def test_fields(self):
fields = collections.OrderedDict([
("int_field", DataTypes.INT()), ("long_field", DataTypes.BIGINT()),
("string_field", DataTypes.STRING()),
("timestamp_field", DataTypes.TIMESTAMP(3)),
("time_field", DataTypes.TIME()), ("date_field", DataTypes.DATE()),
("double_field", DataTypes.DOUBLE()),
("float_field", DataTypes.FLOAT()),
("byte_field", DataTypes.TINYINT()),
("short_field", DataTypes.SMALLINT()),
("boolean_field", DataTypes.BOOLEAN())
])
schema = Schema().fields(fields)
properties = schema.to_properties()
expected = {
'schema.0.name': 'int_field',
'schema.0.data-type': 'INT',
'schema.1.name': 'long_field',
'schema.1.data-type': 'BIGINT',
'schema.2.name': 'string_field',
'schema.2.data-type': 'VARCHAR(2147483647)',
'schema.3.name': 'timestamp_field',
'schema.3.data-type': 'TIMESTAMP(3)',
'schema.4.name': 'time_field',
'schema.4.data-type': 'TIME(0)',
'schema.5.name': 'date_field',
'schema.5.data-type': 'DATE',
'schema.6.name': 'double_field',
'schema.6.data-type': 'DOUBLE',
'schema.7.name': 'float_field',
'schema.7.data-type': 'FLOAT',
'schema.8.name': 'byte_field',
'schema.8.data-type': 'TINYINT',
'schema.9.name': 'short_field',
'schema.9.data-type': 'SMALLINT',
'schema.10.name': 'boolean_field',
'schema.10.data-type': 'BOOLEAN'
}
self.assertEqual(expected, properties)
if sys.version_info[:2] <= (3, 5):
fields = {
"int_field": DataTypes.INT(),
"long_field": DataTypes.BIGINT(),
"string_field": DataTypes.STRING(),
"timestamp_field": DataTypes.TIMESTAMP(3),
"time_field": DataTypes.TIME(),
"date_field": DataTypes.DATE(),
"double_field": DataTypes.DOUBLE(),
"float_field": DataTypes.FLOAT(),
"byte_field": DataTypes.TINYINT(),
"short_field": DataTypes.SMALLINT(),
"boolean_field": DataTypes.BOOLEAN()
}
self.assertRaises(TypeError, Schema().fields, fields)
def sql_type(cls):
return DataTypes.ARRAY(DataTypes.DOUBLE(False))
def test_decimal_type(self):
t1 = DataTypes.DECIMAL(10, 0)
t2 = DataTypes.DECIMAL(10, 2)
self.assertTrue(t2 is not t1)
self.assertNotEqual(t1, t2)
def test_array_type(self):
# nullable/not_null flag will be lost during the conversion.
test_types = [
DataTypes.ARRAY(DataTypes.BIGINT()),
DataTypes.ARRAY(DataTypes.BIGINT()),
DataTypes.ARRAY(DataTypes.STRING()),
DataTypes.ARRAY(DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.ARRAY(DataTypes.ARRAY(DataTypes.STRING()))
]
java_types = [_to_java_type(item) for item in test_types]
converted_python_types = [_from_java_type(item) for item in java_types]
self.assertEqual(test_types, converted_python_types)
def test_struct_type(self):
row1 = DataTypes.ROW().add("f1", DataTypes.STRING(nullable=True)) \
.add("f2", DataTypes.STRING(nullable=True))
row2 = DataTypes.ROW([
DataTypes.FIELD("f1", DataTypes.STRING(nullable=True)),
DataTypes.FIELD("f2", DataTypes.STRING(nullable=True), None)
])
self.assertEqual(row1.field_names(), row2.names)
self.assertEqual(row1, row2)
row1 = DataTypes.ROW().add("f1", DataTypes.STRING(nullable=True)) \
.add("f2", DataTypes.STRING(nullable=True))
row2 = DataTypes.ROW(
[DataTypes.FIELD("f1", DataTypes.STRING(nullable=True))])
self.assertNotEqual(row1.field_names(), row2.names)
self.assertNotEqual(row1, row2)
row1 = (DataTypes.ROW().add(
DataTypes.FIELD("f1", DataTypes.STRING(nullable=True))).add(
"f2", DataTypes.STRING(nullable=True)))
row2 = DataTypes.ROW([
DataTypes.FIELD("f1", DataTypes.STRING(nullable=True)),
DataTypes.FIELD("f2", DataTypes.STRING(nullable=True))
])
self.assertEqual(row1.field_names(), row2.names)
self.assertEqual(row1, row2)
row1 = (DataTypes.ROW().add(
DataTypes.FIELD("f1", DataTypes.STRING(nullable=True))).add(
"f2", DataTypes.STRING(nullable=True)))
row2 = DataTypes.ROW(
[DataTypes.FIELD("f1", DataTypes.STRING(nullable=True))])
self.assertNotEqual(row1.field_names(), row2.names)
self.assertNotEqual(row1, row2)
# Catch exception raised during improper construction
self.assertRaises(ValueError, lambda: DataTypes.ROW().add("name"))
row1 = DataTypes.ROW().add("f1", DataTypes.STRING(nullable=True)) \
.add("f2", DataTypes.STRING(nullable=True))
for field in row1:
self.assertIsInstance(field, RowField)
