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.

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

fix: Fixing pinot query generation for date format conversion from python datetime format to java simple date format #13163

Merged
merged 3 commits into from
Feb 20, 2021
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
26 changes: 16 additions & 10 deletions superset/db_engine_specs/pinot.py
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,6 @@
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
import datetime
from typing import Dict, List, Optional

from sqlalchemy.sql.expression import ColumnClause, ColumnElement
Expand Down Expand Up @@ -69,38 +68,45 @@ def get_timestamp_expr(
time_grain: Optional[str],
type_: Optional[str] = None,
) -> TimestampExpression:
if not pdf:
raise NotImplementedError(f"Empty date format for '{col}'")
is_epoch = pdf in ("epoch_s", "epoch_ms")

# The DATETIMECONVERT pinot udf is documented at
# Per https://github.com/apache/incubator-pinot/wiki/dateTimeConvert-UDF
# We are not really converting any time units, just bucketing them.
tf = ""
java_date_format = ""
if not is_epoch:
try:
today = datetime.datetime.today()
today.strftime(str(pdf))
except ValueError:
raise ValueError(f"Invalid column datetime format:{str(pdf)}")
java_date_format = str(pdf)
java_date_format = pdf
for (
python_pattern,
java_pattern,
) in cls._python_to_java_time_patterns.items():
java_date_format.replace(python_pattern, java_pattern)
java_date_format = java_date_format.replace(
python_pattern, java_pattern
)
tf = f"1:SECONDS:SIMPLE_DATE_FORMAT:{java_date_format}"
else:
seconds_or_ms = "MILLISECONDS" if pdf == "epoch_ms" else "SECONDS"
tf = f"1:{seconds_or_ms}:EPOCH"
if time_grain:
granularity = cls.get_time_grain_expressions().get(time_grain)
if not granularity:
raise NotImplementedError("No pinot grain spec for " + str(time_grain))
raise NotImplementedError(f"No pinot grain spec for '{time_grain}'")
else:
return TimestampExpression("{{col}}", col)

# In pinot the output is a string since there is no timestamp column like pg
if cls._use_date_trunc_function.get(time_grain):
time_expr = f"DATETRUNC('{granularity}', {{col}}, '{seconds_or_ms}')"
if is_epoch:
time_expr = f"DATETRUNC('{granularity}', {{col}}, '{seconds_or_ms}')"
else:
time_expr = (
f"ToDateTime(DATETRUNC('{granularity}', "
+ f"FromDateTime({{col}}, '{java_date_format}'), "
+ f"'MILLISECONDS'), '{java_date_format}')"
)
else:
time_expr = f"DATETIMECONVERT({{col}}, '{tf}', '{tf}', '{granularity}')"

Expand Down
38 changes: 36 additions & 2 deletions tests/db_engine_specs/pinot_tests.py
Original file line number Diff line number Diff line change
Expand Up @@ -30,12 +30,46 @@ def test_pinot_time_expression_sec_one_1d_grain(self):
self.assertEqual(
result,
"DATETIMECONVERT(tstamp, '1:SECONDS:EPOCH', '1:SECONDS:EPOCH', '1:DAYS')",
) # noqa
)

def test_pinot_time_expression_simple_date_format_1d_grain(self):
col = column("tstamp")
expr = PinotEngineSpec.get_timestamp_expr(col, "%Y-%m-%d %H:%M:%S", "P1D")
result = str(expr.compile())
self.assertEqual(
result,
(
"DATETIMECONVERT(tstamp, "
+ "'1:SECONDS:SIMPLE_DATE_FORMAT:yyyy-MM-dd HH:mm:ss', "
+ "'1:SECONDS:SIMPLE_DATE_FORMAT:yyyy-MM-dd HH:mm:ss', '1:DAYS')"
),
)

def test_pinot_time_expression_simple_date_format_1w_grain(self):
col = column("tstamp")
expr = PinotEngineSpec.get_timestamp_expr(col, "%Y-%m-%d %H:%M:%S", "P1W")
result = str(expr.compile())
self.assertEqual(
result,
(
"ToDateTime(DATETRUNC('week', FromDateTime(tstamp, "
+ "'yyyy-MM-dd HH:mm:ss'), 'MILLISECONDS'), 'yyyy-MM-dd HH:mm:ss')"
),
)

def test_pinot_time_expression_sec_one_1m_grain(self):
col = column("tstamp")
expr = PinotEngineSpec.get_timestamp_expr(col, "epoch_s", "P1M")
result = str(expr.compile())
self.assertEqual(
result, "DATETRUNC('month', tstamp, 'SECONDS')",
) # noqa
)

def test_invalid_get_time_expression_arguments(self):
with self.assertRaises(NotImplementedError):
PinotEngineSpec.get_timestamp_expr(column("tstamp"), None, "P1M")

with self.assertRaises(NotImplementedError):
PinotEngineSpec.get_timestamp_expr(
column("tstamp"), "epoch_s", "invalid_grain"
)