serp-url

URL parsing and encoding for Serpentine. Pure Serpentine ([P]) — the same source dual-runs under CPython.

from serp_url import url_parse, quote, urlencode, parse_query

u = url_parse("https://example.com:8443/a/b?x=1&y=two#frag")
print(u.scheme, u.host, u.port, u.path, u.query, u.fragment)
print(quote("a b/c?"))                 # a%20b/c%3F
print(parse_query("x=1&name=J%C3%BCrgen"))

API

• Url — dataclass: scheme, host, port (-1 when absent), path, query, fragment
• url_parse(url) -> Url — handles IPv6 bracket hosts; no userinfo support
• default_port(scheme) -> int, effective_port(u) -> int
• quote(s) / unquote(s) — RFC 3986 percent-encoding of UTF-8 ('/' kept safe in quote)
• quote_plus(s) / unquote_plus(s) — form encoding (space ↔ '+')
• urlencode(params: dict[str, str]) -> str
• parse_query(q) -> dict[str, str] — last value wins for duplicate keys