Re: [boost] Can you please elaborate on what makes a non-header-only library so very Re: But it would Re: [gil] [announcement] New IO lib

From: Jonathan Franklin (franklin.jonathan_at_[hidden])
Date: 2008-04-26 18:25:42

On Sat, Apr 26, 2008 at 3:51 PM, Tom Brinkman <reportbase_at_[hidden]> wrote:

> Its probably more a UNIX thing, I'm not sure if windows developers
> could care one way or another.

For the record, I use Solaris 10, Linux (ubuntu Gutsy), and Windows w/ vc8.

But on LINUX/UNIX, there is a central repository of pre-built
> libraries that we syncronize our
> builds against.

Synchronize builds?  Not sure exactly what you mean.

> Uncecessarily adding new libraries makes the process
> that much more to manage.

Are you referring to the extra work it takes to build the boost libs for
each supported os+compiler, and add them to your "repository" of 3rd party
libraries?

> Compilers are very smart and very fast, and thats how I pefer to code.

Compilers are indeed smart and fast.  And it still takes a non-zero amount
of time to compile the boost.FileSystem code.  I still prefer to compile
that once up front, and not tack that on to every successive build of my own
code that I do.  It takes a couple of hours to do a full rebuild of the code
base that I work on.  I prefer shaving anything off that I can.

If a library is entirely inlined or template code, then it makes perfect
sense to make the library header only.  It also strikes me as silly to force
a particular library to be something that its not.

Jon