Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:124583
Feedback-ID: ifab94697:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-582-g5a02f8850-fm-20240719.002-g5a02f885
Precedence: bulk
MIME-Version: 1.0
Message-ID: <4703efb9-75c2-4088-b660-f43a82aa7879@app.fastmail.com>
In-Reply-To: 
 <CAOWwgp=apEVtNTr-3_oT1ndTTCuKmkvhsctn7QNqWVKPuZuqVg@mail.gmail.com>
References: 
 <CAOWwgpmbq5VRrZQvUXDsKiNK4r6+bFA4VxnjQ_U=h8T9r0o3DA@mail.gmail.com>
 <ab83af79-0669-47dd-a3cb-ab72327ae174@bastelstu.be>
 <CAP1Jc13JWVg99AULgzrGXcoCWJtJZGz+nr_VoyUocw8n5n5sfQ@mail.gmail.com>
 <1118bbcd-a7b4-47bf-bf35-1a36ab4628e1@bastelstu.be>
 <CAP1Jc11vNiazMRL8Y-4NcJhsEkg6kgFNKLymTGfEPYRv114oTA@mail.gmail.com>
 <45847b93-02bf-459f-bcd2-81ba35a12c24@bastelstu.be>
 <CAOWwgpnt77-Sybzc3hcnFEui9t8wAv03NAov4xCK9ib_VRZ0JA@mail.gmail.com>
 <46bd4098-2936-4e46-98e9-fe55118325c2@bastelstu.be>
 <CAOWwgpkCy++W1C2t0PnkZ71S8vq1B734ZNPGw2-OHjmrwO3L4A@mail.gmail.com>
 <61ab36bc-b045-452a-84e0-87367d4c680e@bastelstu.be>
 <CAEPJduk6m1gdPOrsXYbsc+RLuHeEBav3pGwQO+PHs4UGGqtzhg@mail.gmail.com>
 <07e065f2-8f64-4bad-9a98-51f4eaf63ddb@app.fastmail.com>
 <CAOWwgpngQtg2NRW6M6RwgSH+OS1iipewHBmM5c36P2MxQGX78A@mail.gmail.com>
 <2a0a4650-c2c5-4c6d-ad3a-25365b3391b2@bastelstu.be>
 <CAOWwgpm9FLy_SNt7HLtJh5sxtfwjZMMN6pOKioDqsJc+GgD-sA@mail.gmail.com>
 <CAOWwgpnwoPH_rkJreh8v+X8WagzcCJ_puT1RWC0c717GMj5Onw@mail.gmail.com>
 <CAOWwgp=uwDebrx-=H8yP1Dgwf9ji=v_qagw9aAxbDJcPvCqKNw@mail.gmail.com>
 <b096b1d3-4414-4b53-a3b0-3a87c8dd7ff3@app.fastmail.com>
 <CAOWwgp=88jWSR-roWTrkkgwVgQvLQYbkTGSGAr6ASPs6s7s_4g@mail.gmail.com>
 <2fd6d80d-a55c-4aad-aeee-6d07d4b930a1@app.fastmail.com>
 <CAOWwgp=zu8y95EyEvjidNuiBOr8-ckek_xFmrcA_JifLy3U2Xg@mail.gmail.com>
 <72131a21-1349-47fe-9267-ebad44575995@app.fastmail.com>
 <CAOWwgp=apEVtNTr-3_oT1ndTTCuKmkvhsctn7QNqWVKPuZuqVg@mail.gmail.com>
Date: Wed, 24 Jul 2024 23:01:43 +0200
To: "Nicolas Grekas" <nicolas.grekas+php@gmail.com>
Cc: "php internals" <internals@lists.php.net>
Subject: Re: [PHP-DEV] [RFC] Lazy Objects
Content-Type: multipart/alternative;
 boundary=d3f67098e4284a52994842800a244cb5
From: rob@bottled.codes ("Rob Landers")

--d3f67098e4284a52994842800a244cb5
Content-Type: text/plain;charset=utf-8
Content-Transfer-Encoding: quoted-printable


On Wed, Jul 24, 2024, at 22:45, Nicolas Grekas wrote:
>=20
>=20
> Le mer. 24 juil. 2024 =C3=A0 21:44, Rob Landers <rob@bottled.codes> a =
=C3=A9crit :
>> __
>>=20
>>=20
>> On Wed, Jul 24, 2024, at 16:28, Nicolas Grekas wrote:
>>>=20
>>> To Rob: proxying by interface can be implemented using regular code =
generation so it's not a blocker. Symfony does it already, and will cont=
inue to do it to cover the use case.
>>=20
>> I'm not sure what you mean, as the RFC makes it look this should be c=
ompletely legal, until you get to the end:
>>=20
>> interface Fancy {
>>   function sayHi(): void;
>> }
>> class A implements Fancy {
>>   function sayHi(): void { echo "hi!"; }
>> }
>>=20
>> $container->register(proxyType: Fancy::class, factory: fn() =3D> new =
A());
>>=20
>> // Now container only knows that it needs to return a FancyClass and =
this is "legal looking":
>>=20
>> $lazy =3D new ReflectionClass($proxyType)->newLazyProxy($factory);
>> // same as
>> $lazy =3D new ReflectionClass(Fancy::class)->newLazyProxy(fn() =3D> n=
ew A());
>>=20
>> This ^ compiles, it's perfectly legal and makes 100% sense that I sho=
uld be able to call $lazy->sayHi() and it calls A::sayHi(), which is wha=
t you would expect. Instead, this is runtime error.
>=20
> The reason why this doesn't work is that the RFC is about state proxie=
s, which means properties are required to hook the lazy behavior.
> The RFC forbids your snippet because the initializer must return a par=
ent of Fancy::class, which A::class isn't.
>=20
> Please have a closer read of the RFC, we tried to describe the reasoni=
ng as much as possible and this is covered.
>=20
> Nicolas =20
>=20

