Re: RFA: Proposed patch to fix incorrect table optimization for GLR

[Akim Demaille]

>>>>> "Paul" == Paul Hilfinger <address@hidden> writes:

Paul> I propose the following patch to fix a problem with GLR parsers.
Paul> When packing tables, the rows for two states may be merged if
Paul> they turn out to be identical (the table construction would not
Paul> normally produce redundant states, but they can result when
Paul> default reductions are substituted for some entries).  The
Paul> matching does not take into account extra information for the
Paul> GLR parser.  Here, I conservatively avoid merging states with
Paul> such information.  The glr-regr1.at test that I just checked in
Paul> shows the problem.

Paul> Comments?

It seems very right to me.

Do you know how the bug exhibited itself?  I mean, what was the
application?