row1 = DataTypes.ROW().add("f1", DataTypes.STRING(nullable=True)) \
.add("f2", DataTypes.STRING(nullable=True))
self.assertEqual(len(row1), 2)
row1 = DataTypes.ROW().add("f1", DataTypes.STRING(nullable=True)) \
.add("f2", DataTypes.STRING(nullable=True))
self.assertIs(row1["f1"], row1.fields[0])
self.assertIs(row1[0], row1.fields[0])
self.assertEqual(row1[0:1], DataTypes.ROW(row1.fields[0:1]))
self.assertRaises(KeyError, lambda: row1["f9"])
self.assertRaises(IndexError, lambda: row1[9])
self.assertRaises(TypeError, lambda: row1[9.9])
def test_group_aggregate_with_aux_group(self):
t = self.t_env.from_elements(
[(1, 2, 3), (3, 2, 3), (2, 1, 3), (1, 5, 4), (1, 8, 6), (2, 3, 4)],
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.SMALLINT()),
DataTypes.FIELD("c", DataTypes.INT())
]))
table_sink = source_sink_utils.TestAppendSink(['a', 'b', 'c', 'd'], [
DataTypes.TINYINT(),
DataTypes.INT(),
DataTypes.FLOAT(),
DataTypes.INT()
])
self.t_env.register_table_sink("Results", table_sink)
self.t_env.get_config().get_configuration().set_string(
'python.metric.enabled', 'true')
self.t_env.register_function(
"max_add",
udaf(MaxAdd(), result_type=DataTypes.INT(), func_type="pandas"))
self.t_env.create_temporary_system_function("mean_udaf", mean_udaf)
t.group_by("a") \
.select("a, a + 1 as b, a + 2 as c") \
.group_by("a, b") \
.select("a, b, mean_udaf(b), max_add(b, c, 1)") \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, ["1,2,2.0,6", "2,3,3.0,8", "3,4,4.0,10"])
def test_map_with_pandas_udf(self):
t = self.t_env.from_elements(
[(1, Row(2, 3)), (2, Row(1, 3)), (1, Row(5, 4)), (1, Row(8, 6)), (2, Row(3, 4))],
DataTypes.ROW(
[DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b",
DataTypes.ROW([DataTypes.FIELD("a", DataTypes.INT()),
DataTypes.FIELD("b", DataTypes.INT())]))]))
table_sink = source_sink_utils.TestAppendSink(
['a', 'b'],
[DataTypes.BIGINT(), DataTypes.BIGINT()])
self.t_env.register_table_sink("Results", table_sink)
def func(x, y):
import pandas as pd
a = (x * 2).rename('b')
res = pd.concat([a, x], axis=1) + y
return res
pandas_udf = udf(func,
result_type=DataTypes.ROW(
[DataTypes.FIELD("c", DataTypes.BIGINT()),
DataTypes.FIELD("d", DataTypes.BIGINT())]),
func_type='pandas')
t.map(pandas_udf(t.a, t.b)).execute_insert("Results").wait()
actual = source_sink_utils.results()
self.assert_equals(actual, ["3,5", "3,7", "6,6", "9,8", "5,8"])
def get_result_type(self) -> DataType:
return DataTypes.FLOAT()
def test_from_pandas_with_incorrect_schema(self):
fields = self.data_type.fields.copy()
fields[0], fields[7] = fields[7], fields[0] # swap str with tinyint
wrong_schema = DataTypes.ROW(fields) # should be DataTypes.STRING()
with self.assertRaisesRegex(Exception, "Expected a string.*got int8"):
self.t_env.from_pandas(self.pdf, schema=wrong_schema)
def get_result_type(self):
return DataTypes.BIGINT()
def get_accumulator_type(self) -> DataType:
return DataTypes.ARRAY(DataTypes.BIGINT())
def test_basic_type(self):
test_types = [
DataTypes.STRING(),
DataTypes.BOOLEAN(),
DataTypes.BYTES(),
DataTypes.TINYINT(),
DataTypes.SMALLINT(),
DataTypes.INT(),
DataTypes.BIGINT(),
DataTypes.FLOAT(),
DataTypes.DOUBLE(),
DataTypes.DATE(),
DataTypes.TIME(),
DataTypes.TIMESTAMP(3)
]
java_types = [_to_java_type(item) for item in test_types]
converted_python_types = [_from_java_type(item) for item in java_types]
self.assertEqual(test_types, converted_python_types)
def test_verify_type_not_nullable(self):
import array
import datetime
import decimal
schema = DataTypes.ROW([
DataTypes.FIELD('s', DataTypes.STRING(nullable=False)),
DataTypes.FIELD('i', DataTypes.INT(True))
])
class MyObj:
def __init__(self, **kwargs):
for k, v in kwargs.items():
setattr(self, k, v)
# obj, data_type
success_spec = [
# String
("", DataTypes.STRING()),
(u"", DataTypes.STRING()),
# UDT
(ExamplePoint(1.0, 2.0), ExamplePointUDT()),
# Boolean
(True, DataTypes.BOOLEAN()),
# TinyInt
(-(2**7), DataTypes.TINYINT()),
(2**7 - 1, DataTypes.TINYINT()),
# SmallInt
(-(2**15), DataTypes.SMALLINT()),
(2**15 - 1, DataTypes.SMALLINT()),
# Int
(-(2**31), DataTypes.INT()),
(2**31 - 1, DataTypes.INT()),
