Chris Mayo

A `?` is a terminating character for a path: https://tools.ietf.org/html/rfc3986#section-3.3 If desired it should be percent encoded to `%3F`. Some quick testing suggests that works in LinkChecker. So I guess...

My recipe for installing master on Windows 10: Install Python 3 from the Windows Store https://docs.microsoft.com/en-us/windows/python/beginners As mentioned in that page, install Git (accepting the defaults) from: https://git-scm.com/download/win Now do:...

Best still the Microsoft page actually says "we recommend installing and using Python via the Windows Subsystem for Linux" (for web development).

v10.1.0 creates an entry point, and on Windows pip does create a linkchecker.exe (albeit in the Scripts directory). I have updated running the tests on Windows using GitHub Actions in...

A [pyoxidizer.bzl](https://github.com/linkchecker/linkchecker/blob/master/pyoxidizer.bzl) is now available, in case that is useful. https://gregoryszorc.com/docs/pyoxidizer/main/

The failures are: ``` tests\checker\test_file.py:85: in test_markdown self.file_test("file.markdown", confargs=confargs) tests\checker\__init__.py:262: in file_test self.fail(msg) tests\checker\test_loginurl.py:56: in test_loginurl self.assertEqual(cookies["login"], "test_user") .tox\py\lib\site-packages\requests\cookies.py:328: in __getitem__ return self._find_no_duplicates(name) .tox\py\lib\site-packages\requests\cookies.py:399: in _find_no_duplicates raise KeyError('name=%r, domain=%r, path=%r'...

A `:` is OK in a path, ...except the first segment of a relative link: https://tools.ietf.org/html/rfc3986#section-3.3

New warning `url-content-type-unparseable` proposed in PR #638.

My guess would be [pikepdf](https://pikepdf.readthedocs.io/en/latest/index.html) might do a better job. That theory might be able to be tested with `qpdf` (that pikepdf uses). If it turns out to be better...

This might be enough though: ```diff --- a/linkcheck/plugins/parsepdf.py +++ b/linkcheck/plugins/parsepdf.py @@ -43,7 +43,7 @@ def search_url(obj, url_data, pageno, seen_objs): obj = obj.resolve() if isinstance(obj, dict): for key, value in obj.items():...