats,
-        }
-
-
-class MildomClipIE(MildomBaseIE):
-    IE_NAME = 'mildom:clip'
-    IE_DESC = 'Clip in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/clip/(?P<id>(?P<user_id>\d+)-[a-zA-Z0-9]+)'
-    _TESTS = [{
-        'url': 'https://www.mildom.com/clip/10042245-63921673e7b147ebb0806d42b5ba5ce9',
-        'info_dict': {
-            'id': '10042245-63921673e7b147ebb0806d42b5ba5ce9',
-            'title': '全然違ったよ',
-            'timestamp': 1619181890,
-            'duration': 59,
-            'thumbnail': r're:https?://.+',
-            'uploader': 'ざきんぽ',
-            'uploader_id': '10042245',
-        },
-    }, {
-        'url': 'https://www.mildom.com/clip/10111524-ebf4036e5aa8411c99fb3a1ae0902864',
-        'info_dict': {
-            'id': '10111524-ebf4036e5aa8411c99fb3a1ae0902864',
-            'title': 'かっこいい',
-            'timestamp': 1621094003,
-            'duration': 59,
-            'thumbnail': r're:https?://.+',
-            'uploader': '(ルーキー',
-            'uploader_id': '10111524',
-        },
-    }, {
-        'url': 'https://www.mildom.com/clip/10660174-2c539e6e277c4aaeb4b1fbe8d22cb902',
-        'info_dict': {
-            'id': '10660174-2c539e6e277c4aaeb4b1fbe8d22cb902',
-            'title': 'あ',
-            'timestamp': 1614769431,
-            'duration': 31,
-            'thumbnail': r're:https?://.+',
-            'uploader': 'ドルゴルスレンギーン＝ダグワドルジ',
-            'uploader_id': '10660174',
-        },
-    }]
-
-    def _real_extract(self, url):
-        user_id, video_id = self._match_valid_url(url).group('user_id', 'id')
-        webpage = self._download_webpage(f'https://www.mildom.com/clip/{video_id}', video_id)
-
-        clip_detail = self._call_api(
-            'https://cloudac-cf-jp.mildom.com/nonolive/videocontent/clip/detail', video_id,
-            note='Downloading playback metadata', query={
-                'clip_id': video_id,
-            })
-
-        return {
-            'id': video_id,
-            'title': self._html_search_meta(
-                ('og:description', 'description'), webpage, default=None) or clip_detail.get('title'),
-            'timestamp': float_or_none(clip_detail.get('create_time')),
-            'duration': float_or_none(clip_detail.get('length')),
-            'thumbnail': clip_detail.get('cover'),
-            'uploader': traverse_obj(clip_detail, ('user_info', 'loginname')),
-            'uploader_id': user_id,
-
-            'url': clip_detail['url'],
-            'ext': determine_ext(clip_detail.get('url'), 'mp4'),
-        }
-
-
-class MildomUserVodIE(MildomBaseIE):
-    IE_NAME = 'mildom:user:vod'
-    IE_DESC = 'Download all VODs from specific user in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/profile/(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'https://www.mildom.com/profile/10093333',
-        'info_dict': {
-            'id': '10093333',
-            'title': 'Uploads from ねこばたけ',
-        },
-        'playlist_mincount': 732,
-    }, {
-        'url': 'https://www.mildom.com/profile/10882672',
-        'info_dict': {
-            'id': '10882672',
-            'title': 'Uploads from kson組長(けいそん)',
-        },
-        'playlist_mincount': 201,
-    }]
-
-    def _fetch_page(self, user_id, page):
-        page += 1
-        reply = self._call_api(
-            'https://cloudac.mildom.com/nonolive/videocontent/profile/playbackList',
-            user_id, note=f'Downloading page {page}', query={
-                'user_id': user_id,
-                'page': page,
-                'limit': '30',
-            })
-        if not reply:
-            return
-        for x in reply:
-            v_id = x.get('v_id')
-            if not v_id:
-                continue
-            yield self.url_result(f'https://www.mildom.com/playback/{user_id}/{v_id}')
-
-    def _real_extract(self, url):
-        user_id = self._match_id(url)
-        self.to_screen(f'This will download all VODs belonging to user. To download ongoing live video, use "https://www.mildom.com/{user_id}" instead')
-
-        profile = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/user/profileV2', user_id,
-            query={'user_id': user_id}, note='Downloading user profile')['user_info']
-
-        return self.playlist_result(
-            OnDemandPagedList(functools.partial(self._fetch_page, user_id), 30),
-            user_id, f'Uploads from {profile["loginname"]}')

