diff --git a/linkcheck/htmlutil/linkparse.py b/linkcheck/htmlutil/linkparse.py
index cfbe5ab3..d8b586aa 100644
--- a/linkcheck/htmlutil/linkparse.py
+++ b/linkcheck/htmlutil/linkparse.py
@@ -18,6 +18,8 @@ Find link tags in HTML text.
 """
 
 import re
+
+from .srcsetparse import parse_srcset
 from .. import strformat, log, LOG_CHECK, url as urlutil
 
 unquote = strformat.unquote
@@ -200,19 +202,9 @@ class LinkFinder:
         elif attr == 'archive':
             for url in value.split(','):
                 self.found_url(url, name, base, lineno, column)
-        elif attr == 'srcset' and not value.startswith('data:'):
-            for img_candidate in value.split(','):
-                try:
-                    url = img_candidate.split()[0]
-                except IndexError:
-                    log.debug(
-                        LOG_CHECK,
-                        _("trailing comma in line: "
-                          "%(line)s srcset attribute: %(value)s")
-                        % {"line": lineno, "value": value}
-                    )
-                else:
-                    self.found_url(url, name, base, lineno, column)
+        elif attr == 'srcset':
+            for url in parse_srcset(value):
+                self.found_url(url, name, base, lineno, column)
         else:
             self.found_url(value, name, base, lineno, column)
 
diff --git a/linkcheck/htmlutil/srcsetparse.py b/linkcheck/htmlutil/srcsetparse.py
new file mode 100644
index 00000000..9bbff15a
--- /dev/null
+++ b/linkcheck/htmlutil/srcsetparse.py
@@ -0,0 +1,91 @@
+# Copyright (C) 2022 Stefan Fisk
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program; if not, write to the Free Software Foundation, Inc.,
+# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+"""
+srcset attribute parser
+"""
+
+
+_TAB = '\u0009'
+_LF = '\u000A'
+_FF = '\u000C'
+_CR = '\u000D'
+_SPACE = '\u0020'
+_COMMA = '\u002C'
+_LEFT_PARENTHESIS = '\u0028'
+_RIGHT_PARENTHESIS = '\u0029'
+
+
+_WHITESPACE = {_TAB,  _LF,  _FF,  _CR,  _SPACE}
+_WHITESPACE_OR_COMMA = _WHITESPACE | {_COMMA}
+
+
+def parse_srcset(input):
+    """
+    Parse HTML srcset
+
+    Based on WhatWG HTML standard § 4.8.4.3.10 Parsing a srcset attribute,
+    but does not parse or validate descriptors.
+
+    https://html.spec.whatwg.org/multipage/images.html#parse-a-srcset-attribute
+    """
+
+    input_end = len(input)
+    position = 0
+    urls = []
+
+    while position < input_end:
+        # 4. Splitting loop: Collect a sequence of code points that are ASCII
+        # whitespace or U+002C COMMA characters from input given position.
+        while position < input_end and input[position] in _WHITESPACE_OR_COMMA:
+            position += 1
+
+        # 5. If position is past the end of input, return candidates.
+        if position >= input_end:
+            return urls
+
+        # 6. Collect a sequence of code points that are not ASCII
+        # whitespace from input given position, and let that be url.
+        url_start = position
+        while position < input_end and input[position] not in _WHITESPACE:
+            position += 1
+        url_end = position
+
+        # 8, If url ends with U+002C (,), then:
+        if input[url_end - 1] == _COMMA:
+            # Remove all trailing U+002C COMMA characters from url.
+            while url_end > url_start and input[url_end - 1] == _COMMA:
+                url_end -= 1
+        else:
+            # This is a shortened version of 1–4 that simply skips the
+            # descriptors
+            while position < input_end:
+                if input[position] == _LEFT_PARENTHESIS:
+                    # Skip until first closing parenthesis
+                    while (position < input_end and input[position] !=
+                            _RIGHT_PARENTHESIS):
+                        position += 1
+                elif input[position] == _COMMA:
+                    break
+
+                position += 1
+
+        # 9-15 is parsing and validation of the descriptors, which we ignore
+
+        # If we found an URL
+        if url_end > url_start:
+            urls.append(input[url_start:url_end])
+
+    return urls
diff --git a/tests/test_srcsetparse.py b/tests/test_srcsetparse.py
new file mode 100644
index 00000000..e1f56720
--- /dev/null
+++ b/tests/test_srcsetparse.py
@@ -0,0 +1,51 @@
+# Copyright (C) 2022 Stefan fisk
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program; if not, write to the Free Software Foundation, Inc.,
+# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+"""
+Test srcset attribute parsing.
+"""
+
+import unittest
+from linkcheck.htmlutil.srcsetparse import parse_srcset
+
+from parameterized import parameterized
+
+
+# list of tuples
+# (<input>, <expected parsed URLs>)
+parsetests = [
+    ('', []),
+    ('   ', []),
+    (',', []),
+    ('\t\n ,,\t\n,,   \t\n', []),
+    ('foo', ['foo']),
+    ('foo,bar, ,foo, bar', ['foo,bar', 'foo', 'bar']),
+    ('https://example.com/1 foo, https://example.com/2 bar',
+        ['https://example.com/1', 'https://example.com/2']),
+    ('   foo   ', ['foo']),
+    (',,,foo,,,', ['foo']),
+    (',foo,bar,baz,', ['foo,bar,baz']),
+    ('foo bar baz', ['foo']),
+    ('foo, bar baz', ['foo', 'bar']),
+    ('foo/1 bar, foo/2', ['foo/1', 'foo/2']),
+    ('foo/1 (foo/2)', ['foo/1']),
+    ('foo/1 (((, foo/2', ['foo/1']),
+]
+
+
+class TestSrcsetParsing(unittest.TestCase):
+    @parameterized.expand(parsetests)
+    def test_parse(self, _in, _urls):
+        self.assertEqual(parse_srcset(_in), _urls)