Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:111511
Message-ID: <tencent_411D3FAF7CD51091A3CDF1F002D086EEA40A@qq.com>
Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.80.23.2.2\))
In-Reply-To: <CA+kxMuTSDVJFXL=Cfx3+XQLpF60ajcSdVy-N5eXsfL43iujjTw@mail.gmail.com>
Date: Fri, 14 Aug 2020 13:30:57 +0800
Cc: =?utf-8?B?TWljaGFlbCBWb8WZw63FoWVrIC0gxIxWVVQgRkVM?= <vorismi3@fel.cvut.cz>,
 PHP internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
References: <6cfb77d4d511b5c591e7d7b5cc80199955123b2cd4276a5bca2327e4b02942db@mahalux.com>
 <CA+kxMuTSDVJFXL=Cfx3+XQLpF60ajcSdVy-N5eXsfL43iujjTw@mail.gmail.com>
To: Dan Ackroyd <Danack@basereality.com>
Subject: Re: [PHP-DEV] Allow sleep() to accept non-integer values
From: twose@qq.com (twosee)



> 2020=E5=B9=B48=E6=9C=8811=E6=97=A5 =E4=B8=8B=E5=8D=886:39=EF=BC=8CDan =
Ackroyd <Danack@basereality.com> =E5=86=99=E9=81=93=EF=BC=9A
>=20
> Michael Vo=C5=99=C3=AD=C5=A1ek wrote:
>=20
>> Another reason is that sleep(0.1); is silently accepted now (even =
with
> strict types enabled),
>=20
> That appears to not be true: https://3v4l.org/7YbqX
>=20
> Rowan wrote:
>> Unless there are problems with the implementation, this seems like a
> straight-forward win.
>=20
> =46rom the PR.
>> Implemented using nanosleep which is not guaranteed to be available =
everywhere.
>=20
> Please just use usleep if you need more accuracy than seconds.
>=20
> Having a function that behaves differently based on different
> platforms is a bad idea.
>=20
> Changing a function to have surprising behaviour just to avoid using a
> different function, that is already available, is a really bad
> tradeoff.
>=20
> cheers
> Dan
> Ack
>=20
> --
> PHP Internals - PHP Runtime Development Mailing List
> To unsubscribe, visit: https://www.php.net/unsub.php
>=20
>=20

+1

I would rather have a new function like this (actually I am using it on =
my project):

```
function msleep(int $milli_seconds) { }
```

int is always a better choice than float, otherwise, we have to do dirty =
conversions and more checks on it
And usually, we don't need high precision, millisecond sleep is enough, =
e.g. the timeout feature provided by epoll, it is the reason why many =
async libraries provide timeout feature in milliseconds.


Regards,
Twosee