yt_dlp/extractor/pialive.py

@@ -0,0 +1,122 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    extract_attributes,
+    get_element_by_class,
+    get_element_html_by_class,
+    multipart_encode,
+    str_or_none,
+    unified_timestamp,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class PiaLiveIE(InfoExtractor):
+    _VALID_URL = r'https?://player\.pia-live\.jp/stream/(?P<id>[\w-]+)'
+    _PLAYER_ROOT_URL = 'https://player.pia-live.jp/'
+    _PIA_LIVE_API_URL = 'https://api.pia-live.jp'
+    _API_KEY = 'kfds)FKFps-dms9e'
+    _TESTS = [{
+        'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krUDqGOwN4d61dCWQYOd6CTxl4hjya9dsfEZGsM4uGOUdax60lEI4twsXGXf7crmz8Gk__GhupTrWxA7RFRVt76',
+        'info_dict': {
+            'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+            'display_id': '2431867_001',
+            'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+            'live_status': 'was_live',
+            'comment_count': int,
+        },
+        'params': {
+            'getcomments': True,
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'skip': 'The video is no longer available',
+    }, {
+        'url': 'https://player.pia-live.jp/stream/4JagFBEIM14s_hK9aXHKf3k3F3bY5eoHFQxu68TC6krJdu0GVBVbVy01IwpJ6J3qBEm3d9TCTt1d0eWpsZGj7DrOjVOmS7GAWGwyscMgiThopJvzgWC4H5b-7XQjAfRZ',
+        'info_dict': {
+            'id': '9ce8b8ba-f6d1-4d1f-83a0-18c3148ded93',
+            'display_id': '2431867_002',
+            'title': 'こながめでたい日２０２４の視聴ページ | PIA LIVE STREAM(ぴあライブストリーム)',
+            'live_status': 'was_live',
+            'comment_count': int,
+        },
+        'params': {
+            'getcomments': True,
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'skip': 'The video is no longer available',
+    }]
+
+    def _extract_var(self, variable, html):
+        return self._search_regex(
+            rf'(?:var|const|let)\s+{variable}\s*=\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
+            html, f'variable {variable}', group='value')
+
+    def _real_extract(self, url):
+        video_key = self._match_id(url)
+        webpage = self._download_webpage(url, video_key)
+
+        program_code = self._extract_var('programCode', webpage)
+        article_code = self._extract_var('articleCode', webpage)
+        title = self._html_extract_title(webpage)
+
+        if get_element_html_by_class('play-end', webpage):
+            raise ExtractorError('The video is no longer available', expected=True, video_id=program_code)
+
+        if start_info := clean_html(get_element_by_class('play-waiting__date', webpage)):
+            date, time = self._search_regex(
+                r'(?P<date>\d{4}/\d{1,2}/\d{1,2})\([月火水木金土日]\)(?P<time>\d{2}:\d{2})',
+                start_info, 'start_info', fatal=False, group=('date', 'time'))
+            if date and time:
+                release_timestamp_str = f'{date} {time} +09:00'
+                release_timestamp = unified_timestamp(release_timestamp_str)
+                self.raise_no_formats(f'The video will be available after {release_timestamp_str}', expected=True)
+                return {
+                    'id': program_code,
+                    'title': title,
+                    'live_status': 'is_upcoming',
+                    'release_timestamp': release_timestamp,
+                }
+
+        payload, content_type = multipart_encode({
+            'play_url': video_key,
+            'api_key': self._API_KEY,
+        })
+        api_data_and_headers = {
+            'data': payload,
+            'headers': {'Content-Type': content_type, 'Referer': self._PLAYER_ROOT_URL},
+        }
+
+        player_tag_list = self._download_json(
+            f'{self._PIA_LIVE_API_URL}/perf/player-tag-list/{program_code}', program_code,
+            'Fetching player tag list', 'Unable to fetch player tag list', **api_data_and_headers)
+
+        return self.url_result(
+            extract_attributes(player_tag_list['data']['movie_one_tag'])['src'],
+            url_transparent=True, title=title, display_id=program_code,
+            __post_extractor=self.extract_comments(program_code, article_code, api_data_and_headers))
+
+    def _get_comments(self, program_code, article_code, api_data_and_headers):
+        chat_room_url = traverse_obj(self._download_json(
+            f'{self._PIA_LIVE_API_URL}/perf/chat-tag-list/{program_code}/{article_code}', program_code,
+            'Fetching chat info', 'Unable to fetch chat info', fatal=False, **api_data_and_headers),
+            ('data', 'chat_one_tag', {extract_attributes}, 'src', {url_or_none}))
+        if not chat_room_url:
+            return
+        comment_page = self._download_webpage(
+            chat_room_url, program_code, 'Fetching comment page', 'Unable to fetch comment page',
+            fatal=False, headers={'Referer': self._PLAYER_ROOT_URL})
+        if not comment_page:
+            return
+        yield from traverse_obj(self._search_json(
+            r'var\s+_history\s*=', comment_page, 'comment list',
+            program_code, contains_pattern=r'\[(?s:.+)\]', fatal=False), (..., {
+                'timestamp': (0, {int}),
+                'author_is_uploader': (1, {lambda x: x == 2}),
+                'author': (2, {str}),
+                'text': (3, {str}),
+                'id': (4, {str_or_none}),
+            }))

yt_dlp/extractor/piaulizaportal.py

@@ -1,70 +0,0 @@
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    int_or_none,
-    parse_qs,
-    time_seconds,
-    traverse_obj,
-)
-
-
-class PIAULIZAPortalIE(InfoExtractor):
-    IE_DESC = 'ulizaportal.jp - PIA LIVE STREAM'
-    _VALID_URL = r'https?://(?:www\.)?ulizaportal\.jp/pages/(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})'
-    _TESTS = [{
-        'url': 'https://ulizaportal.jp/pages/005f18b7-e810-5618-cb82-0987c5755d44',
-        'info_dict': {
-            'id': '005f18b7-e810-5618-cb82-0987c5755d44',
-            'title': 'プレゼンテーションプレイヤーのサンプル',
-            'live_status': 'not_live',
-        },
-        'params': {
-            'skip_download': True,
-            'ignore_no_formats_error': True,
-        },
-    }, {
-        'url': 'https://ulizaportal.jp/pages/005e1b23-fe93-5780-19a0-98e917cc4b7d?expires=4102412400&signature=f422a993b683e1068f946caf406d211c17d1ef17da8bef3df4a519502155aa91&version=1',
-        'info_dict': {
-            'id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
-            'title': '【確認用】視聴サンプルページ（ULIZA）',
-            'live_status': 'not_live',
-        },
-        'params': {
-            'skip_download': True,
-            'ignore_no_formats_error': True,
-        },
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        expires = int_or_none(traverse_obj(parse_qs(url), ('expires', 0)))
-        if expires and expires <= time_seconds():
-            raise ExtractorError('The link is expired.', video_id=video_id, expected=True)
-
-        webpage = self._download_webpage(url, video_id)
-
-        player_data = self._download_webpage(
-            self._search_regex(
-                r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
-                webpage, 'player data url'),
-            video_id, headers={'Referer': 'https://ulizaportal.jp/'},
-            note='Fetching player data', errnote='Unable to fetch player data')
-
-        formats = self._extract_m3u8_formats(
-            self._search_regex(
-                r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data,
-                'm3u8 url', default=None),
-            video_id, fatal=False)
-        m3u8_type = self._search_regex(
-            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
-
-        return {
-            'id': video_id,
-            'title': self._html_extract_title(webpage),
-            'formats': formats,
-            'live_status': {
-                'video': 'is_live',
-                'dvr': 'was_live',  # short-term archives
-            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
-        }

