Re: Re: [boost] Serialization Formal Review #2

From: Robert Ramey (ramey_at_[hidden])
Date: 2004-04-17 10:22:59

Matthias Troyer
> On Apr 13, 2004, at 9:42 PM, Robert Ramey wrote:

>> Making a new archive class is described in the section "New Archives -
>> Implementation".  The model for an archive class is:
>
>> ////////////////////////////////////////////////////////////////////// 
>> ///
>> // class trivial_iarchive - read serialized objects from a input text  
>> stream
>> class trivial_iarchive
>
>> : public boost::archive::common_iarchive<trivial_iarchive>
>> {
>>     // permit serialization system priledged access to permit
>>     // implementation of inline templates for maximum speed.
>>     friend class boost::serialization::load_access;
>
>>     // member template for loading primitive types.
>>     // Override for any types/templates that special treatment
>>     template<class T>
>>     void load(T & t);
>
>> public:
>>     //////////////////////////////////////////////////////////
>>     // public interface used by programs that use the
>>     // serialization library
>
>>     // the >>> operator
>>     template<class T>
>>     trivial_iarchive & operator>>(T & t){
>>         boost::serialization::load(* This(), t);
>>         return * This();
>>     }
>
>>     // archives are expected to support this function
>>     void load_binary(void *address, size_t count);
>
>> };
>
>> So in theory one has to implement the two undefined functions.
>
>> For each primitive data type, int, float, etc. or any type designated  
>> as primitive define a "load" template for function to read them from the

>> file.
>
>> For each data type that you want to treat specially - rather than  
>> allowing  the serialization library to do the job - define an overload
>> for operator.

> Just a quick question: what should the default  load  (I mean   
> template<class T>> void load(T & t);) function do? Do I guess correctly  
> that it should never be called and is thus undefined to cause link-time  
> errors?

It should handle all primitives not otherwise handled specifically.  For
example the default load for text archives resolves to input_stream >> t
While for binary archives it resolves to load_binary(sizeof(t), t).

> Looking through the documentation I find references to virtual  
> functions:

And?

>> class trivial_iarchive

>> // all input archives must be derived from this base class.  This base
>> // class defines the virtual functions necessary for the serialization
>> // library to handle special types for the serialization system.

> I guess that these are incorrect comments left over from an old version?

Correct.

Robert Ramey