# BigInt
(2**64, DataTypes.BIGINT()),
# Float & Double
(1.0, DataTypes.FLOAT()),
(1.0, DataTypes.DOUBLE()),
# Decimal
(decimal.Decimal("1.0"), DataTypes.DECIMAL(10, 0)),
# Binary
(bytearray([1]), DataTypes.BINARY(1)),
# Date/Time/Timestamp
(datetime.date(2000, 1, 2), DataTypes.DATE()),
(datetime.datetime(2000, 1, 2, 3, 4), DataTypes.DATE()),
(datetime.time(1, 1, 2), DataTypes.TIME()),
(datetime.datetime(2000, 1, 2, 3, 4), DataTypes.TIMESTAMP()),
# Array
([], DataTypes.ARRAY(DataTypes.INT())),
(["1", None], DataTypes.ARRAY(DataTypes.STRING(nullable=True))),
([1, 2], DataTypes.ARRAY(DataTypes.INT())),
((1, 2), DataTypes.ARRAY(DataTypes.INT())),
(array.array('h', [1, 2]), DataTypes.ARRAY(DataTypes.INT())),
# Map
({}, DataTypes.MAP(DataTypes.STRING(), DataTypes.INT())),
({
"a": 1
}, DataTypes.MAP(DataTypes.STRING(), DataTypes.INT())),
({
"a": None
},
DataTypes.MAP(DataTypes.STRING(nullable=False),
DataTypes.INT(True))),
# Struct
({
"s": "a",
"i": 1
}, schema),
({
"s": "a",
"i": None
}, schema),
({
"s": "a"
}, schema),
({
"s": "a",
"f": 1.0
}, schema),
(Row(s="a", i=1), schema),
(Row(s="a", i=None), schema),
(Row(s="a", i=1, f=1.0), schema),
(["a", 1], schema),
(["a", None], schema),
(("a", 1), schema),
(MyObj(s="a", i=1), schema),
(MyObj(s="a", i=None), schema),
(MyObj(s="a"), schema),
]
# obj, data_type, exception class
failure_spec = [
# Char/VarChar (match anything but None)
(None, DataTypes.VARCHAR(1), ValueError),
(None, DataTypes.CHAR(1), ValueError),
# VarChar (length exceeds maximum length)
("abc", DataTypes.VARCHAR(1), ValueError),
# Char (length exceeds length)
("abc", DataTypes.CHAR(1), ValueError),
# UDT
(ExamplePoint(1.0, 2.0), PythonOnlyUDT(), ValueError),
# Boolean
(1, DataTypes.BOOLEAN(), TypeError),
("True", DataTypes.BOOLEAN(), TypeError),
([1], DataTypes.BOOLEAN(), TypeError),
# TinyInt
(-(2**7) - 1, DataTypes.TINYINT(), ValueError),
(2**7, DataTypes.TINYINT(), ValueError),
("1", DataTypes.TINYINT(), TypeError),
(1.0, DataTypes.TINYINT(), TypeError),
# SmallInt
(-(2**15) - 1, DataTypes.SMALLINT(), ValueError),
(2**15, DataTypes.SMALLINT(), ValueError),
# Int
(-(2**31) - 1, DataTypes.INT(), ValueError),
(2**31, DataTypes.INT(), ValueError),
# Float & Double
(1, DataTypes.FLOAT(), TypeError),
(1, DataTypes.DOUBLE(), TypeError),
# Decimal
(1.0, DataTypes.DECIMAL(10, 0), TypeError),
(1, DataTypes.DECIMAL(10, 0), TypeError),
("1.0", DataTypes.DECIMAL(10, 0), TypeError),
# Binary
(1, DataTypes.BINARY(1), TypeError),
# VarBinary (length exceeds maximum length)
(bytearray([1, 2]), DataTypes.VARBINARY(1), ValueError),
# Char (length exceeds length)
(bytearray([1, 2]), DataTypes.BINARY(1), ValueError),
# Date/Time/Timestamp
("2000-01-02", DataTypes.DATE(), TypeError),
("10:01:02", DataTypes.TIME(), TypeError),
(946811040, DataTypes.TIMESTAMP(), TypeError),
# Array
(["1", None], DataTypes.ARRAY(DataTypes.VARCHAR(1,
nullable=False)),
ValueError),
([1, "2"], DataTypes.ARRAY(DataTypes.INT()), TypeError),
# Map
({
"a": 1
}, DataTypes.MAP(DataTypes.INT(), DataTypes.INT()), TypeError),
({
"a": "1"
}, DataTypes.MAP(DataTypes.VARCHAR(1),
DataTypes.INT()), TypeError),
({
"a": None
}, DataTypes.MAP(DataTypes.VARCHAR(1),
DataTypes.INT(False)), ValueError),
# Struct
({
"s": "a",
"i": "1"
}, schema, TypeError),
(Row(s="a"), schema, ValueError), # Row can't have missing field
(Row(s="a", i="1"), schema, TypeError),
(["a"], schema, ValueError),
(["a", "1"], schema, TypeError),
(MyObj(s="a", i="1"), schema, TypeError),
(MyObj(s=None, i="1"), schema, ValueError),
]
# Check success cases
for obj, data_type in success_spec:
try:
_create_type_verifier(data_type.not_null())(obj)
except (TypeError, ValueError):
self.fail("verify_type(%s, %s, nullable=False)" %
(obj, data_type))
# Check failure cases
for obj, data_type, exp in failure_spec:
msg = "verify_type(%s, %s, nullable=False) == %s" % (
obj, data_type, exp)
with self.assertRaises(exp, msg=msg):
_create_type_verifier(data_type.not_null())(obj)
def test_timestamp_microsecond(self):
tst = DataTypes.TIMESTAMP()
self.assertEqual(
tst.to_sql_type(datetime.datetime.max) % 1000000, 999999)
def test_sliding_group_window_over_time(self):