yt_dlp/extractor/pokemon.py

@@ -1,136 +0,0 @@
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    extract_attributes,
-    int_or_none,
-    js_to_json,
-    merge_dicts,
-)
-
-
-class PokemonIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pokemon\.com/[a-z]{2}(?:.*?play=(?P<id>[a-z0-9]{32})|/(?:[^/]+/)+(?P<display_id>[^/?#&]+))'
-    _TESTS = [{
-        'url': 'https://www.pokemon.com/us/pokemon-episodes/20_30-the-ol-raise-and-switch/',
-        'md5': '2fe8eaec69768b25ef898cda9c43062e',
-        'info_dict': {
-            'id': 'afe22e30f01c41f49d4f1d9eab5cd9a4',
-            'ext': 'mp4',
-            'title': 'The Ol’ Raise and Switch!',
-            'description': 'md5:7db77f7107f98ba88401d3adc80ff7af',
-        },
-        'add_id': ['LimelightMedia'],
-    }, {
-        # no data-video-title
-        'url': 'https://www.pokemon.com/fr/episodes-pokemon/films-pokemon/pokemon-lascension-de-darkrai-2008',
-        'info_dict': {
-            'id': 'dfbaf830d7e54e179837c50c0c6cc0e1',
-            'ext': 'mp4',
-            'title': "Pokémon : L'ascension de Darkrai",
-            'description': 'md5:d1dbc9e206070c3e14a06ff557659fb5',
-        },
-        'add_id': ['LimelightMedia'],
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://www.pokemon.com/uk/pokemon-episodes/?play=2e8b5c761f1d4a9286165d7748c1ece2',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.pokemon.com/fr/episodes-pokemon/18_09-un-hiver-inattendu/',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.pokemon.com/de/pokemon-folgen/01_20-bye-bye-smettbo/',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        video_id, display_id = self._match_valid_url(url).groups()
-        webpage = self._download_webpage(url, video_id or display_id)
-        video_data = extract_attributes(self._search_regex(
-            r'(<[^>]+data-video-id="{}"[^>]*>)'.format(video_id if video_id else '[a-z0-9]{32}'),
-            webpage, 'video data element'))
-        video_id = video_data['data-video-id']
-        title = video_data.get('data-video-title') or self._html_search_meta(
-            'pkm-title', webpage, ' title', default=None) or self._search_regex(
-            r'<h1[^>]+\bclass=["\']us-title[^>]+>([^<]+)', webpage, 'title')
-        return {
-            '_type': 'url_transparent',
-            'id': video_id,
-            'url': f'limelight:media:{video_id}',
-            'title': title,
-            'description': video_data.get('data-video-summary'),
-            'thumbnail': video_data.get('data-video-poster'),
-            'series': 'Pokémon',
-            'season_number': int_or_none(video_data.get('data-video-season')),
-            'episode': title,
-            'episode_number': int_or_none(video_data.get('data-video-episode')),
-            'ie_key': 'LimelightMedia',
-        }
-
-
-class PokemonWatchIE(InfoExtractor):
-    _VALID_URL = r'https?://watch\.pokemon\.com/[a-z]{2}-[a-z]{2}/(?:#/)?player(?:\.html)?\?id=(?P<id>[a-z0-9]{32})'
-    _API_URL = 'https://www.pokemon.com/api/pokemontv/v2/channels/{0:}'
-    _TESTS = [{
-        'url': 'https://watch.pokemon.com/en-us/player.html?id=8309a40969894a8e8d5bc1311e9c5667',
-        'md5': '62833938a31e61ab49ada92f524c42ff',
-        'info_dict': {
-            'id': '8309a40969894a8e8d5bc1311e9c5667',
-            'ext': 'mp4',
-            'title': 'Lillier and the Staff!',
-            'description': 'md5:338841b8c21b283d24bdc9b568849f04',
-        },
-    }, {
-        'url': 'https://watch.pokemon.com/en-us/#/player?id=3fe7752ba09141f0b0f7756d1981c6b2',
-        'only_matching': True,
-    }, {
-        'url': 'https://watch.pokemon.com/de-de/player.html?id=b3c402e111a4459eb47e12160ab0ba07',
-        'only_matching': True,
-    }]
-
-    def _extract_media(self, channel_array, video_id):
-        for channel in channel_array:
-            for media in channel.get('media'):
-                if media.get('id') == video_id:
-                    return media
-        return None
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        info = {
-            '_type': 'url',
-            'id': video_id,
-            'url': f'limelight:media:{video_id}',
-            'ie_key': 'LimelightMedia',
-        }
-
-        # API call can be avoided entirely if we are listing formats
-        if self.get_param('listformats', False):
-            return info
-
-        webpage = self._download_webpage(url, video_id)
-        build_vars = self._parse_json(self._search_regex(
-            r'(?s)buildVars\s*=\s*({.*?})', webpage, 'build vars'),
-            video_id, transform_source=js_to_json)
-        region = build_vars.get('region')
-        channel_array = self._download_json(self._API_URL.format(region), video_id)
-        video_data = self._extract_media(channel_array, video_id)
-
-        if video_data is None:
-            raise ExtractorError(
-                f'Video {video_id} does not exist', expected=True)
-
-        info['_type'] = 'url_transparent'
-        images = video_data.get('images')
-
-        return merge_dicts(info, {
-            'title': video_data.get('title'),
-            'description': video_data.get('description'),
-            'thumbnail': images.get('medium') or images.get('small'),
-            'series': 'Pokémon',
-            'season_number': int_or_none(video_data.get('season')),
-            'episode': video_data.get('title'),
-            'episode_number': int_or_none(video_data.get('episode')),
-        })

yt_dlp/extractor/sonyliv.py

@@ -199,8 +199,9 @@ class SonyLIVSeriesIE(InfoExtractor):
         },
     }]
     _API_BASE = 'https://apiv2.sonyliv.com/AGL'
+    _SORT_ORDERS = ('asc', 'desc')
 
