fix spurious failure of ltlcross4.test

Reported by Yuri Victorovich.

* tests/core/ltlcross4.test: Drop the 'formula' column before
computing aggregates.  It causes warnings in some Pandas versions, and
errors in others.
This commit is contained in:
Alexandre Duret-Lutz 2023-06-12 11:01:30 +02:00
parent 7868115a8b
commit 8369663380

View file

@ -1,7 +1,7 @@
#!/bin/sh
# -*- coding: utf-8 -*-
# Copyright (C) 2012-2014, 2017, 2020 Laboratoire de Recherche et
# Développement de l'Epita (LRDE).
# Copyright (C) 2012-2014, 2017, 2020, 2023 Laboratoire de Recherche
# et Développement de l'Epita (LRDE).
#
# This file is part of Spot, a model checking library.
#
@ -54,8 +54,7 @@ x = pandas.read_csv("output.csv")
# We used to call describe() instead of agg(),
# but the output of this function was changed
# in pandas 0.20.
print(x.filter(('formula', 'tool',
'states', 'transitions')).\
print(x.filter(('tool', 'states', 'transitions')).\
groupby('tool').\
agg([np.mean, np.std, np.min, np.max]))
EOF