# create source file path
import tempfile
import os
tmp_dir = tempfile.gettempdir()
data = [
'1,1,2,2018-03-11 03:10:00', '3,3,2,2018-03-11 03:10:00',
'2,2,1,2018-03-11 03:10:00', '1,1,3,2018-03-11 03:40:00',
'1,1,8,2018-03-11 04:20:00', '2,2,3,2018-03-11 03:30:00'
]
source_path = tmp_dir + '/test_sliding_group_window_over_time.csv'
with open(source_path, 'w') as fd:
for ele in data:
fd.write(ele + '\n')
from pyflink.table.window import Slide
self.env.set_stream_time_characteristic(TimeCharacteristic.EventTime)
self.t_env.register_function("mean_udaf", mean_udaf)
source_table = """
create table source_table(
a TINYINT,
b SMALLINT,
c SMALLINT,
rowtime TIMESTAMP(3),
WATERMARK FOR rowtime AS rowtime - INTERVAL '60' MINUTE
) with(
'connector.type' = 'filesystem',
'format.type' = 'csv',
'connector.path' = '%s',
'format.ignore-first-line' = 'false',
'format.field-delimiter' = ','
)
""" % source_path
self.t_env.execute_sql(source_table)
t = self.t_env.from_path("source_table")
table_sink = source_sink_utils.TestAppendSink(['a', 'b', 'c', 'd'], [
DataTypes.TINYINT(),
DataTypes.TIMESTAMP(3),
DataTypes.TIMESTAMP(3),
DataTypes.FLOAT()
])
self.t_env.register_table_sink("Results", table_sink)
t.window(Slide.over("1.hours").every("30.minutes").on("rowtime").alias("w")) \
.group_by("a, b, w") \
.select("a, w.start, w.end, mean_udaf(c) as b") \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, [
"1,2018-03-11 02:30:00.0,2018-03-11 03:30:00.0,2.0",
"1,2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.5",
"1,2018-03-11 03:30:00.0,2018-03-11 04:30:00.0,5.5",
"1,2018-03-11 04:00:00.0,2018-03-11 05:00:00.0,8.0",
"2,2018-03-11 02:30:00.0,2018-03-11 03:30:00.0,1.0",
"2,2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.0",
"2,2018-03-11 03:30:00.0,2018-03-11 04:30:00.0,3.0",
"3,2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.0",
"3,2018-03-11 02:30:00.0,2018-03-11 03:30:00.0,2.0"
])
os.remove(source_path)
def setUpClass(cls):
super(PandasConversionTestBase, cls).setUpClass()
cls.data = [
(1, 1, 1, 1, True, 1.1, 1.2, 'hello', bytearray(b"aaa"),
decimal.Decimal('1000000000000000000.01'),
datetime.date(2014, 9,
13), datetime.time(hour=1, minute=0, second=1),
datetime.datetime(1970, 1, 1, 0, 0, 0, 123000), ['hello', 'ä¸æ–‡'],
Row(a=1,
b='hello',
c=datetime.datetime(1970, 1, 1, 0, 0, 0, 123000),
d=[1, 2])),
(1, 2, 2, 2, False, 2.1, 2.2, 'world', bytearray(b"bbb"),
decimal.Decimal('1000000000000000000.02'),
datetime.date(2014, 9,
13), datetime.time(hour=1, minute=0, second=1),
datetime.datetime(1970, 1, 1, 0, 0, 0, 123000), ['hello', 'ä¸æ–‡'],
Row(a=1,
b='hello',
c=datetime.datetime(1970, 1, 1, 0, 0, 0, 123000),
d=[1, 2]))
]
cls.data_type = DataTypes.ROW([
DataTypes.FIELD("f1", DataTypes.TINYINT()),
DataTypes.FIELD("f2", DataTypes.SMALLINT()),
DataTypes.FIELD("f3", DataTypes.INT()),
DataTypes.FIELD("f4", DataTypes.BIGINT()),
DataTypes.FIELD("f5", DataTypes.BOOLEAN()),
DataTypes.FIELD("f6", DataTypes.FLOAT()),
DataTypes.FIELD("f7", DataTypes.DOUBLE()),
DataTypes.FIELD("f8", DataTypes.STRING()),
DataTypes.FIELD("f9", DataTypes.BYTES()),
DataTypes.FIELD("f10", DataTypes.DECIMAL(38, 18)),
DataTypes.FIELD("f11", DataTypes.DATE()),
DataTypes.FIELD("f12", DataTypes.TIME()),
DataTypes.FIELD("f13", DataTypes.TIMESTAMP(3)),
DataTypes.FIELD("f14", DataTypes.ARRAY(DataTypes.STRING())),
DataTypes.FIELD(
"f15",
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.INT()),
DataTypes.FIELD("b", DataTypes.STRING()),
DataTypes.FIELD("c", DataTypes.TIMESTAMP(3)),
DataTypes.FIELD("d", DataTypes.ARRAY(DataTypes.INT()))
]))
], False)
cls.pdf = cls.create_pandas_data_frame()
def test_group_aggregate_function(self):
t = self.t_env.from_elements(
[(1, 2, 3), (3, 2, 3), (2, 1, 3), (1, 5, 4), (1, 8, 6), (2, 3, 4)],
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.SMALLINT()),
DataTypes.FIELD("c", DataTypes.INT())
]))
table_sink = source_sink_utils.TestAppendSink(['a', 'b', 'c'], [
DataTypes.TINYINT(),
DataTypes.FLOAT(),
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.INT()),
DataTypes.FIELD("b", DataTypes.INT())
])
])
