ZBateson\MailMimeParser\Header\Consumer\QuotedStringMimeLiteralPartTokenSplitPatternTrait

Complexity

Total Complexity

1

Size/Duplication

Total Lines	18
Duplicated Lines	0 %

Test Coverage

Coverage

100%

Importance

Changes	1
Bugs	0	Features	0

1 Method

Rating	Name	Duplication	Size	Complexity
A	getTokenSplitPattern()	0	5	1

<?php
/**
 * This file is part of the ZBateson\MailMimeParser project.
 *
 * @license http://opensource.org/licenses/bsd-license.php BSD
 */

namespace ZBateson\MailMimeParser\Header\Consumer;

use ZBateson\MailMimeParser\Header\Part\MimeToken;

/**
 * Provides a getTokenSplitPattern for consumers that could have quoted parts
 * that are mime-header-encoded.
 *
 * @author Zaahid Bateson
 */
trait QuotedStringMimeLiteralPartTokenSplitPatternTrait
{
    /**
     * Overridden to use a specialized regex for finding mime-encoded parts
     * (RFC 2047).
     *
     * Some implementations seem to place mime-encoded parts within quoted
     * parameters, and split the mime-encoded parts across multiple split
     * parameters.  The specialized regex doesn't allow double quotes inside a
     * mime encoded part, so it can be "continued" in another parameter.
     *
     * @return string the regex pattern
     */
    protected function getTokenSplitPattern() : string
    {
        $sChars = \implode('|', $this->getAllTokenSeparators());

It seems like getAllTokenSeparators() must be provided by classes using this trait. How about adding it as abstract method to this trait?

        $mimePartPattern = MimeToken::MIME_PART_PATTERN_NO_QUOTES;
        return '~(' . $mimePartPattern . '|\\\\\r\n|\\\\.|' . $sChars . ')~ms';
    }
}