includes/json/FormatJson.php

   1 <?php
   2 /**
   3  * Wrapper for json_encode and json_decode.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  */
  22
  23 /**
  24  * JSON formatter wrapper class
  25  */
  26 class FormatJson {
  27
  28         /**
  29          * Skip escaping most characters above U+007F for readability and compactness.
  30          * This encoding option saves 3 to 8 bytes (uncompressed) for each such character;
  31          * however, it could break compatibility with systems that incorrectly handle UTF-8.
  32          *
  33          * @since 1.21
  34          */
  35         const UTF8_OK = 1;
  36
  37         /**
  38          * Skip escaping the characters '<', '>', and '&', which have special meanings in
  39          * HTML and XML.
  40          *
  41          * @warning Do not use this option for JSON that could end up in inline scripts.
  42          * - HTML5, §4.3.1.2 Restrictions for contents of script elements
  43          * - XML 1.0 (5th Ed.), §2.4 Character Data and Markup
  44          *
  45          * @since 1.21
  46          */
  47         const XMLMETA_OK = 2;
  48
  49         /**
  50          * Skip escaping as many characters as reasonably possible.
  51          *
  52          * @warning When generating inline script blocks, use FormatJson::UTF8_OK instead.
  53          *
  54          * @since 1.21
  55          */
  56         const ALL_OK = 3;
  57
  58         /**
  59          * Characters problematic in JavaScript.
  60          *
  61          * @note These are listed in ECMA-262 (5.1 Ed.), §7.3 Line Terminators along with U+000A (LF)
  62          *       and U+000D (CR). However, PHP already escapes LF and CR according to RFC 4627.
  63          */
  64         private static $badChars = array(
  65                 "\xe2\x80\xa8", // U+2028 LINE SEPARATOR
  66                 "\xe2\x80\xa9", // U+2029 PARAGRAPH SEPARATOR
  67         );
  68
  69         /**
  70          * Escape sequences for characters listed in FormatJson::$badChars.
  71          */
  72         private static $badCharsEscaped = array(
  73                 '\u2028', // U+2028 LINE SEPARATOR
  74                 '\u2029', // U+2029 PARAGRAPH SEPARATOR
  75         );
  76
  77         /**
  78          * Returns the JSON representation of a value.
  79          *
  80          * @note Empty arrays are encoded as numeric arrays, not as objects, so cast any associative
  81          *       array that might be empty to an object before encoding it.
  82          *
  83          * @note In pre-1.21 versions of MediaWiki, using this function for generating inline script
  84          *       blocks may result in an XSS vulnerability, and quite likely will in XML documents
  85          *       (cf. FormatJson::XMLMETA_OK). Use Xml::encodeJsVar() instead in such cases.
  86          *
  87          * @param mixed $value The value to encode. Can be any type except a resource.
  88          * @param bool $pretty If true, add non-significant whitespace to improve readability.
  89          * @param int $escaping Bitfield consisting of _OK class constants
  90          * @return string|bool: String if successful; false upon failure
  91          */
  92         public static function encode( $value, $pretty = false, $escaping = 0 ) {
  93                 if ( version_compare( PHP_VERSION, '5.4.0', '<' ) ) {
  94                         return self::encode53( $value, $pretty, $escaping );
  95                 }
  96                 return self::encode54( $value, $pretty, $escaping );
  97         }
  98
  99         /**
 100          * Decodes a JSON string.
 101          *
 102          * @param string $value The JSON string being decoded
 103          * @param bool $assoc When true, returned objects will be converted into associative arrays.
 104          *
 105          * @return mixed: the value encoded in JSON in appropriate PHP type.
 106          * Values `"true"`, `"false"`, and `"null"` (case-insensitive) are returned as `true`, `false`
 107          * and `null` respectively. `null` is returned if the JSON cannot be
 108          * decoded or if the encoded data is deeper than the recursion limit.
 109          */
 110         public static function decode( $value, $assoc = false ) {
 111                 return json_decode( $value, $assoc );
 112         }
 113
 114         /**
 115          * JSON encoder wrapper for PHP >= 5.4, which supports useful encoding options.
 116          *
 117          * @param mixed $value
 118          * @param bool $pretty
 119          * @param int $escaping
 120          * @return string|bool
 121          */
 122         private static function encode54( $value, $pretty, $escaping ) {
 123                 // PHP escapes '/' to prevent breaking out of inline script blocks using '</script>',
 124                 // which is hardly useful when '<' and '>' are escaped, and such escaping negatively
 125                 // impacts the human readability of URLs and similar strings.
