Re: [boost] [modularization] Extract xml_archive from serialization

Subject: Re: [boost] [modularization] Extract xml_archive from serialization
From: Robert Ramey (ramey_at_[hidden])
Date: 2014-09-18 15:39:10

Emil Dotchevski-3 wrote
> On Wed, Sep 17, 2014 at 8:14 PM, Andrey Semashev <

> andrey.semashev@

> >
> wrote:
> 
>> On Wednesday 17 September 2014 14:24:28 Robert Ramey wrote:
>> > Stephen Kelly-2 wrote
>> >
>> > > Robert Ramey wrote:
>> > >> The "correct" solution to the above is for date-time to build
>> > >> two modules:  date-time and date-time-serialization.
>> > >
>> > > Is this "at the expense of everyone who wants to ship datetime with
>> > > support
>> > > for serialization in the package"? Is that 'non-obvious' too? Is this
>> a
>> > > net-
>> > > positive?
>> >
>> > I think its a much smaller number of people.
>> >
>> > anyone who explicitly includes date-time/serialization.hpp will know
>> that
>> > he has to ship the data-time-serialization.dll.
>>
>> Just to be closer to reality, serialization support in DateTime is
>> header-
>> only.
>>
> 
> In principle there should be no reason for e.g. time_serialize.hpp to
> include Boost Serialization headers in order to define the "load" or"save"
> function templates needed for serialization.
> 
> For example, to define:
> 
> template
> <class Archive>
> void save( Archive & ar, const posix_time::time_duration& td, unsigned int
> /*version*/)
> {
>    ....
> }
> 
> one doesn't need to include any serialization headers.
> 
> Users of time_serialize.hpp who need to save posix_time::time_duration
> objects should include the necessary serialization headers themselves.

My original concern was that we really need to spend some time reaching
a consensus on what would like future deployment of boost to look like,
what goals it should fulfill, and what policies/requirements we want to
formulate
to achieve this.  I think this should get this done before we start  we
start just moving 
things around in order to make some dependency graph look simpler.

Emils idea is an example of the kind of idea which looks very interesting to
me and I think it needs exploring.  It would be great to leverage on our
modularization effort to create flexibility to in deployment of boost.  Only
if we do this will we be able to reach 500 libraries in the next 10 years.

Robert Ramey

--
View this message in context: http://boost.2283326.n4.nabble.com/modularization-Extract-xml-archive-from-serialization-tp4667615p4667726.html
Sent from the Boost - Dev mailing list archive at Nabble.com.