url_decode: Best-effort host rewriting in a URL-shaped string (ASCII...

In punycoder: Unicode and Punycode Domain Name Processing

Best-effort host rewriting in a URL-shaped string (ASCII punycode to Unicode)

Description

Locates the host portion of a URL-shaped string with a hand-rolled splitter, decodes that host from ASCII punycode to Unicode, and substitutes it back, leaving the rest of the string untouched.

Usage

url_decode(url, strict = getOption("punycoder.strict", TRUE))

Arguments

url	Character vector of URL-shaped strings with ASCII punycode hosts
strict	Logical; whether to apply strict validation. Defaults to getOption("punycoder.strict", TRUE).

Details

Like url_encode(), this is best-effort host extraction and rewriting, not URL parsing or canonicalization, and is not RFC 3986 / WHATWG URL conformant (no percent encoding/decoding, scheme/port/path semantics, full IPv6, or serialization). Those concerns live upstack in rurl.

Value

A character vector the same length as url, with each element containing the URL with its host portion decoded to Unicode. Only the domain component is transformed; scheme, path, query, and fragment are preserved. Elements corresponding to NA inputs are NA_character_.

Deprecated

This function is deprecated and slated for removal in a future release. For URL parsing and canonicalization use a dedicated URL package (e.g. rurl); for host-only decoding pass the host alone to puny_decode().

See Also

url_encode for the reverse operation, puny_decode for domain-only decoding, parse_url for URL component extraction.

Examples

# Basic URL decoding
url_decode("https://xn--caf-dma.example.com/path")
url_decode("https://xn--80adxhks.xn--p1ai/page")

# Vectorized URL decoding
ascii_urls <- c(
  "https://xn--caf-dma.com/menu",
  "https://xn--1qqw23a.xn--55qx5d/info"
)
url_decode(ascii_urls)

punycoder documentation built on July 20, 2026, 1:07 a.m.