Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site watdaisy.UUCP
Path: utzoo!watmath!watdaisy!ndiamond
From: ndiamond@watdaisy.UUCP (Norman Diamond)
Newsgroups: net.lang.c
Subject: external names that aren't valid C identifiers
Message-ID: <6969@watdaisy.UUCP>
Date: Fri, 15-Feb-85 17:03:33 EST
Article-I.D.: watdaisy.6969
Posted: Fri Feb 15 17:03:33 1985
Date-Received: Sat, 16-Feb-85 05:51:18 EST
Distribution: net
Organization: U of Waterloo, Ontario
Lines: 41

Here is a suggestion that is only moderately ugly, but unambiguous and
won't break any programs.

(1)  To refer to an external name (data or function) whose name isn't
a valid C identifier:

extern "external$name" [maybe a type] c_identifier [maybe ()]

(The external name is in a quoted string.  The brackets are meta-syntax
to indicate optional tokens.)

For that matter, the external name could also be a valid c identifier,
but old or ported code still wants to use the old name.  Anyway, the
external name would not be available to the source code; it would have
to use the C identifier.  However, the compiler would generate object
code using the external name.

(2)  To define some data or a function with an external name that isn't
a valid C identifer:

"external$name" [maybe a type] c_identifer [maybe () or initializer, etc.]

Same criteria as for the first suggestion.  This one's a little harder on
a parser, because the string might look like the beginning of an expression
instead of a declaration.  However, it really isn't ambiguous, and it is
surrounded by semicolons (unless it's the first declaration).

The second suggestion is only of interest to anyone who wants C stuff to
be callable from other languages.  Admitted, fewer C programmers would
accept this kind of requirement, than for their own need to issue system
calls and such stuff in non-unix environments :-) .

-- 

   Norman Diamond

UUCP:  {decvax|utzoo|ihnp4|allegra|clyde}!watmath!watdaisy!ndiamond
CSNET: ndiamond%watdaisy@waterloo.csnet
ARPA:  ndiamond%watdaisy%waterloo.csnet@csnet-relay.arpa

"Opinions are those of the keyboard, and do not reflect on me or higher-ups."