mmucklo / email-parse Goto Github PK

According to RFC 5322, page 12, ! is in the list of allowed characters for atext, but the library doesn't allow it. There's text in the source, but this may apply to the address-spec part:

// Note: check for Exim-banned characters
// See Bug #18749 - Unhandled Exception: 550 Restricted characters in address
if ($curChar == '%' || $curChar == '!')

but this makes perfectly valid display names fail.

According to RFC822 this is a valid email address

The usage section of the readme is out of date.

parseEmailAddresses is not a method.

Update dependencies to support PHP >=7.1 and remove the composer.lock file which shouldn't be checked in.

Also update PHPUnit, and other packages that are depended upon.

It would be great if the parser showed comments as a specific attribute, see https://tools.ietf.org/html/rfc5322#section-3.2.2

Informal example:
comments are allowed with parentheses at either end of the local-part; e.g. john.smith(comment)@example.com and (comment)[email protected] are both equivalent to [email protected].

$sendingAddress = "Example | Example <[email protected]>";
$result = \Email\Parse::getInstance()->parse($sendingAddress, false);
$sendingDomain = strtolower($result['domain']);
print '$sendingDomain = ' . $sendingDomain;
$sendingAddress = strtolower($result['local_part']) . '@' . $sendingDomain;
print ': $sendingAddress = ' . $sendingAddress;

$sendingDomain =
$sendingAddress = @

Stripping out the '|' character before parsing resolves the issue.

It looks that the underlying zendframework/zend-validator package is using the container-interop/container-interop package which is now deprecated. The zend validatior package is also now abandoned https://github.com/zendframework/zend-validator

mmucklo/email-parse 2.0.0 email-parse a (reasonably) RFC822 / RF2822-compliant library for batch parsing multiple (and single) email addresses
├──php >=5.6.0
├──psr/log ~1.0
│  └──php >=5.3.0
├──true/punycode ~2.0
│  ├──php >=5.3.0
│  └──symfony/polyfill-mbstring ^1.3
│     └──php >=5.3.3
└──zendframework/zend-validator >=2.0,<=3.0
   ├──container-interop/container-interop ^1.1
   │  └──psr/container ^1.0
   │     └──php >=5.3.0
   ├──php ^7.1
   └──zendframework/zend-stdlib ^3.2.1
      └──php ^5.6 || ^7.0

Right now scrutinizer is reporting 72% code coverage.

It would be good to increase that to above 90%

Please fix it i use this character and fatal error "_"

please add to your tests: "Supports (E-mail)" <[email protected]>

zendframework/zend-mail#148

ps: it parses correctly here.

Fails to install with composer:

oodle/email-parse dev-master requires pear-pear/net_idna2 * -> no matching package found.

[email protected]; [email protected] doesn't seem to be parsed here.
can you add support?

also reported in zend-mail:
zendframework/zend-mail#147

not sure why this is an invalid email

Dogster dogster_351A7140D2F608917B168757AC6E2E47911F5D180AD881092669F00B7081D5F5@reply.dogster.com

This is coming directly from an email header. Is it the code or is it indeed invalid?

It would be good to do a little refactoring to improve our scrutinizer rating.

One concern of mine is if I modularize things too much that the performance of the code may suffer as a result due to things like function-call overhead in tight loops, nevertheless it's probably less than simple to maintain at this point.

Although there is plenty of complex code in things like say "gcc" or other tools that are the bread and butter of our industry, so it may make sense to let it be and focus more on test coverage...