Skip to content

python(fix): remove client side detection for missing parquet time column #544

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
wei-qlu merged 1 commit into from
Apr 27, 2026
Merged

python(fix): remove client side detection for missing parquet time column #544

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
41 changes: 1 addition & 40 deletions python/lib/sift_client/resources/data_imports.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,21 +10,17 @@
from sift_client._internal.util.tdms import detect_tdms_config
from sift_client.resources._base import ResourceBase
from sift_client.sift_types.asset import Asset
from sift_client.sift_types.channel import ChannelDataType
from sift_client.sift_types.data_import import (
EXTENSION_TO_DATA_TYPE_KEY,
CsvImportConfig,
DataTypeKey,
ImportConfig,
ParquetFlatDatasetImportConfig,
ParquetSingleChannelPerRowImportConfig,
ParquetTimeColumn,
)
from sift_client.sift_types.run import Run

if TYPE_CHECKING:
from collections.abc import Iterable

from sift_client.client import SiftClient
from sift_client.sift_types.job import Job

Expand Down Expand Up @@ -312,24 +308,6 @@ def _parse_csv_detect_response(proto) -> CsvImportConfig:
return csv_config


def _infer_time_column(columns: Iterable[tuple[str, ChannelDataType, str]]) -> str | None:
"""Find a likely time column from a sequence of (name, data_type, path) tuples.

The backend only detects arrow timestamp types. This falls back to the first
integer column whose name starts with "time".
"""
_integer_types = {
ChannelDataType.INT_32,
ChannelDataType.INT_64,
ChannelDataType.UINT_32,
ChannelDataType.UINT_64,
}
for name, data_type, path in columns:
if data_type in _integer_types and name.lower().startswith("time"):
return path
return None


def _parse_parquet_detect_response(
proto, filename: str, footer_offset: int, footer_length: int
) -> ParquetFlatDatasetImportConfig | ParquetSingleChannelPerRowImportConfig:
Expand All @@ -343,28 +321,11 @@ def _parse_parquet_detect_response(
parquet_config.data_columns = [
dc for dc in parquet_config.data_columns if dc.path != time_path
]
else:
inferred = _infer_time_column(
(dc.name, dc.data_type, dc.path) for dc in parquet_config.data_columns
)
if inferred is not None:
parquet_config.time_column = ParquetTimeColumn(path=inferred)
parquet_config.data_columns = [
c for c in parquet_config.data_columns if c.path != inferred
]
return parquet_config
elif proto.HasField("single_channel_per_row"):
parquet_scpr_config = ParquetSingleChannelPerRowImportConfig._from_proto(
return ParquetSingleChannelPerRowImportConfig._from_proto(
proto, footer_offset=footer_offset, footer_length=footer_length
)
if not parquet_scpr_config.time_column.path:
inferred = _infer_time_column(
(col.column_config.name, ChannelDataType(col.column_config.data_type), col.path)
for col in proto.single_channel_per_row.columns
)
if inferred is not None:
parquet_scpr_config.time_column = ParquetTimeColumn(path=inferred)
return parquet_scpr_config
raise ValueError(f"Unsupported parquet layout in DetectConfig response for '{filename}'.")


Expand Down
Loading