Skip to content

processone/stringprep

Fast Stringprep implementation for Erlang / Elixir

CI Coverage Status Hex version

Stringprep is a framework for preparing Unicode test strings in order to increase the likelihood that string input and string comparison work.

The principle are defined in RFC-3454: Preparation of Internationalized Strings .

This library is leverage Erlang native NIF mechanism to provide extremely fast and efficient processing.

The library includes support for several Stringprep profiles used in XMPP protocole like:

  • Nodeprep
  • Nameprep
  • Resourceprep

For those profiles, the rules are applied according to RFC6122. The various functions perform check on the allowed / forbidden chars for a given profile or prevent combining left-to-right and right-to-left chars.

It the binary string passed to a function of the API is valid, it will return its normalized version according to Stringprep profile. Otherwise, if the binary string is invalid, for example because it contains invalid chars, the function will return error.

The library is heavily used in XMPP string processing. However, the library is more generally useful in code that need to manipulate and compare Unicode strings.

Building

Fast Stringprep processing tool can be build as follow:

./configure && make

Configure script recognizes one flag - pass --enable-gcov to enable gcov coverage reporting.

It is a rebar-compatible OTP application. Alternatively, you can build it with rebar:

rebar get-deps compile

Usage

You can start the application with the command:

$ erl -pa ebin/
Erlang/OTP 18 [erts-7.1] [source] [64-bit] [smp:4:4] [async-threads:10] [hipe] [kernel-poll:false] [dtrace]

Eshell V7.1  (abort with ^G)
1> application:start(p1_stringprep).

You can then call any of the stringprep function to apply a profile:

stringprep:nodeprep(<<>>).
stringprep:nameprep(<<>>).
stringprep:resourceprep(<<>>).
stringprep:tolower(<<>>)).

Development

Test

Unit test

You can run eunit test with the command:

$ make test