Re: [boost] But it would Re: [gil] [announcement] New IO lib

From: Christian Henning (chhenning_at_[hidden])
Date: 2008-04-29 11:06:57

Hi guys, sorry for being so unresponsive. I took a short vacation to
recharge my batteries.

On Sat, Apr 26, 2008 at 6:20 AM, John Femiani <JOHN.FEMIANI_at_[hidden]> wrote:
>
> > >> But it would introduce the dependency to boost::filesystem
> >
> > That would be a show-stopper for shur.  I definately could not use
> > your IO library if it had a boost::filesystem dependency.  PLEASE -
> > keep it header only.
>
>
> I assume Christian was referring to the extension function in boost
> filesystem -- I can't really think of why else filesystem would be used
> to provide the feature I was asking about, which was a read_image
> function that would choose the appropriate file format based on the name
> or first bytes of the file.

Yes, if I would have to implement such a feature, I would use
boost::filesystem. What's the point of having such a lib when nobody
is using it? Also, I have some dependencies to boost::filesystem
already in io_new. But they are all wrapped by ADD_FS_PATH_SUPPORT.

On the other hand I don't get why boost::filesystem isn't header only?
Asio managed to do that, why not filesystem? I think both libs are
more or less thin wrappers around operation system's SDKs. Please
correct me if I'm wrong.

Christian