|
Boost : |
From: Robert Ramey (ramey_at_[hidden])
Date: 2006-08-01 11:03:41
brass goowy wrote:
> John Reid wrote:
>>
>> I'm getting a runtime check error from MSVC8 in the following code
>> checked out from cvs_head:
>>
>> serialization/collections_load_imp.hpp
>> line 116
>>
>> ...
>>
>> template<class Archive, class Container, class InputFunction, class
>> R> inline void load_collection(Archive & ar, Container &s)
>> {
>> s.clear();
>> // retrieve number of elements
>> unsigned int count;
>> unsigned int item_version;
>> ar >> BOOST_SERIALIZATION_NVP(count);
>> if(3 < ar.get_library_version()){
>> ar >> BOOST_SERIALIZATION_NVP(item_version);
>> }
>> R rx;
>> rx(s, count);
>> InputFunction ifunc;
>> while(count-- > 0){
>> ifunc(ar, s, item_version);
>> }
>> }
>>
>
> I think this function would be more flexible if the clear
> call was removed. Users could decide if they want to call
> clear() or not. If you want append-like behaviour you
> wouldn't call clear(). If you know the container is
> already empty you also wouldn't have to call clear().
I couldn't be convinced that this is a good idea. The "fixed idea"
of the serialization library is that the >> operator reconstructs
things as they were when the << operator was called. Having
the operation of >> depend upon some non-obvious state variable
would result in very error prone code - as it always does.
Robert Ramey
>
> Regards,
> Brian Wood
> www.webEbenezer.net
> _______________________________________________
> Unsubscribe & other changes:
> http://lists.boost.org/mailman/listinfo.cgi/boost
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk