This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: Standard header format.
"Steven T. Hatton" <hattons@globalsymmetry.com> writes:
| There are many C++ programmers who have suggested to me the best way
| to learn what the standard headers probide is to look a them.
That definitely is not the best way. It is one of the worst way.
| > | interface - a set of declarations that defines how a part of a program
| > | can be accessed.
| >
| > An interface need not excusively consist in declarations. It may
| > consist in directives too.
|
| The definition is not mine. It's from Stroustrup's glossary.
And note that he used that term in very specific contexts. He did not
mean it to apply automatically to broader contexts, specifically when
it comes to how C++ implementations done. And, reading carefully
chapter §24 reveals that he left room for broader meaning of that word
and he did not mean people pick the narrowest meaning and use it in
wider contexts.
| It's not the
| best overall definition of 'interface' possible, but it seemed the most
| appropriate.
No, a meaning that includes non-delarations is more appropriate.
[...]
| My objective is to establish a means of increasing the usability of libstdc++
| by providing a more comprehensive IDE than currently exists (to my
| knowledge).
I support the abstract goal; but as I repeatedly told you, you're in a
dead end. V3 won't expose to user any implementation details about
where internal files get installed. User has no business in that.
If you want integration with an IDE, work with the compiler. It knows
better than V3.
[...]
| I have yet to see that question addressed constructively.
when your goals are confused with misrepresentations, it is hard to
see where facts end and fiction begins. I've tried to address your
question as constructively as possible given that fuzziness.
-- Gaby