RFC 3986 URL Parser

UrlParser is PHP library that provides a RFC 3986 compliant URL parser and a PSR-7 compatible URI component. The purpose of this library is to provide a parser that accurately implements the RFC specification unlike the built in function parse_url(), which differs from the specification in some subtle ways.

This library has two main purposes. The first to provide information from the parsed URLs. To achieve this, the library implements the standard URI handling interface from the PSR-7 and also provides additional methods that make it easier to retrieve commonly used information from the URLs. The second purpose is to also permit the modification of said URLs using the interface from the PSR-7 standard in addition to few extra methods that make some tasks more straightforward.

While this library is mainly intended for parsing URLs, the parsing is simply based on the generic URI syntax. Thus, it is possible to use this library to validate and parse any other types of URIs against the generic syntax. The library does not perform any scheme specific validation for the URLs.

In addition to the default RFC 3986 compliant mode, the library also offers options that allow parsing of URLs that contain UTF-8 characters in different components of the URL while converting them to the appropriate percent encoded and IDN ascii formats.

The API documentation is available at: http://kit.riimu.net/api/urlparser/

Requirements

• The minimum supported PHP version is 5.6
• The library depends on the following external PHP libraries:
  • psr/http-message (^1.0)
• The library depends on the following PHP Extensions
  • intl (only required IDN support)

Installation

Installation with Composer

The easiest way to install this library is to use Composer to handle your dependencies. In order to install this library via Composer, simply follow these two steps:

1. Acquire the composer.phar by running the Composer Command-line installation in your project root.

2. Once you have run the installation script, you should have the composer.phar file in you project root and you can run the following command:

   php composer.phar require "riimu/kit-urlparser:^2.1"
   

After installing this library via Composer, you can load the library by including the vendor/autoload.php file that was generated by Composer during the installation.

Adding the library as a dependency

If you are already familiar with how to use Composer, you may alternatively add the library as a dependency by adding the following composer.json file to your project and running the composer install command:

{
    "require": {
        "riimu/kit-urlparser": "^2.1"
    }
}

Manual installation

If you do not wish to use Composer to load the library, you may also download the library manually by downloading the latest release and extracting the src folder to your project. You may then include the provided src/autoload.php file to load the library classes.

Please note that using Composer will also automatically download the other required PHP libraries. If you install this library manually, you will also need to make those other required libraries available.

Usage

Using this library is relatively straightforward. The library provides a URL parsing class UriParser and an immutable value object class Uri that represents the URL. To parse an URL, you could simply provide the URL as a string to the parse() method in UriParser which returns an instance of Uri that has been generated from the parsed URL.

For example:

<?php

require 'vendor/autoload.php';

$parser = new \Riimu\Kit\UrlParser\UriParser();
$uri = $parser->parse('http://www.example.com');

echo $uri->getHost(); // Outputs 'www.example.com'

Alternatively, you can just skip using the UriParser completely and simply provide the URL as a constructor parameter to the Uri:

<?php

require 'vendor/autoload.php';
$uri = new \Riimu\Kit\UrlParser\Uri('http://www.example.com');
echo $uri->getHost(); // Outputs 'www.example.com'

The main difference between using the parse() method and the constructor is that the parse() method will return a null if the provided URL is not a valid url, while the constructor will throw an InvalidArgumentException.

To retrieve different types of information from the URL, the Uri class provides various different methods to help you. Here is a simple example as an overview of the different available methods:

<?php

require 'vendor/autoload.php';

$parser = new \Riimu\Kit\UrlParser\UriParser();
$uri = $parser->parse('http://jane:pass123@www.example.com:8080/site/index.php?action=login&prev=index#form');

echo $uri->getScheme() . PHP_EOL;         // outputs: http
echo $uri->getUsername() . PHP_EOL;       // outputs: jane
echo $uri->getPassword() . PHP_EOL;       // outputs: pass123
echo $uri->getHost() . PHP_EOL;           // outputs: www.example.com
echo $uri->getTopLevelDomain() . PHP_EOL; // outputs: com
echo $uri->getPort() . PHP_EOL;           // outputs: 8080
echo $uri->getStandardPort() . PHP_EOL;   // outputs: 80
echo $uri->getPath() . PHP_EOL;           // outputs: /site/index.php
echo $uri->getPathExtension() . PHP_EOL;  // outputs: php
echo $uri->getQuery() . PHP_EOL;          // outputs: action=login&prev=index
echo $uri->getFragment() . PHP_EOL;       // outputs: form

print_r($uri->getPathSegments());    // [0 => 'site', 1 => 'index.php']
print_r($uri->getQueryParameters()); // ['action' => 'login', 'prev' => 'index']