self.t_env.register_table_sink("Results", table_sink)
# general udf
add = udf(lambda a: a + 1, result_type=DataTypes.INT())
# pandas udf
substract = udf(lambda a: a - 1,
result_type=DataTypes.INT(),
func_type="pandas")
max_udaf = udaf(lambda a: (a.max(), a.min()),
result_type=DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.INT()),
DataTypes.FIELD("b", DataTypes.INT())
]),
func_type="pandas")
t.group_by("a") \
.select(t.a, mean_udaf(add(t.b)), max_udaf(substract(t.c))) \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, ["1,6.0,5,2", "2,3.0,3,2", "3,3.0,2,2"])
def test_register_table_source_and_sink(self):
source_path = os.path.join(self.tempdir + '/streaming.csv')
field_names = ["a", "b", "c"]
field_types = [DataTypes.INT(), DataTypes.STRING(), DataTypes.STRING()]
data = [(1, "Hi", "Hello"), (2, "Hello", "Hello")]
self.prepare_csv_source(source_path, data, field_types, field_names)
sink_path = os.path.join(self.tempdir + '/streaming2.csv')
if os.path.isfile(sink_path):
os.remove(sink_path)
t_env = self.t_env
t_env.connect(FileSystem().path(source_path))\
.with_format(OldCsv()
.field_delimiter(',')
.field("a", DataTypes.INT())
.field("b", DataTypes.STRING())
.field("c", DataTypes.STRING()))\
.with_schema(Schema()
.field("a", DataTypes.INT())
.field("b", DataTypes.STRING())
.field("c", DataTypes.STRING()))\
.register_table_source_and_sink("source")
t_env.connect(FileSystem().path(sink_path))\
.with_format(OldCsv()
.field_delimiter(',')
.field("a", DataTypes.INT())
.field("b", DataTypes.STRING())
.field("c", DataTypes.STRING()))\
.with_schema(Schema()
.field("a", DataTypes.INT())
.field("b", DataTypes.STRING())
.field("c", DataTypes.STRING()))\
.register_table_source_and_sink("sink")
t_env.scan("source") \
.select("a + 1, b, c") \
.insert_into("sink")
self.env.execute()
with open(sink_path, 'r') as f:
lines = f.read()
assert lines == '2,Hi,Hello\n' + "3,Hello,Hello\n"
def test_map(self):
t = self.t_env.from_elements(
[(1, 2, 3), (2, 1, 3), (1, 5, 4), (1, 8, 6), (2, 3, 4)],
DataTypes.ROW(
[DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.SMALLINT()),
DataTypes.FIELD("c", DataTypes.INT())]))
table_sink = source_sink_utils.TestAppendSink(
['a', 'b'],
[DataTypes.BIGINT(), DataTypes.BIGINT()])
self.t_env.register_table_sink("Results", table_sink)
func = udf(lambda x: Row(x + 1, x * x), result_type=DataTypes.ROW(
[DataTypes.FIELD("a", DataTypes.BIGINT()),
DataTypes.FIELD("b", DataTypes.BIGINT())]))
t.map(func(t.b)).alias("a", "b") \
.map(func(t.a)).alias("a", "b") \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, ["4,9", "3,4", "7,36", "10,81", "5,16"])
def test_field(self):
schema = Schema()\
.field("int_field", DataTypes.INT())\
.field("long_field", DataTypes.BIGINT())\
.field("string_field", DataTypes.STRING())\
.field("timestamp_field", DataTypes.TIMESTAMP())\
.field("time_field", DataTypes.TIME())\
.field("date_field", DataTypes.DATE())\
.field("double_field", DataTypes.DOUBLE())\
.field("float_field", DataTypes.FLOAT())\
.field("byte_field", DataTypes.TINYINT())\
.field("short_field", DataTypes.SMALLINT())\
.field("boolean_field", DataTypes.BOOLEAN())
properties = schema.to_properties()
expected = {
'schema.0.name': 'int_field',
'schema.0.type': 'INT',
'schema.1.name': 'long_field',
'schema.1.type': 'BIGINT',
'schema.2.name': 'string_field',
'schema.2.type': 'VARCHAR',
'schema.3.name': 'timestamp_field',
'schema.3.type': 'TIMESTAMP',
'schema.4.name': 'time_field',
'schema.4.type': 'TIME',
'schema.5.name': 'date_field',
'schema.5.type': 'DATE',
'schema.6.name': 'double_field',
'schema.6.type': 'DOUBLE',
'schema.7.name': 'float_field',
'schema.7.type': 'FLOAT',
'schema.8.name': 'byte_field',
'schema.8.type': 'TINYINT',
'schema.9.name': 'short_field',
'schema.9.type': 'SMALLINT',
'schema.10.name': 'boolean_field',
'schema.10.type': 'BOOLEAN'
}
self.assertEqual(expected, properties)
def test_flat_map(self):
t = self.t_env.from_elements(
[(1, "2,3", 3), (2, "1", 3), (1, "5,6,7", 4)],
DataTypes.ROW(
[DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.STRING()),
DataTypes.FIELD("c", DataTypes.INT())]))
table_sink = source_sink_utils.TestAppendSink(