My point was that this can=E2=80=99t be used for any DI container on the=
 planet (unless you parse the code and statically figure out what concre=
te type is returned =E2=80=94 hope it is only one type =E2=80=94 and wil=
l only work for the simplest of cases). There is nothing here that can b=
e used for DI unless people are injecting concrete types and not using i=
nheritance, which goes against the =E2=80=9Ccode to the interface=E2=80=9D=
 rule we=E2=80=99ve been using for the last 20-40 years=E2=80=A6

=E2=80=94 Rob
--d3f67098e4284a52994842800a244cb5
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>=
On Wed, Jul 24, 2024, at 22:45, Nicolas Grekas wrote:<br></div><blockquo=
te type=3D"cite" id=3D"qt" style=3D""><div dir=3D"auto"><div><div><br></=
div><div><br></div><div class=3D"qt-gmail_quote"><div dir=3D"ltr" class=3D=
"qt-gmail_attr">Le mer. 24 juil. 2024 =C3=A0 21:44, Rob Landers &lt;rob@=
bottled.codes&gt; a =C3=A9crit&nbsp;:<br></div><blockquote class=3D"qt-g=
mail_quote" style=3D"margin-top:0px;margin-right:0px;margin-bottom:0px;m=
argin-left:0.8ex;border-left-width:1px;border-left-style:solid;border-le=
ft-color:rgb(204, 204, 204);padding-left:1ex;"><div><u></u><br></div><di=
v><div><br></div><div><br></div><div>On Wed, Jul 24, 2024, at 16:28, Nic=
olas Grekas wrote:<br></div><blockquote type=3D"cite" id=3D"qt-m_3612589=
975186624774qt"><div dir=3D"ltr"><div dir=3D"ltr"><br></div><div><div>To=
 Rob: proxying by interface can be implemented using regular code genera=
tion so it's not a blocker. Symfony does it already, and will continue t=
o do it to cover the use case.<br></div></div></div></blockquote><div><b=
r></div><div>I'm not sure what you mean, as the RFC makes it look this s=
hould be completely legal, until you get to the end:<br></div><div><br><=
/div><div>interface Fancy {<br></div><div>&nbsp; function sayHi(): void;=
<br></div><div>}<br></div><div>class A implements Fancy {<br></div><div>=
&nbsp; function sayHi(): void { echo "hi!"; }<br></div><div>}<br></div><=
div><br></div><div>$container-&gt;register(proxyType: Fancy::class, fact=
ory: fn() =3D&gt; new A());<br></div><div><br></div><div>// Now containe=
r only knows that it needs to return a FancyClass and this is "legal loo=
king":<br></div><div><br></div><div>$lazy =3D new ReflectionClass($proxy=
Type)-&gt;newLazyProxy($factory);<br></div><div>// same as<br></div><div=
>$lazy =3D new ReflectionClass(Fancy::class)-&gt;newLazyProxy(fn() =3D&g=
t; new A());<br></div><div><br></div><div>This ^ compiles, it's perfectl=
y legal and makes 100% sense that I should be able to call $lazy-&gt;say=
Hi() and it calls A::sayHi(), which is what you would expect. Instead, t=
his is runtime error.<br></div></div></blockquote></div></div><div dir=3D=
"auto"><br></div><div dir=3D"auto">The reason why this doesn't work is t=
hat the RFC is about state proxies, which means properties are required =
to hook the lazy behavior.<br></div><div dir=3D"auto">The RFC forbids yo=
ur snippet because the initializer must return a parent of Fancy::class,=
 which A::class isn't.<br></div><div dir=3D"auto"><br></div><div dir=3D"=
auto">Please have a closer read of the RFC, we tried to describe the rea=
soning as much as possible and this is covered.<br></div><div dir=3D"aut=
o"><br></div><div dir=3D"auto">Nicolas&nbsp;&nbsp;<br></div><div dir=3D"=
auto"><br></div></div></blockquote><div><br></div><div>My point was that=
 this can=E2=80=99t be used for any DI container on the planet (unless y=
ou parse the code and statically figure out what concrete type is return=
ed =E2=80=94 hope it is only one type =E2=80=94 and will only work for t=
he simplest of cases). There is nothing here that can be used for DI unl=
ess people are injecting concrete types and not using inheritance, which=
 goes against the =E2=80=9Ccode to the interface=E2=80=9D rule we=E2=80=99=
ve been using for the last 20-40 years=E2=80=A6<br></div><div><br></div>=
<div id=3D"sig121229152">=E2=80=94 Rob<br></div></body></html>
--d3f67098e4284a52994842800a244cb5--