p variables seems to be rejected by Utool

[arademaker]

See delph-in/erg#39

The MRS seems to be valid. But Utool rejected it:

psoa(h0,e2,[rel('_be_v_id',h1,[attrval('ARG0',e2),attrval('ARG1',p3),attrval('ARG2',x4)]),
	    rel('udef_q',h5,[attrval('ARG0',x4),attrval('RSTR',h6),attrval('BODY',h7)]),
	    rel('compound',h8,[attrval('ARG0',e9),attrval('ARG1',x4),attrval('ARG2',x10)]),
	    rel('udef_q',h11,[attrval('ARG0',x10),attrval('RSTR',h12),attrval('BODY',h13)]),
	    rel('_candidate_n_1',h14,[attrval('ARG0',x10)]),
	    rel('_photoactive/nn_u_unknown',h8,[attrval('ARG0',x4)])],
     hcons([qeq(h0,h1),qeq(h6,h8),qeq(h12,h14)]))

Traceback (most recent call last):
  File "MrsPrologInputCodec.java", line 35, in de.saar.chorus.domgraph.codec.mrs.MrsPrologInputCodec.decode
de.saar.chorus.domgraph.codec.mrs.de.saar.chorus.domgraph.codec.mrs.TokenMgrError: de.saar.chorus.domgraph.codec.mrs.TokenMgrError: Lexical error at line 1, column 67.  Encountered: "3" (51), after : "p"

Any idea?