2011/11/2 Etienne Kneuss <col...@php.net>:
> Hi,
>
> On Wed, Nov 2, 2011 at 17:27, Stas Malyshev <smalys...@sugarcrm.com> wrote:
>> Hi!
>>
>>> What about array_diff(array("Array"), array(array()))? To me the sole
>>> act of having an array in it is an almost sure indication that the code
>>> is wrong somewhere... IMO it deserves a notice.
>>
>> That one would have the arrays different.
>
> Well, they are not different elements w.r.t array_diff, so it does
> return array() since (string)"Array" === (string)array().
>
> My point is: there is an implicit usage of string representations in
> those functions due to the cast. And it can yield various unexpected
> results.
>
> Another example is:
>
> array_intersect(array(array(1)), array(array(2))) === array(array(1)).
> Even though array(1) != array(2).
>
> To me it falls exactly within what the patch tries to achieve: report
> potential unintentional misuses of arrays.
That was indeed my exact intention when doing that patch.
>> Having multi-dimensonal arrays is
>> not an error, comparing arrays of different structure is not an error
>> either. Consider expression: "a" == array(). Is it an error? No, it's just a
>> false expression.
>
> Right, but that's not the question here. We are not normally comparing
> the elements, arrays or not, we are comparing their string
> representations, which in the case of arrays makes almost no sense to
> compare.
Exactly.
No more no less than you expect:
array_diff( array( new stdClass ), array( "stdClass" ) );
to produce:
PHP Catchable fatal error: Object of class stdClass could not be
converted to string in ...
>> In the same way all comparisons where you compare string
>> to array should just say it's not equal, since we know they're not equal.
>
> I surely don't expect it to issue a Notice if I do $var == "asd" and
> $var turns out to be an array. But then again, it's not the question
> here.
>
>>
>> --
>> Stanislav Malyshev, Software Architect
>> SugarCRM: http://www.sugarcrm.com/
>> (408)454-6900 ext. 227
>>
>
>
>
> --
> Etienne Kneuss
> http://www.colder.ch
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
