FormatJson.php

Go to the documentation of this file.

<?php
class FormatJson {

    const UTF8_OK = 1;

    const XMLMETA_OK = 2;

    const ALL_OK = 3;

    const WS_CLEANUP_REGEX = '/(?<=[\[{])\n\s*+(?=[\]}])/';

    private static $badChars = array(
        "\xe2\x80\xa8", // U+2028 LINE SEPARATOR
        "\xe2\x80\xa9", // U+2029 PARAGRAPH SEPARATOR
    );

    private static $badCharsEscaped = array(
        '\u2028', // U+2028 LINE SEPARATOR
        '\u2029', // U+2029 PARAGRAPH SEPARATOR
    );

    public static function encode( $value, $pretty = false, $escaping = 0 ) {
        if ( defined( 'JSON_UNESCAPED_UNICODE' ) ) {
            return self::encode54( $value, $pretty, $escaping );
        }
        return self::encode53( $value, $pretty, $escaping );
    }

    public static function decode( $value, $assoc = false ) {
        return json_decode( $value, $assoc );
    }

    private static function encode54( $value, $pretty, $escaping ) {
        // PHP escapes '/' to prevent breaking out of inline script blocks using '</script>',
        // which is hardly useful when '<' and '>' are escaped (and inadequate), and such
        // escaping negatively impacts the human readability of URLs and similar strings.
        $options = JSON_UNESCAPED_SLASHES;
        $options |= $pretty ? JSON_PRETTY_PRINT : 0;
        $options |= ( $escaping & self::UTF8_OK ) ? JSON_UNESCAPED_UNICODE : 0;
        $options |= ( $escaping & self::XMLMETA_OK ) ? 0 : ( JSON_HEX_TAG | JSON_HEX_AMP );
        $json = json_encode( $value, $options );
        if ( $json === false ) {
            return false;
        }

        if ( $pretty ) {
            // Remove whitespace inside empty arrays/objects; different JSON encoders
            // vary on this, and we want our output to be consistent across implementations.
            $json = preg_replace( self::WS_CLEANUP_REGEX, '', $json );
        }
        if ( $escaping & self::UTF8_OK ) {
            $json = str_replace( self::$badChars, self::$badCharsEscaped, $json );
        }
        return $json;
    }

    private static function encode53( $value, $pretty, $escaping ) {
        $options = ( $escaping & self::XMLMETA_OK ) ? 0 : ( JSON_HEX_TAG | JSON_HEX_AMP );
        $json = json_encode( $value, $options );
        if ( $json === false ) {
            return false;
        }

        // Emulate JSON_UNESCAPED_SLASHES. Because the JSON contains no unescaped slashes
        // (only escaped slashes), a simple string replacement works fine.
        $json = str_replace( '\/', '/', $json );

        if ( $escaping & self::UTF8_OK ) {
            // JSON hex escape sequences follow the format \uDDDD, where DDDD is four hex digits
            // indicating the equivalent UTF-16 code unit's value. To most efficiently unescape
            // them, we exploit the JSON extension's built-in decoder.
            // * We escape the input a second time, so any such sequence becomes \\uDDDD.
            // * To avoid interpreting escape sequences that were in the original input,
            //   each double-escaped backslash (\\\\) is replaced with \\\u005c.
            // * We strip one of the backslashes from each of the escape sequences to unescape.
            // * Then the JSON decoder can perform the actual unescaping.
            $json = str_replace( "\\\\\\\\", "\\\\\\u005c", addcslashes( $json, '\"' ) );
            $json = json_decode( preg_replace( "/\\\\\\\\u(?!00[0-7])/", "\\\\u", "\"$json\"" ) );
            $json = str_replace( self::$badChars, self::$badCharsEscaped, $json );
        }

        if ( $pretty ) {
            return self::prettyPrint( $json );
        }
        return $json;
    }

    private static function prettyPrint( $json ) {
        $buf = '';
        $indent = 0;
        $json = strtr( $json, array( '\\\\' => '\\\\', '\"' => "\x01" ) );
        for ( $i = 0, $n = strlen( $json ); $i < $n; $i += $skip ) {
            $skip = 1;
            switch ( $json[$i] ) {
                case ':':
                    $buf .= ': ';
                    break;
                case '[':
                case '{':
                    ++$indent;
                    // falls through
                case ',':
                    $buf .= $json[$i] . "\n" . str_repeat( '    ', $indent );
                    break;
                case ']':
                case '}':
                    $buf .= "\n" . str_repeat( '    ', --$indent ) . $json[$i];
                    break;
                case '"':
                    $skip = strcspn( $json, '"', $i + 1 ) + 2;
                    $buf .= substr( $json, $i, $skip );
                    break;
                default:
                    $skip = strcspn( $json, ',]}"', $i + 1 ) + 1;
                    $buf .= substr( $json, $i, $skip );
            }
        }
        $buf = preg_replace( self::WS_CLEANUP_REGEX, '', $buf );
        return str_replace( "\x01", '\"', $buf );
    }
}