Skip to content

fix(bigquery): Prefer query over table in get_table_query_string #6360

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
ntkathole merged 2 commits into from
May 3, 2026
Merged

fix(bigquery): Prefer query over table in get_table_query_string #6360

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
cb5b03b
fix(bigquery): prefer query over table in get_table_query_string
Jwrede May 2, 2026
7db33e0
docs(bigquery): update docstring to reflect both table and query support
Jwrede May 3, 2026
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: 13 additions & 13 deletions sdk/python/feast/infra/offline_stores/bigquery_source.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,15 +47,15 @@ def __init__(
case the table must be specified.
timestamp_field (optional): Event timestamp field used for point in time
joins of feature values.
table (optional): BigQuery table where the features are stored. Exactly one of 'table'
and 'query' must be specified.
table (optional): The BigQuery table where features can be found.
table (optional): BigQuery table where the features are stored. At least one of 'table'
and 'query' must be specified. When both are set, 'query' is used for reads and
'table' is used as the write destination.
created_timestamp_column (optional): Timestamp column when row was created, used for deduplicating rows.
field_mapping (optional): A dictionary mapping of column names in this data source to feature names in a feature table
or view. Only used for feature columns, not entities or timestamp columns.
date_partition_column (optional): Timestamp column used for partitioning.
query (optional): The query to be executed to obtain the features. Exactly one of 'table'
and 'query' must be specified.
query (optional): The query to be executed to obtain the features. When both 'table'
and 'query' are provided, 'query' takes priority for reads.
description (optional): A human-readable description.
tags (optional): A dictionary of key-value pairs to store arbitrary metadata.
owner (optional): The owner of the bigquery source, typically the email of the primary
Expand Down Expand Up @@ -156,10 +156,10 @@ def validate(self, config: RepoConfig):

def get_table_query_string(self) -> str:
"""Returns a string that can directly be used to reference this table in SQL"""
if self.table:
return f"`{self.table}`"
else:
if self.query:
return f"({self.query})"
else:
return f"`{self.table}`"

@staticmethod
def source_datatype_to_feast_value_type() -> Callable[[str], ValueType]:
Expand All @@ -185,14 +185,14 @@ def get_table_column_names_and_types(
location=config.offline_store.location,
client_info=http_client_info.ClientInfo(user_agent=get_user_agent()),
)
if self.table:
schema = client.get_table(self.table).schema
if not isinstance(schema[0], bigquery.schema.SchemaField):
raise TypeError("Could not parse BigQuery table schema.")
else:
if self.query:
bq_columns_query = f"SELECT * FROM ({self.query}) LIMIT 0"
query_res = client.query(bq_columns_query).result()
schema = query_res.schema
else:
schema = client.get_table(self.table).schema
if not isinstance(schema[0], bigquery.schema.SchemaField):
raise TypeError("Could not parse BigQuery table schema.")

name_type_pairs: List[Tuple[str, str]] = []
for field in schema:
Expand Down
48 changes: 48 additions & 0 deletions sdk/python/tests/unit/infra/offline_stores/test_bigquery.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -152,3 +152,51 @@ def test_pull_all_from_table_or_query_partition_pruning(mock_get_bigquery_client
)
assert "partition_date >= '2021-01-01'" in actual_query
assert "partition_date <= '2021-01-02'" in actual_query


class TestBigQuerySourceGetTableQueryString:
def test_table_only(self):
source = BigQuerySource(
name="test",
table="project.dataset.table",
timestamp_field="ts",
)
assert source.get_table_query_string() == "`project.dataset.table`"

def test_query_only(self):
source = BigQuerySource(
name="test",
query="SELECT * FROM `project.dataset.table` WHERE active = TRUE",
timestamp_field="ts",
)
assert (
source.get_table_query_string()
== "(SELECT * FROM `project.dataset.table` WHERE active = TRUE)"
)

def test_both_table_and_query_prefers_query(self):
"""When both table and query are set, query takes priority for reads."""
query = (
"SELECT * FROM `project.dataset.table`"
" QUALIFY ROW_NUMBER() OVER (PARTITION BY entity_id, event_time) = 1"
)
source = BigQuerySource(
name="test",
table="project.dataset.table",
query=query,
timestamp_field="ts",
)
result = source.get_table_query_string()
assert result.startswith("(")
assert "QUALIFY" in result
assert result != "`project.dataset.table`"

def test_table_property_unaffected_by_query_priority(self):
"""The .table property is still accessible for write paths."""
source = BigQuerySource(
name="test",
table="project.dataset.write_target",
query="SELECT * FROM `project.dataset.write_target` WHERE deduped",
timestamp_field="ts",
)
assert source.table == "project.dataset.write_target"
Loading