[review] conversion - Kevlin's comments

Hi Vicente, all - I got this comment from Kevlin Henney, author of lexical_cast: On Sep 1, 2011, at 11:45 AM, Kevlin Henney <kevlin@curbralan.com> wrote:
I was looking through the docs and they seem to offer a solution to a common family of conversion problems. Based on casual reading, it feels a little more complex and requires a higher-entry level of knowledge than would be appropriate for many of the people who actually need this functionality. These are initial impressions only, so don't read too much into them.
The eternal battle between power and simplicity. :-) Cheers, Gordon

Le 01/09/11 19:30, Gordon Woodhull a écrit :
Hi Vicente, all -
I got this comment from Kevlin Henney, author of lexical_cast:
On Sep 1, 2011, at 11:45 AM, Kevlin Henney<kevlin@curbralan.com> wrote:
I was looking through the docs and they seem to offer a solution to a common family of conversion problems. Based on casual reading, it feels a little more complex and requires a higher-entry level of knowledge than would be appropriate for many of the people who actually need this functionality. These are initial impressions only, so don't read too much into them.
The eternal battle between power and simplicity. :-)
Thanks Gordon for forwarding the Kevlin comments. Loocking forward a simple and power library making easier conversions ;-) Vicente

On Sep 1, 2011, at 1:43 PM, Vicente J. Botet Escriba wrote:
Thanks Gordon for forwarding the Kevlin comments.
Looking forward a simple and power library making easier conversions ;-)
If taking this further, I would just find or write a library capable of glueing together function/functor overloads. The overloads would still be operator()(Source const&, type_tag<Target>) and there would be some convenience class which defines member function template<typename Target, typename Source> Target cast(Source const&); But of course I won't get around to it unless I find some other reason why I need such a thing! :-) Still hoping someone will chime in and say Phoenix (or Egg?) already does this. Jeff has suggested a plausible design using Eric's can_be_called in the other thread. Cheers, Gordon
participants (2)
-
Gordon Woodhull
-
Vicente J. Botet Escriba