Skip to content

test: Expand tests for built-in functions #129

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
andygrove merged 5 commits into from
Jan 24, 2023
Merged

test: Expand tests for built-in functions #129

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
02cf871
Expand math function tests
simicd Jan 10, 2023
45d5d2a
Expand string function tests
simicd Jan 12, 2023
e8402e3
Expand hash function tests
simicd Jan 12, 2023
6804cf4
Add temporal function tests
simicd Jan 15, 2023
8d0a2e4
Run formatter and fix failing tests after branch update
simicd Jan 19, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
199 changes: 190 additions & 9 deletions datafusion/tests/test_functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,7 @@
import numpy as np
import pyarrow as pa
import pytest
from datetime import datetime

from datafusion import SessionContext, column
from datafusion import functions as f
Expand All @@ -29,8 +30,19 @@ def df():
ctx = SessionContext()
# create a RecordBatch and a new DataFrame from it
batch = pa.RecordBatch.from_arrays(
[pa.array(["Hello", "World", "!"]), pa.array([4, 5, 6])],
names=["a", "b"],
[
pa.array(["Hello", "World", "!"]),
pa.array([4, 5, 6]),
pa.array(["hello ", " world ", " !"]),
pa.array(
[
datetime(2022, 12, 31),
datetime(2027, 6, 26),
datetime(2020, 7, 2),
]
),
],
names=["a", "b", "c", "d"],
)
return ctx.create_dataframe([[batch]])

Expand Down Expand Up @@ -91,6 +103,16 @@ def test_math_functions():
f.log2(col_v + literal(pa.scalar(1))),
f.log10(col_v + literal(pa.scalar(1))),
f.random(),
f.atan(col_v),
f.atan2(col_v, literal(pa.scalar(1.1))),
f.ceil(col_v),
f.floor(col_v),
f.power(col_v, literal(pa.scalar(3))),
f.pow(col_v, literal(pa.scalar(4))),
f.round(col_v),
f.sqrt(col_v),
f.signum(col_v),
f.trunc(col_v),
)
batches = df.collect()
assert len(batches) == 1
Expand All @@ -113,29 +135,116 @@ def test_math_functions():
result.column(9), np.log10(values + 1.0)
)
np.testing.assert_array_less(result.column(10), np.ones_like(values))
np.testing.assert_array_almost_equal(result.column(11), np.arctan(values))
np.testing.assert_array_almost_equal(
result.column(12), np.arctan2(values, 1.1)
)
np.testing.assert_array_almost_equal(result.column(13), np.ceil(values))
np.testing.assert_array_almost_equal(result.column(14), np.floor(values))
np.testing.assert_array_almost_equal(
result.column(15), np.power(values, 3)
)
np.testing.assert_array_almost_equal(
result.column(16), np.power(values, 4)
)
np.testing.assert_array_almost_equal(result.column(17), np.round(values))
np.testing.assert_array_almost_equal(result.column(18), np.sqrt(values))
np.testing.assert_array_almost_equal(result.column(19), np.sign(values))
np.testing.assert_array_almost_equal(result.column(20), np.trunc(values))


def test_string_functions(df):
df = df.select(f.md5(column("a")), f.lower(column("a")))
df = df.select(
f.ascii(column("a")),
f.bit_length(column("a")),
f.btrim(literal(" World ")),
f.character_length(column("a")),
f.chr(literal(68)),
f.concat_ws("-", column("a"), literal("test")),
f.concat(column("a"), literal("?")),
f.initcap(column("c")),
f.left(column("a"), literal(3)),
f.length(column("c")),
f.lower(column("a")),
f.lpad(column("a"), literal(7)),
f.ltrim(column("c")),
f.md5(column("a")),
f.octet_length(column("a")),
f.repeat(column("a"), literal(2)),
f.replace(column("a"), literal("l"), literal("?")),
f.reverse(column("a")),
f.right(column("a"), literal(4)),
f.rpad(column("a"), literal(8)),
f.rtrim(column("c")),
f.split_part(column("a"), literal("l"), literal(1)),
f.starts_with(column("a"), literal("Wor")),
f.strpos(column("a"), literal("o")),
f.substr(column("a"), literal(3)),
f.translate(column("a"), literal("or"), literal("ld")),
f.trim(column("c")),
f.upper(column("c")),
)
result = df.collect()
assert len(result) == 1
result = result[0]
assert result.column(0) == pa.array(
[72, 87, 33], type=pa.int32()
) # H = 72; W = 87; ! = 33
assert result.column(1) == pa.array([40, 40, 8], type=pa.int32())
assert result.column(2) == pa.array(["World", "World", "World"])
assert result.column(3) == pa.array([5, 5, 1], type=pa.int32())
assert result.column(4) == pa.array(["D", "D", "D"])
assert result.column(5) == pa.array(["Hello-test", "World-test", "!-test"])
assert result.column(6) == pa.array(["Hello?", "World?", "!?"])
assert result.column(7) == pa.array(["Hello ", " World ", " !"])
assert result.column(8) == pa.array(["Hel", "Wor", "!"])
assert result.column(9) == pa.array([6, 7, 2], type=pa.int32())
assert result.column(10) == pa.array(["hello", "world", "!"])
assert result.column(11) == pa.array([" Hello", " World", " !"])
assert result.column(12) == pa.array(["hello ", "world ", "!"])
assert result.column(13) == pa.array(
[
"8b1a9953c4611296a827abf8c47804d7",
"f5a7924e621e84c9280a9a27e1bcb7f6",
"9033e0e305f247c0c3c80d0c7848c8b3",
]
)
assert result.column(1) == pa.array(["hello", "world", "!"])
assert result.column(14) == pa.array([5, 5, 1], type=pa.int32())
assert result.column(15) == pa.array(["HelloHello", "WorldWorld", "!!"])
assert result.column(16) == pa.array(["He??o", "Wor?d", "!"])
assert result.column(17) == pa.array(["olleH", "dlroW", "!"])
assert result.column(18) == pa.array(["ello", "orld", "!"])
assert result.column(19) == pa.array(["Hello ", "World ", "! "])
assert result.column(20) == pa.array(["hello", " world", " !"])
assert result.column(21) == pa.array(["He", "Wor", "!"])
assert result.column(22) == pa.array([False, True, False])
assert result.column(23) == pa.array([5, 2, 0], type=pa.int32())
assert result.column(24) == pa.array(["llo", "rld", ""])
assert result.column(25) == pa.array(["Helll", "Wldld", "!"])
assert result.column(26) == pa.array(["hello", "world", "!"])
assert result.column(27) == pa.array(["HELLO ", " WORLD ", " !"])


