PHP
6 days ago
LICENSE.md
6 days ago
README.md
6 days ago
class-xmlattributetoken.php
6 days ago
class-xmldecoder.php
6 days ago
class-xmlelement.php
6 days ago
class-xmlnativecursorprocessor.php
6 days ago
class-xmlprocessor.php
6 days ago
class-xmlunsupportedexception.php
6 days ago
composer.json
6 days ago
class-xmldecoder.php
227 lines
| 1 | <?php |
| 2 | |
| 3 | namespace WordPress\XML; |
| 4 | |
| 5 | use function WordPress\Encoding\codepoint_to_utf8_bytes; |
| 6 | |
| 7 | /** |
| 8 | * XML API: WP_XML_Decoder class |
| 9 | * |
| 10 | * Decodes spans of raw text found inside XML content, |
| 11 | * whether found in an attribute or in a text node. |
| 12 | * |
| 13 | * Do not use this function on the contents of a CDATA section, |
| 14 | * as those sections are not encoded with the XML rules unless |
| 15 | * they are embedded XML content. |
| 16 | * |
| 17 | * @package WordPress |
| 18 | * @subpackage HTML-API |
| 19 | * @since WP_VERSION |
| 20 | */ |
| 21 | class XMLDecoder { |
| 22 | /** |
| 23 | * Decodes a span of XML text. |
| 24 | * |
| 25 | * Example: |
| 26 | * |
| 27 | * '&' = WP_XML_Decoder::decode( '&' ); |
| 28 | * '…' = WP_XML_Decoder::decode( '…' ); |
| 29 | * |
| 30 | * @todo Add examples of parse failures, and decide if it should fail or not. |
| 31 | * |
| 32 | * @since WP_VERSION |
| 33 | * |
| 34 | * @access private |
| 35 | * |
| 36 | * @param string $text Text document containing span of text to decode. |
| 37 | * @return string Decoded UTF-8 string. |
| 38 | */ |
| 39 | public static function decode( $text ) { |
| 40 | $decoded = ''; |
| 41 | $end = strlen( $text ); |
| 42 | $at = 0; |
| 43 | $was_at = 0; |
| 44 | |
| 45 | while ( $at < $end ) { |
| 46 | $next_character_reference_at = strpos( $text, '&', $at ); |
| 47 | if ( false === $next_character_reference_at || $next_character_reference_at >= $end ) { |
| 48 | break; |
| 49 | } |
| 50 | |
| 51 | $start_of_potential_reference_at = $next_character_reference_at + 1; |
| 52 | if ( $start_of_potential_reference_at >= $end ) { |
| 53 | // @todo This is an error. The document ended too early; consume the rest as plaintext, which is wrong. |
| 54 | break; |
| 55 | } |
| 56 | |
| 57 | /** |
| 58 | * First character after the opening `&`. |
| 59 | */ |
| 60 | $start_of_potential_reference = $text[ $start_of_potential_reference_at ]; |
| 61 | |
| 62 | /* |
| 63 | * If it's a named character reference, it will be one of the five mandated references. |
| 64 | * |
| 65 | * - `&` |
| 66 | * - `'` |
| 67 | * - `>` |
| 68 | * - `<` |
| 69 | * - `"` |
| 70 | * |
| 71 | * These all must be found within the five successive characters from the `&`. |
| 72 | * |
| 73 | * Example: |
| 74 | * |
| 75 | * ╭ ampersand at 9 = $end - 6 |
| 76 | * 'XML' ($end = 15) |
| 77 | * ╰───┴─ this length must be at least 5 long, |
| 78 | * which is $end - 5. |
| 79 | */ |
| 80 | if ( |
| 81 | $next_character_reference_at < $end - 5 && |
| 82 | ( |
| 83 | 'a' === $start_of_potential_reference || |
| 84 | 'g' === $start_of_potential_reference || |
| 85 | 'l' === $start_of_potential_reference || |
| 86 | 'q' === $start_of_potential_reference |
| 87 | ) |
| 88 | ) { |
| 89 | foreach ( array( |
| 90 | 'amp;' => '&', |
| 91 | 'apos;' => "'", |
| 92 | 'lt;' => '<', |
| 93 | 'gt;' => '>', |
| 94 | 'quot;' => '"', |
| 95 | ) as $name => $substitution ) { |
| 96 | if ( 0 === substr_compare( $text, $name, $start_of_potential_reference_at, strlen( $name ) ) ) { |
| 97 | $decoded .= substr( $text, $was_at, $next_character_reference_at - $was_at ) . $substitution; |
| 98 | $at = $start_of_potential_reference_at + strlen( $name ); |
| 99 | $was_at = $at; |
| 100 | continue 2; |
| 101 | } |
| 102 | } |
| 103 | |
| 104 | // @todo This is an invalid document. It should be communicated. Treat as plaintext and continue. |
| 105 | ++$at; |
| 106 | continue; |
| 107 | } |
| 108 | |
| 109 | /* |
| 110 | * The shortest numerical character reference is four characters. |
| 111 | * |
| 112 | * Example: |
| 113 | * |
| 114 | * 	 |
| 115 | */ |
| 116 | if ( '#' !== $start_of_potential_reference || $next_character_reference_at + 4 >= $end ) { |
| 117 | // @todo This is an error. This ampersand _must_ be encoded. Treat as plaintext and move on. |
| 118 | ++$at; |
| 119 | continue; |
| 120 | } |
| 121 | |
