Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:91734
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: error (pb1.pair.com: domain mprelu.de from 74.201.84.163 cause and error)
To: internals@lists.php.net
References: <CABDNdAHwgx+LJRc_gi9citiypxxWVGFMB+6hrmU+vVYDWCDV_g@mail.gmail.com>
Message-ID: <56EA8EA9.80905@mprelu.de>
Date: Thu, 17 Mar 2016 11:02:01 +0000
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101
 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <CABDNdAHwgx+LJRc_gi9citiypxxWVGFMB+6hrmU+vVYDWCDV_g@mail.gmail.com>
Content-Type: multipart/alternative;
 boundary="------------090300080107040305040901"
Subject: Re: [PHP-DEV] [RFC Discussion] Typed Properties
From: me@mprelu.de (Matt Prelude)

--------------090300080107040305040901
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit

Hi,

On 16/03/16 16:36, Phil Sturgeon wrote:
> 2. This whole temporary nullability situation, where unset properties
> will error on attempted usage if not set. Should they instead error
> after the constructor has been called if they are still not holding a
> value?

I'd have the error at the time of attempted use, if only because it's
easier for people trying to introduce typing to existing code.

I'd support borrowing the "?" nullable annotation from HackLang for
people who want a less strict behavior:

public ?string $name;

This means that $name can either be a string or the NULL value. It could
be argued that this encourages bad practices (using NULL to represent
errors, for example) but I think the benefits in providing easier adoption
outweigh the negatives.

Massive thumbs up from me on typed properties. Might be worth looking at
Hack's implementation as they're already doing this.

- Matt.

--------------090300080107040305040901--