- RFC 3986 6.2.3 (always slash on empty path and non-empty host, not just when query string is present)
- Polish logo
- Get coverage up
- Switch off ctypes/socket for IP validation
- rebase method for path (prepends to path)
- sibling() should be maximal=False like child()
- make subencoding an exposed parameter
- unquote all unreserved characters (RFC 3986 2.3)
- lowercase scheme
- lowercase host (including lowercase ipv6 addresses)
- resolve path parts (RFC 3986 6.2.2)
- percent encoding (RFC 3986 2.1) to uppercase (add a
normalize_case=Falseto_percent_decode, with appropriate upward propagation)
- docstrings
- Parse with regular expression
- Parse/emit IPv6
- Add netloc detection logic
- Speed up percent encoding with urlutils approach
- More default ports
- resolve dots on (empty) click
- better error on URL constructor (single string argument leads to successful instantiation with invalid scheme)
- pct encode userinfo
__hash__- README
- CI
- Get Mark to fix the git history
- Python 2.6 support
- Full read the docs
- What's the deal with twisted.python.urlpath?
- What's the deal with sip.URL?
- Should we take the alternative IPv6 literal approach and just interpret anything with colons in the host as requiring square brackets? Or should we try to validate with a regex?
- Do we need a separate .normalize()? Resolves all dots, lowercases scheme and host.
- Is whitespace really allowed in IRIs?
- Can we drop the ':' when password isn't displayed?
- Why is the constructor's default scheme "http" when host is not None?
- What should the default port of the "s3" scheme be? premature?
- Should _url be lifted?