Skip to content
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

Add support for decimal256 and allow rounding decimal to scale #439

Merged
merged 6 commits into from
Jan 3, 2025
Merged

Add support for decimal256 and allow rounding decimal to scale #439

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
e55ae73
Initial commit
raghumdani Jan 1, 2025
c77d8de
refactor
raghumdani Jan 2, 2025
e62a6a1
add more tests
raghumdani Jan 2, 2025
9e6a93d
Seek to the beginning of the file when handling exceptions
raghumdani Jan 3, 2025
4fed5ba
add test case for negative scale
raghumdani Jan 3, 2025
d38f327
Rename scale
raghumdani Jan 3, 2025
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
3 changes: 3 additions & 0 deletions deltacat/tests/utils/data/overflowing_decimal_precision.csv
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
Y 322236.65
"" 32.33
N 0.40
3 changes: 3 additions & 0 deletions deltacat/tests/utils/data/overflowing_decimal_scale.csv
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
Y 322236.65883738
"" 32.33
N 0.40
255 changes: 255 additions & 0 deletions deltacat/tests/utils/test_pyarrow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,9 @@
s3_file_to_table,
ReadKwargsProviderPyArrowSchemaOverride,
RAISE_ON_EMPTY_CSV_KWARG,
RAISE_ON_DECIMAL_OVERFLOW,
)
import decimal
from deltacat.types.media import ContentEncoding, ContentType
from deltacat.types.partial_download import PartialParquetParameters
from pyarrow.parquet import ParquetFile
Expand All @@ -16,6 +18,12 @@
PARQUET_FILE_PATH = "deltacat/tests/utils/data/test_file.parquet"
EMPTY_UTSV_PATH = "deltacat/tests/utils/data/empty.csv"
NON_EMPTY_VALID_UTSV_PATH = "deltacat/tests/utils/data/non_empty_valid.csv"
OVERFLOWING_DECIMAL_PRECISION_UTSV_PATH = (
"deltacat/tests/utils/data/overflowing_decimal_precision.csv"
)
OVERFLOWING_DECIMAL_SCALE_UTSV_PATH = (
"deltacat/tests/utils/data/overflowing_decimal_scale.csv"
)
GZIP_COMPRESSED_FILE_UTSV_PATH = "deltacat/tests/utils/data/non_empty_compressed.gz"
BZ2_COMPRESSED_FILE_UTSV_PATH = "deltacat/tests/utils/data/non_empty_compressed.bz2"

Expand Down Expand Up @@ -407,6 +415,253 @@ def test_read_csv_when_empty_csv_and_raise_on_empty_passed(self):
),
)

def test_read_csv_when_decimal_precision_overflows_and_raise_kwarg_specified(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(4, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)
read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)
self.assertRaises(
pa.lib.ArrowInvalid,
lambda: pyarrow_read_csv(
OVERFLOWING_DECIMAL_PRECISION_UTSV_PATH,
**{**kwargs, RAISE_ON_DECIMAL_OVERFLOW: True}
),
)

def test_read_csv_when_decimal_precision_overflows_sanity(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(4, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

self.assertRaises(
pa.lib.ArrowInvalid,
lambda: pyarrow_read_csv(OVERFLOWING_DECIMAL_PRECISION_UTSV_PATH, **kwargs),
)

def test_read_csv_when_decimal_scale_overflows_and_raise_kwarg_specified(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(20, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)
read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

self.assertRaises(
pa.lib.ArrowInvalid,
lambda: pyarrow_read_csv(
OVERFLOWING_DECIMAL_SCALE_UTSV_PATH,
**{**kwargs, RAISE_ON_DECIMAL_OVERFLOW: True}
),
)

def test_read_csv_when_decimal_scale_overflows_sanity(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(20, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

result = pyarrow_read_csv(OVERFLOWING_DECIMAL_SCALE_UTSV_PATH, **kwargs)

self.assertEqual(len(result), 3)
self.assertEqual(
result[1][0].as_py(), decimal.Decimal("322236.66")
) # rounding decimal
self.assertEqual(result[1][1].as_py(), decimal.Decimal("32.33")) # not rounded
self.assertEqual(len(result.column_names), 2)
result_schema = result.schema
self.assertEqual(result_schema.field(0).type, "string")
self.assertEqual(result_schema.field(1).type, pa.decimal128(20, 2))

