e ) ) { $name = ''; $value = ''; $whole = ''; return false; } $whole = str_replace( $value, $new_value, $whole ); $value = $new_value; } if ( is_array( $allowed_attr[ $name_low ] ) ) { // There are some checks. foreach ( $allowed_attr[ $name_low ] as $currkey => $currval ) { if ( ! wp_kses_check_attr_val( $value, $vless, $currkey, $currval ) ) { $name = ''; $value = ''; $whole = ''; return false; } } } return true; } /** * 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'`). * * @since 1.0.0 * * @param string $attr Attribute list from HTML element to closing HTML element tag. * @param string[] $allowed_protocols Array of allowed URL protocols. * @return array[] Array of attribute information after parsing. */ 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; } /** * Finds all attributes of an HTML element. * * Does not modify input. May return "evil" output. * * Based on `wp_kses_split2()` and `wp_kses_attr()`. * * @since 4.2.3 * * @param string $element HTML element. * @return array|false List of attributes found in the element. Returns false on failure. */ function wp_kses_attr_parse( $element ) { $valid = preg_match( '%^(<\s*)(/\s*)?([a-zA-Z0-9]+\s*)([^>]*)(>?)$%', $element, $matches ); if ( 1 !== $valid ) { return false; } $begin = $matches[1]; $slash = $matches[2]; $elname = $matches[3]; $attr = $matches[4]; $end = $matches[5]; if ( '' !== $slash ) { // Closing elements do not get parsed. return false; } // Is there a closing XHTML slash at the end of the attributes? if ( 1 === preg_match( '%\s*/\s*$%', $attr, $matches ) ) { $xhtml_slash = $matches[0]; $attr = substr( $attr, 0, -strlen( $xhtml_slash ) ); } else { $xhtml_slash = ''; } // Split it. $attrarr = wp_kses_hair_parse( $attr ); if ( false === $attrarr ) { return false; } // Make sure all input is returned by adding front and back matter. array_unshift( $attrarr, $begin . $slash . $elname ); array_push( $attrarr, $xhtml_slash . $end ); return $attrarr; } /** * Builds an attribute list from string containing attributes. * * Does not modify input. May return "evil" output. * In case of unexpected input, returns false instead of stripping things. * * Based on `wp_kses_hair()` but does not return a multi-dimensional array. * * @since 4.2.3 * * @param string $attr Attribute list from HTML element to closing HTML element tag. * @return array|false List of attributes found in $attr. Returns false on failure. */ function wp_kses_hair_parse( $attr ) { if ( '' === $attr ) { return array(); } // phpcs:disable Squiz.Strings.ConcatenationSpacing.PaddingFound -- don't remove regex indentation $regex = '(?:' . '[_a-zA-Z][-_a-zA-Z0-9:.]*' // Attribute name. . '|' . '\[\[?[^\[\]]+\]\]?' // Shortcode in the name position implies unfiltered_html. . ')' . '(?:' // Attribute value. . '\s*=\s*' // All values begin with '='. . '(?:' . '"[^"]*"' // Double-quoted. . '|' . "'[^']*'" // Single-quoted. . '|' . '[^\s"\']+' // Non-quoted. . '(?:\s|$)' // Must have a space. . ')' . '|' . '(?:\s|$)' // If attribute has no value, space is required. . ')' . '\s*'; // Trailing space is optional except as mentioned above. // phpcs:enable // Although it is possible to reduce this procedure to a single regexp, // we must run that regexp twice to get exactly the expected result. $validation = "%^($regex)+$%"; $extraction = "%$regex%"; if ( 1 === preg_match( $validation, $attr ) ) { preg_match_all( $extraction, $attr, $attrarr ); return $attrarr[0]; } else { return false; } } /** * Performs different checks for attribute values. * * The currently implemented checks are "maxlen", "minlen", "maxval", "minval", * and "valueless". * * @since 1.0.0 * * @param string $value Attribute value. * @param string $vless Whether the attribute is valueless. Use 'y' or 'n'. * @param string $checkname What $checkvalue is checking for. * @param mixed $checkvalue What constraint the value should pass. * @return bool Whether check passes. */ function wp_kses_check_attr_val( $value, $vless, $checkname, $checkvalue ) { $ok = true; switch ( strtolower( $checkname ) ) { case 'maxlen': /* * The maxlen check makes sure that the attribute value has a length not * greater than the given value. This can be used to avoid Buffer Overflows * in WWW clients and various Internet servers. */ if ( strlen( $value ) > $checkvalue ) { $ok = false; } break; case 'minlen': /* * The minlen check makes sure that the attribute value has a length not * smaller than the given value. */ if ( strlen( $value ) < $checkvalue ) { $ok = false; } break; case 'maxval': /* * The maxval check does two things: it checks that the attribute value is * an integer from 0 and up, without an excessive amount of zeroes or * whitespace (to avoid Buffer Overflows). It also checks that the attribute * value is not greater than the given value. * This check can be used to avoid Denial of Service attacks. */ if ( ! preg_match( '/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value ) ) { $ok = false; } if ( $value > $checkvalue ) { $ok = false; } break; case 'minval': /* * The minval check makes sure that the attribute value is a positive integer, * and that it is not smaller than the given value. */ if ( ! preg_match( '/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value ) ) { $ok = false; } if ( $value < $checkvalue ) { $ok = false; } break; case 'valueless': /* * The valueless check makes sure if the attribute has a value * (like ``) or not (``). If the given value * is a "y" or a "Y", the attribute must not have a value. * If the given value is an "n" or an "N", the attribute must have a value. */ if ( strtolower( $checkvalue ) != $vless ) { $ok = false; } break; case 'values': /* * The values check is used when you want to make sure that the attribute * has one of the given values. */ if ( false === array_search( strtolower( $value ), $checkvalue, true ) ) { $ok = false; } break; case 'value_callback': /* * The value_callback check is used when you want to make sure that the attribute * value is accepted by the callback function. */ if ( ! call_user_func( $checkvalue, $value ) ) { $ok = false; } break; } // End switch. return $ok; } /** * Sanitizes a string and removed disallowed URL protocols. * * This function removes all non-allowed protocols from the beginning of the * string. It ignores whitespace and the case of the letters, and it does * understand HTML entities. It does its work recursively, so it won't be * fooled by a string like `javascript:javascript:alert(57)`. * * @since 1.0.0 * * @param string $content Content to filter bad protocols from. * @param string[] $allowed_protocols Array of allowed URL protocols. * @return string Filtered content. */ function wp_kses_bad_protocol( $content, $allowed_protocols ) { $content = wp_kses_no_null( $content ); // Short-circuit if the string starts with `https://` or `http://`. Most common cases. if ( ( str_starts_with( $content, 'https://' ) && in_array( 'https', $allowed_protocols, true ) ) || ( str_starts_with( $content, 'http://' ) && in_array( 'http', $allowed_protocols, true ) ) ) { return $content; } $iterations = 0; do { $original_content = $content; $content = wp_kses_bad_protocol_once( $content, $allowed_protocols ); } while ( $original_content != $content && ++$iterations < 6 ); if ( $original_content != $content ) { return ''; } return $content; } /** * Removes any invalid control characters in a text string. * * Also removes any instance of the `\0` string. * * @since 1.0.0 * * @param string $content Content to filter null characters from. * @param array $options Set 'slash_zero' => 'keep' when '\0' is allowed. Default is 'remove'. * @return string Filtered content. */ function wp_kses_no_null( $content, $options = null ) { if ( ! isset( $options['slash_zero'] ) ) { $options = array( 'slash_zero' => 'remove' ); } $content = preg_replace( '/[\x00-\x08\x0B\x0C\x0E-\x1F]/', '', $content ); if ( 'remove' === $options['slash_zero'] ) { $content = preg_replace( '/\\\\+0+/', '', $content ); } return $content; } /** * Strips slashes from in front of quotes. * * This function changes the character sequence `\"` to just `"`. It leaves all other * slashes alone. The quoting from `preg_replace(//e)` requires this. * * @since 1.0.0 * * @param string $content String to strip slashes from. * @return string Fixed string with quoted slashes. */ function wp_kses_stripslashes( $content ) { return preg_replace( '%\\\\"%', '"', $content ); } /** * Converts the keys of an array to lowercase. * * @since 1.0.0 * * @param array $inarray Unfiltered array. * @return array Fixed array with all lowercase keys. */ function wp_kses_array_lc( $inarray ) { $outarray = array(); foreach ( (array) $inarray as $inkey => $inval ) { $outkey = strtolower( $inkey ); $outarray[ $outkey ] = array(); foreach ( (array) $inval as $inkey2 => $inval2 ) { $outkey2 = strtolower( $inkey2 ); $outarray[ $outkey ][ $outkey2 ] = $inval2; } } return $outarray; } /** * Handles parsing errors in `wp_kses_hair()`. * * The general plan is to remove everything to and including some whitespace, * but it deals with quotes and apostrophes as well. * * @since 1.0.0 * * @param string $attr * @return string */ function wp_kses_html_error( $attr ) { return preg_replace( '/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $attr ); } /** * Sanitizes content from bad protocols and other characters. * * This function searches for URL protocols at the beginning of the string, while * handling whitespace and HTML entities. * * @since 1.0.0 * * @param string $content Content to check for bad protocols. * @param string[] $allowed_protocols Array of allowed URL protocols. * @param int $count Depth of call recursion to this function. * @return string Sanitized content. */ function wp_kses_bad_protocol_once( $content, $allowed_protocols, $count = 1 ) { $content = preg_replace( '/(�*58(?![;0-9])|�*3a(?![;a-f0-9]))/i', '$1;', $content ); $content2 = preg_split( '/:|�*58;|�*3a;|:/i', $content, 2 ); if ( isset( $content2[1] ) && ! preg_match( '%/\?%', $content2[0] ) ) { $content = trim( $content2[1] ); $protocol = wp_kses_bad_protocol_once2( $content2[0], $allowed_protocols ); if ( 'feed:' === $protocol ) { if ( $count > 2 ) { return ''; } $content = wp_kses_bad_protocol_once( $content, $allowed_protocols, ++$count ); if ( empty( $content ) ) { return $content; } } $content = $protocol . $content; } return $content; } /** * Callback for `wp_kses_bad_protocol_once()` regular expression. * * This function processes URL protocols, checks to see if they're in the * list of allowed protocols or not, and returns different data depending * on the answer. * * @access private * @ignore * @since 1.0.0 * * @param string $scheme URI scheme to check against the list of allowed protocols. * @param string[] $allowed_protocols Array of allowed URL protocols. * @return string Sanitized content. */ function wp_kses_bad_protocol_once2( $scheme, $allowed_protocols ) { $scheme = wp_kses_decode_entities( $scheme ); $scheme = preg_replace( '/\s/', '', $scheme ); $scheme = wp_kses_no_null( $scheme ); $scheme = strtolower( $scheme ); $allowed = false; foreach ( (array) $allowed_protocols as $one_protocol ) { if ( strtolower( $one_protocol ) == $scheme ) { $allowed = true; break; } } if ( $allowed ) { return "$scheme:"; } else { return ''; } } /** * Converts and fixes HTML entities. * * This function normalizes HTML entities. It will convert `AT&T` to the correct * `AT&T`, `:` to `:`, `&#XYZZY;` to `&#XYZZY;` and so on. * * When `$context` is set to 'xml', HTML entities are converted to their code points. For * example, `AT&T…&#XYZZY;` is converted to `AT&T…&#XYZZY;`. * * @since 1.0.0 * @since 5.5.0 Added `$context` parameter. * * @param string $content Content to normalize entities. * @param string $context Context for normalization. Can be either 'html' or 'xml'. * Default 'html'. * @return string Content with normalized entities. */ function wp_kses_normalize_entities( $content, $context = 'html' ) { // Disarm all entities by converting & to & $content = str_replace( '&', '&', $content ); // Change back the allowed entities in our list of allowed entities. if ( 'xml' === $context ) { $content = preg_replace_callback( '/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_xml_named_entities', $content ); } else { $content = preg_replace_callback( '/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $content ); } $content = preg_replace_callback( '/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $content ); $content = preg_replace_callback( '/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $content ); return $content; } /** * Callback for `wp_kses_normalize_entities()` regular expression. * * This function only accepts valid named entity references, which are finite, * case-sensitive, and highly scrutinized by HTML and XML validators. * * @since 3.0.0 * * @global array $allowedentitynames * * @param array $matches preg_replace_callback() matches array. * @return string Correctly encoded entity. */ function wp_kses_named_entities( $matches ) { global $allowedentitynames; if ( empty( $matches[1] ) ) { return ''; } $i = $matches[1]; return ( ! in_array( $i, $allowedentitynames, true ) ) ? "&$i;" : "&$i;"; } /** * Callback for `wp_kses_normalize_entities()` regular expression. * * This function only accepts valid named entity references, which are finite, * case-sensitive, and highly scrutinized by XML validators. HTML named entity * references are converted to their code points. * * @since 5.5.0 * * @global array $allowedentitynames * @global array $allowedxmlentitynames * * @param array $matches preg_replace_callback() matches array. * @return string Correctly encoded entity. */ function wp_kses_xml_named_entities( $matches ) { global $allowedentitynames, $allowedxmlentitynames; if ( empty( $matches[1] ) ) { return ''; } $i = $matches[1]; if ( in_array( $i, $allowedxmlentitynames, true ) ) { return "&$i;"; } elseif ( in_array( $i, $allowedentitynames, true ) ) { return html_entity_decode( "&$i;", ENT_HTML5 ); } return "&$i;"; } /** * Callback for `wp_kses_normalize_entities()` regular expression. * * This function helps `wp_kses_normalize_entities()` to only accept 16-bit * values and nothing more for `&#number;` entities. * * @access private * @ignore * @since 1.0.0 * * @param array $matches `preg_replace_callback()` matches array. * @return string Correctly encoded entity. */ function wp_kses_normalize_entities2( $matches ) { if ( empty( $matches[1] ) ) { return ''; } $i = $matches[1]; if ( valid_unicode( $i ) ) { $i = str_pad( ltrim( $i, '0' ), 3, '0', STR_PAD_LEFT ); $i = "&#$i;"; } else { $i = "&#$i;"; } return $i; } /** * Callback for `wp_kses_normalize_entities()` for regular expression. * * This function helps `wp_kses_normalize_entities()` to only accept valid Unicode * numeric entities in hex form. * * @since 2.7.0 * @access private * @ignore * * @param array $matches `preg_replace_callback()` matches array. * @return string Correctly encoded entity. */ function wp_kses_normalize_entities3( $matches ) { if ( empty( $matches[1] ) ) { return ''; } $hexchars = $matches[1]; return ( ! valid_unicode( hexdec( $hexchars ) ) ) ? "&#x$hexchars;" : '&#x' . ltrim( $hexchars, '0' ) . ';'; } /** * Determines if a Unicode codepoint is valid. * * @since 2.7.0 * * @param int $i Unicode codepoint. * @return bool Whether or not the codepoint is a valid Unicode codepoint. */ function valid_unicode( $i ) { return ( 0x9 == $i || 0xa == $i || 0xd == $i || ( 0x20 <= $i && $i <= 0xd7ff ) || ( 0xe000 <= $i && $i <= 0xfffd ) || ( 0x10000 <= $i && $i <= 0x10ffff ) ); } /** * Converts all numeric HTML entities to their named counterparts. * * This function decodes numeric HTML entities (`A` and `A`). * It doesn't do anything with named entities like `ä`, but we don't * need them in the allowed URL protocols system anyway. * * @since 1.0.0 * * @param string $content Content to