Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:127019
Date: Tue, 01 Apr 2025 21:17:05 +0100
To: internals@lists.php.net, Larry Garfield <larry@garfieldtech.com>
Subject: Re: [PHP-DEV] [RFC brainstorm] Approximately equals operator
User-Agent: K-9 Mail for Android
In-Reply-To: <799bd54a-2136-4410-823a-9233ccbeb16a@app.fastmail.com>
References: <4a3c6ce7-102d-4cfe-a7a8-35630715b870@gmail.com> <799bd54a-2136-4410-823a-9233ccbeb16a@app.fastmail.com>
Message-ID: <89495A22-BCBD-4779-835D-56B90257FF51@php.net>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
From: derick@php.net (Derick Rethans)

On 1 April 2025 20:52:32 BST, Larry Garfield <larry@garfieldtech=2Ecom> wro=
te:
>On Mon, Mar 31, 2025, at 5:03 PM, Niels Dossche wrote:
>> Hi internals!
>>
>> I'm excited to share what I've been working on!
>> I had an epiphany=2E I realized what we truly need to revolutionize PHP=
:=20
>> a new operator=2E
>>
>> Hear me out=2E
>> We live in an imperfect world, and we often approximate data, but=20
>> neither `=3D=3D` nor `=3D=3D=3D` are ideal comparison operators to deal=
 with=20
>> these kinds of data=2E
>>
>> Introducing: the "approximately equal" (or "approx-equal") operator=20
>> `~=3D` (to immitate the maths symbol =E2=89=83)=2E
>> This combines the power of type coercion with approximating equality=2E
>> Who cares if things are actually equal, close enough amirite?
>>
>> First of all, if `$a =3D=3D $b` holds, then `$a ~=3D $b` obviously=2E
>> The true power lies where the data is not exactly the same, but "close =
enough"!
>>
>> Here are some examples:
>>
>> We all had situations where we wanted to compare two floating point=20
>> numbers and it turns out that due to the non-exact representation,=20
>> seemingly-equal numbers don't match! Gone are those days because the=20
>> `~=3D` operator nicely rounds the numbers for you before comparing them=
=2E
>> This also means that the "Fundamental Theorem of Engineering" now holds=
!
>> i=2Ee=2E 2=2E7 ~=3D 3 and 3=2E14 ~=3D 3=2E Of course also 2=2E7 ~=3D 3=
=2E14=2E But this is=20
>> false obviously: 2 ~=3D 1=2E
>>
>> Ever had trouble with users mistyping something? Say no more!
>> "This is a tpyo" ~=3D "This is a typo"=2E It's typo-resistant!
>> However, if the strings are too different, then they're not=20
>> approx-equal=2E
>> For example: "vanilla" ~=3D "strawberry" gives false=2E
>> How does this work?
>> * The strings are equal if their levenshtein ratio is <=3D 50%, so it's=
=20
>> adaptive to the length=2E
>> * If the ratio is > 50%, then the shortest string comes first in the=20
>> comparison, such that if we ever get a `~<` operator, then "vanilla" ~<=
=20
>> "strawberry"=2E
>>
>> There is of course a PoC implementation available at:=20
>> https://github=2Ecom/php/php-src/pull/18214
>> You can see more examples on GitHub in the tests, here is a copy:
>> ```php
>> // Number compares
>> var_dump(2 ~=3D 1); // false
>> var_dump(1=2E4 ~=3D 1); // true
>> var_dump(-1=2E4 ~=3D -1); // true
>> var_dump(-1=2E5 ~=3D -1=2E8); // true
>> var_dump(random_int(1, 1) ~=3D 1=2E1); // true
>>
>> // Array compares (just compares the lengths)
>> var_dump([1, 2, 3] ~=3D [2, 3, 4]); // true
>> var_dump([1, 2, 3] ~=3D [2, 3, 4, 5]); // false
>>
>> // String / string compares
>> var_dump("This is a tpyo" ~=3D "This is a typo"); // true
>> var_dump("something" ~=3D "different"); // false
>> var_dump("Wtf bro" ~=3D "Wtf sis"); // true
>>
>> // String / different type compares
>> var_dump(-1=2E5 ~=3D "-1=2Ea"); // true
>> var_dump(-1=2E5 ~=3D "-1=2Eaaaaaaa"); // false
>> var_dump(NULL ~=3D "blablabla"); // false
>> ```
>>
>> Note that this does not support all possible Opcache optimizations=20
>> _yet_, nor does it support the JIT yet=2E
>> However, there are no real blockers to add support for that=2E
>>
>> I look forward to hearing you!
>>
>> Have a nice first day of the month ;)
>> Kind regards
>> Niels
>
>Naturally, the degree of closeness for strings or for floats should be co=
ntrolled by an ini setting=2E  Maximum flexibility!
>
>--Larry Garfield

You got to be joking! Everybody knows ini settings make things unportable=
=2E I suggest we introduce AI to determine the closeness instead=2E

cheers
Derick