Problem with named parameters in routes

[theDisco]

I think router has problem with curly bracket. This one works correctly:

$router->add( 
    '/activate/{token:[a-zA-Z0-9]+}', 
    array (
        'module' => 'frontend',
        'controller' => 'user',
        'action'     => 'activate'
    )
)->setName( 'frontend-user-activate' );

but this one does not match:

$router->add( 
    '/activate/{token:[a-zA-Z0-9]{32}}', 
    array (
        'module' => 'frontend',
        'controller' => 'user',
        'action'     => 'activate'
    )
)->setName( 'frontend-user-activate' );

Because curly brackets are used to enclose parameters by name, use them as repetition quantifier may confuse the router. A raw regular expression can be used to avoid that:

$router->add( 
    '#^/activate/([a-zA-Z0-9]{32})$#', 
    array (
        'module' => 'frontend',
        'controller' => 'user',
        'action'     => 'activate',
        'token' => 1
    )
)->setName( 'frontend-user-activate' );

[trsrm]

OMG! In this case we are able to use short syntax in route pattern without chars limit or pattern with chars limit without short syntax. But we can't use both. I think this is a bug, because there are cases when we are forced to use both (for example - URLs generation using Phalcon\Mvc\Url).

Yep, this is a bug, this will be addressed in 0.6.1

[trsrm]

OK, thank you!

There is no an easy fix for this since named parameters are delimited with { } and quantifiers too. The router uses a regular expression to extract the contents of the named parameters, but the quantifier tells to the router that the named parameter has ended producing this behavior.

Something like this is done by Phalcon internally:

preg_match_all(
    '#{([a-zA-Z][a-zA-Z0-9\_\-]*)(:([^}]+))*}#', 
    '/activate/{token:[a-zA-Z0-9]{32}}', 
    $matches, 
    PREG_SET_ORDER
);
print_r($matches);

If anyone can make a regular expression that doesn't have this problem I could port it to C.

[theDisco]

would this one do the trick?

{([a-zA-Z0-9\_\-]*)(:(.*)}?)}

preg_match_all(
    '#{([a-zA-Z0-9\_\-]*)(:(.*)}?)}#',
    '/activate/{token:[a-zA-Z0-9]{32}}',
    $matches,
    PREG_SET_ORDER
);
print_r($matches);

preg_match_all(
    '#{([a-zA-Z0-9\_\-]*)(:(.*)}?)}#',
    '/activate/{token:[a-zA-Z0-9]+}',
    $matches,
    PREG_SET_ORDER
);
print_r($matches);

it works! thanks :), I'm adding this to 0.6.1

I'm trying with this another:

preg_match_all(
    '#{([a-zA-Z][a-zA-Z0-9\_\-]*)(:(.*))?}#',
    '/manual/{language:[a-z]{2}}/{file:[a-z]+}.html',
    $matches,
    PREG_SET_ORDER
);

Printing $matches returns:

Array
(
    [0] => Array
        (
            [0] => {language:[a-z]{2}}/{file:[a-z]+}
            [1] => language
            [2] => :[a-z]{2}}/{file:[a-z]+
            [3] => [a-z]{2}}/{file:[a-z]+
        )

)

The .* is stopping until the last } :(

[theDisco]

What would be the expected result in this case?

Edit:
I just used the old one to check the expected result

A match for each named variable:

Array
(
    [0] => Array
        (
            [0] => {language:[a-z]{2}}
            [1] => language
            [2] => :[a-z]{2}
            [3] => [a-z]{2}
        )

    [1] => Array
        (
            [0] => {file:[a-z]+}
            [1] => file
            [2] => :[a-z]+
            [3] => [a-z]+
        )

)

[theDisco]

ok, second try:

{([a-zA-Z0-9\_\-]*)(:([^}]+}?))*}

[niden]

This one actually returns null with the previous preg_match

On Wed, Nov 7, 2012 at 3:40 PM, Wojtek Gancarczyk
notifications@github.comwrote:

{([a-zA-Z0-9_-])(:([^}]+}?))}

[theDisco]

Well here is my env:

$ /tmp php -v
PHP 5.4.7 (cli) (built: Oct  1 2012 09:56:15) 
Copyright (c) 1997-2012 The PHP Group
Zend Engine v2.4.0, Copyright (c) 1998-2012 Zend Technologies

and here the scripts I have used to test this regex:

<?php
preg_match_all(
    '#{([a-zA-Z0-9\_\-]*)(:([^}]+}?))*}#',
    '/manual/{language:[a-z]{2}}/{file:[a-z]+}.html',
    $matches,
    PREG_SET_ORDER
);
print_r($matches);

Array
(
    [0] => Array
        (
            [0] => {language:[a-z]{2}}
            [1] => language
            [2] => :[a-z]{2}
            [3] => [a-z]{2}
        )

    [1] => Array
        (
            [0] => {file:[a-z]+}
            [1] => file
            [2] => :[a-z]+
            [3] => [a-z]+
        )

)

<?php
preg_match_all(
    '#{([a-zA-Z0-9\_\-]*)(:([^}]+}?))*}#',
    '/activate/{token:[a-zA-Z0-9]{32}}',
    $matches,
    PREG_SET_ORDER
);
print_r($matches);

Array
(
    [0] => Array
        (
            [0] => {token:[a-zA-Z0-9]{32}}
            [1] => token
            [2] => :[a-zA-Z0-9]{32}
            [3] => [a-zA-Z0-9]{32}
        )

)

<?php
preg_match_all(
    '#{([a-zA-Z0-9\_\-]*)(:([^}]+}?))*}#',
    '/activate/{token:[a-zA-Z0-9]+}',
    $matches,
    PREG_SET_ORDER
);
print_r($matches);

Array
(
    [0] => Array
        (
            [0] => {token:[a-zA-Z0-9]+}
            [1] => token
            [2] => :[a-zA-Z0-9]+
            [3] => [a-zA-Z0-9]+
        )

)

Am I missing something?