The patch looks fine. Thanks!

Thank you Malte, patch works perfectly.

Hi kori,

the problem is that the translator detects that the goal violates a mutex and is
hence unreachable, but there is no code to handle this case. I've made a small
change to fix this.

Our public repository is currently frozen because of the IPC, but I have
attached a patch. Can you apply it, or do you need help with that?

Gabi, can you review the patch? Because I think it is likely to be fine, it is
already merged into default in your bitbucket repository; you can see it with:

    hg diff -c df1f36aa9cf3

If you like the change, you can mark this as resolved. I'll also save the
problem as a test case for the future.

Hi to all,
 when I attempt to run following very simple blocksworld problem (in attachment, domain is in previous ), which in fact has no solution 
(goal is to have one box put on two different boxes) it fails with following output:

Traceback (most recent call last):
  File "./src/translate/translate.py", line 665, in <module>
    main()
  File "./src/translate/translate.py", line 655, in main
    sas_task = pddl_to_sas(task)
  File "./src/translate/translate.py", line 518, in pddl_to_sas
    implied_facts)
  File "./src/translate/translate.py", line 435, in translate_task
    assert len(goal_dict_list) == 1, "Negative goal not supported"
TypeError: object of type 'NoneType' has no len()

It is intentionally unsolvable, but I hoped to receive "Completely explored state space -- no solution!" in the output.
How should I interpret this failure? Would it be possible to fix it?
Thanks