['a', 'b'],
[DataTypes.BIGINT(), DataTypes.STRING()])
self.t_env.register_table_sink("Results", table_sink)
@udtf(result_types=[DataTypes.INT(), DataTypes.STRING()])
def split(x, string):
for s in string.split(","):
yield x, s
t.flat_map(split(t.a, t.b)) \
.alias("a, b") \
.flat_map(split(t.a, t.b)) \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, ["1,2", "1,3", "2,1", "1,5", "1,6", "1,7"])
mean_udaf(b)
over (PARTITION BY a ORDER BY proctime
ROWS BETWEEN 1 PRECEDING AND CURRENT ROW),
max_add_min_udaf(b)
over (PARTITION BY a ORDER BY proctime
ROWS BETWEEN 1 PRECEDING AND CURRENT ROW)
from source_table
""").wait()
actual = source_sink_utils.results()
self.assert_equals(actual, [
"1,1.0,2", "1,3.0,6", "1,6.5,13", "2,1.0,2", "2,2.0,4", "3,2.0,4"
])
os.remove(source_path)
@udaf(result_type=DataTypes.FLOAT(), func_type="pandas")
def mean_udaf(v):
return v.mean()
class MaxAdd(AggregateFunction, unittest.TestCase):
def open(self, function_context):
mg = function_context.get_metric_group()
self.counter = mg.add_group("key", "value").counter("my_counter")
self.counter_sum = 0
def get_value(self, accumulator):
# counter
self.counter.inc(10)
self.counter_sum += 10
return accumulator[0]
def test_map_type(self):
test_types = [
DataTypes.MAP(DataTypes.BIGINT(), DataTypes.BIGINT()),
DataTypes.MAP(DataTypes.STRING(), DataTypes.STRING()),
DataTypes.MAP(
DataTypes.STRING(),
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT())),
DataTypes.MAP(
DataTypes.STRING(),
DataTypes.MAP(DataTypes.STRING(), DataTypes.STRING()))
]
java_types = [_to_java_type(item) for item in test_types]
converted_python_types = [_from_java_type(item) for item in java_types]
self.assertEqual(test_types, converted_python_types)
def test_tumble_group_window_aggregate_function(self):
import datetime
from pyflink.table.window import Tumble
t = self.t_env.from_elements(
[(1, 2, 3, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(3, 2, 4, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(2, 1, 2, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(1, 3, 1, datetime.datetime(2018, 3, 11, 3, 40, 0, 0)),
(1, 8, 5, datetime.datetime(2018, 3, 11, 4, 20, 0, 0)),
(2, 3, 6, datetime.datetime(2018, 3, 11, 3, 30, 0, 0))],
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.SMALLINT()),
DataTypes.FIELD("c", DataTypes.INT()),
DataTypes.FIELD("rowtime", DataTypes.TIMESTAMP(3))
]))
table_sink = source_sink_utils.TestAppendSink(['a', 'b', 'c'], [
DataTypes.TIMESTAMP(3),
DataTypes.TIMESTAMP(3),
DataTypes.FLOAT()
])
self.t_env.register_table_sink("Results", table_sink)
self.t_env.create_temporary_system_function("mean_udaf", mean_udaf)
tumble_window = Tumble.over(expr.lit(1).hours) \
.on(expr.col("rowtime")) \
.alias("w")
t.window(tumble_window) \
.group_by("w") \
.select("w.start, w.end, mean_udaf(b)") \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, [
"2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.2",
"2018-03-11 04:00:00.0,2018-03-11 05:00:00.0,8.0"
])
def test_merge_type(self):
self.assertEqual(_merge_type(DataTypes.BIGINT(), DataTypes.NULL()),
DataTypes.BIGINT())
self.assertEqual(_merge_type(DataTypes.NULL(), DataTypes.BIGINT()),
DataTypes.BIGINT())
self.assertEqual(_merge_type(DataTypes.BIGINT(), DataTypes.BIGINT()),
DataTypes.BIGINT())
self.assertEqual(
_merge_type(DataTypes.ARRAY(DataTypes.BIGINT()),
DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.ARRAY(DataTypes.BIGINT()))
with self.assertRaises(TypeError):
_merge_type(DataTypes.ARRAY(DataTypes.BIGINT()),
DataTypes.ARRAY(DataTypes.DOUBLE()))
self.assertEqual(
_merge_type(DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT()),
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT())),
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT()))
with self.assertRaises(TypeError):
_merge_type(DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT()),
DataTypes.MAP(DataTypes.DOUBLE(), DataTypes.BIGINT()))
with self.assertRaises(TypeError):
_merge_type(DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT()),
DataTypes.MAP(DataTypes.STRING(), DataTypes.DOUBLE()))
self.assertEqual(
_merge_type(
