Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:61226
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 209.85.212.182 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <CAKxcST8Xc5F=NtYWnFTXhfKB8+KwG8gB_V0VjVSW9w0QuKDXww@mail.gmail.com>
References: <CAN+gjSeiE-0S4hpwZkZfyc93QQgGPtAy+SbdQZ=GmYB4ZcFcDQ@mail.gmail.com>
	<CAN+gjSfHNFXF==gMpu3fLOooMepC+1bn5XJimtTcVVhRHyJP7Q@mail.gmail.com>
	<CAL9B3uxU0vyW4_Q3y-4Rt5g837GX9dq2fQQYuCEOMZ1zRhHGLQ@mail.gmail.com>
	<5000BF44.3080907@sugarcrm.com>
	<CAAyV7nGTQHgFLyeiXavdY0x5wVh=kBr75vp_FWdHgcgU1TAxNQ@mail.gmail.com>
	<CAKxcST8Xc5F=NtYWnFTXhfKB8+KwG8gB_V0VjVSW9w0QuKDXww@mail.gmail.com>
Date: Sat, 14 Jul 2012 09:28:40 -0600
Message-ID: <CAGsnKxYsAdcwDfgz-pC+tRpXi44CHqrji_aGi2mSjgR-01rHJA@mail.gmail.com>
To: Paul Dragoonis <dragoonis@gmail.com>
Cc: Anthony Ferrara <ircmaxell@gmail.com>, Stas Malyshev <smalyshev@sugarcrm.com>, 
	Brandon Wamboldt <brandon.wamboldt@gmail.com>, Andrew Faulds <ajfweb@googlemail.com>, 
	internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary=f46d04428104ee32ae04c4cbd950
Subject: Re: [PHP-DEV] Re: array_last_key() function
From: john.lesueur@gmail.com (John LeSueur)

--f46d04428104ee32ae04c4cbd950
Content-Type: text/plain; charset=ISO-8859-1

On Sat, Jul 14, 2012 at 4:34 AM, Paul Dragoonis <dragoonis@gmail.com> wrote:

> On Sat, Jul 14, 2012 at 3:04 AM, Anthony Ferrara <ircmaxell@gmail.com>
> wrote:
> > Stas,
> >
> >> I like this idea. array_first_key would be nice too
> >>
> >> I am probably missing something, but what those would allow to do that
> >> rewind/end+key() doesn't do?
> >>
> >
> > The big thing that it does that end()+key() doesn't do is really what it
> > doesn't do (update the internal pointer).
> >
> > end() modifies the array pointer. So if you try this in user-land:
> >
> > function array_last_key(array $array) {
> >     end($array);
> >     return key($array);
> > }
> >
> > It will always force a full copy of the array (at least the hash table).
> > Implementing this as a core function however would eliminate that one
> > side-effect...
>
> and it beats doing $var = null; if(isset($array[0])) { $var = $array[0]; }
>
> Not sure that does the trick(you're pulling values, not keys, and assuming
0-based array), but...

What about reset(array_keys($array)); or end(array_keys($array)); which
only makes copies of the keys?
or even better key(array_slice($array, 0, 1, true)); or
key(array_slice($array,-1,1,true)); which makes a copy of one element of
the array?

I'm of the opinion that we mostly have enough array_* functions, and new
ones have to provide a significant advantage to be accepted. If these
functions are significantly faster than the alternatives, then perhaps they
have merit, but they seem like convenience functions, rather than
fundamentals.

John

--f46d04428104ee32ae04c4cbd950--