[developers] Bug in mt/vpm.lisp
oe at csli.Stanford.EDU
Thu Mar 22 00:24:44 CET 2007
> Today had me victim of a nasty bug in map-mrs():
> Probably map-mrs should check for the type of MRS coming in, or else
> presnece of these features.
i am afraid your suffering is related to parallel universes. this bug
is present only in the LinGO version of that code, and the copy in the
LOGON tree has the checking you suggest.
while i have yet to merge in the last four months of LOGON development
to the LinGO repository, ben (needing characterization support) put in
a patch (in december), which results in the behavior you saw.
but there is good news about parallel universes too. i was planning to
do my quaterly code synchronization in the second half of next week, so
that this problem should soon disappear.
best - oe
nb: and we should probably re-evaluate our compiler settings, at least
for the run-time binaries it might make sense to produce safer code at
a moderate loss in efficiency.
+++ Universitetet i Oslo (IFI); Boks 1080 Blindern; 0316 Oslo; (+47) 2284 0125
+++ CSLI Stanford; Ventura Hall; Stanford, CA 94305; (+1 650) 723 0515
+++ --- oe at csli.stanford.edu; oe at ifi.uio.no; oepen at idi.ntnu.no ---
More information about the developers