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.BIGINT()),
DataTypes.FIELD('f2', DataTypes.STRING())
]),
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.BIGINT()),
DataTypes.FIELD('f2', DataTypes.STRING())
])),
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.BIGINT()),
DataTypes.FIELD('f2', DataTypes.STRING())
]))
with self.assertRaises(TypeError):
_merge_type(
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.BIGINT()),
DataTypes.FIELD('f2', DataTypes.STRING())
]),
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.DOUBLE()),
DataTypes.FIELD('f2', DataTypes.STRING())
]))
self.assertEqual(
_merge_type(
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ROW(
[DataTypes.FIELD('f2', DataTypes.BIGINT())]))
]),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ROW(
[DataTypes.FIELD('f2', DataTypes.BIGINT())]))
])),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ROW([DataTypes.FIELD('f2', DataTypes.BIGINT())]))
]))
with self.assertRaises(TypeError):
_merge_type(
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ROW(
[DataTypes.FIELD('f2', DataTypes.BIGINT())]))
]),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ROW(
[DataTypes.FIELD('f2', DataTypes.STRING())]))
]))
self.assertEqual(
_merge_type(
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
]),
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
])),
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
]))
with self.assertRaises(TypeError):
_merge_type(
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
]),
DataTypes.ROW([
DataTypes.FIELD('f1', DataTypes.ARRAY(DataTypes.DOUBLE())),
DataTypes.FIELD('f2', DataTypes.STRING())
]))
self.assertEqual(
_merge_type(
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
]),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
])),
DataTypes.ROW([
DataTypes.FIELD(
'f1', DataTypes.MAP(DataTypes.STRING(),
DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
]))
with self.assertRaises(TypeError):
_merge_type(
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT())),
DataTypes.FIELD('f2', DataTypes.STRING())
]),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.MAP(DataTypes.STRING(), DataTypes.DOUBLE())),
DataTypes.FIELD('f2', DataTypes.STRING())
]))
self.assertEqual(
_merge_type(
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ARRAY(
DataTypes.MAP(DataTypes.STRING(),
DataTypes.BIGINT())))
]),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ARRAY(
DataTypes.MAP(DataTypes.STRING(),
DataTypes.BIGINT())))
])),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ARRAY(
DataTypes.MAP(DataTypes.STRING(), DataTypes.BIGINT())))
]))
with self.assertRaises(TypeError):
_merge_type(
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ARRAY(
DataTypes.MAP(DataTypes.STRING(),
DataTypes.BIGINT())))
]),
DataTypes.ROW([
DataTypes.FIELD(
'f1',
DataTypes.ARRAY(
DataTypes.MAP(DataTypes.DOUBLE(),
DataTypes.BIGINT())))
]))
def test_slide_group_window_aggregate_function(self):
import datetime
from pyflink.table.window import Slide
t = self.t_env.from_elements(
[(1, 2, 3, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(3, 2, 4, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(2, 1, 2, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(1, 3, 1, datetime.datetime(2018, 3, 11, 3, 40, 0, 0)),
(1, 8, 5, datetime.datetime(2018, 3, 11, 4, 20, 0, 0)),
(2, 3, 6, datetime.datetime(2018, 3, 11, 3, 30, 0, 0))],
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.SMALLINT()),
DataTypes.FIELD("c", DataTypes.INT()),
DataTypes.FIELD("rowtime", DataTypes.TIMESTAMP(3))
]))
table_sink = source_sink_utils.TestAppendSink(
['a', 'b', 'c', 'd', 'e'], [
DataTypes.TINYINT(),
DataTypes.TIMESTAMP(3),
DataTypes.TIMESTAMP(3),
DataTypes.FLOAT(),
DataTypes.INT()
])
self.t_env.register_table_sink("Results", table_sink)
self.t_env.register_function(
"max_add",
udaf(MaxAdd(), result_type=DataTypes.INT(), func_type="pandas"))
self.t_env.create_temporary_system_function("mean_udaf", mean_udaf)
slide_window = Slide.over(expr.lit(1).hours) \
.every(expr.lit(30).minutes) \
.on(expr.col("rowtime")) \
.alias("w")
t.window(slide_window) \
