Issue280

Title translator: failed assertion about normalized conditions
Priority bug Status resolved
Superseder Nosy List gabi, malte
Assigned To malte Keywords
Optional summary

Created on 2011-09-14.15:34:23 by gabi, last changed by malte.

Files
File name Uploaded Type Edit Remove
domain.pddl gabi, 2011-09-14.15:34:40 application/octet-stream
problem-new.pddl malte, 2014-08-08.20:01:36 application/octet-stream
problem.pddl gabi, 2011-09-14.15:34:22 application/octet-stream
Messages
msg3289 (view) Author: malte Date: 2014-08-08.21:57:36
Merged.
msg3288 (view) Author: gabi Date: 2014-08-08.21:55:23
Thanks, Malte! The patch looks good. Do you merge it?
msg3287 (view) Author: malte Date: 2014-08-08.21:23:28
Looks ready to be merged to me. Gabi, do you think it's OK?

https://bitbucket.org/malte/downward/commits/branch/issue280

With the change, the translator succeeds on the given problem. However, the
search component then complains that there is no goal condition. This is
actually correct for this task, and the question is how to best handle these
things. I think this is a separate issue, though. I'll bring it up on the
mailing list.
msg3286 (view) Author: malte Date: 2014-08-08.20:01:36
Uploading the new, modified problem to reproduce the issue (to be used together
with the original domain.pddl).
msg3285 (view) Author: malte Date: 2014-08-08.19:59:43
This is fixed with issue335. However, if you change the goal in the example
problem to

  (:goal (forall (?x)
                 (and
;                  (pred ?x)
;                  (ok ?x)
                  )))

(i.e., comment out the two atoms in the conjunction), the translator fails with
an assertion error:

AssertionError: Condition not normalized

I'm changing the title of this issue to address this other bug.
msg1768 (view) Author: malte Date: 2011-09-14.15:38:58
I don't think it has to be a release blocker, so doesn't have to be marked 1.0.
But of course it would be good to fix it.
msg1767 (view) Author: gabi Date: 2011-09-14.15:34:22
The attached task comes from issue163. It
throws an AssertionError ("Negated axiom impossible; cannot deal with that")
which should easily be avoidable for constant predicates.

Should this be marked as a bug for 1.0?
History
Date User Action Args
2014-08-08 21:57:36maltesetstatus: reviewing -> resolved
messages: + msg3289
2014-08-08 21:55:23gabisetmessages: + msg3288
2014-08-08 21:23:32maltesetstatus: chatting -> reviewing
2014-08-08 21:23:28maltesetmessages: + msg3287
2014-08-08 20:01:36maltesetfiles: + problem-new.pddl
messages: + msg3286
2014-08-08 19:59:43maltesetassignedto: malte
messages: + msg3285
title: negated axioms with constant predicates not handled properly -> translator: failed assertion about normalized conditions
2011-09-14 15:38:59maltesetstatus: unread -> chatting
messages: + msg1768
2011-09-14 15:34:40gabisetfiles: + domain.pddl
2011-09-14 15:34:23gabicreate