Subject: [ggl] Definition of exception types
From: Hartmut Kaiser (hartmut.kaiser)
Date: 2009-06-15 09:40:13
> > However, as it introduces significant changes and would require to
> > update many places in the library, I'm not the right person to
> > and decide on further applicability of this solution.
> > So, I would leave this aspect of the discussion to Barend and Bruno.
> I Like the approach used in FileSystem, I didn't know it and was
> impressed by its simplicify and efficiency to produce a perfect blend
> of 2 very different error handling technics. So I'd be very interested
> in having something similar in GGL.
> However, when reading the second proposition of Hartmut about
> extending Boost.System, I didn't see the relation with the first one.
> Especially, I can't see what a GGL algorithm would finally look like,
> and what it'd give from a user's point of view. Does it still look
> like the "some_function" function in the first example?
Yes. The code I presented was there to define the error_type usable as the
some_function(..., error_type& e = throws);
Blending this with Boost.System integrates the error handling nicely with
the C++0x standard and the way other libraries are handling errors (i.e.
But there might be other ways of achieving similar goals.
Geometry list run by mateusz at loskot.net