1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
|
# -*- coding: utf-8 -*-
"""test_converted_types.py - tests for decoding data to their logical data types."""
import datetime
import os.path
import numpy as np
import pandas as pd
import pytest
from fastparquet import parquet_thrift as pt
from fastparquet.converted_types import convert
def test_int32():
"""Test decimal data stored as int32."""
schema = pt.SchemaElement(
type=pt.Type.INT32,
name="test",
converted_type=pt.ConvertedType.DECIMAL,
scale=10,
precision=9
)
assert (convert(pd.Series([9876543210]), schema)[0] - 9.87654321) < 0.01
def test_date():
"""Test int32 encoding a date."""
schema = pt.SchemaElement(
type=pt.Type.INT32,
name="test",
converted_type=pt.ConvertedType.DATE,
)
days = (datetime.date(2004, 11, 3) - datetime.date(1970, 1, 1)).days
data = pd.Series([days]).to_numpy()
data.flags.writeable = False
assert (convert(data, schema)[0] ==
pd.to_datetime([datetime.date(2004, 11, 3)]))
def test_time_millis():
"""Test int32 encoding a timedelta in millis."""
schema = pt.SchemaElement(
type=pt.Type.INT32,
name="test",
converted_type=pt.ConvertedType.TIME_MILLIS,
)
assert (convert(np.array([731888], dtype='int32'), schema)[0] ==
np.array([731888], dtype='timedelta64[ms]'))
def test_timestamp_millis():
"""Test int64 encoding a datetime."""
schema = pt.SchemaElement(
type=pt.Type.INT64,
name="test",
converted_type=pt.ConvertedType.TIMESTAMP_MILLIS,
)
assert (convert(np.array([1099511625014], dtype='int64'), schema)[0] ==
np.array(datetime.datetime(2004, 11, 3, 19, 53, 45, 14 * 1000),
dtype='datetime64[ns]'))
def test_utf8():
"""Test bytes representing utf-8 string."""
schema = pt.SchemaElement(
type=pt.Type.BYTE_ARRAY,
name="test",
converted_type=pt.ConvertedType.UTF8
)
data = u"Ördög" # conversion now happens on read
assert convert(pd.Series([data]), schema)[0] == u"Ördög"
def test_json():
"""Test bytes representing json."""
schema = pt.SchemaElement(
type=pt.Type.BYTE_ARRAY,
name="test",
converted_type=pt.ConvertedType.JSON
)
assert convert(pd.Series([b'{"foo": ["bar", "\\ud83d\\udc7e"]}']),
schema)[0] == {'foo': ['bar', u'👾']}
def test_bson():
"""Test bytes representing bson."""
bson = pytest.importorskip('bson')
schema = pt.SchemaElement(
type=pt.Type.BYTE_ARRAY,
name="test",
converted_type=pt.ConvertedType.BSON
)
assert convert(pd.Series(
[b'&\x00\x00\x00\x04foo\x00\x1c\x00\x00\x00\x020'
b'\x00\x04\x00\x00\x00bar\x00\x021\x00\x05\x00'
b'\x00\x00\xf0\x9f\x91\xbe\x00\x00\x00']),
schema)[0] == {'foo': ['bar', u'👾']}
def test_uint16():
"""Test decoding int32 as uint16."""
schema = pt.SchemaElement(
type=pt.Type.INT32,
name="test",
converted_type=pt.ConvertedType.UINT_16
)
assert convert(pd.Series([-3]), schema)[0] == 65533
def test_uint32():
"""Test decoding int32 as uint32."""
schema = pt.SchemaElement(
type=pt.Type.INT32,
name="test",
converted_type=pt.ConvertedType.UINT_32
)
assert convert(pd.Series([-6884376]), schema)[0] == 4288082920
def test_uint64():
"""Test decoding int64 as uint64."""
schema = pt.SchemaElement(
type=pt.Type.INT64,
name="test",
converted_type=pt.ConvertedType.UINT_64
)
assert convert(pd.Series([-6884376]), schema)[0] == 18446744073702667240
def test_big_decimal():
schema = pt.SchemaElement(
type=pt.Type.FIXED_LEN_BYTE_ARRAY,
name="test",
converted_type=pt.ConvertedType.DECIMAL,
type_length=32,
scale=1,
precision=38
)
pad = b'\x00' * 16
data = np.array([
pad, pad + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x1e\\',
pad + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x1d\\',
pad + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\r{',
pad + b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x19)'],
dtype='|S32')
assert np.isclose(convert(data, schema),
np.array([0., 777.2, 751.6, 345.1, 644.1])).all()
def test_tz_nonstring(tmpdir):
# https://github.com/dask/fastparquet/issues/578
import uuid
event = {}
event_id = str(uuid.uuid4())
event['id'] = [event_id]
event['site_name'] = ['TestString']
event['start_time'] = ['2021-01-01T14:58:19.3677-05:00']
event['end_time'] = ['2021-01-01T14:59:50.5272-05:00']
event_df = pd.DataFrame(event)
event_df['start_time'] = pd.to_datetime(event_df['start_time'])
event_df['end_time'] = pd.to_datetime(event_df['end_time'])
fn = '{}/{}.parquet'.format(tmpdir, event_id)
event_df.to_parquet(fn, compression='uncompressed', engine='fastparquet')
round = pd.read_parquet(fn, engine="fastparquet")
assert (event_df == round).all().all()
def test_pandas_simple_type(tmpdir):
import pandas as pd
fn = os.path.join(tmpdir, "out.parquet")
df = pd.DataFrame({"a": [1, 2, 3]}, dtype='uint8')
df.to_parquet(fn, engine="fastparquet")
df2 = pd.read_parquet(fn, engine="fastparquet")
assert df2.a.dtype == "uint8"
assert not(isinstance(df2.a.dtype, pd.UInt8Dtype))
|