The Uri component also provides various methods for modifying the URL, which allows you to construct new URLs from separate components or modify existing ones. Note that the Uri component is an immutable value object, which means that each of the modifying methods return a new Uri instance instead of modifying the existing one. Here is a simple example of constructing an URL from it's components:

<?php

require 'vendor/autoload.php';

$uri = (new \Riimu\Kit\UrlParser\Uri())
    ->withScheme('http')
    ->withUserInfo('jane', 'pass123')
    ->withHost('www.example.com')
    ->withPort(8080)
    ->withPath('/site/index.php')
    ->withQueryParameters(['action' => 'login', 'prev' => 'index'])
    ->withFragment('form');

// Outputs: http://jane:pass123@www.example.com:8080/site/index.php?action=login&prev=index#form
echo $uri;

As can be seen from the previous example, the Uri component also provides a __toString() method that provides the URL as a string.

Retrieving Information

Here is the list of methods that the Uri component provides for retrieving information from the URL:

• getScheme() returns the scheme from the URL or an empty string if the URL has no scheme.

• getAuthority() returns the component from the URL that consists of the username, password, hostname and port in the format user-info@hostname:port

• getUserInfo() returns the component from the URL that contains the username and password separated by a colon.

• getUsername() returns the decoded username from the URL or an empty string if there is no username present in the URL.

• getPassword() returns the decoded password from the URL or an empty string if there is no password present in the URL.

• getHost() return the hostname from the URL or an empty string if the URL has no host.

• getIpAddress() returns the IP address from the host, if the host is an IP address. Otherwise this method will return null. If an IPv6 address was provided, the address is returned without the surrounding braces.

• getTopLevelDomain() returns the top level domain from the host. If there is no host or the host is an IP address, an empty string will be returned instead.

• getPort() returns the port from the URL or a null if there is no port present in the url. This method will also return a null if the port is the standard port for the current scheme (e.g. 80 for http).

• getStandardPort() returns the standard port for the current scheme. If there is no scheme or the standard port for the scheme is not known, a null will be returned instead.

• getPath() returns the path from the URL or an empty string if the URL has no path.

• getPathSegments() returns an array of decoded path segments (i.e. the path split by each forward slash). Empty path segments are discarded and not included in the returned array.

• getPathExtension() returns the file extension from the path or an empty string if the URL has no path.

• getQuery() returns the query string from the URL or an empty string if the URL has no query string.

• getQueryParameters() parses the query string from the URL using the parse_str() function and returns the array of parsed values.

• getFragment() returns the fragment from the URL or an empty string if the URL has no fragment.

• __toString() returns the URL as a string.

Modifying the URL

The Uri component provides various methods that can be used to modify URLs and construct new ones. Note that since the Uri class is an immutable value object, each method returns a new instance of Uri rather than modifying the existing one.

• withScheme($scheme) returns a new instance with the given scheme. An empty scheme can be used to remove the scheme from the URL. Note that any provided scheme is normalized to lowercase.

• withUserInfo($user, $password = null) returns a new instance with the given username and password. Note that the password is ignored unless an username is provided. Empty username can be used to remove the username and password from the URL. Any character that cannot be inserted in the URL by itself will be percent encoded.

• withHost($host) returns a new instance with the given host. An empty host can be used to remove the host from the URL. Note that this method does not accept international domain names. Note that this method will also normalize the host to lowercase.

• withPort($port) returns a new instance with the given port. A null can be used to remove the port from the URL.

