Revert "Add a batch_mode flag to VCF scalable queries (#683)"

This reverts commit 9dd8695.

src/tiledb/cloud/vcf/query.py

@@ -267,7 +267,6 @@ def build_read_dag(
     namespace: Optional[str] = None,
     resource_class: Optional[str] = None,
     verbose: bool = False,
-    batch_mode: bool = False,
 ) -> Tuple[tiledb.cloud.dag.DAG, tiledb.cloud.dag.Node]:
     """
     Build the DAG for a distributed read on a TileDB-VCF dataset.
@@ -290,7 +289,6 @@ def build_read_dag(
     :param namespace: TileDB-Cloud namespace, defaults to None
     :param resource_class: TileDB-Cloud resource class for UDFs, defaults to None
     :param verbose: verbose logging, defaults to False
-    :param batch_mode: run the query with batch UDFs, defaults to False
     :return: DAG and result Node
     """
 
@@ -340,13 +338,10 @@ def build_read_dag(
     logger.debug("num_sample_partitions=%d", num_sample_partitions)
     logger.debug("num_region_partitions=%d", num_region_partitions)
 
-    mode = tiledb.cloud.dag.Mode.BATCH if batch_mode else tiledb.cloud.dag.Mode.REALTIME
-
     dag = tiledb.cloud.dag.DAG(
         namespace=namespace,
         name="VCF-Distributed-Query",
         max_workers=max_workers,
-        mode=mode,
     )
 
     # If `regions` is a Delayed object, we set the parent nodes to `dag` so the
@@ -356,12 +351,6 @@ def build_read_dag(
     ):
         regions._DelayedBase__set_all_parent_nodes_same_dag(dag)
 
-    result_format = (
-        tiledb.cloud.UDFResultType.NATIVE
-        if batch_mode
-        else tiledb.cloud.UDFResultType.ARROW
-    )
-
     tables = []
     for region in range(num_region_partitions):
         for sample in range(num_sample_partitions):
@@ -384,14 +373,12 @@ def build_read_dag(
                     name=f"VCF Query - Region {region+1}/{num_region_partitions},"
                     f" Sample {sample+1}/{num_sample_partitions}",
                     resource_class=resource_class,
-                    result_format=result_format,
+                    result_format=tiledb.cloud.UDFResultType.ARROW,
                 )
             )
 
     if len(tables) > 1:
-        submit = dag.submit if batch_mode else dag.submit_local
-
-        table = submit(
+        table = dag.submit_local(
             concat_tables_udf,
             tables,
             config=config,
@@ -442,7 +429,6 @@ def read(
     namespace: Optional[str] = None,
     resource_class: Optional[str] = None,
     verbose: bool = False,
-    batch_mode: bool = False,
 ) -> pa.Table:
     """
     Run a distributed read on a TileDB-VCF dataset.
@@ -465,7 +451,6 @@ def read(
     :param namespace: TileDB-Cloud namespace, defaults to None
     :param resource_class: TileDB-Cloud resource class for UDFs, defaults to None
     :param verbose: verbose logging, defaults to False
-    :param batch_mode: run the query with batch UDFs, defaults to False
     :return: Arrow table containing the query results
     """
 
@@ -486,7 +471,6 @@ def read(
         namespace=namespace,
         resource_class=resource_class,
         verbose=verbose,
-        batch_mode=batch_mode,
     )
 
     run_dag(dag, debug=verbose)

tests/vcf/test_vcf.py

@@ -1,26 +1,17 @@
-import cloudpickle
 import numpy as np
 import pytest
 
 import tiledb.cloud.vcf as vcf
 import tiledb.cloud.vcf.vcf_toolbox as vtb
 
-# Pickle the vcf module by value, so tests run on the latest code.
-cloudpickle.register_pickle_by_value(vcf)
-
-
-# Run VCF tests with:
-#   pytest -m vcf --run-vcf -n 8
-
 
 @vtb.df_transform
 def filter_vcf(df, *, filter=None):
     return df.query(filter)
 
 
 @pytest.mark.vcf
-@pytest.mark.parametrize("batch_mode", [False, True])
-def test_vcf_transform(batch_mode):
+def test_vcf_transform():
     vcf_uri = "tiledb://TileDB-Inc/vcf-1kg-dragen-v376"
 
     regions = [
@@ -48,7 +39,6 @@ def test_vcf_transform(batch_mode):
         regions=regions,
         samples="NA12878",
         transform_result=filter_vcf(filter=filter),
-        batch_mode=batch_mode,
     )
 
     assert vcf_table.num_rows == 336