| 122 | $is_hex = 'x' === $text[ $start_of_potential_reference_at + 1 ]; |
| 123 | if ( $is_hex ) { |
| 124 | $zeros_at = $start_of_potential_reference_at + 2; |
| 125 | $base = 16; |
| 126 | $digit_chars = '0123456789abcdefABCDEF'; |
| 127 | $max_digits = 6; // ``. |
| 128 | } else { |
| 129 | $zeros_at = $start_of_potential_reference_at + 1; |
| 130 | $base = 10; |
| 131 | $digit_chars = '0123456789'; |
| 132 | $max_digits = 7; // ``. |
| 133 | } |
| 134 | |
| 135 | $zero_count = strspn( $text, '0', $zeros_at ); |
| 136 | $digits_at = $zeros_at + $zero_count; |
| 137 | $digit_count = strspn( $text, $digit_chars, $digits_at, $max_digits ); |
| 138 | $semi_at = $digits_at + $digit_count; |
| 139 | |
| 140 | if ( 0 === $digit_count || $semi_at >= $end || ';' !== $text[ $semi_at ] ) { |
| 141 | // @todo This is an error. Treat as plaintext and move on. |
| 142 | ++$at; |
| 143 | continue; |
| 144 | } |
| 145 | |
| 146 | $codepoint = intval( substr( $text, $digits_at, $digit_count ), $base ); |
| 147 | $character_reference = codepoint_to_utf8_bytes( $codepoint ); |
| 148 | if ( '�' === $character_reference && 0xFFFD !== $codepoint ) { |
| 149 | /* |
| 150 | * Stop processing if we got an invalid character AND the reference does not |
| 151 | * specifically refer code point FFFD (�). |
| 152 | * |
| 153 | * > It is a fatal error when an XML processor encounters an entity with an |
| 154 | * > encoding that it is unable to process. It is a fatal error if an XML entity |
| 155 | * > is determined (via default, encoding declaration, or higher-level protocol) |
| 156 | * > to be in a certain encoding but contains byte sequences that are not legal |
| 157 | * > in that encoding. Specifically, it is a fatal error if an entity encoded in |
| 158 | * > UTF-8 contains any ill-formed code unit sequences, as defined in section |
| 159 | * > 3.9 of Unicode [Unicode]. Unless an encoding is determined by a higher-level |
| 160 | * > protocol, it is also a fatal error if an XML entity contains no encoding |
| 161 | * > declaration and its content is not legal UTF-8 or UTF-16. |
| 162 | * |
| 163 | * See https://www.w3.org/TR/xml/#charencoding |
| 164 | */ |
| 165 | // @todo This is an error. Treat as plaintext and continue, which is wrong. |
| 166 | ++$at; |
| 167 | continue; |
| 168 | } |
| 169 | |
| 170 | $decoded .= substr( $text, $was_at, $at - $was_at ); |
| 171 | $decoded .= $character_reference; |
| 172 | $at = $semi_at + 1; |
| 173 | $was_at = $at; |
| 174 | } |
| 175 | |
| 176 | if ( 0 === $was_at ) { |
| 177 | return $text; |
| 178 | } |
| 179 | |
| 180 | if ( $was_at < $end ) { |
| 181 | $decoded .= substr( $text, $was_at, $end - $was_at ); |
| 182 | } |
| 183 | |
| 184 | return $decoded; |
| 185 | } |
| 186 | |
| 187 | /** |
| 188 | * Finds and parses the next entity in a given text starting after the |
| 189 | * given byte offset, and being entirely found within the given max length. |
| 190 | * |
| 191 | * @since {WP_VERSION} |
| 192 | * |
| 193 | * // @todo Implement this function. |
| 194 | * |
| 195 | * @param string $text Text in which to search for an XML entity. |
| 196 | * @param int $starting_byte_offset Start looking after this byte offset. |
| 197 | * @param int $ending_byte_offset Stop looking if entity is not fully contained before this byte offset. |
| 198 | * @param int|null $entity_at Optional. If provided, will be set to byte offset where entity was |
| 199 | * found, if found. Otherwise, will not be set. |
| 200 | * |
| 201 | * @return string|null Parsed entity, if parsed, otherwise `null`. |
| 202 | */ |
| 203 | public static function next_entity( string $text, int $starting_byte_offset, int $ending_byte_offset, ?int &$entity_at = null ): ?string { |
| 204 | $at = $starting_byte_offset; |
| 205 | $end = $ending_byte_offset; |
| 206 | |
| 207 | while ( $at < $end ) { |
| 208 | $remaining = $end - $at; |
| 209 | $amp_after = strcspn( $text, '&', $at, $remaining ); |
| 210 | |
| 211 | // There are no more possible entities. |
| 212 | if ( $amp_after === $remaining ) { |
| 213 | return null; |
| 214 | } |
| 215 | |
| 216 | /* |
| 217 | * @todo Move the decoding logic from `decode()` above into here, |
| 218 | * then call this function in a loop from `decode()`. |
| 219 | */ |
| 220 | |
| 221 | ++$at; |
| 222 | } |
| 223 | |
| 224 | return null; |
| 225 | } |
| 226 | } |
| 227 |