wp_kses_hair

Builds an attribute list from string containing attributes.

This function does a lot of work. It parses an attribute list into an array
with attribute data, and tries to do the right thing even if it gets weird
input. It will add quotes around attribute values that don’t have any quotes
or apostrophes around them, to make it easier to produce HTML code that will
conform to W3C’s HTML specification. It will also remove bad URL protocols
from attribute values. It also reduces duplicate attributes by using the
attribute defined first (`foo=’bar’ foo=’baz’` will result in `foo=’bar’`).

function wp_kses_hair( $attr, $allowed_protocols ) {
	$attrarr  = array();
	$mode     = 0;
	$attrname = '';
	$uris     = wp_kses_uri_attributes();

	// Loop through the whole attribute list.

	while ( strlen( $attr ) != 0 ) {
		$working = 0; // Was the last operation successful?

		switch ( $mode ) {
			case 0:
				if ( preg_match( '/^([_a-zA-Z][-_a-zA-Z0-9:.]*)/', $attr, $match ) ) {
					$attrname = $match[1];
					$working  = 1;
					$mode     = 1;
					$attr     = preg_replace( '/^[_a-zA-Z][-_a-zA-Z0-9:.]*/', '', $attr );
				}

				break;

			case 1:
				if ( preg_match( '/^s*=s*/', $attr ) ) { // Equals sign.
					$working = 1;
					$mode    = 2;
					$attr    = preg_replace( '/^s*=s*/', '', $attr );
					break;
				}

				if ( preg_match( '/^s+/', $attr ) ) { // Valueless.
					$working = 1;
					$mode    = 0;
					if ( false === array_key_exists( $attrname, $attrarr ) ) {
						$attrarr[ $attrname ] = array(
							'name'  => $attrname,
							'value' => '',
							'whole' => $attrname,
							'vless' => 'y',
						);
					}
					$attr = preg_replace( '/^s+/', '', $attr );
				}

				break;

			case 2:
				if ( preg_match( '%^"([^"]*)"(s+|/?$)%', $attr, $match ) ) {
					// "value"
					$thisval = $match[1];
					if ( in_array( strtolower( $attrname ), $uris, true ) ) {
						$thisval = wp_kses_bad_protocol( $thisval, $allowed_protocols );
					}

					if ( false === array_key_exists( $attrname, $attrarr ) ) {
						$attrarr[ $attrname ] = array(
							'name'  => $attrname,
							'value' => $thisval,
							'whole' => "$attrname="$thisval"",
							'vless' => 'n',
						);
					}
					$working = 1;
					$mode    = 0;
					$attr    = preg_replace( '/^"[^"]*"(s+|$)/', '', $attr );
					break;
				}

				if ( preg_match( "%^'([^']*)'(s+|/?$)%", $attr, $match ) ) {
					// 'value'
					$thisval = $match[1];
					if ( in_array( strtolower( $attrname ), $uris, true ) ) {
						$thisval = wp_kses_bad_protocol( $thisval, $allowed_protocols );
					}

					if ( false === array_key_exists( $attrname, $attrarr ) ) {
						$attrarr[ $attrname ] = array(
							'name'  => $attrname,
							'value' => $thisval,
							'whole' => "$attrname='$thisval'",
							'vless' => 'n',
						);
					}
					$working = 1;
					$mode    = 0;
					$attr    = preg_replace( "/^'[^']*'(s+|$)/", '', $attr );
					break;
				}

				if ( preg_match( "%^([^s"']+)(s+|/?$)%", $attr, $match ) ) {
					// value
					$thisval = $match[1];
					if ( in_array( strtolower( $attrname ), $uris, true ) ) {
						$thisval = wp_kses_bad_protocol( $thisval, $allowed_protocols );
					}

					if ( false === array_key_exists( $attrname, $attrarr ) ) {
						$attrarr[ $attrname ] = array(
							'name'  => $attrname,
							'value' => $thisval,
							'whole' => "$attrname="$thisval"",
							'vless' => 'n',
						);
					}
					// We add quotes to conform to W3C's HTML spec.
					$working = 1;
					$mode    = 0;
					$attr    = preg_replace( "%^[^s"']+(s+|$)%", '', $attr );
				}

				break;
		} // End switch.

		if ( 0 == $working ) { // Not well-formed, remove and try again.
			$attr = wp_kses_html_error( $attr );
			$mode = 0;
		}
	} // End while.

	if ( 1 == $mode && false === array_key_exists( $attrname, $attrarr ) ) {
		// Special case, for when the attribute list ends with a valueless
		// attribute like "selected".
		$attrarr[ $attrname ] = array(
			'name'  => $attrname,
			'value' => '',
			'whole' => $attrname,
			'vless' => 'y',
		);
	}

	return $attrarr;
}