On Tuesday 20 November 2012, Felix Salfelder wrote:
> when implementing the filters i must have missed that
> COMMON_COMPONENT::tr_{accept,review}(COMPONENT*) are not
> const. as it seems, the COMMON_COMPONENT is intended to be
> abused for device states (in that case it's no longer
> 'common').

No.  commons must not store device states.

those functions should be const.

Thanks for the report, from now on they are.

The change must be propagated to anything derived from 
COMMON_COMPONENT, including bm_pwl 160-163 where "iterator" 
changes to "const_iterator".

When you see things like this, most likely it is a remnant of 
something from the past, or something that is not quite what it 
should be.

_______________________________________________
Gnucap-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/gnucap-devel

Reply via email to