On Sun, 2 Nov 2008, Yang Tse wrote:

> Since there seems none of us have yet thought of something 'more automatic'
> it is true that initially it does not matter much if the 'canonical host'
> identification is provided inside a comment or given on a preprocessor
> macro.
>
> If it is provided inside a comment, the format of the line would have to be
> 'fixed' and kept equal for successive releases, as a library packager might
> set up some scripts to automate the job. So in the end it would be something
> that people would expect to be reliable.

I would say that makes a #define CURL_HOST_TRIPLET a better choice since it is
easier and more intuitive to keep the same in future releases.

> Obviously if we add something to the public API, we must be real sure that
> we want to do so.

We _could_ first add it and mention it being a test (like in a comment above
it) to start with so that we _can_ modify it slightly before we consider it
carved in stone.

The "triplet" isn't really libtool's. It is made by config.guess and I'm
pretty sure we used it in the build for the curl tool even before we used
libtool for the lib build. The tool 'curl' already provides a replacement
define for those systems that don't run configure and I figure we would need a
similar system to that in the curlbuild.h for non-configure systems.

But the question is of course quite simply: what value does it really add?
We're only guessing here that it would be useful to someone (of course we
usually do that in a lot of other API-related issues so it's not like it is
the fist time) without any packager/user/distributer saying it is a good idea
or even something that would help them.

I think we can wait with this until someone sees an actual specific use case
or demand. Especially when being this close to the next release...

-- 
  / daniel.haxx.se