-    def _entries(self, show_id):
+    def _entries(self, show_id, sort_order):
         headers = {
             'Accept': 'application/json, text/plain, */*',
             'Referer': 'https://www.sonyliv.com',
@@ -215,6 +216,9 @@ class SonyLIVSeriesIE(InfoExtractor):
                 'from': '0',
                 'to': '49',
             }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
+
+        if sort_order == 'desc':
+            seasons = reversed(seasons)
         for season in seasons:
             season_id = str(season['id'])
             note = traverse_obj(season, ('metadata', 'title', {str})) or 'season'
@@ -226,7 +230,7 @@ class SonyLIVSeriesIE(InfoExtractor):
                         'from': str(cursor),
                         'to': str(cursor + 99),
                         'orderBy': 'episodeNumber',
-                        'sortOrder': 'asc',
+                        'sortOrder': sort_order,
                     }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
                 if not episodes:
                     break
@@ -237,4 +241,10 @@ class SonyLIVSeriesIE(InfoExtractor):
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
-        return self.playlist_result(self._entries(show_id), playlist_id=show_id)
+
+        sort_order = self._configuration_arg('sort_order', [self._SORT_ORDERS[0]])[0]
+        if sort_order not in self._SORT_ORDERS:
+            raise ValueError(
+                f'Invalid sort order "{sort_order}". Allowed values are: {", ".join(self._SORT_ORDERS)}')
+
+        return self.playlist_result(self._entries(show_id, sort_order), playlist_id=show_id)

yt_dlp/extractor/uliza.py

@@ -0,0 +1,113 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    make_archive_id,
+    parse_qs,
+    time_seconds,
+)
+from ..utils.traversal import traverse_obj
+
+
+class UlizaPlayerIE(InfoExtractor):
+    _VALID_URL = r'https://player-api\.p\.uliza\.jp/v1/players/[^?#]+\?(?:[^#]*&)?name=(?P<id>[^#&]+)'
+    _TESTS = [{
+        'url': 'https://player-api.p.uliza.jp/v1/players/timeshift-disabled/pia/admin?type=normal&playerobjectname=ulizaPlayer&name=livestream01_dvr&repeatable=true',
+        'info_dict': {
+            'id': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+            'ext': 'mp4',
+            'title': '88f3109a-f503-4d0f-a9f7-9f39ac745d84',
+            'live_status': 'was_live',
+            '_old_archive_ids': ['piaulizaportal 88f3109a-f503-4d0f-a9f7-9f39ac745d84'],
+        },
+    }, {
+        'url': 'https://player-api.p.uliza.jp/v1/players/uliza_jp_gallery_normal/promotion/admin?type=presentation&name=cookings&targetid=player1',
+        'info_dict': {
+            'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'ext': 'mp4',
+            'title': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal ae350126-5e22-4a7f-a8ac-8d0fd448b800'],
+        },
+    }, {
+        'url': 'https://player-api.p.uliza.jp/v1/players/default-player/pia/admin?type=normal&name=pia_movie_uliza_fix&targetid=ulizahtml5&repeatable=true',
+        'info_dict': {
+            'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'ext': 'mp4',
+            'title': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal 0644ecc8-e354-41b4-b957-3b08a2d63df1'],
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        player_data = self._download_webpage(
+            url, display_id, headers={'Referer': 'https://player-api.p.uliza.jp/'},
+            note='Fetching player data', errnote='Unable to fetch player data')
+
+        m3u8_url = self._search_regex(
+            r'["\'](https://vms-api\.p\.uliza\.jp/v1/prog-index\.m3u8[^"\']+)', player_data, 'm3u8 url')
+        video_id = parse_qs(m3u8_url).get('ss', [display_id])[0]
+
+        formats = self._extract_m3u8_formats(m3u8_url, video_id)
+        m3u8_type = self._search_regex(
+            r'/hls/(dvr|video)/', traverse_obj(formats, (0, 'url')), 'm3u8 type', default=None)
+        return {
+            'id': video_id,
+            'title': video_id,
+            'formats': formats,
+            'live_status': {
+                'video': 'is_live',
+                'dvr': 'was_live',  # short-term archives
+            }.get(m3u8_type, 'not_live'),  # VOD or long-term archives
+            '_old_archive_ids': [make_archive_id('PIAULIZAPortal', video_id)],
+        }
+
+
+class UlizaPortalIE(InfoExtractor):
+    IE_DESC = 'ulizaportal.jp'
+    _VALID_URL = r'https?://(?:www\.)?ulizaportal\.jp/pages/(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})'
+    _TESTS = [{
+        'url': 'https://ulizaportal.jp/pages/005f18b7-e810-5618-cb82-0987c5755d44',
+        'info_dict': {
+            'id': 'ae350126-5e22-4a7f-a8ac-8d0fd448b800',
+            'display_id': '005f18b7-e810-5618-cb82-0987c5755d44',
+            'title': 'プレゼンテーションプレイヤーのサンプル',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal ae350126-5e22-4a7f-a8ac-8d0fd448b800'],
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+    }, {
+        'url': 'https://ulizaportal.jp/pages/005e1b23-fe93-5780-19a0-98e917cc4b7d?expires=4102412400&signature=f422a993b683e1068f946caf406d211c17d1ef17da8bef3df4a519502155aa91&version=1',
+        'info_dict': {
+            'id': '0644ecc8-e354-41b4-b957-3b08a2d63df1',
+            'display_id': '005e1b23-fe93-5780-19a0-98e917cc4b7d',
+            'title': '【確認用】視聴サンプルページ（ULIZA）',
+            'live_status': 'not_live',
+            '_old_archive_ids': ['piaulizaportal 0644ecc8-e354-41b4-b957-3b08a2d63df1'],
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        expires = int_or_none(traverse_obj(parse_qs(url), ('expires', 0)))
+        if expires and expires <= time_seconds():
+            raise ExtractorError('The link is expired', video_id=video_id, expected=True)
+
+        webpage = self._download_webpage(url, video_id)
+
+        player_data_url = self._search_regex(
+            r'<script [^>]*\bsrc="(https://player-api\.p\.uliza\.jp/v1/players/[^"]+)"',
+            webpage, 'player data url')
+        return self.url_result(
+            player_data_url, UlizaPlayerIE, url_transparent=True,
+            display_id=video_id, video_title=self._html_extract_title(webpage))

