Re: Berkeley yacc, was Looking for a Lex/Yacc free ebook

Thomas Dickey wrote:
> toby <toby@telegraphics.com.au> wrote:
> > Thomas Dickey wrote:
> >> > [Agree about flex, but Berkeley Yacc hasn't changed in at least a decade.
> >> > It doesn't need support. -John]
> >>
> >> see
> >> http://invisible-island.net/byacc
>
> > The aspersion on that page seems unfounded. It's getting difficult to
>
> At the time I wrote it, bison insisted on using alloca(). But see below -
>
> > find a mainstream UNIX that has not adopted gcc as its standard
> > compiler, but bison-2.1 builds with gcc/g++ in "-ansi -pedantic" mode,
>
> yes - it does, but watching bug reports, I notice that it doesn't
> necessarily accept grammars that yacc did (whether this is a good
> thing or not appears to depend on your attitude).
>
> > as well as with Intel icc "-ansi" and lcc (which is bare bones ANSI),
> > so I expect it is portable to pretty much any ANSI compiler.
>
> It would be nice if "-ansi" actually did what it's advertised.


In which of those three compilers? (One of which is all-ANSI, all the
time.) Why would a random ANSI compiler be more or less ANSI than gcc
or icc's "-ansi"?


(Bison 2.1 also builds with tcc, but the binary segfaults, which might
be my problem. :)


--Toby


>
> --
> Thomas E. Dickey
> http://invisible-island.net
> ftp://invisible-island.net

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