I've been poking around in Burton's dstrip stuff, and I'm left in a bit of a quandary. Whilst the stuff seems good, and I've no reason to doubt that it's up-to-date with the latest version of D, it seems like the wrong approach. What would be better, surely, is to have this as a built in part of the compiler itself. I don't mean necessarily part of dmd.exe, but rather that the stripper should be using the same lexing, tokenisation, etc. subs-systems as the compiler. That way is the only way to guarantee that it doesn't get out of step.

Also, having a D-library tool would be very useful. dlib.exe, or whatever, would compile, link (to lib or DLL) and prepare a stripped version of source from a specified set of source. I've not looked into the compiler source, but I assume it would be reasonably easy to get borrow the requisite functionality from it. (Of course I could be wrong ...)

"Matthew Wilson" <dmd@synesis.com.au> wrote in message news:b6e02j$kq7$1@digitaldaemon.com...
> I've been poking around in Burton's dstrip stuff, and I'm left in a bit of
a
> quandary. Whilst the stuff seems good, and I've no reason to doubt that
it's
> up-to-date with the latest version of D, it seems like the wrong approach. What would be better, surely, is to have this as a built in part of the compiler itself. I don't mean necessarily part of dmd.exe, but rather that the stripper should be using the same lexing, tokenisation, etc. subs-systems as the compiler. That way is the only way to guarantee that
it
> doesn't get out of step.
>
> Also, having a D-library tool would be very useful. dlib.exe, or whatever, would compile, link (to lib or DLL) and prepare a stripped version of
source
> from a specified set of source. I've not looked into the compiler source, but I assume it would be reasonably easy to get borrow the requisite functionality from it. (Of course I could be wrong ...)


Definitely.

BTW, how does Eiffel manage this "short form" thing,
for an example? (I played with that too long ago.)

Luna Kid