please, use bottom posting.

On 23 March 2015 at 18:47, Marcos Cardinot <[email protected]> wrote:
> Hi Lubomir,
>
> It's not about a dangerous lack of initialization - it's more about code
> consistence. So, the question should be why not initialize all members as we
> should do?

the answer is in the requests to change the patches.
because it looks horrible and it's prone to order warnings for more members.

i would do it if there is a compiler warning (e.g. requirement -
const, references etc..) or a potentially dangerous action with an
uninitialized member.
i agree that it's it's safer, the language recommends it and there is
a performance boost in for some non-POD members, but in the end it's a
matter of a preference.

>
> In addition, most of these are not following the new coding style (which we
> discussed in another thread), so if you want to apply them, I could send new
> patches fixing it (and join all changes in a single patch if you like)...
>

if the project maintainer and others are OK with your patches, i'm
certainly OK as well.

lubomir
--
_______________________________________________
subsurface mailing list
[email protected]
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Reply via email to