Re: Lookahead vs. Scanner Feedback

hjelm+@cs.cmu.edu (Mark Hjelm) writes:


>How do I KNOW that there are no situations
>which, due to parser lookahead, would cause the scanner to return an
>incorrect token type for an identifier (i.e. return "identifier", even
>though the identifier was just/will be made into a "typedef name")?


The way to KNOW would be to prove that the scanner and parser conform to
specifications (not easy, since we have neither an exact spec or a
semantics of C :-), and to prove properties of one's microsyntax or
grammar.


Another way to handle the identifier/keyword problem is to introduce a
token class <name> = <letter> {<letter> | <digit>} or whatever, and then
either insert a "filtering pass" that checks each name against a list of
reserved words and replaces the token with <identifier> or the appropriate
keyword token (I feel the "but-that's-grossly-inefficient" flames already,
but I'd rather be right than fast :-), or else to insert a blub of code at
the point that the name token has been recognized to differentiate between
the two.


--
Debora Weber-Wulff dww@inf.fu-berlin.de
Institut fuer Informatik +49 30 89691 124
Nestorstr. 8-9
D-W-1000 Berlin 31
--

Return to the comp.compilers page.
Search the comp.compilers archives again.