Merge c49aa772cc into 74e90dd9b8

test/test_utils.py

@@ -1787,6 +1787,10 @@ Line 1
         <div itemprop="author" itemscope>foo</div>
     '''
 
+    GET_ELEMENT_BY_ATTRIBUTE_TEST_STRING_UPPERCASE = '''
+        <DIV itemprop="author" itemscope>foo</DIV>
+    '''
+
     def test_get_element_by_attribute(self):
         html = self.GET_ELEMENT_BY_CLASS_TEST_STRING
 
@@ -1798,6 +1802,10 @@ Line 1
 
         self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo')
 
+        html = self.GET_ELEMENT_BY_ATTRIBUTE_TEST_STRING_UPPERCASE
+
+        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
 
@@ -1856,7 +1864,7 @@ Line 1
     random text lorem ipsum</p>
     <div>
         this should be returned
-        <span>this should also be returned</span>
+        <SPAN>this should also be returned</SPAN>
         <div>
             this should also be returned
         </div>
@@ -1878,6 +1886,10 @@ Line 1
         self.assertEqual(
             get_element_text_and_html_by_tag('span', html),
             (self.GET_ELEMENT_BY_TAG_RES_INNERSPAN_TEXT, self.GET_ELEMENT_BY_TAG_RES_INNERSPAN_HTML))
+        self.assertEqual(
+            get_element_text_and_html_by_tag('SPAN', html),
+            (self.GET_ELEMENT_BY_TAG_RES_INNERSPAN_TEXT, self.GET_ELEMENT_BY_TAG_RES_INNERSPAN_HTML))
+
         self.assertRaises(compat_HTMLParseError, get_element_text_and_html_by_tag, 'article', html)
 
     def test_iri_to_uri(self):

yt_dlp/utils/_utils.py

@@ -432,10 +432,14 @@ def get_element_text_and_html_by_tag(tag, html):
     return its' content (text) and the whole element (html)
     """
     def find_or_raise(haystack, needle, exc):
-        try:
+        with contextlib.suppress(ValueError):
             return haystack.index(needle)
-        except ValueError:
-            raise exc
+
+        with contextlib.suppress(ValueError):
+            return haystack.index(needle.upper())
+
+        raise exc
+
     closing_tag = f'</{tag}>'
     whole_start = find_or_raise(
         html, f'<{tag}', compat_HTMLParseError(f'opening {tag} tag not found'))
@@ -444,7 +448,7 @@ def get_element_text_and_html_by_tag(tag, html):
     content_start += whole_start + 1
     with HTMLBreakOnClosingTagParser() as parser:
         parser.feed(html[whole_start:content_start])
-        if not parser.tagstack or parser.tagstack[0] != tag:
+        if not parser.tagstack or parser.tagstack[0] != tag.lower():
             raise compat_HTMLParseError(f'parser did not match opening {tag} tag')
         offset = content_start
         while offset < len(html):