change entities. * @return string Content after decoded entities. */ function wp_kses_decode_entities( $content ) { $content = preg_replace_callback( '/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $content ); $content = preg_replace_callback( '/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $content ); return $content; } /** * Regex callback for `wp_kses_decode_entities()`. * * @since 2.9.0 * @access private * @ignore * * @param array $matches preg match * @return string */ function _wp_kses_decode_entities_chr( $matches ) { return chr( $matches[1] ); } /** * Regex callback for `wp_kses_decode_entities()`. * * @since 2.9.0 * @access private * @ignore * * @param array $matches preg match * @return string */ function _wp_kses_decode_entities_chr_hexdec( $matches ) { return chr( hexdec( $matches[1] ) ); } /** * Sanitize content with allowed HTML KSES rules. * * This function expects slashed data. * * @since 1.0.0 * * @param string $data Content to filter, expected to be escaped with slashes. * @return string Filtered content. */ function wp_filter_kses( $data ) { return addslashes( wp_kses( stripslashes( $data ), current_filter() ) ); } /** * Sanitize content with allowed HTML KSES rules. * * This function expects unslashed data. * * @since 2.9.0 * * @param string $data Content to filter, expected to not be escaped. * @return string Filtered content. */ function wp_kses_data( $data ) { return wp_kses( $data, current_filter() ); } /** * Sanitizes content for allowed HTML tags for post content. * * Post content refers to the page contents of the 'post' type and not `$_POST` * data from forms. * * This function expects slashed data. * * @since 2.0.0 * * @param string $data Post content to filter, expected to be escaped with slashes. * @return string Filtered post content with allowed HTML tags and attributes intact. */ function wp_filter_post_kses( $data ) { return addslashes( wp_kses( stripslashes( $data ), 'post' ) ); } /** * Sanitizes global styles user content removing unsafe rules. * * @since 5.9.0 * * @param string $data Post content to filter. * @return string Filtered post content with unsafe rules removed. */ function wp_filter_global_styles_post( $data ) { $decoded_data = json_decode( wp_unslash( $data ), true ); $json_decoding_error = json_last_error(); if ( JSON_ERROR_NONE === $json_decoding_error && is_array( $decoded_data ) && isset( $decoded_data['isGlobalStylesUserThemeJSON'] ) && $decoded_data['isGlobalStylesUserThemeJSON'] ) { unset( $decoded_data['isGlobalStylesUserThemeJSON'] ); $data_to_encode = WP_Theme_JSON::remove_insecure_properties( $decoded_data ); $data_to_encode['isGlobalStylesUserThemeJSON'] = true; return wp_slash( wp_json_encode( $data_to_encode ) ); } return $data; } /** * Sanitizes content for allowed HTML tags for post content. * * Post content refers to the page contents of the 'post' type and not `$_POST` * data from forms. * * This function expects unslashed data. * * @since 2.9.0 * * @param string $data Post content to filter. * @return string Filtered post content with allowed HTML tags and attributes intact. */ function wp_kses_post( $data ) { return wp_kses( $data, 'post' ); } /** * Navigates through an array, object, or scalar, and sanitizes content for * allowed HTML tags for post content. * * @since 4.4.2 * * @see map_deep() * * @param mixed $data The array, object, or scalar value to inspect. * @return mixed The filtered content. */ function wp_kses_post_deep( $data ) { return map_deep( $data, 'wp_kses_post' ); } /** * Strips all HTML from a text string. * * This function expects slashed data. * * @since 2.1.0 * * @param string $data Content to strip all HTML from. * @return string Filtered content without any HTML. */ function wp_filter_nohtml_kses( $data ) { return addslashes( wp_kses( stripslashes( $data ), 'strip' ) ); } /** * Adds all KSES input form content filters. * * All hooks have default priority. The `wp_filter_kses()` function is added to * the 'pre_comment_content' and 'title_save_pre' hooks. * * The `wp_filter_post_kses()` function is added to the 'content_save_pre', * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks. * * @since 2.0.0 */ function kses_init_filters() { // Normal filtering. add_filter( 'title_save_pre', 'wp_filter_kses' ); // Comment filtering. if ( current_user_can( 'unfiltered_html' ) ) { add_filter( 'pre_comment_content', 'wp_filter_post_kses' ); } else { add_filter( 'pre_comment_content', 'wp_filter_kses' ); } // Global Styles filtering: Global Styles filters should be executed before normal post_kses HTML filters. add_filter( 'content_save_pre', 'wp_filter_global_styles_post', 9 ); add_filter( 'content_filtered_save_pre', 'wp_filter_global_styles_post', 9 ); // Post filtering. add_filter( 'content_save_pre', 'wp_filter_post_kses' ); add_filter( 'excerpt_save_pre', 'wp_filter_post_kses' ); add_filter( 'content_filtered_save_pre', 'wp_filter_post_kses' ); } /** * Removes all KSES input form content filters. * * A quick procedural method to removing all of the filters that KSES uses for * content in WordPress Loop. * * Does not remove the `kses_init()` function from {@see 'init'} hook (priority is * default). Also does not remove `kses_init()` function from {@see 'set_current_user'} * hook (priority is also default). * * @since 2.0.6 */ function kses_remove_filters() { // Normal filtering. remove_filter( 'title_save_pre', 'wp_filter_kses' ); // Comment filtering. remove_filter( 'pre_comment_content', 'wp_filter_post_kses' ); remove_filter( 'pre_comment_content', 'wp_filter_kses' ); // Global Styles filtering. remove_filter( 'content_save_pre', 'wp_filter_global_styles_post', 9 ); remove_filter( 'content_filtered_save_pre', 'wp_filter_global_styles_post', 9 ); // Post filtering. remove_filter( 'content_save_pre', 'wp_filter_post_kses' ); remove_filter( 'excerpt_save_pre', 'wp_filter_post_kses' ); remove_filter( 'content_filtered_save_pre', 'wp_filter_post_kses' ); } /** * Sets up most of the KSES filters for input form content. * * First removes all of the KSES filters in case the current user does not need * to have KSES filter the content. If the user does not have `unfiltered_html` * capability, then KSES filters are added. * * @since 2.0.0 */ function kses_init() { kses_remove_filters(); if ( ! current_user_can( 'unfiltered_html' ) ) { kses_init_filters(); } } /** * Filters an inline style attribute and removes disallowed rules. * * @since 2.8.1 * @since 4.4.0 Added support for `min-height`, `max-height`, `min-width`, and `max-width`. * @since 4.6.0 Added support for `list-style-type`. * @since 5.0.0 Added support for `background-image`. * @since 5.1.0 Added support for `text-transform`. * @since 5.2.0 Added support for `background-position` and `grid-template-columns`. * @since 5.3.0 Added support for `grid`, `flex` and `column` layout properties. * Extended `background-*` support for individual properties. * @since 5.3.1 Added support for gradient backgrounds. * @since 5.7.1 Added support for `object-position`. * @since 5.8.0 Added support for `calc()` and `var()` values. * @since 6.1.0 Added support for `min()`, `max()`, `minmax()`, `clamp()`, * nested `var()` values, and assigning values to CSS variables. * Added support for `object-fit`, `gap`, `column-gap`, `row-gap`, and `flex-wrap`. * Extended `margin-*` and `padding-*` support for logical properties. * @since 6.2.0 Added support for `aspect-ratio`, `position`, `top`, `right`, `bottom`, `left`, * and `z-index` CSS properties. * * @param string $css A string of CSS rules. * @param string $deprecated Not used. * @return string Filtered string of CSS rules. */ function safecss_filter_attr( $css, $deprecated = '' ) { if ( ! empty( $deprecated ) ) { _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented. } $css = wp_kses_no_null( $css ); $css = str_replace( array( "\n", "\r", "\t" ), '', $css ); $allowed_protocols = wp_allowed_protocols(); $css_array = explode( ';', trim( $css ) ); /** * Filters the list of allowed CSS attributes. * * @since 2.8.1 * * @param string[] $attr Array of allowed CSS attributes. */ $allowed_attr = apply_filters( 'safe_style_css', array( 'background', 'background-color', 'background-image', 'background-position', 'background-size', 'background-attachment', 'background-blend-mode', 'border', 'border-radius', 'border-width', 'border-color', 'border-style', 'border-right', 'border-right-color', 'border-right-style', 'border-right-width', 'border-bottom', 'border-bottom-color', 'border-bottom-left-radius', 'border-bottom-right-radius', 'border-bottom-style', 'border-bottom-width', 'border-bottom-right-radius', 'border-bottom-left-radius', 'border-left', 'border-left-color', 'border-left-style', 'border-left-width', 'border-top', 'border-top-color', 'border-top-left-radius', 'border-top-right-radius', 'border-top-style', 'border-top-width', 'border-top-left-radius', 'border-top-right-radius', 'border-spacing', 'border-collapse', 'caption-side', 'columns', 'column-count', 'column-fill', 'column-gap', 'column-rule', 'column-span', 'column-width', 'color', 'filter', 'font', 'font-family', 'font-size', 'font-style', 'font-variant', 'font-weight', 'letter-spacing', 'line-height', 'text-align', 'text-decoration', 'text-indent', 'text-transform', 'height', 'min-height', 'max-height', 'width', 'min-width', 'max-width', 'margin', 'margin-right', 'margin-bottom', 'margin-left', 'margin-top', 'margin-block-start', 'margin-block-end', 'margin-inline-start', 'margin-inline-end', 'padding', 'padding-right', 'padding-bottom', 'padding-left', 'padding-top', 'padding-block-start', 'padding-block-end', 'padding-inline-start', 'padding-inline-end', 'flex', 'flex-basis', 'flex-direction', 'flex-flow', 'flex-grow', 'flex-shrink', 'flex-wrap', 'gap', 'column-gap', 'row-gap', 'grid-template-columns', 'grid-auto-columns', 'grid-column-start', 'grid-column-end', 'grid-column-gap', 'grid-template-rows', 'grid-auto-rows', 'grid-row-start', 'grid-row-end', 'grid-row-gap', 'grid-gap', 'justify-content', 'justify-items', 'justify-self', 'align-content', 'align-items', 'align-self', 'clear', 'cursor', 'direction', 'float', 'list-style-type', 'object-fit', 'object-position', 'overflow', 'vertical-align', 'position', 'top', 'right', 'bottom', 'left', 'z-index', 'aspect-ratio', // Custom CSS properties. '--*', ) ); /* * CSS attributes that accept URL data types. * * This is in accordance to the CSS spec and unrelated to * the sub-set of supported attributes above. * * See: https://developer.mozilla.org/en-US/docs/Web/CSS/url */ $css_url_data_types = array( 'background', 'background-image', 'cursor', 'list-style', 'list-style-image', ); /* * CSS attributes that accept gradient data types. * */ $css_gradient_data_types = array( 'background', 'background-image', ); if ( empty( $allowed_attr ) ) { return $css; } $css = ''; foreach ( $css_array as $css_item ) { if ( '' === $css_item ) { continue; } $css_item = trim( $css_item ); $css_test_string = $css_item; $found = false; $url_attr = false; $gradient_attr = false; $is_custom_var = false; if ( strpos( $css_item, ':' ) === false ) { $found = true; } else { $parts = explode( ':', $css_item, 2 ); $css_selector = trim( $parts[0] ); // Allow assigning values to CSS variables. if ( in_array( '--*', $allowed_attr, true ) && preg_match( '/^--[a-zA-Z0-9-_]+$/', $css_selector ) ) { $allowed_attr[] = $css_selector; $is_custom_var = true; } if ( in_array( $css_selector, $allowed_attr, true ) ) { $found = true; $url_attr = in_array( $css_selector, $css_url_data_types, true ); $gradient_attr = in_array( $css_selector, $css_gradient_data_types, true ); } if ( $is_custom_var ) { $css_value = trim( $parts[1] ); $url_attr = str_starts_with( $css_value, 'url(' ); $gradient_attr = str_contains( $css_value, '-gradient(' ); } } if ( $found && $url_attr ) { // Simplified: matches the sequence `url(*)`. preg_match_all( '/url\([^)]+\)/', $parts[1], $url_matches ); foreach ( $url_matches[0] as $url_match ) { // Clean up the URL from each of the matches above. preg_match( '/^url\(\s*([\'\"]?)