 126                 $options = JSON_UNESCAPED_SLASHES;
 127                 $options |= $pretty ? JSON_PRETTY_PRINT : 0;
 128                 $options |= ( $escaping & self::UTF8_OK ) ? JSON_UNESCAPED_UNICODE : 0;
 129                 $options |= ( $escaping & self::XMLMETA_OK ) ? 0 : ( JSON_HEX_TAG | JSON_HEX_AMP );
 130                 $json = json_encode( $value, $options );
 131                 if ( $json === false ) {
 132                         return false;
 133                 }
 134                 if ( $escaping & self::UTF8_OK ) {
 135                         $json = str_replace( self::$badChars, self::$badCharsEscaped, $json );
 136                 }
 137                 return $json;
 138         }
 139
 140         /**
 141          * JSON encoder wrapper for PHP 5.3, which lacks native support for some encoding options.
 142          * Therefore, the missing options are implemented here purely in PHP code.
 143          *
 144          * @param mixed $value
 145          * @param bool $pretty
 146          * @param int $escaping
 147          * @return string|bool
 148          */
 149         private static function encode53( $value, $pretty, $escaping ) {
 150                 $options = ( $escaping & self::XMLMETA_OK ) ? 0 : ( JSON_HEX_TAG | JSON_HEX_AMP );
 151                 $json = json_encode( $value, $options );
 152                 if ( $json === false ) {
 153                         return false;
 154                 }
 155                 $json = str_replace( '\\/', '/', $json ); // emulate JSON_UNESCAPED_SLASHES
 156                 if ( $escaping & self::UTF8_OK ) {
 157                         // JSON hex escape sequences follow the format \uDDDD, where DDDD is four hex digits
 158                         // indicating the equivalent UTF-16 code unit's value. To most efficiently unescape
 159                         // them, we exploit the JSON extension's built-in decoder.
 160                         // * We escape the input a second time, so any such sequence becomes \\uDDDD.
 161                         // * To avoid interpreting escape sequences that were in the original input,
 162                         //   each double-escaped backslash (\\\\) is replaced with \\\u005c.
 163                         // * We strip one of the backslashes from each of the escape sequences to unescape.
 164                         // * Then the JSON decoder can perform the actual unescaping.
 165                         $json = str_replace( "\\\\\\\\", "\\\\\\u005c", addcslashes( $json, '\"' ) );
 166                         $json = json_decode( preg_replace( "/\\\\\\\\u(?!00[0-7])/", "\\\\u", "\"$json\"" ) );
 167                         $json = str_replace( self::$badChars, self::$badCharsEscaped, $json );
 168                 }
 169                 return $pretty ? self::prettyPrint( $json ) : $json;
 170         }
 171
 172         /**
 173          * Adds non-significant whitespace to an existing JSON representation of an object.
 174          * Only needed for PHP < 5.4, which lacks the JSON_PRETTY_PRINT option.
 175          *
 176          * @param string $json
 177          * @return string
 178          */
 179         private static function prettyPrint( $json ) {
 180                 $buf = '';
 181                 $indent = 0;
 182                 $json = strtr( $json, array( '\\\\' => '\\\\', '\"' => "\x01" ) );
 183                 for ( $i = 0, $n = strlen( $json ); $i < $n; $i += $skip ) {
 184                         $skip = 1;
 185                         switch ( $json[$i] ) {
 186                                 case ':':
 187                                         $buf .= ': ';
 188                                         break;
 189                                 case '[':
 190                                 case '{':
 191                                         $indent++; // falls through
 192                                 case ',':
 193                                         $buf .= $json[$i] . "\n" . str_repeat( '    ', $indent );
 194                                         break;
 195                                 case ']':
 196                                 case '}':
 197                                         $indent--;
 198                                         $buf .= "\n" . str_repeat( '    ', $indent ) . $json[$i];
 199                                         break;
 200                                 case '"':
 201                                         $skip = strcspn( $json, '"', $i + 1 ) + 2;
 202                                         $buf .= substr( $json, $i, $skip );
 203                                         break;
 204                                 default:
 205                                         $skip = strcspn( $json, ',]}"', $i + 1 ) + 1;
 206                                         $buf .= substr( $json, $i, $skip );
 207                         }
 208                 }
 209                 return str_replace( "\x01", '\"', preg_replace( '/ +$/m', '', $buf ) );
 210         }
 211 }