.group_by("a, w") \
.select("a, w.start, w.end, mean_udaf(b), max_add(b, c, 1)") \
.execute_insert("Results") \
.wait()
actual = source_sink_utils.results()
self.assert_equals(actual, [
"1,2018-03-11 02:30:00.0,2018-03-11 03:30:00.0,2.0,6",
"1,2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.5,7",
"1,2018-03-11 03:30:00.0,2018-03-11 04:30:00.0,5.5,14",
"1,2018-03-11 04:00:00.0,2018-03-11 05:00:00.0,8.0,14",
"2,2018-03-11 02:30:00.0,2018-03-11 03:30:00.0,1.0,4",
"2,2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.0,10",
"2,2018-03-11 03:30:00.0,2018-03-11 04:30:00.0,3.0,10",
"3,2018-03-11 03:00:00.0,2018-03-11 04:00:00.0,2.0,7",
"3,2018-03-11 02:30:00.0,2018-03-11 03:30:00.0,2.0,7"
])
def test_data_type_eq(self):
lt = DataTypes.BIGINT()
lt2 = pickle.loads(pickle.dumps(DataTypes.BIGINT()))
self.assertEqual(lt, lt2)
def test_over_window_aggregate_function(self):
import datetime
t = self.t_env.from_elements(
[(1, 2, 3, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(3, 2, 1, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(2, 1, 2, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(1, 3, 1, datetime.datetime(2018, 3, 11, 3, 10, 0, 0)),
(1, 8, 5, datetime.datetime(2018, 3, 11, 4, 20, 0, 0)),
(2, 3, 6, datetime.datetime(2018, 3, 11, 3, 30, 0, 0))],
DataTypes.ROW([
DataTypes.FIELD("a", DataTypes.TINYINT()),
DataTypes.FIELD("b", DataTypes.SMALLINT()),
DataTypes.FIELD("c", DataTypes.INT()),
DataTypes.FIELD("rowtime", DataTypes.TIMESTAMP(3))
]))
table_sink = source_sink_utils.TestAppendSink(
['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j'], [
DataTypes.TINYINT(),
DataTypes.FLOAT(),
DataTypes.INT(),
DataTypes.FLOAT(),
DataTypes.FLOAT(),
DataTypes.FLOAT(),
DataTypes.FLOAT(),
DataTypes.FLOAT(),
DataTypes.FLOAT(),
DataTypes.FLOAT()
])
self.t_env.register_table_sink("Results", table_sink)
self.t_env.create_temporary_system_function("mean_udaf", mean_udaf)
self.t_env.register_function(
"max_add",
udaf(MaxAdd(), result_type=DataTypes.INT(), func_type="pandas"))
self.t_env.register_table("T", t)
self.t_env.execute_sql("""
insert into Results
select a,
mean_udaf(b)
over (PARTITION BY a ORDER BY rowtime
ROWS BETWEEN UNBOUNDED preceding AND UNBOUNDED FOLLOWING),
max_add(b, c)
over (PARTITION BY a ORDER BY rowtime
ROWS BETWEEN UNBOUNDED preceding AND 0 FOLLOWING),
mean_udaf(b)
over (PARTITION BY a ORDER BY rowtime
ROWS BETWEEN 1 PRECEDING AND UNBOUNDED FOLLOWING),
mean_udaf(c)
over (PARTITION BY a ORDER BY rowtime
ROWS BETWEEN 1 PRECEDING AND 0 FOLLOWING),
mean_udaf(c)
over (PARTITION BY a ORDER BY rowtime
RANGE BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING),
mean_udaf(b)
over (PARTITION BY a ORDER BY rowtime
RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW),
mean_udaf(b)
over (PARTITION BY a ORDER BY rowtime
RANGE BETWEEN INTERVAL '20' MINUTE PRECEDING AND UNBOUNDED FOLLOWING),
mean_udaf(c)
over (PARTITION BY a ORDER BY rowtime
RANGE BETWEEN INTERVAL '20' MINUTE PRECEDING AND UNBOUNDED FOLLOWING),
mean_udaf(c)
over (PARTITION BY a ORDER BY rowtime
RANGE BETWEEN INTERVAL '20' MINUTE PRECEDING AND CURRENT ROW)
from T
""").wait()
actual = source_sink_utils.results()
self.assert_equals(actual, [
"1,4.3333335,5,4.3333335,3.0,3.0,2.5,4.3333335,3.0,2.0",
"1,4.3333335,13,5.5,3.0,3.0,4.3333335,8.0,5.0,5.0",
"1,4.3333335,6,4.3333335,2.0,3.0,2.5,4.3333335,3.0,2.0",
"2,2.0,9,2.0,4.0,4.0,2.0,2.0,4.0,4.0",
"2,2.0,3,2.0,2.0,4.0,1.0,2.0,4.0,2.0",
"3,2.0,3,2.0,1.0,1.0,2.0,2.0,1.0,1.0"
])
def test_datetype_equal_zero(self):
dt = DataTypes.DATE()
self.assertEqual(dt.from_sql_type(0), datetime.date(1970, 1, 1))
def test_array_type(self):
test_types = [
DataTypes.ARRAY(DataTypes.BIGINT()),
# array type with not null basic data type means primitive array
DataTypes.ARRAY(DataTypes.BIGINT().not_null()),
DataTypes.ARRAY(DataTypes.STRING()),
DataTypes.ARRAY(DataTypes.ARRAY(DataTypes.BIGINT())),
DataTypes.ARRAY(DataTypes.ARRAY(DataTypes.STRING()))
]
java_types = [_to_java_type(item) for item in test_types]
converted_python_types = [_from_java_type(item) for item in java_types]
self.assertEqual(test_types, converted_python_types)
