Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:126036
Feedback-ID: id5114917:Fastmail
Content-Type: multipart/alternative;
 boundary="------------4bOMJ0ijVE7C00GtJtjbnVyY"
Message-ID: <2c286462-a9ac-4b12-89bf-aea66680dea0@rwec.co.uk>
Date: Sat, 23 Nov 2024 17:34:11 +0000
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PHP-DEV] [RFC] Data Classes
To: internals@lists.php.net
References: <18b85ba5-5f1c-489c-9096-3ae203977fbe@app.fastmail.com>
 <163a08c1-2279-4a3f-a1bb-8cdfe406a4ba@rwec.co.uk>
 <0b87a3eb-7e63-4e28-a11d-25824e240a73@app.fastmail.com>
Content-Language: en-GB
In-Reply-To: <0b87a3eb-7e63-4e28-a11d-25824e240a73@app.fastmail.com>
From: imsop.php@rwec.co.uk ("Rowan Tommins [IMSoP]")

This is a multi-part message in MIME format.
--------------4bOMJ0ijVE7C00GtJtjbnVyY
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

On 23/11/2024 16:05, Rob Landers wrote:
>> Your RFC doesn't discuss this - the changeName example shows behaviour
>> *inside* the method, but not behaviour when *calling* it
>
> An interesting observation, can you explain more as to what you mean?

Looking closer, there's a hint at what you expect to happen in your 
Rectangle example:

$bigRectangle = $rectangle->resize(10, 20);
assert($bigRectangle !== $rectangle); // true

It seems that modifications to $this aren't visible outside the method, 
creating a purely local clone, which would be discarded if it wasn't 
returned (or saved somewhere).

I can see the logic, but the result is a bit unintuitive:

data class Example {
    public function __construct(public int $x) {}
    public function inc(): void {
      $this->x++;
   }
}
$foo = new Example(0);
$foo->x++;
$foo->inc();
echo $foo->x; // 1, not 2


I think it would be clearer to prevent direct modification of $this:

data class Example {
    public function __construct(public int $x) {}
    public function inc(): void {
       $this->x++; // ERROR: Can not mutate $this in data class
    }
    public function withInc(): static {
       $new = $this; // explicitly make a local copy of $this
       $new->x++; // copy-on-write separates $new from $this
       return $new;
    }
}


That would still be compatible with Ilija's suggestion, which was to add 
special "mutating methods":

data class Example {
     public function __construct(public int $x) {}
     public mutating function inc(): void {
         $this->x++;
     }
}
$foo = new Example(0);
$foo->x++;
$foo->inc!(); // copy-on-write triggered *before* the method is called
echo $foo->x; // 2


-- 
Rowan Tommins
[IMSoP]

--------------4bOMJ0ijVE7C00GtJtjbnVyY
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<!DOCTYPE html>
<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <div class="moz-cite-prefix">On 23/11/2024 16:05, Rob Landers wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:0b87a3eb-7e63-4e28-a11d-25824e240a73@app.fastmail.com">
      <blockquote type="cite">
        <div>Your RFC doesn't discuss this - the changeName example
          shows behaviour<br style="max-width:100%;height:auto;">
        </div>
        <div>*inside* the method, but not behaviour when *calling* it<br>
        </div>
      </blockquote>
      <div><br>
      </div>
      <div>An interesting observation, can you explain more as to what
        you mean? </div>
    </blockquote>
    <br>
    <p>Looking closer, there's a hint at what you expect to happen in
      your Rectangle example:</p>
    <p>$bigRectangle = $rectangle-&gt;resize(10, 20);<br>
      assert($bigRectangle !== $rectangle); // true</p>
    <p>It seems that modifications to $this aren't visible outside the
      method, creating a purely local clone, which would be discarded if
      it wasn't returned (or saved somewhere).<br>
    </p>
    <p>I can see the logic, but the result is a bit unintuitive:</p>
    <p>data class Example { <br>
         public function __construct(public int $x) {}<br>
         public function inc(): void {<br>
           $this-&gt;x++;<br>
        }<br>
      }<br>
      $foo = new Example(0);<br>
      $foo-&gt;x++;<br>
      $foo-&gt;inc();<br>
      echo $foo-&gt;x; // 1, not 2</p>
    <p><br>
    </p>
    <p>I think it would be clearer to prevent direct modification of
      $this:<br>
    </p>
    <p>data class Example { <br>
         public function __construct(public int $x) {}<br>
         public function inc(): void {<br>
            $this-&gt;x++; // ERROR: Can not mutate $this in data class<br>
         }<br>
         public function withInc(): static {<br>
            $new = $this; // explicitly make a local copy of $this<br>
            $new-&gt;x++; // copy-on-write separates $new from $this<br>
            return $new;<br>
         }<br>
      }</p>
    <p></p>
    <p><br>
    </p>
    <p>That would still be compatible with Ilija's suggestion, which was
      to add special "mutating methods":<br>
    </p>
    <p>data class Example {<br>
          public function __construct(public int $x) {}<br>
          public mutating function inc(): void {<br>
              $this-&gt;x++;<br>
          }<br>
      }<br>
      $foo = new Example(0);<br>
      $foo-&gt;x++;<br>
      $foo-&gt;inc!(); // copy-on-write triggered *before* the method is
      called<br>
      echo $foo-&gt;x; // 2<br>
    </p>
    <p><br>
    </p>
    <p></p>
    <pre class="moz-signature" cols="72">-- 
Rowan Tommins
[IMSoP]</pre>
  </body>
</html>

--------------4bOMJ0ijVE7C00GtJtjbnVyY--