Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:125306
Precedence: bulk
MIME-Version: 1.0
References: <CAJp_myXPpgmxkJautxUiADLz_fgVNv4RJSkpz_duAao9acSojA@mail.gmail.com>
 <C9BD8302-0F81-485E-B306-D865A702DFBC@getmailspring.com>
In-Reply-To: <C9BD8302-0F81-485E-B306-D865A702DFBC@getmailspring.com>
Date: Tue, 27 Aug 2024 09:46:37 +0200
Message-ID: <CAJmjfvUuSxZKmmEr_izLrLOgNdZcJeSsxHXtfjuzT9n7itbcuQ@mail.gmail.com>
Subject: Re: [PHP-DEV] [RFC] Default expression
To: John Coggeshall <john@coggeshall.org>
Cc: "Matthew Weier O'Phinney" <mweierophinney@gmail.com>, Larry Garfield <larry@garfieldtech.com>, 
	php internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary="000000000000aa16b90620a5710b"
From: jakob@givoni.dk (Jakob Givoni)

--000000000000aa16b90620a5710b
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Aug 27, 2024 at 6:23=E2=80=AFAM John Coggeshall <john@coggeshall.or=
g> wrote:

>
> But let's not just say "no expressions" - I think there's been some
> demonstrated value from a lot of these.
>
>
> Honest to god I don't think I've read from one person to come out and say
> "no expressions".
>

Well...
I think it's causing more confusion than insight when we talk about blanket
"expressions" in the context of "default" here.

Expressions in function calls were always permitted:

foo(1+2); // Evaluating an expression and sending the result as the value
of the first parameter

The question is whether to allow OPERATIONS on "default".
Notice the difference here:

1.
foo($userValue ?? default); // No operations performed on default, very
useful feature! This is still an expression involving the keyword default.

2.
foo(default + 2); // "default" must evaluate to a value so we can perform
operations on it. This is also an expression, but I would not be in favor
of this behavior.

So the real 3 options are:
1. No operations on "default". Treat "default" as a placeholder and send it
as such to the function. Any expression that evaluates to "default" is
fine. My opinion: Useful.
2. Allow any operations on "default". Extract the value and treat "default"
as any other "variable". My opinion: Dirty, possibly useful to some, but I
wouldn't use it, and would rather not find it in my code.
3. Allow SOME operations on "default", and block others. My opinion:
Nightmare. Both discussing which operations and why, documenting it and
maintaining it. I'm out :-)

So let's STOP talking about yes or no to EXPRESSIONS and START talking
about yes or no to OPERATIONS.

Best,
Jakob

--000000000000aa16b90620a5710b
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">On Tue, Aug 27, 2024 at 6:23=E2=80=AF=
AM John Coggeshall &lt;<a href=3D"mailto:john@coggeshall.org">john@coggesha=
ll.org</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"m=
argin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left=
:1ex"><div><div class=3D"gmail_quote"><div><br></div><blockquote>But let=
9;s not just say &quot;no expressions&quot; - I think there&#39;s been some=
 demonstrated value from a lot of these.</blockquote><br><div>Honest to god=
 I don&#39;t think I&#39;ve read from one person to come out and say &quot;=
no expressions&quot;.=C2=A0</div></div></div></blockquote><div><br></div><d=
iv>Well...</div><div>I think it&#39;s causing more confusion than insight w=
hen we talk about blanket &quot;expressions&quot; in the context of &quot;d=
efault&quot; here.</div><div><br></div><div>Expressions in function calls w=
ere always permitted:</div><div><br></div><div>foo(1+2); // Evaluating an e=
xpression and sending the result as the value of the first parameter</div><=
div><br></div><div>The question is whether to allow OPERATIONS on &quot;def=
ault&quot;.</div><div>Notice the difference here:</div><div><br></div><div>=
1.</div><div>foo($userValue ?? default); // No operations performed on defa=
ult, very useful feature! This is still an expression involving the keyword=
 default.</div><div><br></div><div>2.</div><div>foo(default=C2=A0+ 2); // &=
quot;default&quot; must evaluate to a value so we can perform operations on=
 it. This is also an=C2=A0expression, but I would not be in favor of this b=
ehavior.</div><div><br></div><div>So the real 3 options are:</div><div>1. N=
o operations on &quot;default&quot;. Treat &quot;default&quot; as a placeho=
lder and send it as such to the function. Any expression that evaluates to =
&quot;default&quot; is fine. My opinion: Useful.</div><div>2. Allow any ope=
rations on &quot;default&quot;. Extract the value and treat &quot;default&q=
uot; as any other &quot;variable&quot;. My opinion: Dirty, possibly useful =
to some, but I wouldn&#39;t use it, and would rather not find it in my code=
.</div><div>3. Allow SOME operations on &quot;default&quot;, and block othe=
rs. My opinion: Nightmare. Both discussing which operations and why, docume=
nting it and maintaining it. I&#39;m out :-)</div><div><br></div><div>So le=
t&#39;s STOP talking about yes or no to EXPRESSIONS and START talking about=
 yes or no to OPERATIONS.</div><div><br></div><div>Best,</div><div>Jakob</d=
iv><div>=C2=A0</div></div></div>

--000000000000aa16b90620a5710b--