(.*)(\g1)\s*\)$/', $url_match, $url_pieces ); if ( empty( $url_pieces[2] ) ) { $found = false; break; } $url = trim( $url_pieces[2] ); if ( empty( $url ) || wp_kses_bad_protocol( $url, $allowed_protocols ) !== $url ) { $found = false; break; } else { // Remove the whole `url(*)` bit that was matched above from the CSS. $css_test_string = str_replace( $url_match, '', $css_test_string ); } } } if ( $found && $gradient_attr ) { $css_value = trim( $parts[1] ); if ( preg_match( '/^(repeating-)?(linear|radial|conic)-gradient\(([^()]|rgb[a]?\([^()]*\))*\)$/', $css_value ) ) { // Remove the whole `gradient` bit that was matched above from the CSS. $css_test_string = str_replace( $css_value, '', $css_test_string ); } } if ( $found ) { /* * Allow CSS functions like var(), calc(), etc. by removing them from the test string. * Nested functions and parentheses are also removed, so long as the parentheses are balanced. */ $css_test_string = preg_replace( '/\b(?:var|calc|min|max|minmax|clamp)(\((?:[^()]|(?1))*\))/', '', $css_test_string ); /* * Disallow CSS containing \ ( & } = or comments, except for within url(), var(), calc(), etc. * which were removed from the test string above. */ $allow_css = ! preg_match( '%[\\\(&=}]|/\*%', $css_test_string ); /** * Filters the check for unsafe CSS in `safecss_filter_attr`. * * Enables developers to determine whether a section of CSS should be allowed or discarded. * By default, the value will be false if the part contains \ ( & } = or comments. * Return true to allow the CSS part to be included in the output. * * @since 5.5.0 * * @param bool $allow_css Whether the CSS in the test string is considered safe. * @param string $css_test_string The CSS string to test. */ $allow_css = apply_filters( 'safecss_filter_attr_allow_css', $allow_css, $css_test_string ); // Only add the CSS part if it passes the regex check. if ( $allow_css ) { if ( '' !== $css ) { $css .= ';'; } $css .= $css_item; } } } return $css; } /** * Helper function to add global attributes to a tag in the allowed HTML list. * * @since 3.5.0 * @since 5.0.0 Added support for `data-*` wildcard attributes. * @since 6.0.0 Added `dir`, `lang`, and `xml:lang` to global attributes. * * @access private * @ignore * * @param array $value An array of attributes. * @return array The array of attributes with global attributes added. */ function _wp_add_global_attributes( $value ) { $global_attributes = array( 'aria-describedby' => true, 'aria-details' => true, 'aria-label' => true, 'aria-labelledby' => true, 'aria-hidden' => true, 'class' => true, 'data-*' => true, 'dir' => true, 'id' => true, 'lang' => true, 'style' => true, 'title' => true, 'role' => true, 'xml:lang' => true, ); if ( true === $value ) { $value = array(); } if ( is_array( $value ) ) { return array_merge( $value, $global_attributes ); } return $value; } /** * Helper function to check if this is a safe PDF URL. * * @since 5.9.0 * @access private * @ignore * * @param string $url The URL to check. * @return bool True if the URL is safe, false otherwise. */ function _wp_kses_allow_pdf_objects( $url ) { // We're not interested in URLs that contain query strings or fragments. if ( str_contains( $url, '?' ) || str_contains( $url, '#' ) ) { return false; } // If it doesn't have a PDF extension, it's not safe. if ( ! str_ends_with( $url, '.pdf' ) ) { return false; } // If the URL host matches the current site's media URL, it's safe. $upload_info = wp_upload_dir( null, false ); $parsed_url = wp_parse_url( $upload_info['url'] ); $upload_host = isset( $parsed_url['host'] ) ? $parsed_url['host'] : ''; $upload_port = isset( $parsed_url['port'] ) ? ':' . $parsed_url['port'] : ''; if ( str_starts_with( $url, "http://$upload_host$upload_port/" ) || str_starts_with( $url, "https://$upload_host$upload_port/" ) ) { return true; } return false; }