• withPath($path) returns a new instance with the given path. An empty path can be used to remove the path from the URL. Note that any character that is not a valid path character will be percent encoded in the URL. Existing percent encoded characters will not be double encoded, however.

• withPathSegments(array $segments) returns a new instance with the path constructed from the array of path segments. All invalid path characters in the segments will be percent encoded, including the forward slash and existing percent encoded characters.

• withQuery($query) returns a new instance with the given query string. An empty query string can be used to remove the path from the URL. Note that any character that is not a valid query string character will be percent encoded in the URL. Existing percent encoded characters will not be double encoded, however.

• withQueryParameters(array $parameters) returns a new instance with the query string constructed from the provided parameters using the http_build_query() function. All invalid query string characters in the parameters will be percent encoded, including the ampersand, equal sign and existing percent encoded characters.

• withFragment($fragment) returns a new instance with the given fragment. An empty string can be used to remove the fragment from the URL. Note that any character that is not a valid fragment character will be percent encoded in the URL. Existing percent encoded characters will not be double encoded, however.

UTF-8 and International Domains Names

By default, this library provides a parser that is RFC 3986 compliant. The RFC specification does not permit the use of UTF-8 characters in the domain name or any other parts of the URL. The correct representation for these in the URL is to use the an IDN standard for domain names and percent encoding the UTF-8 characters in other parts.

However, to help you deal with UTF-8 encoded characters, many of the methods in the Uri component will automatically percent encode any characters that cannot be inserted in the URL on their own, including UTF-8 characters. Due to complexities involved, however, the withHost() method does not allow UTF-8 encoded characters.

By default, the parser also does not parse any URLs that include UTF-8 encoded characters because that would be against the RFC specification. However, the parser does provide two additional parsing modes that allows these characters whenever possible.

If you wish to parse URLs that may contain UTF-8 characters in the user information (i.e. the username or password), path, query or fragment components of the URL, you can simply use the UTF-8 parsing mode. For example:

<?php

require 'vendor/autoload.php';

$parser = new \Riimu\Kit\UrlParser\UriParser();
$parser->setMode(\Riimu\Kit\UrlParser\UriParser::MODE_UTF8);

$uri = $parser->parse('http://www.example.com/föö/bär.html');
echo $uri->getPath(); // Outputs: /f%C3%B6%C3%B6/b%C3%A4r.html

UTF-8 characters in the domain name, however, are a bit more complex issue. The parser, however, does provide a rudimentary support for parsing these domain names using the IDNA mode. For example:

<?php

require 'vendor/autoload.php';

$parser = new \Riimu\Kit\UrlParser\UriParser();
$parser->setMode(\Riimu\Kit\UrlParser\UriParser::MODE_IDNA);

$uri = $parser->parse('http://www.fööbär.com');
echo $uri->getHost(); // Outputs: www.xn--fbr-rla2ga.com

Note that using this parsing mode requires the PHP extension intl to be enabled. The appropriate parsing mode can also be provided to the constructor of the Uri component using the second constructor parameter.

While support for parsing these UTF-8 characters is available, this library does not provide any methods for the reverse operations since the purpose of this library is to deal with RFC 3986 compliant URIs.

URL Normalization

Due to the fact that the RFC 3986 specification defines some URLs as equivalent despite having some slight differences, this library does some minimal normalization to the provided values. You may encounter these instances when, for example, parsing URLs provided by users. The most notable normalizations you may encounter are as follows:

• The scheme and host components are considered case insensitive. Thus, these components will always be normalized to lower case.
• The port number will not be included in the strings returned by getAuthority() and __toString() if the port is the standard port for the current scheme.
• Percent encodings are treated in case insensitive manner. Thus, this library will normalize the hexadecimal characters to upper case.
• The number of forward slashes in the beginning of the path in the string returned by __toString() may change depending on whether the URL has an authority component or not.
• Percent encoded characters in parsed and generated URIs in the userinfo component may differ due to the fact that the UriParser works with the PSR-7 specification which does not provide a way to provide encoded username or password.

Credits

This library is Copyright (c) 2013-2022 Riikka Kalliomäki.

See LICENSE for license and copying information.