]> git.openstreetmap.org Git - nominatim.git/blob - test/python/tools/test_tiger_data.py
ICU: better letter identification in normalization
[nominatim.git] / test / python / tools / test_tiger_data.py
1 # SPDX-License-Identifier: GPL-2.0-only
2 #
3 # This file is part of Nominatim. (https://nominatim.org)
4 #
5 # Copyright (C) 2022 by the Nominatim developer community.
6 # For a full list of authors see the git log.
7 """
8 Test for tiger data function
9 """
10 import tarfile
11 from textwrap import dedent
12
13 import pytest
14
15 from nominatim.tools import tiger_data
16 from nominatim.errors import UsageError
17
18 class MockTigerTable:
19
20     def __init__(self, conn):
21         self.conn = conn
22         with conn.cursor() as cur:
23             cur.execute("""CREATE TABLE tiger (linegeo GEOMETRY,
24                                                start INTEGER,
25                                                stop INTEGER,
26                                                interpol TEXT,
27                                                token_info JSONB,
28                                                postcode TEXT)""")
29
30     def count(self):
31         with self.conn.cursor() as cur:
32             return cur.scalar("SELECT count(*) FROM tiger")
33
34     def row(self):
35         with self.conn.cursor() as cur:
36             cur.execute("SELECT * FROM tiger LIMIT 1")
37             return cur.fetchone()
38
39 @pytest.fixture
40 def tiger_table(def_config, temp_db_conn, sql_preprocessor,
41                 temp_db_with_extensions, tmp_path):
42     def_config.lib_dir.sql = tmp_path / 'sql'
43     def_config.lib_dir.sql.mkdir()
44
45     (def_config.lib_dir.sql / 'tiger_import_start.sql').write_text(
46         """CREATE OR REPLACE FUNCTION tiger_line_import(linegeo GEOMETRY, start INTEGER,
47                                                         stop INTEGER, interpol TEXT,
48                                                         token_info JSONB, postcode TEXT)
49            RETURNS INTEGER AS $$
50             INSERT INTO tiger VALUES(linegeo, start, stop, interpol, token_info, postcode)
51             RETURNING 1
52            $$ LANGUAGE SQL;""")
53     (def_config.lib_dir.sql / 'tiger_import_finish.sql').write_text(
54         """DROP FUNCTION tiger_line_import (linegeo GEOMETRY, in_startnumber INTEGER,
55                                  in_endnumber INTEGER, interpolationtype TEXT,
56                                  token_info JSONB, in_postcode TEXT);""")
57
58     return MockTigerTable(temp_db_conn)
59
60
61 @pytest.fixture
62 def csv_factory(tmp_path):
63     def _mk_file(fname, hnr_from=1, hnr_to=9, interpol='odd', street='Main St',
64                  city='Newtown', state='AL', postcode='12345',
65                  geometry='LINESTRING(-86.466995 32.428956,-86.466923 32.428933)'):
66         (tmp_path / (fname + '.csv')).write_text(dedent("""\
67         from;to;interpolation;street;city;state;postcode;geometry
68         {};{};{};{};{};{};{};{}
69         """.format(hnr_from, hnr_to, interpol, street, city, state,
70                    postcode, geometry)))
71
72     return _mk_file
73
74
75 @pytest.mark.parametrize("threads", (1, 5))
76 def test_add_tiger_data(def_config, src_dir, tiger_table, tokenizer_mock, threads):
77     tiger_data.add_tiger_data(str(src_dir / 'test' / 'testdb' / 'tiger'),
78                               def_config, threads, tokenizer_mock())
79
80     assert tiger_table.count() == 6213
81
82
83 def test_add_tiger_data_no_files(def_config, tiger_table, tokenizer_mock,
84                                  tmp_path):
85     tiger_data.add_tiger_data(str(tmp_path), def_config, 1, tokenizer_mock())
86
87     assert tiger_table.count() == 0
88
89
90 def test_add_tiger_data_bad_file(def_config, tiger_table, tokenizer_mock,
91                                  tmp_path):
92     sqlfile = tmp_path / '1010.csv'
93     sqlfile.write_text("""Random text""")
94
95     tiger_data.add_tiger_data(str(tmp_path), def_config, 1, tokenizer_mock())
96
97     assert tiger_table.count() == 0
98
99
100 def test_add_tiger_data_hnr_nan(def_config, tiger_table, tokenizer_mock,
101                                 csv_factory, tmp_path):
102     csv_factory('file1', hnr_from=99)
103     csv_factory('file2', hnr_from='L12')
104     csv_factory('file3', hnr_to='12.4')
105
106     tiger_data.add_tiger_data(str(tmp_path), def_config, 1, tokenizer_mock())
107
108     assert tiger_table.count() == 1
109     assert tiger_table.row()['start'] == 99
110
111
112 @pytest.mark.parametrize("threads", (1, 5))
113 def test_add_tiger_data_tarfile(def_config, tiger_table, tokenizer_mock,
114                                 tmp_path, src_dir, threads):
115     tar = tarfile.open(str(tmp_path / 'sample.tar.gz'), "w:gz")
116     tar.add(str(src_dir / 'test' / 'testdb' / 'tiger' / '01001.csv'))
117     tar.close()
118
119     tiger_data.add_tiger_data(str(tmp_path / 'sample.tar.gz'), def_config, threads,
120                               tokenizer_mock())
121
122     assert tiger_table.count() == 6213
123
124
125 def test_add_tiger_data_bad_tarfile(def_config, tiger_table, tokenizer_mock,
126                                     tmp_path):
127     tarfile = tmp_path / 'sample.tar.gz'
128     tarfile.write_text("""Random text""")
129
130     with pytest.raises(UsageError):
131         tiger_data.add_tiger_data(str(tarfile), def_config, 1, tokenizer_mock())
132
133
134 def test_add_tiger_data_empty_tarfile(def_config, tiger_table, tokenizer_mock,
135                                       tmp_path):
136     tar = tarfile.open(str(tmp_path / 'sample.tar.gz'), "w:gz")
137     tar.add(__file__)
138     tar.close()
139
140     tiger_data.add_tiger_data(str(tmp_path / 'sample.tar.gz'), def_config, 1,
141                               tokenizer_mock())
142
143     assert tiger_table.count() == 0