Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:125254
Feedback-ID: id5114917:Fastmail
Date: Mon, 26 Aug 2024 08:28:33 +0100
To: internals@lists.php.net
Subject: Re: [PHP-DEV] [RFC] Default expression
User-Agent: K-9 Mail for Android
In-Reply-To: <99846F78-38F8-4549-B0A6-6310933AA78D@newclarity.net>
References: <792a811e-edf8-4a82-8422-e11cc302a6db@scriptfusion.com> <71F1A1D0-9958-4C86-B8C4-50B502B7560D@getmailspring.com> <99846F78-38F8-4549-B0A6-6310933AA78D@newclarity.net>
Message-ID: <88BCC47D-C8FF-44DB-8816-17CB899CD887@rwec.co.uk>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
From: imsop.php@rwec.co.uk ("Rowan Tommins [IMSoP]")



On 26 August 2024 05:44:44 BST, Mike Schinkel <mike@newclarity=2Enet> wrot=
e:

>> On Aug 25, 2024, at 12:21 PM, Rowan Tommins [IMSoP] <imsop=2Ephp@rwec=
=2Eco=2Euk> wrote:
>> The Reflection API is a bit like the Advanced Settings panel in a piece=
 of software, it comes with a big "Proceed with Caution" warning=2E You onl=
y move something from that Advanced Settings panel to the main UI when it's=
 going to be commonly used, and generally safe to use=2E I don't think allo=
wing arbitrary operations on a value that's declared as the default of some=
 other function passes that test=2E
>
>You analogy is faulty=2E You are conflating a complex API =E2=80=94 Refle=
ction =E2=80=94 with one use-case of that API which =E2=80=94 per the RFC =
=E2=80=94 has well-defined syntax and semantics and purely due to its simpl=
icity is more likely to be used than the Reflection API and far less likely=
 to be used incorrectly than the Reflection API=2E

You are misunderstanding the analogy=2E The analogy is that the Reflection=
 API as a whole is like the Advanced Settings page, with its big "Proceed w=
ith Caution" sign=2E Accessing the default value of someone else's function=
 is like a setting which you can only currently access on that advanced set=
ting screen, but people are proposing we move to a big tempting tickbox on =
the main UI=2E

I was responding to someone justifying anything and everything the proposa=
l allows, because Reflection already allows it=2E If the feature was "first=
 class syntax to access private methods of a class", I don't think it would=
 be controversial to challenge it=2E Saying "Reflection can already do it" =
would be a poor defence, because part of Reflection's job is to break the n=
ormal rules of the language=2E



>> On Aug 25, 2024, at 11:31 AM, Rowan Tommins [IMSoP] <imsop=2Ephp@rwec=
=2Eco=2Euk> wrote:
>Can you give a specific example from code we are likely to find in a prod=
uction application =E2=80=94 vs=2E a forum debate =E2=80=94 where someone i=
s likely to use `default+1` AND where it would be problematic for the autho=
r?

The overriding rule, in my head, is that the caller shouldn't need to know=
, and shouldn't be able to find out, what a particular implementation has c=
hosen as the default=2E So the particularly problematic operations are thin=
gs like this:=20

foo($whatWasTheDefault=3Ddefault)=20
foo(logAndReturn(default))
foo(doSomethingUnrelated(default) && false?: default)=20

I can see people inventing use cases for these as soon as they're availabl=
e, but by doing so they will be completely changing the meaning of default =
parameters, which currently mean "I trust the implementation to substitute =
something valid here, and have no interest in what it is"=2E

Note that under inheritance, the default value may even change type:

class A { public function foo(int $bar=3D42) { =2E=2E=2E } }
class B extends A { public function foo(int|string $bar=3D'hello') { =2E=
=2E=2E } }

Perhaps that would be more obvious if the declaration happened to look lik=
e this:=20

function foo(optional int $bar) {
    default for $bar is 42;
    // =2E=2E=2E
}

As far as I can see, nobody has actually justified reading values out in t=
his way, only said it's a side-effect of the current implementation=2E


You ask how a library can provide access to that default, and the answer i=
s generally pretty trivial: define a public constant, and refer to it in th=
e parameter definition=2E I sometimes do that with private constants anyway=
, just to make the value more easily documented, or use it in a couple of d=
ifferent contexts=2E

The only exception I think is "new in initializer", where you would have t=
o provide a function/method instead of a constant, and couldn't currently r=
euse it in the actual signature=2E



Aside: one of those examples brings up an interesting question: is the val=
ue pulled out by "default" calculated only once, or each time it's mentione=
d? In other words, would this create 3 pointers to the same object, or 3 di=
fferent objects?=20

foo(array|Something $x=3Dnew Something);
foo([default, default, default]);



Rowan Tommins
[IMSoP]