IdnaJS
Download source code: ZIP file
A JavaScript implementation of IDNA 2008 and Unicode normalization.
Installation
Run the following code, assuming you have npm
installed:
npm install peteroupc/IdnaJS
Usage
var idna=require("idnajs"), Normalizer=idna.Normalizer, Idna=idna.Idna;
Normalizer.IsNormalized(string[, form])
Returns whether the given string is in a Unicode normalization form.
The form can be Normalization.NFC
, Normalization.NFD
, Normalization.NFKC
,
or Normalization.NFKD
. If not given, NFC is assumed. Returns false if the string
is null or contains an unpaired surrogate code point.
Normalizer.Normalize(string[, form])
Converts the given string is in a Unicode normalization form.
The form can be Normalization.NFC
, Normalization.NFD
, Normalization.NFKC
,
or Normalization.NFKD
. If not given, NFC is assumed. Throws an error if the
string is null. Replaces unpaired surrogate code points in the string with
U+FFFD (the replacement character).
Idna.EncodeDomainName(string)
Tries to encode each label of the given domain name into Punycode. Labels where this is not possible remain unchanged. The labels are not checked for validity beyond the Punycode checks.
Idna.IsValidDomainName(string[, lookupRules])
Checks whether a domain name is valid under IDNA 2008 (the second version
of Internationalized Domain Names in Applications). If lookupRules
is true,
uses somewhat more relaxed rules to be followed before looking up a domain
name in the DNS; if lookupRules
is omitted, the value false
is assumed.
Source Code
Source code is available in the project page.
About
Written by Peter O.
See the LICENSE file for license information.