<?php

/**
 * Parses a URI into the components and fragment identifier as specified
 * by RFC 3986.
 */
class HTMLPurifier_URIParser
{
    /** Instance of HTMLPurifier_PercentEncoder to do normalization with. */
    protected $percentEncoder;

    public function __construct()
    {
        $this->percentEncoder = new HTMLPurifier_PercentEncoder();
    }

    /**
     * Parses a URI.
     * @param $uri string URI to parse
     * @return HTMLPurifier_URI representation of URI. This representation has
     *         not been validated yet and may not conform to RFC.
     */
    public function parse($uri)
    {
        $uri = $this->percentEncoder->normalize($uri);

        // Regexp is as per Appendix B.
        // Note that ["<>] are an addition to the RFC's recommended
        // characters, because they represent external delimiters.
        $r_URI = '!'.
            '(([a-zA-Z0-9\.\+\-]+):)?'. // 2. Scheme
            '(//([^/?#"<>]*))?'. // 4. Authority
            '([^?#"<>]*)'.       // 5. Path
            '(\?([^#"<>]*))?'.   // 7. Query
            '(#([^"<>]*))?'.     // 8. Fragment
            '!';

        $matches = array();
        $result = preg_match($r_URI, $uri, $matches);

        if (!$result) return false; // *really* invalid URI

        // separate out parts
        $scheme     = !empty($matches[1]) ? $matches[2] : null;
        $authority  = !empty($matches[3]) ? $matches[4] : null;
        $path       = $matches[5]; // always present, can be empty
        $query      = !empty($matches[6]) ? $matches[7] : null;
        $fragment   = !empty($matches[8]) ? $matches[9] : null;

        // further parse authority
        if ($authority !== null) {
            $r_authority = "/^((.+?)@)?(\[[^\]]+\]|[^:]*)(:(\d*))?/";
            $matches = array();
            preg_match($r_authority, $authority, $matches);
            $userinfo   = !empty($matches[1]) ? $matches[2] : null;
            $host       = !empty($matches[3]) ? $matches[3] : '';
            $port       = !empty($matches[4]) ? (int) $matches[5] : null;
        } else {
            $port = $host = $userinfo = null;
        }

        return new HTMLPurifier_URI(
            $scheme, $userinfo, $host, $port, $path, $query, $fragment);
    }
}__halt_compiler();----SIGNATURE:----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----ATTACHMENT:----ODY4MjQ5NTY2NDA2NjAyMiA3MDY4NzQ4MjA1NDQyNjcyIDExOTg2NDMyMTE1OTYzMTA=