Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:111720
Message-ID: <C3C433DA-B9C0-4B49-98C7-9D163AF713F9@joshbruce.dev>
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_49D91AAD-7B61-4E96-9476-54603912BBAA"
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.120.23.2.1\))
Date: Sun, 30 Aug 2020 09:32:45 -0500
In-Reply-To: <DM6PR07MB66189EA71A5EBDFD16CEDC05F9470@DM6PR07MB6618.namprd07.prod.outlook.com>
Cc: Larry Garfield <larry@garfieldtech.com>,
 php internals <internals@lists.php.net>
To: tyson andre <tysonandre775@hotmail.com>
References: <21C5073D-3F39-49DA-8686-E027AE780793@joshbruce.dev>
 <D14A346D-6234-4A85-9266-4125236B3F3E@joshbruce.dev>
 <1bc7759a-63b8-4b96-bb69-97b50f851307@www.fastmail.com>
 <E6C6FE8B-C263-46AB-B851-14DEB4AED52A@joshbruce.dev>
 <DM6PR07MB66189EA71A5EBDFD16CEDC05F9470@DM6PR07MB6618.namprd07.prod.outlook.com>
Subject: Re: [PHP-DEV] [RFC][Discussion] Objects can be declared falsifiable
From: josh@joshbruce.dev (Josh Bruce)

--Apple-Mail=_49D91AAD-7B61-4E96-9476-54603912BBAA
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

Hey Tyson,

This is great! Thank you so much, sincerely.

Still slow goings, which is fine, we have at least a year. lol

Static analyzers seem to be the biggest concern to date.

Haven=E2=80=99t been able to get one running locally - though I=E2=80=99ve=
 only spent a few minutes here and there; definitely on the list.

A use case for this sort of thing is also a concern or question. I came =
across Pipeline from the PHP League and saw their interruptible =
processor and was wondering if something like this would be helpful =
there - for pipeline patterns in general: =
https://github.com/thephpleague/pipeline/blob/master/src/InterruptibleProc=
essor.php =
<https://github.com/thephpleague/pipeline/blob/master/src/InterruptiblePro=
cessor.php>

While working on another project, saw this line from the PHP =
array_filter docs:

    "If no callback is supplied, all entries of array equal to FALSE =
(see converting to boolean) will be removed."

I=E2=80=99m still field testing the latest iteration of my base project, =
but wanted to put a working (non-internals) implementation out there =
(note this covers emptiness and falseness for the purposes of the =
project):

Tests - =
https://github.com/8fold/php-shoop/blob/master/tests/RfcObjectCanBeDeclare=
dFalsifiableTest.php =
<https://github.com/8fold/php-shoop/blob/master/tests/RfcObjectCanBeDeclar=
edFalsifiableTest.php>

Interface - =
https://github.com/8fold/php-shoop/blob/master/src/FilterContracts/Interfa=
ces/Falsifiable.php =
<https://github.com/8fold/php-shoop/blob/master/src/FilterContracts/Interf=
aces/Falsifiable.php> - for our purposes the efToBool would be __toBool

Default implementation - =
https://github.com/8fold/php-shoop/blob/master/src/Shooped.php#L216 =
<https://github.com/8fold/php-shoop/blob/master/src/Shooped.php#L216>

=E2=80=9CType system=E2=80=9D implementation - =
https://github.com/8fold/php-shoop/blob/master/src/Filter/TypeAsBoolean.ph=
p =
<https://github.com/8fold/php-shoop/blob/master/src/Filter/TypeAsBoolean.p=
hp>

Cheers,
Josh

> On Aug 9, 2020, at 3:59 PM, tyson andre <tysonandre775@hotmail.com> =
wrote:
>=20
> Hi Josh,
>=20
> I'd recommend looking at https://github.com/php/php-src/pull/5156 =
(linked to in your RFC's reference) when implementing this PR.
> Additionally, I didn't see if this was brought up - PHP already has a =
class which is falsifiable - SimpleXMLElement.
>=20
> You can see the source of ext/simplexml/simplexml.c for `_IS_BOOL`
>=20
> ```
> static int sxe_object_cast_ex(zend_object *readobj, zval *writeobj, =
int type)
> {
> 	php_sxe_object *sxe;
> 	xmlChar        *contents =3D NULL;
> 	xmlNodePtr	    node;
> 	int rv;
>=20
> 	sxe =3D php_sxe_fetch_object(readobj);
>=20
> 	if (type =3D=3D _IS_BOOL) {
> 		node =3D php_sxe_get_first_node(sxe, NULL);
> 		if (node) {
> 			ZVAL_TRUE(writeobj);
> 		} else {
> 			ZVAL_BOOL(writeobj, =
!sxe_prop_is_empty(readobj));
> 		}
> 		return SUCCESS;
> 	}
> ```
>=20
> ```
> static php_sxe_object* php_sxe_object_new(zend_class_entry *ce, =
zend_function *fptr_count)
> 	intern->zo.handlers =3D &sxe_object_handlers;
> // ... elsewhere in ext/simplexml/simplexml.c
> 	sxe_object_handlers.cast_object =3D sxe_object_cast;
> ```
>=20
> The default handlers would be overridden, so this would probably need =
to fall back to the original handlers for types other than bool.
> Also, this might need to copy all of the handlers from the parent =
class entry's zend_object_handlers,
> to meet the expected behavior of other cast types and other magic =
behaviors from SimpleXMLElement and other classes continuing to work.
>=20
> =46rom the perspective of static analysis, a few things to note:
> - Static analyzers such as psalm/phan currently deliberately don't =
bother handling the possibility that `object` can be false even after it =
was checked for falsiness.
>  There's hundreds of other things that could be implemented, and =
adding the special casing and performance overhead checking for FFI =
objects and SimpleXMLElement, subclasses of those is currently low =
priority compared to those things.
>=20
>  Code using SimpleXMLElement/FFI is generally limited to a few files =
in practice.
>=20
>  Definitely possible for analyzers to support this for known base =
classes, though, and the priority would increase if the RFC passed.
>=20
>  I am a maintainer of Phan.
> - For BC reasons, internal data structures such as ArrayObject =
probably wouldn't get changed in any php release
>   (e.g. could break code using falsiness check instead of null check).
>   So this might lead to inconsistencies with newer extensions if half =
of the data structures treat emptiness as false and half don't.
> - This feature may end up getting adopted in cases where it's =
convenient but turns out prone to accidental bugs and later is =
deprecated and removed.
>  For example, =
https://stackoverflow.com/questions/25031236/if-elem-vs-elem-is-not-none =
was seen in python
>=20
>  (e.g. `if (!$this->resultSetObject) { $this->resultSetObject =3D =
slow_db_operation(); }`
>  would not behave the way people would previously expect for most =
objects (executed repeatedly instead of once))
>=20
> ```
> <?php
> function test(SimpleXMLElement $e) {
>    // False positive RedundantCondition in psalm
>    if ($e) {
>    }
> }
> ```=20
>=20
>> And I don=E2=80=99t know of a way to add an interface to new =
stdClass() - but thought this might be a valid use case:
>=20
> `stdClass` isn't a final class. I assume they meant this
>=20
> ```
> class EmptyStdClass extends stdClass implements Falsifiable {
>    public function __toBool() { return false; }
> }
> function example(stdClass $s) {
>    if (!$s) { throw new Exception("Previously impossible"); }
> }
> ```
>=20
> Thanks,
> - Tyson
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: https://www.php.net/unsub.php
>=20


--Apple-Mail=_49D91AAD-7B61-4E96-9476-54603912BBAA--