yt_dlp/extractor/veoh.py

@@ -1,189 +0,0 @@
-import functools
-import json
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    OnDemandPagedList,
-    int_or_none,
-    parse_duration,
-    qualities,
-    remove_start,
-    strip_or_none,
-)
-
-
-class VeohIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|videos|embed|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
-
-    _TESTS = [{
-        'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
-        'md5': '620e68e6a3cff80086df3348426c9ca3',
-        'info_dict': {
-            'id': 'v56314296nk7Zdmz3',
-            'ext': 'mp4',
-            'title': 'Straight Backs Are Stronger',
-            'description': 'md5:203f976279939a6dc664d4001e13f5f4',
-            'thumbnail': 're:https://fcache\\.veoh\\.com/file/f/th56314296\\.jpg(\\?.*)?',
-            'uploader': 'LUMOback',
-            'duration': 46,
-            'view_count': int,
-            'average_rating': int,
-            'comment_count': int,
-            'age_limit': 0,
-            'categories': ['technology_and_gaming'],
-            'tags': ['posture', 'posture', 'sensor', 'back', 'pain', 'wearable', 'tech', 'lumo'],
-        },
-    }, {
-        'url': 'http://www.veoh.com/embed/v56314296nk7Zdmz3',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
-        'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
-        'info_dict': {
-            'id': '27701988',
-            'ext': 'mp4',
-            'title': 'Chile workers cover up to avoid skin damage',
-            'description': 'md5:2bd151625a60a32822873efc246ba20d',
-            'uploader': 'afp-news',
-            'duration': 123,
-        },
-        'skip': 'This video has been deleted.',
-    }, {
-        'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
-        'md5': '4fde7b9e33577bab2f2f8f260e30e979',
-        'note': 'Embedded ooyala video',
-        'info_dict': {
-            'id': '69525809',
-            'ext': 'mp4',
-            'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
-            'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
-            'uploader': 'newsy-videos',
-        },
-        'skip': 'This video has been deleted.',
-    }, {
-        'url': 'http://www.veoh.com/watch/e152215AJxZktGS',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.veoh.com/videos/v16374379WA437rMH',
-        'md5': 'cceb73f3909063d64f4b93d4defca1b3',
-        'info_dict': {
-            'id': 'v16374379WA437rMH',
-            'ext': 'mp4',
-            'title': 'Phantasmagoria 2, pt. 1-3',
-            'description': 'Phantasmagoria: a Puzzle of Flesh',
-            'thumbnail': 're:https://fcache\\.veoh\\.com/file/f/th16374379\\.jpg(\\?.*)?',
-            'uploader': 'davidspackage',
-            'duration': 968,
-            'view_count': int,
-            'average_rating': int,
-            'comment_count': int,
-            'age_limit': 18,
-            'categories': ['technology_and_gaming', 'gaming'],
-            'tags': ['puzzle', 'of', 'flesh'],
-        },
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        metadata = self._download_json(
-            'https://www.veoh.com/watch/getVideo/' + video_id,
-            video_id)
-        video = metadata['video']
-        title = video['title']
-
-        thumbnail_url = None
-        q = qualities(['Regular', 'HQ'])
-        formats = []
-        for f_id, f_url in video.get('src', {}).items():
-            if not f_url:
-                continue
-            if f_id == 'poster':
-                thumbnail_url = f_url
-            else:
-                formats.append({
-                    'format_id': f_id,
-                    'quality': q(f_id),
-                    'url': f_url,
-                })
-
-        categories = metadata.get('categoryPath')
-        if not categories:
-            category = remove_start(strip_or_none(video.get('category')), 'category_')
-            categories = [category] if category else None
-        tags = video.get('tags')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': video.get('description'),
-            'thumbnail': thumbnail_url,
-            'uploader': video.get('author', {}).get('nickname'),
-            'duration': int_or_none(video.get('lengthBySec')) or parse_duration(video.get('length')),
-            'view_count': int_or_none(video.get('views')),
-            'formats': formats,
-            'average_rating': int_or_none(video.get('rating')),
-            'comment_count': int_or_none(video.get('numOfComments')),
-            'age_limit': 18 if video.get('contentRatingId') == 2 else 0,
-            'categories': categories,
-            'tags': tags.split(', ') if tags else None,
-        }
-
-
-class VeohUserIE(VeohIE):  # XXX: Do not subclass from concrete IE
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/users/(?P<id>[\w-]+)'
-    IE_NAME = 'veoh:user'
-
-    _TESTS = [
-        {
-            'url': 'https://www.veoh.com/users/valentinazoe',
-            'info_dict': {
-                'id': 'valentinazoe',
-                'title': 'valentinazoe (Uploads)',
-            },
-            'playlist_mincount': 75,
-        },
-        {
-            'url': 'https://www.veoh.com/users/PiensaLibre',
-            'info_dict': {
-                'id': 'PiensaLibre',
-                'title': 'PiensaLibre (Uploads)',
-            },
-            'playlist_mincount': 2,
-        }]
-
-    _PAGE_SIZE = 16
-
-    def _fetch_page(self, uploader, page):
-        response = self._download_json(
-            'https://www.veoh.com/users/published/videos', uploader,
-            note=f'Downloading videos page {page + 1}',
-            headers={
-                'x-csrf-token': self._TOKEN,
-                'content-type': 'application/json;charset=UTF-8',
-            },
-            data=json.dumps({
-                'username': uploader,
-                'maxResults': self._PAGE_SIZE,
-                'page': page + 1,
-                'requestName': 'userPage',
-            }).encode())
-        if not response.get('success'):
-            raise ExtractorError(response['message'])
-
-        for video in response['videos']:
-            yield self.url_result(f'https://www.veoh.com/watch/{video["permalinkId"]}', VeohIE,
-                                  video['permalinkId'], video.get('title'))
-
-    def _real_initialize(self):
-        webpage = self._download_webpage(
-            'https://www.veoh.com', None, note='Downloading authorization token')
-        self._TOKEN = self._search_regex(
-            r'csrfToken:\s*(["\'])(?P<token>[0-9a-zA-Z]{40})\1', webpage,
-            'request token', group='token')
-
-    def _real_extract(self, url):
-        uploader = self._match_id(url)
-        return self.playlist_result(OnDemandPagedList(
-            functools.partial(self._fetch_page, uploader),
-            self._PAGE_SIZE), uploader, f'{uploader} (Uploads)')

