Repository navigation
Expand file tree
/
Copy pathsql-using-python-udaf.py
More file actions
86 lines (72 loc) · 2.56 KB
/
Copy pathsql-using-python-udaf.py
File metadata and controls
86 lines (72 loc) · 2.56 KB
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
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
import pyarrow as pa
from datafusion import Accumulator, SessionContext, udaf
# Define a user-defined aggregation function (UDAF)
class MyAccumulator(Accumulator):
"""
Interface of a user-defined accumulation.
"""
def __init__(self) -> None:
self._sum = pa.scalar(0.0)
def update(self, values: list[pa.Array]) -> None:
# not nice since pyarrow scalars can't be summed yet. This breaks on `None`
self._sum = pa.scalar(self._sum.as_py() + pa.compute.sum(values).as_py())
def merge(self, states: pa.Array) -> None:
# not nice since pyarrow scalars can't be summed yet. This breaks on `None`
self._sum = pa.scalar(self._sum.as_py() + pa.compute.sum(states[0]).as_py())
def state(self) -> list[pa.Array]:
return [self._sum]
def evaluate(self) -> pa.Scalar:
return self._sum
my_udaf = udaf(
MyAccumulator,
pa.float64(),
pa.float64(),
[pa.float64()],
"stable",
# This will be the name of the UDAF in SQL
# If not specified it will by default the same as accumulator class name
name="my_accumulator",
)
# Create a context
ctx = SessionContext()
# Create a datafusion DataFrame from a Python dictionary
source_df = ctx.from_pydict({"a": [1, 1, 3], "b": [4, 5, 6]}, name="t")
# Dataframe:
# +---+---+
# | a | b |
# +---+---+
# | 1 | 4 |
# | 1 | 5 |
# | 3 | 6 |
# +---+---+
# Register UDF for use in SQL
ctx.register_udaf(my_udaf)
# Query the DataFrame using SQL
result_df = ctx.sql(
"select a, my_accumulator(b) as b_aggregated from t group by a order by a"
)
# Dataframe:
# +---+--------------+
# | a | b_aggregated |
# +---+--------------+
# | 1 | 9 |
# | 3 | 6 |
# +---+--------------+
assert result_df.to_pydict()["a"] == [1, 3]
assert result_df.to_pydict()["b_aggregated"] == [9, 6]