Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:99472
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain gmail.com designates 209.85.214.52 as permitted sender)
MIME-Version: 1.0
Sender: morrison.levi@gmail.com
In-Reply-To: <1E0ECC7E-B960-46EB-9953-538DF874EA9B@gmail.com>
References: <CAFMT4Npoo_ZrRk-kk4C6KHYEtV93pU7Wmh0FhaPi6Z-9qkObaQ@mail.gmail.com>
 <CADqTB_hWrhBBQetfp=-LFR=OYALHwhs=mYd3vPQOYGGng26UsA@mail.gmail.com>
 <4077c099-2247-c3c1-7dee-02a6b25f8586@fleshgrinder.com> <CAObuZdtHXp=icU_aB8dWyjbzW-yZ0sSTW6ObmEsxTuvR679-Fw@mail.gmail.com>
 <CAObuZdvck_w5iEeH5Sz23pDS1T727gtN7pn8XCiPWBvqVWR5Ag@mail.gmail.com>
 <CAObuZdvYX8Yr4KQnzJNeF8CSJVwpxuDevV+6=w_J5J9gx5R8HQ@mail.gmail.com>
 <559b73f7-c40f-e593-f0e1-51b56210d7cb@telia.com> <2053A608-6B66-42FB-A6D7-7CC64E94DA6A@koalephant.com>
 <4FBDE643-1EBC-485A-A39F-7E28C5D4807A@koalephant.com> <e1c608a0-d046-d3a8-a360-d33c3aa75666@telia.com>
 <034f3ebf-95b8-a7cf-1279-77c7cf0424d7@gmail.com> <46143e7c-ef75-7c05-eaa9-98c8f3b405f1@telia.com>
 <CAESVnVrLG4nRrcm7MYAQutb4umLq=FkGo2CWHYqHw+bYo72Bsg@mail.gmail.com>
 <d45a3327-59d4-5c93-1425-70cc2d08a87e@telia.com> <CAESVnVpQtLXhKnUn2EE4s7rZn11r5Kpj2zOSOKRCWFaxaeOrjg@mail.gmail.com>
 <CANUQDCg_x8VHbjsm9BjXeC5vOPuasRe7cWWZP3zszRzFiR3HaQ@mail.gmail.com> <1E0ECC7E-B960-46EB-9953-538DF874EA9B@gmail.com>
Date: Fri, 9 Jun 2017 17:57:57 -0600
Message-ID: <CAFMT4Npo3Ar3BpKh7OkKKCfGU7atMvUh91Oqu=Cw+d44=r4dYA@mail.gmail.com>
To: Rowan Collins <rowan.collins@gmail.com>
Cc: PHP internals <internals@lists.php.net>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Subject: Re: [PHP-DEV] [RFC]Discuss] Syntax for Arrow Functions
From: levim@php.net (Levi Morrison)

On Fri, Jun 9, 2017 at 4:43 PM, Rowan Collins <rowan.collins@gmail.com> wro=
te:
> On 9 June 2017 21:00:48 BST, Niklas Keller <me@kelunik.com> wrote:
>>2017-06-09 15:44 GMT+02:00 Sara Golemon <pollita@php.net>:
>>
>>> On Fri, Jun 9, 2017 at 7:23 AM, Bj=C3=B6rn Larsson
>><bjorn.x.larsson@telia.com>
>>> wrote:
>>> > If I take the liberty in using the example above on our option
>>list:
>>> > 1. $someDict->map(fn($v) =3D> $v * 2)->filter(fn($v) =3D> $v % 3);
>>> > 2. $someDict->map(function($v) =3D> $v * 2)->filter(function($v) =3D>
>>$v %
>>> 3);
>>> > 3. $someDict->map($v =3D=3D> $v * 2)->filter($v =3D=3D> $v % 3);
>>> > 4. $someDict->map(($v) =3D> $v * 2)->filter(($v) =3D> $v % 3);
>>//
>>> > Ambiguous
>>> > 5. $someDict->map([]($v) =3D> $v * 2)->filter([]($v) =3D> $v % 3);
>>> >
>>> > Old proposals:
>>> > 6. $someDict->map($v ~> $v * 2)->filter($v ~> $v % 3);
>>> > 7. $someDict->map(lambda($v) =3D> $v * 2)->filter(lambda($v) =3D> $v =
%
>>3);
>>> >
>>> Something else which really pops in these examples is the effect of
>>> not needing to use parentheses when embedding a single-arg short
>>> lambda.   3 and 6 in your list read cleaner to me (due to the lack of
>>> parenthesis clutter).  Sadly ~> has the same hacky implementation
>>> issues as =3D=3D>, but I think that shows a little bit of why the
>>HackLang
>>> team decided the messy lexer was worth the clearer syntax.
>>
>>
>>Another possible syntax (dunno whether this has already been suggested
>>on-list):
>>
>>$function =3D { $x =3D> 2 * $x };
>>$function =3D { ($x) =3D> 2 * $x };
>>$function =3D | $x =3D> 2 * $x |;
>>$function =3D | ($x) =3D> 2 * $x |;
>>
>>Nikita and Levi prefer it with parenthesis, I prefer it without,
>>because I
>>think it's unnecessary clutter.
>
> I already suggested the brace version, and Levi said it would conflict in=
 the parser: https://externals.io/thread/911#email-15331 I don't know enoug=
h about parsing to comment on what workarounds would be possible.


It ends up not conflicting; when at top level you can keep existing
behavior and otherwise make it a closure. Given that creating a
closure without assigning it to variable or passing it to a function
is useless this is fine.