Fix preprocessor/5899

[Zack Weinberg]

On Sun, Mar 10, 2002 at 12:11:08AM +0000, Neil Booth wrote:
> This patch fixes this PR (suppression of debug output when generating
> dependencies), and updates the docs.
> 
> I was getting fed up of duplicating the options documentation of CPP
> between CPP and GCC manuals.  I also promised to add environment
> variable docs to the CPP manual.  So I've split these common sections
> out into two new texi files, and documented SUNPRO_DEPENDENCIES while
> I was at it.

This seems like a good change in principle.  I'll address Joseph's
concerns in a separate message.  Here is some commentary on the new
documentation.

Suggest you go ahead and apply the code change now, so that the glibc
folks are happy.

> +@c This file should be included with a flag ``cppmanual'' set or clear
> +@c if the CPP manual is including this file or not.

Awkward.  Suggest instead

@c If this file is included with ``cppmanual'' set, format for
@c inclusion in the CPP manual; otherwise, the main GCC manual.


> +@table @env

If you use @ftable, you don't need all the @findex lines.

> +@c @itemx OBJCPLUS_INCLUDE_PATH

Insert
@c comment out until ObjC++ is merged:
just before, or people will wonder why.

> +These environment variables pertain to particular languages.  Each
> +variable's value is a colon-separated list of directories, much like
> +@env{PATH}.  When GCC searches for header files, it tries the
> +directories listed in the variable for the language you are using, after
> +the directories specified with @option{-I} but before the standard header
> +file directories.

Should say whether these directories are treated as containing system
headers.

Do we honor PATH_SEPARATOR?

...
> +@c This file should be included with a flag ``cppmanual'' set or clear
> +@c if the CPP manual is including this file or not.

Same as before.

I think the only changes you made to the option documentation were
structural; did I miss something?

zw