Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:129553
Precedence: list
MIME-Version: 1.0
References: <e906b61e-b951-4eb4-aa0f-700e76dae6a9@app.fastmail.com>
 <CAOWwgp=L=YaZDdLmFr+paOVEotU1Lt6S4vAuU+HFTNAo0S0hSQ@mail.gmail.com>
 <d221df4a-cb0c-4729-a7b3-156b0db52ff6@app.fastmail.com> <CAOWwgpk8+18HzhsK235h35b2a2unKH=k+q9ch+Y+j=A+yWegcQ@mail.gmail.com>
 <b95d10f5-df03-483a-b80f-a0b5a478c2bc@app.fastmail.com>
In-Reply-To: <b95d10f5-df03-483a-b80f-a0b5a478c2bc@app.fastmail.com>
Date: Thu, 4 Dec 2025 19:00:29 +0100
Message-ID: <CAOWwgpnQNsB5v--+nny5x2iWA028QNioo1hoA5r_H3QxjGxh4A@mail.gmail.com>
Subject: Re: [PHP-DEV] [RFC] isReadable/isWriteable property reflection
To: Larry Garfield <larry@garfieldtech.com>
Cc: php internals <internals@lists.php.net>
Content-Type: multipart/alternative; boundary="0000000000005c1e110645241b9e"
From: nicolas.grekas+php@gmail.com (Nicolas Grekas)

--0000000000005c1e110645241b9e
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Le jeu. 4 d=C3=A9c. 2025 =C3=A0 18:03, Larry Garfield <larry@garfieldtech.c=
om> a
=C3=A9crit :

> On Thu, Dec 4, 2025, at 10:47 AM, Nicolas Grekas wrote:
> > Le jeu. 4 d=C3=A9c. 2025 =C3=A0 17:39, Larry Garfield <larry@garfieldte=
ch.com> a
> =C3=A9crit :
> >> On Thu, Dec 4, 2025, at 8:05 AM, Nicolas Grekas wrote:
> >> > Le jeu. 6 nov. 2025 =C3=A0 00:28, Larry Garfield <larry@garfieldtech=
.com>
> a =C3=A9crit :
> >> >> In other news, Ilija and I said a year ago that we'd take a swing a=
t
> adding isReadable/isWriteable methods to ReflectionProperty.  Took a whil=
e,
> but here we are.  A strangely small RFC from us:
> >> >>
> >> >> https://wiki.php.net/rfc/isreadable-iswriteable
> >> >>
> >> >
> >> >
> >> > Thanks for this.
> >> > I also think the auto-scope is [NOT] a good idea. You state that 90%
> of use
> >> > cases will need this but my experience doesn't back this claim. The
> >> > only cases where I had to check for read/writeable were out of the
> >> > local scope, so I'd say 100% of my experience goes against that 90%
> >> > number ;) Joke aside, it'd be just fine to let ppl be explicit.
> That's
> >> > better than "oops I forgot to give the correct scope" bugs.
> >> >
> >> > About magic methods, one unsets a property only to have __get/__set
> >> > called. Existing code works with this assumption. This means we have
> to
> >> > return true IMHO. Magic methods are just generic hooks also. Which
> >> > means they should behave the same.
> >> >
> >> > Nicolas
> >>
> >> Well, the only people who seem to have an opinion don't like "static",
> so we've removed it.  RFC updated.
> >>
> >> As for __get/__set, that's so far one vote for ignore (Tim), and one
> for always-true (Nicolas).  Not a consensus. :-)
> >>
> >> Nicolas, can you clarify with an example if/how ignore would break
> things?
> >>
> >> I think once we settle that question and the cooldown passes we're
> ready for a vote, though at this point that means January.
> >>
> >
> > I can try to build a synthetic example but the gist is:
> > A class that starts with only properties and no __get() should be able
> > to move to a __get()-based hooking in a later version without breaking
> > code that uses isReadable().
> > That on its own should be enough to settle the desired behavior :)
>
> So you want to be able to transition from:
>
> class Foo
> {
>   public private(set) string $name;
> }
>
> To
>
> class Foo
> {
>   private array $vals =3D [];
>
>   public function __get(string $name)
>   {
>     return $this->vals[$name] ?? throw new Exception();
>   }
> }
>
> Is that right?  Because both approaches would result in changes in some
> situations there.
>
> If __get is ignored and the value is set, then a global isReadable check
> will go from true to false in that transition.
>
> If __get always returns true and the value is still uninitialized, a
> global isReadable check will go from false to true in that transition.
>
> Either way, that's not a fully safe transition to make.
>
> --Larry Garfield
>

