Re: The g++ preprocessor: Include Guards

[gutmant]

On Jun 20, 4:16 pm, Bernd Strieder <strie...@informatik.uni-kl.de>
wrote:
> Hello,
>
>  gutm...@gmail.com wrote:
> > Yes, what you suggest will work, but is impractical when you have an
> > amount or code that takes hours to compile you need to fix... :/
> > I'm looking for a solution along the lines of adding a flag, or using
> > a different preprocessor maybe...
> > Do you know anything like that?
>
> No, sorry. Perhaps you could create a header file containing only a
> comment, and prepend #include "new_header.h" to all files in the
> project, should not be too hard with a script. Perhaps you can even
> copy that file to all directories to avoid -I problems.
>
> Perhaps you can tell what you really want. This seems pretty strange to
> me.
>
> Bernd Strieder

Hi Bernd,
What I need is to build (script wise) a reliable inclusion tree.
I already have the code (py) to do that, but that relies on the
preprocessor's output, so the result ends up partial.
While exploring this issue, I realized that the problem is this
optimization, and now I'm wondering if I can work around it without
changing the source files.
Thanks again for your help! :)
 - Tali