def test_counts_dump_2(germaparl):
strategy = 2
corpus = Corpus(germaparl['corpus_name'],
registry_path=germaparl['registry_path'])
dump = corpus.dump_from_query('[lemma="Helmut"%cd] [lemma="Kohl"%cd]')
df = corpus.counts.dump(dump,
p_atts=['word'],
split=True,
strategy=strategy)
assert (df["freq"]["Helmut"] == 6)
df = corpus.counts.dump(dump,
p_atts=['word', 'pos'],
split=True,
strategy=strategy)
assert (df["freq"][("Helmut", "NE")] == 6)
df = corpus.counts.dump(dump,
p_atts=['word'],
split=False,
strategy=strategy)
assert ("Helmut Kohl" in df.index)
assert (df["freq"].iloc[0] == 6)
df = corpus.counts.dump(dump,
p_atts=['word', 'pos'],
split=False,
strategy=strategy)
assert (("Helmut Kohl", "NE NE") in df.index)
assert (df["freq"].iloc[0] == 6)
def test_counts_dump_1_split(germaparl):
strategy = 1
corpus = Corpus(germaparl['corpus_name'],
registry_path=germaparl['registry_path'])
dump = corpus.dump_from_query('[lemma="die" %cd] [pos="N.*"]')
df = corpus.counts.dump(dump,
p_atts=['word'],
split=True,
strategy=strategy)
assert (int(df["freq"]["der"]) == 3775)
df = corpus.counts.dump(dump,
p_atts=['word', 'lemma'],
split=True,
strategy=strategy)
assert (int(df["freq"][("der", "die")]) == 3775)
def test_counts_dump_1_no_split(germaparl):
strategy = 1
corpus = Corpus(germaparl['corpus_name'],
registry_path=germaparl['registry_path'])
dump = corpus.dump_from_query('[lemma="Helmut"%cd] [lemma="Kohl"%cd]')
# no split
df = corpus.counts.dump(dump,
p_atts=['word'],
split=False,
strategy=strategy)
assert ("Helmut Kohl" in df.index)
df = corpus.counts.dump(dump,
p_atts=['word', 'pos'],
split=False,
strategy=strategy)
assert (("Helmut Kohl", "NE NE") in df.index)