yt_dlp/extractor/youtube.py

@@ -5087,7 +5087,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
     def _rich_entries(self, rich_grid_renderer):
         renderer = traverse_obj(
             rich_grid_renderer,
-            ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer', 'shortsLockupViewModel'), any)) or {}
+            ('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer', 'shortsLockupViewModel', 'lockupViewModel'), any)) or {}
         video_id = renderer.get('videoId')
         if video_id:
             yield self._extract_video(renderer)
@@ -5114,6 +5114,18 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
                 })),
                 thumbnails=self._extract_thumbnails(renderer, 'thumbnail', final_key='sources'))
             return
+        # lockupViewModel extraction
+        content_id = renderer.get('contentId')
+        if content_id and renderer.get('contentType') == 'LOCKUP_CONTENT_TYPE_PODCAST':
+            yield self.url_result(
+                f'https://www.youtube.com/playlist?list={content_id}',
+                ie=YoutubeTabIE, video_id=content_id,
+                **traverse_obj(renderer, {
+                    'title': ('metadata', 'lockupMetadataViewModel', 'title', 'content', {str}),
+                }),
+                thumbnails=self._extract_thumbnails(renderer, (
+                    'contentImage', 'collectionThumbnailViewModel', 'primaryThumbnail', 'thumbnailViewModel', 'image'), final_key='sources'))
+            return
 
     def _video_entry(self, video_renderer):
         video_id = video_renderer.get('videoId')
@@ -6706,22 +6718,22 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
         },
         'playlist_count': 0,
     }, {
-        # Podcasts tab, with rich entry playlistRenderers
+        # Podcasts tab, with rich entry lockupViewModel
         'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
         'info_dict': {
             'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
             'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
             'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
             'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
-            'title': '99 Percent Invisible - Podcasts',
-            'uploader': '99 Percent Invisible',
+            'title': '99% Invisible - Podcasts',
+            'uploader': '99% Invisible',
             'channel_follower_count': int,
             'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
             'tags': [],
-            'channel': '99 Percent Invisible',
+            'channel': '99% Invisible',
             'uploader_id': '@99percentinvisiblepodcast',
         },
-        'playlist_count': 0,
+        'playlist_count': 5,
     }, {
         # Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
         'url': 'https://www.youtube.com/@AHimitsu/releases',

