PATH_EMPTY regex does not actually match empty path

[milo-minderbinder]

rfc3986/src/rfc3986/abnf_regexp.py

Line 153 in 1640734

PATH_EMPTY = "^$"

The PATH_EMPTY regex won't ever match anything when embedded within another regex, as it is in several places like here:

rfc3986/src/rfc3986/abnf_regexp.py

Lines 184 to 190 in 1640734

	HIER_PART_RE = "(//{}{}|{}|{}|{})".format(
	COMPONENT_PATTERN_DICT["authority"],
	PATH_ABEMPTY,
	PATH_ABSOLUTE,
	PATH_ROOTLESS,
	PATH_EMPTY,
	)

This causes the URIReference.is_absolute() method to return an incorrect result for a URI with a scheme but no path. For example, http: (or even http:?q=v) is an absolute URI according to section-4.3, but URIReference.from_string('http:').is_absolute() returns False.

There are likely other places where this causes intended matches to fail, but I have not tried to investigate further. It's a little odd looking, but a fix that would probably work is just to change the value of PATH_EMPTY to an empty string.

[handrews]

While http and https URIs MUST have an authority component, the about URI scheme does not use an authority, allows an empty path, and allows a query string:

about-uri = "about:" about-token [ about-query ] [ about-fragment ]
     about-token = *pchar
     about-query = "?" query
     about-fragment = "#" fragment
     pchar     = <as specified in [RFC 3986, Appendix A](https://datatracker.ietf.org/doc/html/rfc3986#appendix-A)>
     query     = <as specified in [RFC 3986, Appendix A](https://datatracker.ietf.org/doc/html/rfc3986#appendix-A)>
     fragment  = <as specified in [RFC 3986, Appendix A](https://datatracker.ietf.org/doc/html/rfc3986#appendix-A)>

So this bug can also be reproduced with a URI scheme that allows these forms, which I think would make these good cases to add to the test suite:

>>> rfc3986.__version__
'2.0.0'
>>> rfc3986.uri_reference('about:').is_absolute()
False
>>> rfc3986.uri_reference('about:?foo=bar').is_absolute()
False

[handrews]

EDIT: This was wrong, see below

Hmm... on further thought, an empty path is only allowed if an authority is present, and the authority cannot be empty (because the hostname cannot be empty):

      URI         = scheme ":" hier-part [ "?" query ] [ "#" fragment ]

      hier-part   = "//" authority path-abempty
                  / path-absolute
                  / path-rootless
                  / path-empty

path-abempty being on the line with "//" authority like that means that they go together, while the other lines below are alternatives to that whole first line.

So I now think this is not a bug, or if there is a bug the bug is that about: and about:?foo=bar should not parse. The about URI scheme's syntax ABNF definitely allows for an empty about-token (which functions as the path), but I think RFC3986's disallowing of empty paths without an authority takes precedence.

I'll leave this for a project member to decide whether to close or not, but I'm going to close my PR for now. It can be re-opened if needed, of course.

[handrews]

Oh wait there's also a path-empty that I forgot. I've clearly been staring at ABNF for too long...

Re-opening my PR.