Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:25367
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Message-ID: <002c01c6c11c$bb024110$0201a8c0@pc1>
To: <internals@lists.php.net>, "Pierre" <pierre.php@gmail.com>
References: <005301c6bec7$783e5c80$0201a8c0@pc1> <fe05d1540608130508n73b36be8qefc087f56898448c@mail.gmail.com> <006d01c6bedd$c86c7a00$0201a8c0@pc1> <fe05d1540608130715h4acd0758t4a690deb34fec3c6@mail.gmail.com> <35405.67.108.68.40.1155598740.squirrel@www.l-i-e.com> <fe05d1540608141644r471745c4i608eab1b904c81fc@mail.gmail.com> <004801c6c040$591276a0$0201a8c0@pc1> <fe05d1540608150558l7828bddaya102644d9f2f6c90@mail.gmail.com> <001e01c6c115$e69bec10$0201a8c0@pc1> <fe05d1540608160247k124a3e7cy44aa1450ad5a5e1d@mail.gmail.com>
Date: Wed, 16 Aug 2006 05:14:16 -0500
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Subject: Re: [PHP-DEV] [PATCH] array_count_values bug: mishandling of numbers with leading whitespace
From: php_lists@realplain.com ("Matt W")

Hi Pierre,

----- Original Message -----
From: "Pierre"
Sent: Wednesday, August 16, 2006


> Hello Matt,
>
> > > Which part of "We need to sit down and come up with a proposal?" did
> > > you not understand? seriously?
> >
> > Nothing. :-)  But the patch is all I know of, so that was my proposal.
> > Maybe with the "sitting down," "if" to change this function would be the
> > decision, but not "what" to change, as I only see 2 options: Do nothing;
or
> > fix the bug, which if there's a change, I'm pretty sure the behavior my
> > patch provides will be the final result.  I hope you guys DO see that
> > there's a bug, and are simply wanting to be careful with changes.
>
> We are runing in circles, I do not have the times for that sorry. I

I know, I'm sorry! :-)

> answered many times the questions regarding your hope, explaining the
> best way to take to solve this specific issue and all other related to
> numeric strings. Andrei did too. have nothing else to say for now :)
>
> I will start the draft next week, let me know if you like to help.

Sure, I could, but I'm not sure how much I'd help. :-/  If it's on the list,
I can offer comments or such regarding anything I think (if this is mostly
is_numeric* stuff you will draft) -- I thought I'd have a quick commented
example of what *I* thought of for is_numeric* by now, but didn't get to it
yet.  I wanted to reply to your message from last week in the is_numeric*
thread...

One last thing about the array function I've not mentioned...  ('Course I
realized after the last mail -- I tried this the other day, but it was my
patched code (oops), so it didn't mess up.)  On 32-bit, using string value
of 2^32-1:

$a['2147483647']=1;
$b=array_count_values(array('2147483647'));
var_dump($a, $b);
array(1) {
  ["2147483647"]=>
  int(1)
}
array(1) {
  [2147483647]=>
  int(1)
}

Again, it's behaving differently by converting that value to number.  (Note:
To be fair, I'd say in this case, it's a tiny bug in regular array key
handling ("symtable" functions), but I'll keep that for later -- super
simple to fix, I think.  And safe. :-))

> --Pierre

Don't need to waste time replying to this. ;-)  Thanks for your time,
Matt