Re: Ada -> C or C++ translator

["David Thompson" <david.thompson1@worldnet.att.net>]

Robert I. Eachus <rieachus@attbi.com> wrote :
...
> There was a project many years ago that had us all laughing when we
> found out what was going on.  You could probably use the same technique...
> [emacs and C macros to take subset-Ada and make it work as C]
...
> I know it sounds hard, but once they got started, the #defines were easy:
>
> #define begin {
> #define end } \/\/
> (Did I get that right?  It has been so long since I did "clever" things
> in cpp.  The intent of course is to make everything following the end
> keyword a comment.)
>
I doubt it, or at least not portably.

At least in any standard-conforming (post 89) implementation,
comments become logical whitespace _before_ macro substitution,
which can't legally "construct" a comment, although many implemenations
generate preprocessor output as text and re-parse it (which is not required,
but often convenient) and support deferring comments to the latter stage
to make the preprocessed output more informative.

Plus double-slash comments didn't exist in original/traditional C or C89;
they were (re)introduced in C++ in the early 1990's and migrated back to C,
standardized only in C99.

But most important, backslash+slash is outright illegal.  Backslash
escapes are legal only in character and string literals, where they are
interpreted only during compilation proper (translation phase 7) after
preprocessing is completed; and inside comments they are ignored.
Backslashes are NOT handled differently in preprocessing directives,
or macros.  Line splicing using backslash+physical-end-of-line is a
separate though similar looking case, technically not an escape, and
is equally legal everywhere -- though it is generally only needed, and
hence used, on preprocessor directives.  (And in C++98 and C99 UCNs,
basically Unicode escapes, are also legal in/as an identifier for
alphanumeric-ish characters only.)

Although a compiler, or possibly preprocessor, could use \/ for some
kind of extension, I have never heard of one doing so, and certainly
not turning it into a comment-introducing token as you want here.

The closest you can rely on is a "function-like" (parameterized) macro
that ignores/discards its argument, but usage is not transparent:
#define end(x) }
...  begin /*foo*/ ... end(foo) ...

--
- David.Thompson 1 now at worldnet.att.net