Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:127073
Feedback-ID: ifab94697:Fastmail
Precedence: bulk
MIME-Version: 1.0
Date: Tue, 08 Apr 2025 18:54:11 +0200
To: Daikaras <webmaster@daikaras.lt>,
 "Niels Dossche" <dossche.niels@gmail.com>,
 "PHP internals" <internals@lists.php.net>
Message-ID: <8071593a-6c6d-4019-b0ab-733207e6c1d1@app.fastmail.com>
In-Reply-To: <e17368e7-1210-4cd2-8213-1fe7b5b3ef13@daikaras.lt>
References: <51fcec92-b2ae-482e-920f-c6c991b526c5@gmail.com>
 <e17368e7-1210-4cd2-8213-1fe7b5b3ef13@daikaras.lt>
Subject: Re: [PHP-DEV] [RFC] [Discussion] array_first() and array_last()
Content-Type: multipart/alternative;
 boundary=cd0249bf901f4384958ffb223aa472d0
From: rob@bottled.codes ("Rob Landers")

--cd0249bf901f4384958ffb223aa472d0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable



On Tue, Apr 8, 2025, at 10:53, Daikaras wrote:
> On 4/5/2025 6:51 PM, Niels Dossche wrote:
> > Hi internals
> >
> > I'm opening the discussion for the RFC "array_first() and array_last=
()".
> > https://wiki.php.net/rfc/array_first_last
> >
> > Kind regards
> > Niels
>=20
>=20
> -1 because returning `null` for empty arrays is still wrong. Whatever=20
> similar behavior exists should be corrected to throw `ValueError` in t=
he=20
> future. Just my 2c.
>=20

I=E2=80=99ve always viewed arrays as an infinite field of nulls. The onl=
y time this isn=E2=80=99t true is when you use array_key_exists() or car=
e about the warnings (which make it less useful, IMHO). When you treat i=
t as an infinite field of nulls, you get some interesting mathematical/c=
omputational properties you can exploit. Trying to treat it as a map/ind=
exed-array is when you start running into weird problems.

=E2=80=94 Rob
--cd0249bf901f4384958ffb223aa472d0
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html><html><head><title></title><style type=3D"text/css">p.Mso=
Normal,p.MsoNoSpacing{margin:0}</style></head><body><div><br></div><div>=
<br></div><div>On Tue, Apr 8, 2025, at 10:53, Daikaras wrote:<br></div><=
blockquote type=3D"cite" id=3D"qt" style=3D""><div>On 4/5/2025 6:51 PM, =
Niels Dossche wrote:<br></div><div>&gt; Hi internals<br></div><div>&gt;<=
br></div><div>&gt; I'm opening the discussion for the RFC "array_first()=
 and array_last()".<br></div><div>&gt;&nbsp;<a href=3D"https://wiki.php.=
net/rfc/array_first_last">https://wiki.php.net/rfc/array_first_last</a><=
br></div><div>&gt;<br></div><div>&gt; Kind regards<br></div><div>&gt; Ni=
els<br></div><div><br></div><div><br></div><div>-1 because returning `nu=
ll` for empty arrays is still wrong. Whatever&nbsp;<br></div><div>simila=
r behavior exists should be corrected to throw `ValueError` in the&nbsp;=
<br></div><div>future. Just my 2c.<br></div><div><br></div></blockquote>=
<div><br></div><div>I=E2=80=99ve always viewed arrays as an infinite fie=
ld of nulls. The only time this isn=E2=80=99t true is when you use array=
_key_exists() or care about the warnings (which make it less useful, IMH=
O). When you treat it as an infinite field of nulls, you get some intere=
sting mathematical/computational properties you can exploit. Trying to t=
reat it as a map/indexed-array is when you start running into weird prob=
lems.</div><div><br></div><div id=3D"sig121229152">=E2=80=94 Rob<br></di=
v></body></html>
--cd0249bf901f4384958ffb223aa472d0--