Re: [boost] Boost and XML-RPC

Subject: Re: [boost] Boost and XML-RPC
From: Matus Chochlik (chochlik_at_[hidden])
Date: 2009-06-15 03:15:41

On Fri, Jun 12, 2009 at 11:53 AM, Dmitriy Lyfar<dlyfar_at_[hidden]> wrote:
> Hello, Matus
>
> Thank you for your answer. I did a quick look at this library, as I
> understand it's more general solution and that lib allow have mechanism
> which packs/unpacks objects data into some storage for transferring between
> hosts.

Yes, among other things

But as I understood you use some network layer for RPC implementation
> in addition to that library?

We are using MPI and POSIX sockets, but we are looking for some higher
level library like Boost.ASIO to replace the sockets (we're also checking
other options) for future versions.

>
> Dmitriy
>
> 2009/6/12 Matus Chochlik <chochlik_at_[hidden]>
>
>> Hi Dmitriy,
>>
>> On Thu, Jun 11, 2009 at 1:43 PM, Dmitriy Lyfar<dlyfar_at_[hidden]> wrote:
>> > Hello boosters,
>> >
>> > What do you think about boost.xmlrpc implementation? Some days ago I need
>> > simple xmlrpc library with https support,
>> > but I cannot find anything for windows. What I've found it's
>> > http://libiqxmlrpc.sourceforge.net/ library. But it has too many
>> > dependencies,
>> > I really don't understand why I should compile half of gnome to have an
>> > ability call xmlrpc methods. So after some libs I realized that
>> > there is no good xmlrpc lib for windows at least (if you know, please let
>> me
>> > know).
>> >
>> > So what are you thoughs about implementation such lib as part of boost
>> (asio
>> > based for example)? Is it needed for community?
>>
>> Concerning the implementation of (not only *xml-based*) RPC, You may want
>> to
>> take a look at the Mirror library in the vault. One of the reasons I'm
>> developing
>> this library is that we are using it for RPCs (besides other things) in
>> several
>> of the projects at our university.
>>
>> It needs some work to improve the compilation time and memory requirements
>> and the docs need to be finished, before it can be submitted for review,
>> but
>> I think it is fairly usable already.
>>
>> >
>> > Thank you, Dmitriy.
>>

regards,

Matus