Re: [Boost-users] boost serialization + boost variant + boost blank

Subject: Re: [Boost-users] boost serialization + boost variant + boost blank
From: Ioannis Papadopoulos (ipapadop_at_[hidden])
Date: 2015-02-02 10:30:50

On 02/01/2015 02:25 PM, Andreas M. Iwanowski wrote:
> If you must serialize the variant then a serialize() will have to be defined for every type.
> 
> Unless I have misunderstood what you mean by "really works":
> It's not about satisfying the compiler, but rather telling it how to serialize this member.
> By telling it (with an empty serialize() ) to not serialize or de-serialize anything, it works appropriately, as long as you read/write the same data in the same order (i.e. nothing :) )
> 

Shouldn't the serialization library recognize empty classes (through
boost:is_empty) and handle those cases automatically?

> 
> Mit freundlichen Grüßen / With best regards
> 
> Andreas Iwanowski - IT Administrator / Software Developer
> www.awato.de | namezero_at_[hidden]
> T: [phone] | F: [phone]
> awato Software GmbH | Salm Reifferscheidt Allee 37 | D-41540 Dormagen
> 
> avisor-Support | T: [phone] | F: [phone]
> 
> Geschäftsführer: Ursula Iwanowski | HRB: Neuss 7208 | VAT-no.: DE 122796158
> 
> 
> -----Original Message-----
> From: Boost-users [mailto:boost-users-bounces_at_[hidden]] On Behalf Of Merrill Cornish
> Sent: Sunday, 01 February, 2015 21:16
> To: boost-users_at_[hidden]
> Subject: Re: [Boost-users] boost serialization + boost variant + boost blank
> 
> If simply defining an empty serialize() function really works, then your suggestion is best.  Next, question, does it really work--or just temporarily satisfy the compiler.
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users
> _______________________________________________
> Boost-users mailing list
> Boost-users_at_[hidden]
> http://lists.boost.org/mailman/listinfo.cgi/boost-users
>