|
1
|
|
|
"""Unit tests for corpus functionality in Annif""" |
|
2
|
|
|
|
|
3
|
|
|
import annif.corpus |
|
4
|
|
|
|
|
5
|
|
|
|
|
6
|
|
|
def test_subjectset_uris(): |
|
7
|
|
|
data = """<http://example.org/dummy>\tdummy |
|
8
|
|
|
<http://example.org/another>\tanother |
|
9
|
|
|
""" |
|
10
|
|
|
|
|
11
|
|
|
sset = annif.corpus.SubjectSet(data) |
|
12
|
|
|
assert sset.has_uris() |
|
13
|
|
|
assert len(sset.subject_uris) == 2 |
|
14
|
|
|
assert "http://example.org/dummy" in sset.subject_uris |
|
15
|
|
|
assert "http://example.org/another" in sset.subject_uris |
|
16
|
|
|
|
|
17
|
|
|
|
|
18
|
|
|
def test_subjectset_labels(): |
|
19
|
|
|
data = """dummy |
|
20
|
|
|
another |
|
21
|
|
|
""" |
|
22
|
|
|
|
|
23
|
|
|
sset = annif.corpus.SubjectSet(data) |
|
24
|
|
|
assert not sset.has_uris() |
|
25
|
|
|
assert len(sset.subject_labels) == 2 |
|
26
|
|
|
assert "dummy" in sset.subject_labels |
|
27
|
|
|
assert "another" in sset.subject_labels |
|
28
|
|
|
|
|
29
|
|
|
|
|
30
|
|
|
def test_docdir_key(tmpdir): |
|
31
|
|
|
tmpdir.join('doc1.txt').write('doc1') |
|
32
|
|
|
tmpdir.join('doc1.key').write('key1') |
|
33
|
|
|
tmpdir.join('doc2.txt').write('doc2') |
|
34
|
|
|
tmpdir.join('doc2.key').write('key2') |
|
35
|
|
|
tmpdir.join('doc3.txt').write('doc3') |
|
36
|
|
|
|
|
37
|
|
|
docdir = annif.corpus.DocumentDirectory(str(tmpdir)) |
|
38
|
|
|
files = sorted(list(docdir)) |
|
39
|
|
|
assert len(files) == 3 |
|
40
|
|
|
assert files[0][0] == str(tmpdir.join('doc1.txt')) |
|
41
|
|
|
assert files[0][1] == str(tmpdir.join('doc1.key')) |
|
42
|
|
|
assert files[1][0] == str(tmpdir.join('doc2.txt')) |
|
43
|
|
|
assert files[1][1] == str(tmpdir.join('doc2.key')) |
|
44
|
|
|
assert files[2][0] == str(tmpdir.join('doc3.txt')) |
|
45
|
|
|
assert files[2][1] is None |
|
46
|
|
|
|
|
47
|
|
|
|
|
48
|
|
|
def test_docdir_tsv(tmpdir): |
|
49
|
|
|
tmpdir.join('doc1.txt').write('doc1') |
|
50
|
|
|
tmpdir.join('doc1.tsv').write('<http://example.org/key1>\tkey1') |
|
51
|
|
|
tmpdir.join('doc2.txt').write('doc2') |
|
52
|
|
|
tmpdir.join('doc2.tsv').write('<http://example.org/key2>\tkey2') |
|
53
|
|
|
tmpdir.join('doc3.txt').write('doc3') |
|
54
|
|
|
|
|
55
|
|
|
docdir = annif.corpus.DocumentDirectory(str(tmpdir)) |
|
56
|
|
|
files = sorted(list(docdir)) |
|
57
|
|
|
assert len(files) == 3 |
|
58
|
|
|
assert files[0][0] == str(tmpdir.join('doc1.txt')) |
|
59
|
|
|
assert files[0][1] == str(tmpdir.join('doc1.tsv')) |
|
60
|
|
|
assert files[1][0] == str(tmpdir.join('doc2.txt')) |
|
61
|
|
|
assert files[1][1] == str(tmpdir.join('doc2.tsv')) |
|
62
|
|
|
assert files[2][0] == str(tmpdir.join('doc3.txt')) |
|
63
|
|
|
assert files[2][1] is None |
|
64
|
|
|
|
|
65
|
|
|
|
|
66
|
|
|
def test_docdir_key_require_subjects(tmpdir): |
|
67
|
|
|
