Path: utzoo!telly!ddsw1!lll-winken!killer!mit-eddie!bloom-beacon!tut.cis.ohio-state.edu!BOURBAKI.MIT.EDU!ghoti
From: ghoti@BOURBAKI.MIT.EDU
Newsgroups: gnu.gcc.bug
Subject: (none)
Message-ID: <8810220516.AA09152@hypatia.mit.edu>
Date: 22 Oct 88 05:16:42 GMT
Sender: daemon@tut.cis.ohio-state.edu
Distribution: gnu
Organization: GNUs Not Usenet
Lines: 14

The info file which accompanies the latest version of bison is most welcome
and on the whole I find it quite well written. One thing that bothers me
is that nothing is really said about how one deals with the fact that C is
not LALR(1) when writing parse.y (see section 5.10.3 of Harbison & Steele).
Strictly speaking, it is not necessary to go into this to explain bison
but it would provide another example of the use of bison and one that
addresses an important aspect of an important application of it. Alternatively,
now that bison has good documentation, maybe the GCC internals file can say
more about how one uses bison to get around this aspect of C. 

I am not raising this point to be pedantic or for the sake of completeness.
I would like to know the answer myself.

Allan Adler
ghoti@cauchy.mit.edu