Nah, I'm thinking about a scenario very close to what you talk about in the
RFC: unset + __get

From:
class Foo
{
    public int $abc;

    public function __construct()
    {
        $this->abc =3D stuff();
    }
}

To:
class Foo
{
    public int $abc;

    public function __construct()
    {
        unset($this->abc);
    }

    public function __get($name)
    {
        if ('abc' =3D=3D=3D $name) {
            return $this->abc =3D stuff();
        }
    }
}

--0000000000005c1e110645241b9e
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 g=
mail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">Le=C2=A0jeu. 4 =
d=C3=A9c. 2025 =C3=A0=C2=A018:03, Larry Garfield &lt;<a href=3D"mailto:larr=
y@garfieldtech.com">larry@garfieldtech.com</a>&gt; a =C3=A9crit=C2=A0:<br><=
/div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;bo=
rder-left:1px solid rgb(204,204,204);padding-left:1ex">On Thu, Dec 4, 2025,=
 at 10:47 AM, Nicolas Grekas wrote:<br>
&gt; Le jeu. 4 d=C3=A9c. 2025 =C3=A0 17:39, Larry Garfield &lt;<a href=3D"m=
ailto:larry@garfieldtech.com" target=3D"_blank">larry@garfieldtech.com</a>&=
gt; a =C3=A9crit :<br>
&gt;&gt; On Thu, Dec 4, 2025, at 8:05 AM, Nicolas Grekas wrote:<br>
&gt;&gt; &gt; Le jeu. 6 nov. 2025 =C3=A0 00:28, Larry Garfield &lt;<a href=
=3D"mailto:larry@garfieldtech.com" target=3D"_blank">larry@garfieldtech.com=
</a>&gt; a =C3=A9crit :<br>
&gt;&gt; &gt;&gt; In other news, Ilija and I said a year ago that we&#39;d =
take a swing at adding isReadable/isWriteable methods to ReflectionProperty=
.=C2=A0 Took a while, but here we are.=C2=A0 A strangely small RFC from us:=
<br>
&gt;&gt; &gt;&gt; <br>
&gt;&gt; &gt;&gt; <a href=3D"https://wiki.php.net/rfc/isreadable-iswriteabl=
e" rel=3D"noreferrer" target=3D"_blank">https://wiki.php.net/rfc/isreadable=
-iswriteable</a><br>
&gt;&gt; &gt;&gt; <br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; Thanks for this.<br>
&gt;&gt; &gt; I also think the auto-scope is [NOT] a good idea. You state t=
hat 90% of use <br>
&gt;&gt; &gt; cases will need this but my experience doesn&#39;t back this =
claim. The <br>
&gt;&gt; &gt; only cases where I had to check for read/writeable were out o=
f the <br>
&gt;&gt; &gt; local scope, so I&#39;d say 100% of my experience goes agains=
t that 90% <br>
&gt;&gt; &gt; number ;) Joke aside, it&#39;d be just fine to let ppl be exp=
licit. That&#39;s <br>
&gt;&gt; &gt; better than &quot;oops I forgot to give the correct scope&quo=
t; bugs.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; About magic methods, one unsets a property only to have __get=
/__set <br>
&gt;&gt; &gt; called. Existing code works with this assumption. This means =
we have to <br>
&gt;&gt; &gt; return true IMHO. Magic methods are just generic hooks also. =
Which <br>
&gt;&gt; &gt; means they should behave the same.<br>
&gt;&gt; &gt;<br>
&gt;&gt; &gt; Nicolas<br>
&gt;&gt; <br>
&gt;&gt; Well, the only people who seem to have an opinion don&#39;t like &=
