ReceivedHeader: incorrectly parse

[mariuszkrzaczkowski]

I found a strange bug, the problem is probably in the word 'id'

My code:

$message = \ZBateson\MailMimeParser\Message::from($contents, false);
foreach ($message->getAllHeadersByName('Received') as $received) {
	echo 'FromName: ' . $received->getFromName() . PHP_EOL;
	echo 'FromHostname: ' . $received->getFromHostname() . PHP_EOL;
	echo 'FromAddress: ' . $received->getFromAddress() . PHP_EOL;
	echo 'ByName: ' . $received->getByName() . PHP_EOL;
	echo 'ByHostname: ' . $received->getByHostname() . PHP_EOL;
	echo 'ByAddress: ' . $received->getByAddress() . PHP_EOL;
	echo 'with: ' . $received->getValueFor('with') . PHP_EOL;
	echo 'getComments: ' . implode(' | ', $received->getComments()) . '<hr>';
}

Example 1 (working properly)
Code:

$contents = 'Received: from mail.uiii.ac.idd ([111.222.333.444])
by mail.yetiforce.com with esmtps (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256
(Exim 4.94.2)
(envelope-from <noreply@mail.uiii.ac.idd>)
id 1mrMVK-00079p-W2
for m.krzaczkowski@yetiforce.com; Sun, 28 Nov 2021 16:54:15 +0100';

Response:

FromName: mail.uiii.ac.idd
FromHostname: 
FromAddress: 111.222.333.444
ByName: 
ByHostname: 
ByAddress: 
with: 
getComments: 

Example 1 (bug)
Code:

$contents = 'Received: from mail.uiii.ac.id ([111.222.333.444])
by mail.yetiforce.com with esmtps (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256
(Exim 4.94.2)
(envelope-from <noreply@mail.uiii.ac.id>)
id 1mrMVK-00079p-W2
for m.krzaczkowski@yetiforce.com; Sun, 28 Nov 2021 16:54:15 +0100';

Response:

FromName: mail.uiii.ac.
FromHostname: 
FromAddress: 
ByName: 
ByHostname: 
ByAddress: 
with: 
getComments: [111.222.333.444]

difference: mail.uiii.ac.idd VS mail.uiii.ac.id

[zbateson]

Aah, it's because there's a section called 'id' so it starts a sub-consumer at that point. Should be an easy enough fix to the regex here:

mail-mime-parser/src/Header/Consumer/Received/GenericReceivedConsumer.php

Line 91 in fc0e05d

$pattern = '/^\s*(' . preg_quote($this->getPartName(), '/') . ')\s*$/i';

In most cases the spaces before/after aren't optional except for 'from' which I think is required and always comes at the start but would have to refresh my memory on it. So it might make more sense to have a 'FromReceivedConsumer' that uses \s*(from)\s+ and the rest have \s+(token)\s+.

Lastly some better testing needs to be added.

Thanks for reporting, @mariuszkrzaczkowski

[mariuszkrzaczkowski]

some solution ?? it blocks my job

[zbateson]

some solution ?? it blocks my job

Feel free to submit a PR with a fix.

[mariuszkrzaczkowski]

I have no idea how to fix it and where

[mariuszkrzaczkowski]

if I knew it, I would have created PR a long time ago

[mariuszkrzaczkowski]

In most cases the spaces before/after aren't optional except for 'from' which I think is required and always comes at the start but would have to refresh my memory on it. So it might make more sense to have a 'FromReceivedConsumer' that uses \s*(from)\s+ and the rest have \s+(token)\s+.

I tried what you wrote but it didn't work

[zbateson]

Fixed in 2.2.0.