url-parser-combinator

A proper url parser and combinator that works with eulalie.

Usage

const parser = require('url-parser-combinator')

var goodUrl = parser.parse('http://localhost:80/foo/bar?key1=value1&key2=value2')
// 'http://localhost:80/foo/bar?key1=value1&key2=value2'

var badUrl = parser.parse('//foo//bar')
// ''

Or you can feed the parser.URL combinator into eulalie to parse a url.

Implementation

The parser is an implementation of the BNF form for URLs based on the RFC 1808 spec.

All of the BNF rules in the RFC spec are provided to you as combinators, which means that you can use this library to get partial aspects of a url, such as relativeURL, params, or query, or create your own custom combinator from the components. For example, the library accepts relative and absolute urls, but you can use the provided absoluteURL combinator to make a parser that consumes only absolute urls.

You can also use the URL combinator as part of another parser. For example, in the eulalie readme there is simple parser for an HTTP request. You could use this combinator in conjunction with this example to make the parser validate that the path is a proper url rather than just an arbitrary string:

const URL = require('url-parser-combinator').URL

const parser = p.seq(function*() {
  const {value: method} = yield p.many1(p.upper);
  yield p.spaces1;
  const {value: path} = yield URL;  // *Parse a URL*
  yield p.spaces1;
  yield p.string("HTTP/");
  const {value: version} = yield p.seq(function*() {
    const {value: left} = yield p.many1(p.digit);
    yield p.char(".");
    const {value: right} = yield p.many1(p.digit);
    return `${left}.${right}`;
  });
  return {method, path, version};
});

License

Licensed under the MIT license.