yt_dlp/parsing.py

@@ -0,0 +1,348 @@
+import collections
+import contextlib
+import itertools
+import re
+from html.parser import HTMLParser
+
+from .compat import compat_HTMLParseError
+from .utils import orderedSet
+
+
+class HTMLIgnoreRanges:
+    """check if an offset is within CDATA content elements (script, style) or XML comments
+
+        note:
+            * given offsets must be in increasing order
+            * no detection of nested constructs (e.g. comments within script tags)
+
+        usage:
+            ranges = HTMLIgnoreRanges(html)
+            if offset in ranges:
+                ...
+    """
+    REGEX = re.compile(r'<!--|--!?>|</?\s*(?:script|style)\b[^>]*>', flags=re.IGNORECASE)
+
+    def __init__(self, html):
+        self.html = html
+        self._last_match = None
+        self._final = False
+
+    def __contains__(self, offset):
+        assert isinstance(offset, int)
+
+        if not self._final and (self._last_match is None or offset >= self._last_match.end()):
+            match = self.REGEX.search(self.html, offset)
+            if match:
+                self._last_match = match
+            else:
+                self._final = True
+
+        if self._last_match is None:
+            return False
+        match_string = self._last_match.group()
+        if match_string.startswith('</') or match_string in ('-->', '--!>'):
+            return offset < self._last_match.start()
+        return offset >= self._last_match.end()
+
+
+class HTMLTagParser(HTMLParser):
+    """HTML parser which returns found elements as instances of 'Tag'
+    when STRICT=True can raise compat_HTMLParseError() on malformed HTML elements
+
+    usage:
+        parser = HTMLTagParser()
+        for tag_obj in parser.taglist(html):
+            tag_obj.text_and_html()
+
+    """
+
+    STRICT = False
+    ANY_TAG_REGEX = re.compile(r'''<(?:"[^"]*"|'[^']*'|[^"'>])*?>''')
+    VOID_TAGS = {
+        'area', 'base', 'br', 'col', 'embed', 'hr', 'img', 'input',
+        'keygen', 'link', 'meta', 'param', 'source', 'track', 'wbr',
+    }
+
+    class Tag:
+        __slots__ = 'name', 'string', 'attrs', '_openrange', '_closerange'
+
+        def __init__(self, name, *, string='', attrs=()):
+            self.name = name
+            self.string = string
+            self.attrs = tuple(attrs)
+            self._openrange = None
+            self._closerange = None
+
+        def __str__(self):
+            return self.name
+
+        def __repr__(self):
+            return f'{self.__class__.__name__}({str(self)!r})'
+
+        def __eq__(self, other):
+            return self.name == other
+
+        def openrange(self, offset, startlen=0):
+            if isinstance(offset, slice):
+                self._openrange = offset
+            else:
+                self._openrange = slice(offset, offset + startlen)
+
+        def closerange(self, offset, stoplen=0):
+            if isinstance(offset, slice):
+                self._closerange = offset
+            else:
+                self._closerange = slice(offset, offset + stoplen)
+
+        def opentag(self):
+            return self.string[self._openrange] if self._openrange else ''
+
+        def html(self):
+            if not self._openrange:
+                return ''
+            if self._closerange:
+                return self.string[self._openrange.start:self._closerange.stop]
+            return self.string[self._openrange]
+
+        def text(self):
+            if self._openrange and self._closerange:
+                return self.string[self._openrange.stop:self._closerange.start]
+            return ''
+
+        def text_and_html(self):
+            return self.text(), self.html()
+
+    class AbortException(Exception):
+        pass
+
+    def __init__(self):
+        self.tagstack = collections.deque()
+        self._nestedtags = [[]]
+        super().__init__()
+        self._offset = self.offset
+
+    def predicate(self, tag, attrs):
+        """ return True for every encountered opening tag that should be processed """
+        return True
+
+    def callback(self, tag_obj):
+        """ this will be called when the requested tag is closed """
+
+    def reset(self):
+        super().reset()
+        self.tagstack.clear()
+
+    def taglist(self, data, reset=True, depth_first=False):
+        """ parse data and return found tag objects
+        @param data:    html string
+        @param reset:   reset state
+        @param depth_first: return order: as opened (False), as closed (True), nested (None)
+        @return: list of Tag objects
+        """
+        def flatten(_list, first=True):
+            rlist = _list if first or not depth_first else itertools.chain(_list[1:], _list[:1])
+            for item in rlist:
+                if isinstance(item, list):
+                    yield from flatten(item, first=False)
+                else:
+                    yield item
+
+        if reset:
+            self.reset()
+        with contextlib.suppress(HTMLTagParser.AbortException):
+            self.feed(data)
+        if self.STRICT and self.tagstack:
+            orphans = ', '.join(map(repr, map(str, orderedSet(self.tagstack, lazy=True))))
+            raise compat_HTMLParseError(f'unclosed tag {orphans}')
+        taglist = self._nestedtags[0] if depth_first is None else list(flatten(self._nestedtags[0]))
+        self._nestedtags = [[]]
+        return taglist
+
+    def updatepos(self, i, j):
+        offset = self._offset = super().updatepos(i, j)
+        return offset
+
+    def handle_starttag(self, tag, attrs):
+        try:
+            # we use internal variable for performance reasons
+            tag_text = getattr(self, '_HTMLParser__starttag_text')
+        except AttributeError:
+            tag_text = HTMLTagParser.ANY_TAG_REGEX.match(self.rawdata[self._offset:]).group()
+
+        tag_obj = tag
+        tag_is_open = not (tag_text.endswith('/>') or tag in self.VOID_TAGS)
+        if self.predicate(tag, attrs):
+            tag_obj = self.Tag(tag, string=self.rawdata, attrs=attrs)
+            tag_obj.openrange(self._offset, len(tag_text))
+            nesting = [tag_obj]
+            self._nestedtags[-1].append(nesting)
+            if tag_is_open:
+                self._nestedtags.append(nesting)
+            else:
+                self.callback(tag_obj)
+        if tag_is_open:
+            self.tagstack.appendleft(tag_obj)
+
+    handle_startendtag = handle_starttag
+
+    def handle_endtag(self, tag):
+        if '<' in tag:
+            if self.STRICT:
+                raise compat_HTMLParseError(f'malformed closing tag {tag!r}')
+            tag = tag[:tag.index('<')]
+
+        try:
+            idx = self.tagstack.index(tag)
+            if self.STRICT and idx:
+                open_tags = ''.join(f'</{tag}>' for tag in itertools.islice(self.tagstack, idx))
+                raise compat_HTMLParseError(
+                    f'malnested closing tag {tag!r}, expected after {open_tags!r}')
+            tag_obj = self.tagstack[idx]
+            self.tagstack.remove(tag)
+            if isinstance(tag_obj, self.Tag):
+                tag_obj.closerange(slice(self._offset, self.rawdata.find('>', self._offset) + 1))
+                self._nestedtags.pop()
+                self.callback(tag_obj)
+        except ValueError as exc:
+            if isinstance(exc, compat_HTMLParseError):
+                raise
+            if self.STRICT:
+                raise compat_HTMLParseError(f'stray closing tag {tag!r}') from exc
+
+
+class MatchingElementParser(HTMLTagParser):
+    """ optimized version of HTMLTagParser
+    """
+    def __init__(self, matchfunc):
+        super().__init__()
+        self.matchfunc = matchfunc
+        self.found_none = True
+
+    def reset(self):
+        super().reset()
+        self.found_none = True
+
+    def callback(self, tag_obj):
+        raise self.AbortException()
+
+    def predicate(self, tag, attrs):
+        if self.found_none and self.matchfunc(tag, attrs):
+            self.found_none = False
+            return True
+        return False
+
+    @staticmethod
+    def class_value_regex(class_name):
+        return rf'[\w\s\-]*(?<![\w\-]){re.escape(class_name)}(?![\w\-])[\w\s\-]*'
+
+    @staticmethod
+    def matching_tag_regex(tag, attribute, value_regex, escape=True):
+        if isinstance(value_regex, re.Pattern):
+            value_regex = value_regex.pattern
+        elif escape:
+            value_regex = re.escape(value_regex)
+
+        return rf'''(?x)
+            <(?i:{tag})
+             (?:\s(?:[^>"'\\]|"[^"\\]*"|'[^'\\]*')*)?
+             \s{re.escape(attribute)}\s*=\s*(?P<_q>['"])(?-x:{value_regex})(?P=_q)
+            '''
+
+    @classmethod
+    def iter_tags(cls, regex, html, *, matchfunc):
+        ignored = HTMLIgnoreRanges(html)
+        parser = cls(matchfunc)
+        for match in re.finditer(regex, html):
+            if match.start() not in ignored:
+                yield from parser.taglist(html[match.start():], reset=True)
+
+    @classmethod
+    def tags_by_name(cls, tag, html):
+        def matchfunc(tag_str, _attrs):
+            return tag_str == tag
+
+        tag_regex = rf'''<\s*(?i:{re.escape(tag)})(?:\s(?:[^>"'\\]|"[^"\\]*"|'[^'\\]*')*)?>'''
+        yield from cls.iter_tags(tag_regex, html, matchfunc=matchfunc)
+
+    @classmethod
+    def tags_by_attribute(cls, attribute, value, html, *, tag=r'[\w:.-]+', escape_value=True):
+        def matchfunc(_tag_str, attrs):
+            return any(attr == attribute and re.fullmatch(value, value_str)
+                       for attr, value_str in attrs)
+
+        tag_regex = cls.matching_tag_regex(tag, attribute, value, escape_value)
+        yield from cls.iter_tags(tag_regex, html, matchfunc=matchfunc)
+
+    @classmethod
+    def extract_attributes(cls, html):
+        attr_dict = {}
+
+        def matchfunc(_tag, attrs):
+            attr_dict.update(attrs)
+            raise cls.AbortException()
+
+        with contextlib.suppress(cls.AbortException):
+            cls(matchfunc).feed(html)
+
+        return attr_dict
+
+    @classmethod
+    def get_elements_text_and_html_by_tag(cls, tag, html):
+        return [tag.text_and_html() for tag in cls.tags_by_name(tag, html)]
+
+    @classmethod
+    def get_element_text_and_html_by_tag(cls, tag, html):
+        tag = next(cls.tags_by_name(tag, html), None)
+        return tag and tag.text_and_html()
+
+    @classmethod
+    def get_elements_text_and_html_by_attribute(cls, *args, **kwargs):
+        return [tag.text_and_html() for tag in cls.tags_by_attribute(*args, **kwargs)]
+
+    @classmethod
+    def get_elements_by_attribute(cls, *args, **kwargs):
+        return [tag.text() for tag in cls.tags_by_attribute(*args, **kwargs)]
+
+    @classmethod
+    def get_elements_html_by_attribute(cls, *args, **kwargs):
+        return [tag.html() for tag in cls.tags_by_attribute(*args, **kwargs)]
+
+    @classmethod
+    def get_element_by_attribute(cls, *args, **kwargs):
+        tag = next(cls.tags_by_attribute(*args, **kwargs), None)
+        return tag and tag.text()
+
+    @classmethod
+    def get_element_html_by_attribute(cls, *args, **kwargs):
+        tag = next(cls.tags_by_attribute(*args, **kwargs), None)
+        return tag and tag.html()
+
+    @classmethod
+    def get_elements_by_class(cls, class_name, html):
+        value = cls.class_value_regex(class_name)
+        return [tag.text() for tag
+                in cls.tags_by_attribute('class', value, html, escape_value=False)]
+
+    @classmethod
+    def get_elements_html_by_class(cls, class_name, html):
+        value = cls.class_value_regex(class_name)
+        return [tag.html() for tag
+                in cls.tags_by_attribute('class', value, html, escape_value=False)]
+
+    @classmethod
+    def get_elements_text_and_html_by_class(cls, class_name, html):
+        value = cls.class_value_regex(class_name)
+        return [tag.text_and_html() for tag
+                in cls.tags_by_attribute('class', value, html, escape_value=False)]
+
+    @classmethod
+    def get_element_html_by_class(cls, class_name, html):
+        value = cls.class_value_regex(class_name)
+        tag = next(cls.tags_by_attribute('class', value, html, escape_value=False), None)
+        return tag and tag.html()
+
+    @classmethod
+    def get_element_by_class(cls, class_name, html):
+        value = cls.class_value_regex(class_name)
+        tag = next(cls.tags_by_attribute('class', value, html, escape_value=False), None)
+        return tag and tag.text()

