|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* This file is part of the ZBateson\MailMimeParser project. |
|
4
|
|
|
* |
|
5
|
|
|
* @license http://opensource.org/licenses/bsd-license.php BSD |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
namespace ZBateson\MailMimeParser\Header\Consumer\Received; |
|
9
|
|
|
|
|
10
|
|
|
use ZBateson\MailMimeParser\Header\Consumer\AbstractGenericConsumerService; |
|
11
|
|
|
use ZBateson\MailMimeParser\Header\Consumer\CommentConsumerService; |
|
12
|
|
|
use ZBateson\MailMimeParser\Header\Part\CommentPart; |
|
13
|
|
|
use ZBateson\MailMimeParser\Header\Part\HeaderPartFactory; |
|
14
|
|
|
|
|
15
|
|
|
/** |
|
16
|
|
|
* Consumes simple literal strings for parts of a Received header. |
|
17
|
|
|
* |
|
18
|
|
|
* Starts consuming when the initialized $partName string is located, for |
|
19
|
|
|
* instance when initialized with "FROM", will start consuming on " FROM" or |
|
20
|
|
|
* "FROM ". |
|
21
|
|
|
* |
|
22
|
|
|
* The consumer ends when any possible "Received" header part is found, namely |
|
23
|
|
|
* on one of the following tokens: from, by, via, with, id, for, or when the |
|
24
|
|
|
* start token for the date stamp is found, ';'. |
|
25
|
|
|
* |
|
26
|
|
|
* The consumer allows comments in and around the consumer... although the |
|
27
|
|
|
* Received header specification only allows them before a part, for example, |
|
28
|
|
|
* technically speaking this is valid: |
|
29
|
|
|
* |
|
30
|
|
|
* "FROM machine (host) (comment) BY machine" |
|
31
|
|
|
* |
|
32
|
|
|
* However, this is not: |
|
33
|
|
|
* |
|
34
|
|
|
* "FROM machine (host) BY machine WITH (comment) ESMTP" |
|
35
|
|
|
* |
|
36
|
|
|
* The consumer will allow both. |
|
37
|
|
|
* |
|
38
|
|
|
* @author Zaahid Bateson |
|
39
|
|
|
*/ |
|
40
|
|
|
class GenericReceivedConsumerService extends AbstractGenericConsumerService |
|
41
|
|
|
{ |
|
42
|
|
|
/** |
|
43
|
|
|
* @var string the current part name being parsed. |
|
44
|
|
|
* |
|
45
|
|
|
* This is always the lower-case name provided to the constructor, not the |
|
46
|
|
|
* actual string that started the consumer, which could be in any case. |
|
47
|
|
|
*/ |
|
48
|
|
|
protected $partName; |
|
49
|
|
|
|
|
50
|
|
|
/** |
|
51
|
|
|
* Constructor overridden to include $partName parameter. |
|
52
|
|
|
* |
|
53
|
|
|
*/ |
|
54
|
16 |
|
public function __construct( |
|
55
|
|
|
HeaderPartFactory $partFactory, |
|
56
|
|
|
CommentConsumerService $commentConsumerService, |
|
57
|
|
|
string $partName |
|
58
|
|
|
) { |
|
59
|
16 |
|
parent::__construct($partFactory, [ $commentConsumerService ]); |
|
60
|
16 |
|
$this->partName = $partName; |
|
61
|
|
|
} |
|
62
|
|
|
|
|
63
|
|
|
/** |
|
64
|
|
|
* Returns true if the passed token matches (case-insensitively) |
|
65
|
|
|
* $this->getPartName() with optional whitespace surrounding it. |
|
66
|
|
|
*/ |
|
67
|
11 |
|
protected function isStartToken(string $token) : bool |
|
68
|
|
|
{ |
|
69
|
11 |
|
$pattern = '/^' . \preg_quote($this->partName, '/') . '$/i'; |
|
70
|
11 |
|
return (\preg_match($pattern, $token) === 1); |
|
71
|
|
|
} |
|
72
|
|
|
|
|
73
|
|
|
/** |
|
74
|
|
|
* Returns true if the token matches (case-insensitively) any of the |
|
75
|
|
|
* following, with optional surrounding whitespace: |
|
76
|
|
|
* |
|
77
|
|
|
* o by |
|
78
|
|
|
* o via |
|
79
|
|
|
* o with |
|
80
|
|
|
* o id |
|
81
|
|
|
* o for |
|
82
|
|
|
* o ; |
|
83
|
|
|
*/ |
|
84
|
14 |
|
protected function isEndToken(string $token) : bool |
|
85
|
|
|
{ |
|
86
|
14 |
|
return (\preg_match('/^(by|via|with|id|for|;)$/i', $token) === 1); |
|
87
|
|
|
} |
|
88
|
|
|
|
|
89
|
|
|
/** |
|
90
|
|
|
* Returns a whitespace separator (for filtering ignorable whitespace |
|
91
|
|
|
* between parts), and a separator matching the current part name as |
|
92
|
|
|
* set on $this->partName. |
|
93
|
|
|
* |
|
94
|
|
|
* @return string[] an array of regex pattern matchers |
|
95
|
|
|
*/ |
|
96
|
16 |
|
protected function getTokenSeparators() : array |
|
97
|
|
|
{ |
|
98
|
16 |
|
return [ |
|
99
|
16 |
|
'\s+', |
|
100
|
16 |
|
'(\A\s*|\s+)(?i)' . \preg_quote($this->partName, '/') . '(?-i)(?=\s+)' |
|
101
|
16 |
|
]; |
|
102
|
|
|
} |
|
103
|
|
|
|
|
104
|
|
|
/** |
|
105
|
|
|
* Overridden to combine all part values into a single string and return it |
|
106
|
|
|
* as the first element, followed by any comment elements as subsequent |
|
107
|
|
|
* elements. |
|
108
|
|
|
* |
|
109
|
|
|
* @param \ZBateson\MailMimeParser\Header\Part\IHeaderPart[] $parts |
|
110
|
|
|
* @return \ZBateson\MailMimeParser\Header\Part\IHeaderPart[] |
|
111
|
|
|
*/ |
|
112
|
10 |
|
protected function processParts(array $parts) : array |
|
113
|
|
|
{ |
|
114
|
10 |
|
$strValue = ''; |
|
115
|
10 |
|
$ret = []; |
|
116
|
10 |
|
$filtered = $this->filterIgnoredSpaces($parts); |
|
117
|
10 |
|
foreach ($filtered as $part) { |
|
118
|
10 |
|
if ($part instanceof CommentPart) { |
|
119
|
6 |
|
$ret[] = $part; |
|
120
|
6 |
|
continue; // getValue() is empty anyway, but for clarity... |
|
121
|
|
|
} |
|
122
|
10 |
|
$strValue .= $part->getValue(); |
|
123
|
|
|
} |
|
124
|
10 |
|
\array_unshift($ret, $this->partFactory->newReceivedPart($this->partName, $strValue)); |
|
125
|
10 |
|
return $ret; |
|
126
|
|
|
} |
|
127
|
|
|
} |
|
128
|
|
|
|
zbateson /
mail-mime-parser