def test_read_csv_when_decimal_scale_overflows_and_negative_scale(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(20, -2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

result = pyarrow_read_csv(OVERFLOWING_DECIMAL_SCALE_UTSV_PATH, **kwargs)

self.assertEqual(len(result), 3)
self.assertEqual(
result[1][0].as_py(),
decimal.Decimal("322200"), # consequence of negative scale
) # rounding decimal
self.assertEqual(result[1][1].as_py(), decimal.Decimal("00"))
self.assertEqual(len(result.column_names), 2)
result_schema = result.schema
self.assertEqual(result_schema.field(0).type, "string")
self.assertEqual(result_schema.field(1).type, pa.decimal128(20, -2))

def test_read_csv_when_decimal_scale_overflows_with_decimal256(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal256(20, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

result = pyarrow_read_csv(OVERFLOWING_DECIMAL_SCALE_UTSV_PATH, **kwargs)

self.assertEqual(len(result), 3)
self.assertEqual(
result[1][0].as_py(), decimal.Decimal("322236.66")
) # rounding decimal
self.assertEqual(result[1][1].as_py(), decimal.Decimal("32.33")) # not rounded
self.assertEqual(len(result.column_names), 2)
result_schema = result.schema
self.assertEqual(result_schema.field(0).type, "string")
self.assertEqual(result_schema.field(1).type, pa.decimal256(20, 2))

def test_read_csv_when_decimal_scale_overflows_with_decimal256_and_raise_on_overflow(
self,
):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal256(20, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

self.assertRaises(
pa.lib.ArrowNotImplementedError,
lambda: pyarrow_read_csv(
OVERFLOWING_DECIMAL_SCALE_UTSV_PATH,
**{**kwargs, RAISE_ON_DECIMAL_OVERFLOW: True}
),
)

def test_read_csv_when_decimal_scale_overflows_without_any_schema_then_infers(self):
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=None)
kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

result = pyarrow_read_csv(OVERFLOWING_DECIMAL_SCALE_UTSV_PATH, **kwargs)

# The default behavior of pyarrow is to invalid skip rows
self.assertEqual(len(result), 2)
self.assertEqual(result[1][0].as_py(), 32.33) # rounding decimal
self.assertEqual(result[1][1].as_py(), 0.4) # not rounded
self.assertEqual(len(result.column_names), 2)
result_schema = result.schema
self.assertEqual(result_schema.field(0).type, "string")
self.assertEqual(result_schema.field(1).type, pa.float64())

def test_read_csv_when_decimal_scale_and_precision_overflow_and_raise_on_overflow(
self,
):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(5, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

self.assertRaises(
pa.lib.ArrowInvalid,
lambda: pyarrow_read_csv(OVERFLOWING_DECIMAL_SCALE_UTSV_PATH, **kwargs),
)

def test_read_csv_when_decimal_scale_overflow_and_file_like_obj_passed(self):
schema = pa.schema(
[("is_active", pa.string()), ("decimal_value", pa.decimal128(15, 2))]
)
kwargs = content_type_to_reader_kwargs(ContentType.UNESCAPED_TSV.value)
_add_column_kwargs(
ContentType.UNESCAPED_TSV.value,
["is_active", "decimal_value"],
["is_active", "decimal_value"],
kwargs,
)

read_kwargs_provider = ReadKwargsProviderPyArrowSchemaOverride(schema=schema)

kwargs = read_kwargs_provider(ContentType.UNESCAPED_TSV.value, kwargs)

with open(OVERFLOWING_DECIMAL_SCALE_UTSV_PATH, "rb") as file:
result = pyarrow_read_csv(file, **kwargs)

self.assertEqual(len(result), 3)
self.assertEqual(
result[1][0].as_py(), decimal.Decimal("322236.66")
) # rounding decimal
self.assertEqual(
result[1][1].as_py(), decimal.Decimal("32.33")
) # not rounded
self.assertEqual(len(result.column_names), 2)
result_schema = result.schema
self.assertEqual(result_schema.field(0).type, "string")
self.assertEqual(result_schema.field(1).type, pa.decimal128(15, 2))


class TestS3FileToTable(TestCase):
def test_s3_file_to_table_identity_sanity(self):
Expand Down
Loading
Loading