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 FORCE_ASSOC = 0x100;

    const TRY_FIXING = 0x200;

    const STRIP_COMMENTS = 0x400;

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

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

    private static $badCharsEscaped = [
        '\u2028', // U+2028 LINE SEPARATOR
        '\u2029', // U+2029 PARAGRAPH SEPARATOR
    ];

    public static function encode( $value, $pretty = false, $escaping = 0 ) {
        if ( !is_string( $pretty ) ) {
            $pretty = $pretty ? '    ' : false;
        }

        static $bug66021;
        if ( $pretty !== false && $bug66021 === null ) {
            $bug66021 = json_encode( [], JSON_PRETTY_PRINT ) !== '[]';
        }

        // 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 !== false ? 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 !== false ) {
            // Workaround for <https://bugs.php.net/bug.php?id=66021>
            if ( $bug66021 ) {
                $json = preg_replace( self::WS_CLEANUP_REGEX, '', $json );
            }
            if ( $pretty !== '    ' ) {
                // Change the four-space indent to a tab indent
                $json = str_replace( "\n    ", "\n\t", $json );
                while ( strpos( $json, "\t    " ) !== false ) {
                    $json = str_replace( "\t    ", "\t\t", $json );
                }

                if ( $pretty !== "\t" ) {
                    // Change the tab indent to the provided indent
                    $json = str_replace( "\t", $pretty, $json );
                }
            }
        }
        if ( $escaping & self::UTF8_OK ) {
            $json = str_replace( self::$badChars, self::$badCharsEscaped, $json );
        }

        return $json;
    }

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

    public static function parse( $value, $options = 0 ) {
        if ( $options & self::STRIP_COMMENTS ) {
            $value = self::stripComments( $value );
        }
        $assoc = ( $options & self::FORCE_ASSOC ) !== 0;
        $result = json_decode( $value, $assoc );
        $code = json_last_error();

        if ( $code === JSON_ERROR_SYNTAX && ( $options & self::TRY_FIXING ) !== 0 ) {
            // The most common error is the trailing comma in a list or an object.
            // We cannot simply replace /,\s*[}\]]/ because it could be inside a string value.
            // But we could use the fact that JSON does not allow multi-line string values,
            // And remove trailing commas if they are et the end of a line.
            // JSON only allows 4 control characters: [ \t\r\n].  So we must not use '\s' for matching.
            // Regex match   ,]<any non-quote chars>\n   or   ,\n]   with optional spaces/tabs.
            $count = 0;
            $value =
                preg_replace( '/,([ \t]*[}\]][^"\r\n]*([\r\n]|$)|[ \t]*[\r\n][ \t\r\n]*[}\]])/', '$1',
                    $value, - 1, $count );
            if ( $count > 0 ) {
                $result = json_decode( $value, $assoc );
                if ( JSON_ERROR_NONE === json_last_error() ) {
                    // Report warning
                    $st = Status::newGood( $result );
                    $st->warning( wfMessage( 'json-warn-trailing-comma' )->numParams( $count ) );
                    return $st;
                }
            }
        }

        switch ( $code ) {
            case JSON_ERROR_NONE:
                return Status::newGood( $result );
            default:
                return Status::newFatal( wfMessage( 'json-error-unknown' )->numParams( $code ) );
            case JSON_ERROR_DEPTH:
                $msg = 'json-error-depth';
                break;
            case JSON_ERROR_STATE_MISMATCH:
                $msg = 'json-error-state-mismatch';
                break;
            case JSON_ERROR_CTRL_CHAR:
                $msg = 'json-error-ctrl-char';
                break;
            case JSON_ERROR_SYNTAX:
                $msg = 'json-error-syntax';
                break;
            case JSON_ERROR_UTF8:
                $msg = 'json-error-utf8';
                break;
            case JSON_ERROR_RECURSION:
                $msg = 'json-error-recursion';
                break;
            case JSON_ERROR_INF_OR_NAN:
                $msg = 'json-error-inf-or-nan';
                break;
            case JSON_ERROR_UNSUPPORTED_TYPE:
                $msg = 'json-error-unsupported-type';
                break;
        }
        return Status::newFatal( $msg );
    }

    public static function stripComments( $json ) {
        // Ensure we have a string
        $str = (string) $json;
        $buffer = '';
        $maxLen = strlen( $str );
        $mark = 0;

        $inString = false;
        $inComment = false;
        $multiline = false;

        for ( $idx = 0; $idx < $maxLen; $idx++ ) {
            switch ( $str[$idx] ) {
                case '"':
                    $lookBehind = ( $idx - 1 >= 0 ) ? $str[$idx - 1] : '';
                    if ( !$inComment && $lookBehind !== '\\' ) {
                        // Either started or ended a string
                        $inString = !$inString;
                    }
                    break;

                case '/':
                    $lookAhead = ( $idx + 1 < $maxLen ) ? $str[$idx + 1] : '';
                    $lookBehind = ( $idx - 1 >= 0 ) ? $str[$idx - 1] : '';
                    if ( $inString ) {
                        continue;

                    } elseif ( !$inComment &&
                        ( $lookAhead === '/' || $lookAhead === '*' )
                    ) {
                        // Transition into a comment
                        // Add characters seen to buffer
                        $buffer .= substr( $str, $mark, $idx - $mark );
                        // Consume the look ahead character
                        $idx++;
                        // Track state
                        $inComment = true;
                        $multiline = $lookAhead === '*';

                    } elseif ( $multiline && $lookBehind === '*' ) {
                        // Found the end of the current comment
                        $mark = $idx + 1;
                        $inComment = false;
                        $multiline = false;
                    }
                    break;

                case "\n":
                    if ( $inComment && !$multiline ) {
                        // Found the end of the current comment
                        $mark = $idx + 1;
                        $inComment = false;
                    }
                    break;
            }
        }
        if ( $inComment ) {
            // Comment ends with input
            // Technically we should check to ensure that we aren't in
            // a multiline comment that hasn't been properly ended, but this
            // is a strip filter, not a validating parser.
            $mark = $maxLen;
        }
        // Add final chunk to buffer before returning
        return $buffer . substr( $str, $mark, $maxLen - $mark );
    }
}