Re: [Boost-users] Issues with boost::process from the sandbox and the boost::process::child semantics

Subject: Re: [Boost-users] Issues with boost::process from the sandbox and the boost::process::child semantics
From: Nathan Crookston (nathan.crookston_at_[hidden])
Date: 2011-03-30 21:52:47

Hi all,

On Mon, Mar 28, 2011 at 6:58 PM, Jan Kundrát <jkt_at_[hidden]> wrote:
> On 03/28/11 19:37, Nat Linden wrote:
>> It seems to me that you're asking for the 'child' class to store a
>> flag: is this a real instance or a placeholder instance? Then every
>> access to that object must check the flag.
>
> Hi Nat and others,
> based on your arguments, it seems that I was wrong in my evaluation.
> Thank you for clarifying the design choices, I now realize that it
> indeed makes sense.

I agree with previous suggestions that boost::optional is probably the
way to go with the process library in its current form.  It's worth
noting that when this issue was brought up a couple months ago, Boris
(the most recent developer to spend some time on this project)
indicated that a default-constructed invalid process wouldn't be too
bad since process objects in the code may reference terminated
processes [1].  I'm not sure if/when that change will happen.

HTH,
Nate

[1] <http://lists.boost.org/Archives/boost/2011/02/176958.php>