def test_hash_functions(df):
exprs = [
f.digest(column("a"), literal(m))
for m in ("md5", "sha256", "sha512", "blake2s", "blake3")
for m in (
"md5",
"sha224",
"sha256",
"sha384",
"sha512",
"blake2s",
"blake3",
)
]
df = df.select(*exprs)
df = df.select(
*exprs,
f.sha224(column("a")),
f.sha256(column("a")),
f.sha384(column("a")),
f.sha512(column("a")),
)
result = df.collect()
assert len(result) == 1
result = result[0]
Expand All @@ -148,6 +257,13 @@ def test_hash_functions(df):
]
)
assert result.column(1) == pa.array(
[
b("4149DA18AA8BFC2B1E382C6C26556D01A92C261B6436DAD5E3BE3FCC"),
b("12972632B6D3B6AA52BD6434552F08C1303D56B817119406466E9236"),
b("6641A7E8278BCD49E476E7ACAE158F4105B2952D22AEB2E0B9A231A0"),
]
)
assert result.column(2) == pa.array(
[
b(
"185F8DB32271FE25F561A6FC938B2E26"
Expand All @@ -163,7 +279,26 @@ def test_hash_functions(df):
),
]
)
assert result.column(2) == pa.array(
assert result.column(3) == pa.array(
[
b(
"3519FE5AD2C596EFE3E276A6F351B8FC"
"0B03DB861782490D45F7598EBD0AB5FD"
"5520ED102F38C4A5EC834E98668035FC"
),
b(
"ED7CED84875773603AF90402E42C65F3"
"B48A5E77F84ADC7A19E8F3E8D3101010"
"22F552AEC70E9E1087B225930C1D260A"
),
b(
"1D0EC8C84EE9521E21F06774DE232367"
"B64DE628474CB5B2E372B699A1F55AE3"
"35CC37193EF823E33324DFD9A70738A6"
),
]
)
assert result.column(4) == pa.array(
[
b(
"3615F80C9D293ED7402687F94B22D58E"
Expand All @@ -185,7 +320,7 @@ def test_hash_functions(df):
),
]
)
assert result.column(3) == pa.array(
assert result.column(5) == pa.array(
[
b(
"F73A5FBF881F89B814871F46E26AD3FA"
Expand All @@ -201,7 +336,7 @@ def test_hash_functions(df):
),
]
)
assert result.column(4) == pa.array(
assert result.column(6) == pa.array(
[
b(
"FBC2B0516EE8744D293B980779178A35"
Expand All @@ -217,3 +352,49 @@ def test_hash_functions(df):
),
]
)
assert result.column(7) == result.column(1) # SHA-224
assert result.column(8) == result.column(2) # SHA-256
assert result.column(9) == result.column(3) # SHA-384
assert result.column(10) == result.column(4) # SHA-512


def test_temporal_functions(df):
df = df.select(
f.date_part(literal("month"), column("d")),
f.datepart(literal("year"), column("d")),
f.date_trunc(literal("month"), column("d")),
f.datetrunc(literal("day"), column("d")),
f.from_unixtime(literal(1673383974)),
f.to_timestamp(literal("2023-09-07 05:06:14.523952")),
f.to_timestamp_seconds(literal("2023-09-07 05:06:14.523952")),
f.to_timestamp_millis(literal("2023-09-07 05:06:14.523952")),
f.to_timestamp_micros(literal("2023-09-07 05:06:14.523952")),
)
result = df.collect()
assert len(result) == 1
result = result[0]
assert result.column(0) == pa.array([12, 6, 7], type=pa.float64())
assert result.column(1) == pa.array([2022, 2027, 2020], type=pa.float64())
assert result.column(2) == pa.array(
[datetime(2022, 12, 1), datetime(2027, 6, 1), datetime(2020, 7, 1)],
type=pa.timestamp("ns"),
)
assert result.column(3) == pa.array(
[datetime(2022, 12, 31), datetime(2027, 6, 26), datetime(2020, 7, 2)],
type=pa.timestamp("ns"),
)
assert result.column(4) == pa.array(
[datetime(2023, 1, 10, 20, 52, 54)] * 3, type=pa.timestamp("s")
)
assert result.column(5) == pa.array(
[datetime(2023, 9, 7, 5, 6, 14, 523952)] * 3, type=pa.timestamp("ns")
)
assert result.column(6) == pa.array(
[datetime(2023, 9, 7, 5, 6, 14)] * 3, type=pa.timestamp("s")
)
assert result.column(7) == pa.array(
[datetime(2023, 9, 7, 5, 6, 14, 523000)] * 3, type=pa.timestamp("ms")
)
assert result.column(8) == pa.array(
[datetime(2023, 9, 7, 5, 6, 14, 523952)] * 3, type=pa.timestamp("us")
)