[boost] Re: Release 1.33 [serialization] [variant]

From: troy d. straszheim (troy_at_[hidden])
Date: 2005-01-17 14:40:28

Robert Ramey wrote:
> I have had requests for this from time to time.  Indeed, I have interest
> myself. However, authors of boost::variant havn't shown an interest in doing
> this.
> So I would suggest:

I think I've got these covered:

> a)    Make sure it has a test.  Ideally it should be in the same form as the
> other serialization tests so it can be just "dropped in" to the current test
> setup.

Naturally.  I started with one of the serialization tests, it gets run 5 
times on the 5 different types of archive like the others.

> b)    Make a simple html page to describe any "gotchas" (if any) or other
> quirks one has to be aware of.

Far as I can see, there really aren't any gotchas, it just serializes 
variants.

> c)    Prepare to receive lots of advice on how you "should" have done it.

Sure, I'd be glad to modify it to please the hordes, but I take neither 
credit nor blame, yet, it is from message 
news://news.gmane.org:119/co291u$[e-mail]
Thread title "Serializing a variant", in which Richard Peters says:
> The code in the previous posting gives a general solution for serializing a
> variant, so that there is no need to make an ad-hoc serializer for every
> different variant.

-t