quot;static&quot;, so we&#39;ve removed it.=C2=A0 RFC updated.<br>
&gt;&gt; <br>
&gt;&gt; As for __get/__set, that&#39;s so far one vote for ignore (Tim), a=
nd one for always-true (Nicolas).=C2=A0 Not a consensus. :-)<br>
&gt;&gt; <br>
&gt;&gt; Nicolas, can you clarify with an example if/how ignore would break=
 things?=C2=A0 <br>
&gt;&gt; <br>
&gt;&gt; I think once we settle that question and the cooldown passes we=
9;re ready for a vote, though at this point that means January.<br>
&gt;&gt; <br>
&gt;<br>
&gt; I can try to build a synthetic example but the gist is:<br>
&gt; A class that starts with only properties and no __get() should be able=
 <br>
&gt; to move to a __get()-based hooking in a later version without breaking=
 <br>
&gt; code that uses isReadable().<br>
&gt; That on its own should be enough to settle the desired behavior :)<br>
<br>
So you want to be able to transition from:<br>
<br>
class Foo<br>
{<br>
=C2=A0 public private(set) string $name;<br>
}<br>
<br>
To<br>
<br>
class Foo<br>
{<br>
=C2=A0 private array $vals =3D [];<br>
<br>
=C2=A0 public function __get(string $name)<br>
=C2=A0 {<br>
=C2=A0 =C2=A0 return $this-&gt;vals[$name] ?? throw new Exception();<br>
=C2=A0 }<br>
}<br>
<br>
Is that right?=C2=A0 Because both approaches would result in changes in som=
e situations there.<br>
<br>
If __get is ignored and the value is set, then a global isReadable check wi=
ll go from true to false in that transition.<br>
<br>
If __get always returns true and the value is still uninitialized, a global=
 isReadable check will go from false to true in that transition.<br>
<br>
Either way, that&#39;s not a fully safe transition to make.<br>
<br>
--Larry Garfield<br></blockquote><div><br></div><div>Nah, I&#39;m thinking=
=C2=A0about a scenario very close to what you talk about in the RFC: unset=
=C2=A0+ __get</div><div><br></div><div>From:</div><div>class Foo</div><div>=
{</div><div>=C2=A0 =C2=A0 public int $abc;</div><div><br></div><div>=C2=A0 =
=C2=A0 public function __construct()</div><div>=C2=A0 =C2=A0 {</div><div>=
=C2=A0 =C2=A0 =C2=A0 =C2=A0 $this-&gt;abc =3D stuff();</div><div>=C2=A0 =C2=
=A0 }</div><div>}</div><div><br></div><div>To:</div><div><div>class Foo</di=
v><div>{</div><div>=C2=A0 =C2=A0 public int $abc;</div><div><br></div><div>=
=C2=A0 =C2=A0 public function __construct()</div><div>=C2=A0 =C2=A0 {</div>=
<div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 unset($this-&gt;abc);</div><div>=C2=A0 =C2=
=A0 }</div><div><br></div><div><div>=C2=A0 =C2=A0 public function __get($na=
me)</div><div>=C2=A0 =C2=A0 {</div><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 if (=
9;abc&#39; =3D=3D=3D $name) {</div><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =
=C2=A0 return $this-&gt;abc =3D stuff();</div><div>=C2=A0 =C2=A0 =C2=A0 =C2=
=A0 }</div><div>=C2=A0 =C2=A0 }</div></div><div>}</div><div><br></div></div=
></div></div>

--0000000000005c1e110645241b9e--