yt_dlp/utils/_utils.py

@@ -408,17 +408,13 @@ class HTMLBreakOnClosingTagParser(html.parser.HTMLParser):
         pass
 
     def handle_starttag(self, tag, _):
-        self.tagstack.append(tag)
+        self.tagstack.appendleft(tag)
 
     def handle_endtag(self, tag):
         if not self.tagstack:
             raise compat_HTMLParseError('no tags in the stack')
-        while self.tagstack:
-            inner_tag = self.tagstack.pop()
-            if inner_tag == tag:
-                break
-        else:
-            raise compat_HTMLParseError(f'matching opening tag for closing {tag} tag not found')
+        with contextlib.suppress(ValueError):
+            self.tagstack.remove(tag)
         if not self.tagstack:
             raise self.HTMLBreakOnClosingTagException
 
@@ -452,6 +448,8 @@ def get_element_text_and_html_by_tag(tag, html):
             next_closing_tag_end = next_closing_tag_start + len(closing_tag)
             try:
                 parser.feed(html[offset:offset + next_closing_tag_end])
+                if tag not in parser.tagstack:
+                    raise HTMLBreakOnClosingTagParser.HTMLBreakOnClosingTagException()
                 offset += next_closing_tag_end
             except HTMLBreakOnClosingTagParser.HTMLBreakOnClosingTagException:
                 return html[content_start:offset + next_closing_tag_start], \