tmpdir.join('doc1.txt').write('doc1') |
|
68
|
|
|
tmpdir.join('doc1.key').write('<http://example.org/key1>\tkey1') |
|
69
|
|
|
tmpdir.join('doc2.txt').write('doc2') |
|
70
|
|
|
tmpdir.join('doc2.key').write('<http://example.org/key2>\tkey2') |
|
71
|
|
|
tmpdir.join('doc3.txt').write('doc3') |
|
72
|
|
|
|
|
73
|
|
|
docdir = annif.corpus.DocumentDirectory(str(tmpdir), require_subjects=True) |
|
74
|
|
|
files = sorted(list(docdir)) |
|
75
|
|
|
assert len(files) == 2 |
|
76
|
|
|
assert files[0][0] == str(tmpdir.join('doc1.txt')) |
|
77
|
|
|
assert files[0][1] == str(tmpdir.join('doc1.key')) |
|
78
|
|
|
assert files[1][0] == str(tmpdir.join('doc2.txt')) |
|
79
|
|
|
assert files[1][1] == str(tmpdir.join('doc2.key')) |
|
80
|
|
|
|
|
81
|
|
|
|
|
82
|
|
|
def test_docdir_tsv_require_subjects(tmpdir): |
|
83
|
|
|
tmpdir.join('doc1.txt').write('doc1') |
|
84
|
|
|
tmpdir.join('doc1.tsv').write('key1') |
|
85
|
|
|
tmpdir.join('doc2.txt').write('doc2') |
|
86
|
|
|
tmpdir.join('doc2.tsv').write('key2') |
|
87
|
|
|
tmpdir.join('doc3.txt').write('doc3') |
|
88
|
|
|
|
|
89
|
|
|
docdir = annif.corpus.DocumentDirectory(str(tmpdir), require_subjects=True) |
|
90
|
|
|
files = sorted(list(docdir)) |
|
91
|
|
|
assert len(files) == 2 |
|
92
|
|
|
assert files[0][0] == str(tmpdir.join('doc1.txt')) |
|
93
|
|
|
assert files[0][1] == str(tmpdir.join('doc1.tsv')) |
|
94
|
|
|
assert files[1][0] == str(tmpdir.join('doc2.txt')) |
|
95
|
|
|
assert files[1][1] == str(tmpdir.join('doc2.tsv')) |
|
96
|
|
|
|
|
97
|
|
|
|
|
98
|
|
|
def test_subjdir(tmpdir): |
|
99
|
|
|
tmpdir.join('subj1.txt').write("""http://example.org/subj1 subject one |
|
100
|
|
|
first subject |
|
101
|
|
|
this is the first thing we know about""") |
|
102
|
|
|
tmpdir.join('subj2.txt').write("""http://example.org/subj2 subject two |
|
103
|
|
|
second subject |
|
104
|
|
|
this is the second thing we know about""") |
|
105
|
|
|
tmpdir.join('subj3.txt').write("""http://example.org/subj3 subject three |
|
106
|
|
|
third subject |
|
107
|
|
|
this is the third thing we know about""") |
|
108
|
|
|
|
|
109
|
|
|
subjdir = annif.corpus.SubjectDirectory(str(tmpdir)) |
|
110
|
|
|
subjects = sorted(list(subjdir), key=lambda subj: subj.uri) |
|
111
|
|
|
assert len(subjects) == 3 |
|
112
|
|
|
assert subjects[0].uri == 'http://example.org/subj1' |
|
113
|
|
|
assert subjects[0].label == 'subject one' |
|
114
|
|
|
assert 'first' in subjects[0].text |
|
115
|
|
|
assert subjects[1].uri == 'http://example.org/subj2' |
|
116
|
|
|
assert subjects[1].label == 'subject two' |
|
117
|
|
|
assert 'second' in subjects[1].text |
|
118
|
|
|
assert subjects[2].uri == 'http://example.org/subj3' |
|
119
|
|
|
assert subjects[2].label == 'subject three' |
|
120
|
|
|
assert 'third' in subjects[2].text |
|
121
|
|
|
|
NatLibFi /
Annif
