Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:73211
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain devilix.net designates 209.85.213.51 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <CAGa2bXb45BizZ8g8zzdneV_27R7zgiX34xoZm+f8YyNQrZxdTQ@mail.gmail.com>
References: <CAPhkiZw5xWRkc=ofh-nB5ZkjfThZFkH_+pK5RZaQFLKhno7u7A@mail.gmail.com>
	<CAGa2bXZiBuw_L_ehvTZb5_UUedtbS0n4GXYUjtKix0S9DX_8KQ@mail.gmail.com>
	<5324FE40.1070704@sugarcrm.com>
	<CAPhkiZy1oG_WHijMknB34dyo0aZRDU89EBafpaEGqGM32DCK0A@mail.gmail.com>
	<532603A0.8060802@sugarcrm.com>
	<CAPhkiZxdrxiKT3cTW9ra=0aFDKKP3-OpjmGdxqzHxu8YEUK2ww@mail.gmail.com>
	<CAGa2bXbH8WFp=8C-ON-1aAHLiyTsemVKMH+dGnqBSJAw5zw-GA@mail.gmail.com>
	<CAPhkiZx52CiKkk385itjEtg99ox3PHzXga9ZWDPbWiB2wRyH4g@mail.gmail.com>
	<CAGa2bXZGun=khdMZHzYoERap7qj7bQDPEfQzx2jJ2MwPs7-p_g@mail.gmail.com>
	<CAPhkiZzmzTqPR4KynBCWijvHa=TGKzTni7JdfuuKdy_i=sBeQg@mail.gmail.com>
	<CAGa2bXb6k8NYyYheCM6dvFWj-jA2GLP7ZiusTgZK-YgkNkeHhQ@mail.gmail.com>
	<CAPhkiZz-_PxvfsCiMDm8Hf9YzVNV+Ch1zpqDut19uV1w1WCWoA@mail.gmail.com>
	<CAGa2bXb45BizZ8g8zzdneV_27R7zgiX34xoZm+f8YyNQrZxdTQ@mail.gmail.com>
Date: Mon, 17 Mar 2014 12:47:45 +0200
Message-ID: <CAPhkiZxi0UxrrUMRNUFDmkzs2pX6y81PPabzLjLF-ChboaB8jg@mail.gmail.com>
To: Yasuo Ohgaki <yohgaki@ohgaki.net>
Cc: Stas Malyshev <smalyshev@sugarcrm.com>, 
	"internals@lists.php.net" <internals@lists.php.net>
Content-Type: text/plain; charset=UTF-8
Subject: Re: [PHP-DEV] [RFC] Revert/extend/postpone original RFC about
 read_only, lazy_write sessions
From: narf@devilix.net (Andrey Andreev)

Hi,

>> > If save handler (regardless of user or C) does not implement its own
>> > update(),
>> > then update() will not be called, instead write() is called.
>> > To do that, address of the dummy function is compared.
>>
>> Ok, so why is 'lazy_write' an option at all then? I don't see a reason
>> why it shouldn't be the default and even mandatory behavior.
>
>
> I would like to make 'lazy_write' default to TRUE, since there would not be
> 'unsafe_lock'.
> I wouldn't try to add 'unsafe_lock' anymore. It's safe to make 'lazy_write'
> default to TRUE.
>
> 'lazy_write' needs a little memory, but it's session. Session data would be
> small enough
> for almost all apps and memory is cheap these days.
> (Note: I've implemented 'lazy_write' by using MD5 hash at first, but
> benchmark revealed
>  simple store and string compare is much faster. It's saving loaded session
> data and
>  compare to check modification with new patch.)
>
> Any objection/comment make this default to TRUE or automatically apply
> 'lazy_write'?
> I think automatically applying 'lazy_write' is good choice.

My quesion was rather why does it need to be optional at all
(regardless of the default value)? If it is somehow handled at all
times, it becomes just a performance improvement. I don't see a reason
why performance improvements should be optional.

>> >> Btw, this should probably be called updateTimestamp() instead of
>> >> update(), for userland at least. write() vs. update() can be somewhat
>> >> confusing.
>> >
>> >
>> > This could be changed. Better name is always good.
>>
>> Well, there you go - updateTimestamp() is a better name. ;)
>
>
> I thought of similar name. I choose update() since it seemed it is matching
> name for read()/write(), but descriptive name may be better.
>
> Any comments for renaming?

I'm all for it (naturally, since I suggested it), it could be
updateTimestamp(), updateTs(), updateTime(), etc. Anything in that
fashion would be more descriptive and less confusing.

Cheers,
Andrey.