|[8 earlier articles]|
|Re: Why Can't We Build a C Compiler? email@example.com (Frode Odegard) (1988-12-29)|
|Re: Why Can't We Build a C Compiler? firstname.lastname@example.org (1989-01-05)|
|Why can't we build a C compiler? think!compass!worley@EDDIE.MIT.EDU (1988-12-19)|
|Re: Why Can't We Build a C Compiler? email@example.com (1989-01-03)|
|Re: Why can't we build a C compiler? uokmax!glcowin@Central.Sun.COM (1989-01-18)|
|Re: Why can't we build a C compiler? firstname.lastname@example.org (1989-01-24)|
|Re: Why can't we build a C compiler? waterloo.edu!cognos!rayt@RELAY.CS.NET (R.) (1989-01-25)|
|Re: Why can't we build a C compiler? email@example.com (1989-01-25)|
|Date:||Wed, 25 Jan 89 17:38:21 EST|
|Organization:||Cognos Inc., Ottawa, Canada|
In article <firstname.lastname@example.org> Greg Cowin, in reference to C,
indicates that semantics is a major problem in ANY language
definition, since the present methods of its specification are
either informal or "complex and arcane"; one can hardly disagree.
However, he continues, "[i]n fact, to some degree the grammar
is insignificant. Formalization of syntax is a simpler problem
than the formalization of semantics." My comment here is that
one method of defining the semantics is THROUGH a grammar; in
my case, Van Wijngaarden grammars, which allow one full recursively
ennumerable capabilities. A major problem with such systems is
efficiency; but techniques are becoming available which allow
the parsing of such grammars in polynomial time. The theoretical
power is available, but it seems the trick is to actually FIND such
a grammar definition for normal programming languages. This is
an area of present research.
Ray Tigg 3755 Riverside Dr.
Cognos Incorporated Ottawa, Ontario
(613) 738-1440 x5013 CANADA K1G 3Z4
Return to the
Search the comp.compilers archives again.