Re: [Boost-users] [serialization] std::shared_ptr with Boost >= 1.53

Subject: Re: [Boost-users] [serialization] std::shared_ptr with Boost >= 1.53
From: Michael Powell (mwpowellhtx_at_[hidden])
Date: 2015-05-04 14:25:33

On Mon, May 4, 2015 at 2:20 PM, beet <r.berlich_at_gemfony.eu> wrote:
> Am 04.05.15 um 20:04 schrieb Michael Powell:
>>> So if I have the choice of  a) forcing the user to always install the
>>> newest Boost version or b) having to go through some pains myself to
>>> make older Boost versions work with our code, I opt for simplicity on
>>> the side of the user.
>>
>> Notwithstanding logistical concerns...

Meaning simply, good point, agreed.

> By which you refer to #ifdef-orgies ? These can be bad (as I had to
> learn when we ported the code to MSVC), agreed.
>
> Supporting multiple Boost-versions was usually rather painless so far,
> though, with only a few minor problems. And our current
> "std::shared_ptr"-problems are triggered solely by my decision to be
> progressive and to support C++11 only in the future (as of the current
> trunk version of Geneva)  :-)  .
>
> And I rather feel that, if I do this, I don't want to go half way. So I
> have replaced all boost::shared_ptr with std::shared_ptr in Geneva
> (1.7x). boost.bind has gone away (mostly replaced by C++11-lambdas) and
> I'm working on std::thread (for which I need to implement the
> interrupt() functionality of boost::thread .

Sometimes the 'do nothing' position is best. But for wanting to be
ahead of the game, doing nothing, letting it ride, at least guarantees
you nothing breaks. It may not 'improve', but it doesn't get worse
than where you started.

Anywho, HTH. Good luck!

> Best Regards,
> Ruediger
>
>
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users