Compare Berkeley yacc to Sun's yacc

I am currently working on a project developing a graphics editor. We are
using yacc to define the command grammar for the program. We have now
developed the program to the point where there are over 50 commands and we
are running into table limits for the version of yacc we are using. We
are using the yacc program provided with Sun's 4.1.x version of Unix.


We are in the process of converting to Sun's Solaris operating system.
The version of yacc provided under Solaris is built with higher table
limits, but still has limitations that our current development is
breaking.


We would like to use the Berkeley public domain version of yacc. When we
tested version 1.9 of Berkeley yacc we no longer had problems with table
limitations.


However, because the Berkeley version is public domain, we need to justify
to my management that the Berkeley version is worth using. I would be
interested in comments from other users on their experience with Berkeley
yacc and how they feel it compares to other implementations of yacc,
particularly Sun's yacc.


I would also be interested in where the best (or a good) ftp site is to
check for updates to Berkeley yacc.


Thanks,


Bryan Heil
[Last question first, byacc hasn't been updated in a long time because it
contains very few bugs. I'd say that compared to commercial versions of
yacc, byacc is probably faster and more reliable. Since you have source,
overflowing fixed size tables is no problem, and support can be found on
the net. Every version of yacc I've seen with a commercial UNIX system is
straight from USL (or even straight from AT&T) so in practice, bugs never
get fixed. -John]
--

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