Partial fix of broken OSS tests.

.github/workflows/pytest.yml

@@ -39,4 +39,4 @@ jobs:
     - run: pip freeze
 
     - name: Run core tests
-      run: pytest -vv -n auto
+      run: pytest -vv -n auto --import-mode=importlib

conftest.py

@@ -0,0 +1,25 @@
+# Copyright 2026 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Pytest configuration and global fixtures."""
+
+import sys
+from absl import flags
+
+# Parse absl flags once before any tests run to avoid UnparsedFlagAccessError
+# when tests use absltest.TestCase.create_tempfile() or create_tempdir().
+try:
+  flags.FLAGS(sys.argv, known_only=True)
+except flags.Error:
+  pass

dpsynth/bin/derive_domain.py

@@ -21,7 +21,6 @@
 from absl import logging
 from dpsynth import domain
 from dpsynth.bin import _read_csv_args
-import fancyflags as ff
 import numpy as np
 import pandas as pd
 
@@ -40,10 +39,20 @@
     'Path to the output directory to write the domain to.',
 )
 
-_CSV_READ_ARGS = ff.DEFINE_auto(
-    'csv_read_args',
-    _read_csv_args.ReadCsvArgs,
-    _read_csv_args.FLAG_HELP,
+_CSV_READ_ARGS_FIELD_SEPARATOR = flags.DEFINE_string(
+    'csv_read_args_field_separator',
+    None,
+    'Field separator for reading CSV files.',
+)
+_CSV_READ_ARGS_COLUMN_NAMES = flags.DEFINE_list(
+    'csv_read_args_column_names',
+    None,
+    'Column names for reading CSV files.',
+)
+_CSV_READ_ARGS_COLUMN_COUNT = flags.DEFINE_integer(
+    'csv_read_args_column_count',
+    None,
+    'Column count for reading CSV files.',
 )
 
 
@@ -120,18 +129,18 @@ def derive_domain_from_data(
   for col in df.columns:
     logging.info('Deriving domain for column: %s', col)
     match df[col].dtype:
-      case 'object':
+      case 'object' | 'str' | 'string' | 'category':
         result[col] = domain.CategoricalAttribute(
             possible_values=sorted(
                 df[col].unique(),
                 key=lambda x: (isinstance(x, str), x),  # sort ints before strs.
             )
         )
-      case 'int':
+      case 'int' | 'int64' | 'int32':
         result[col] = _create_numerical_attribute(
             df[col], 'int', numerical_sentinel_value
         )
-      case 'float':
+      case 'float' | 'float64' | 'float32':
         result[col] = _create_numerical_attribute(
             df[col], 'float', numerical_sentinel_value
         )
@@ -145,7 +154,12 @@ def _get_yaml_filename(dataset_path: PathType) -> str:
 
 
 def main(_) -> None:
-  read_csv_kwargs = _CSV_READ_ARGS.value().to_read_csv_kwargs()
+  csv_read_args = _read_csv_args.ReadCsvArgs(
+      field_separator=_CSV_READ_ARGS_FIELD_SEPARATOR.value,
+      column_names=_CSV_READ_ARGS_COLUMN_NAMES.value,
+      column_count=_CSV_READ_ARGS_COLUMN_COUNT.value,
+  )
+  read_csv_kwargs = csv_read_args.to_read_csv_kwargs()
 
   # If output_dir is not set, use the parent directory of the dataset path.
   output_dir = _OUTPUT_DIR.value

dpsynth/local_mode/primitives.py

@@ -277,6 +277,12 @@ def select_partitions_sips(
       rem_user_ids = rem_user_ids[mask]
       rem_partitions = rem_partitions[mask]
 
+  if not selected_partitions:
+    return (
+        np.empty(0, dtype=data.dtype),
+        np.empty(0, dtype=float),
+        max_sigma,
+    )
   selected_partitions = np.concatenate(selected_partitions)
   selected_counts = np.concatenate(selected_counts)
   return selected_partitions, selected_counts, max_sigma

dpsynth/pipeline_transformations/diagnostic_info.py

@@ -15,6 +15,9 @@
 """Module for updating diagnostic information."""
 
 import copy
+from dpsynth.pipeline_transformations import types
+import pipeline_dp
+
 from dataclasses import dataclass, field
 from typing import Optional
 

pyproject.toml

@@ -24,7 +24,7 @@ dependencies = [
     "pipeline-dp",
     "tensorflow",
     "apache-beam",
-    "dp-accounting",
+    "dp_accounting @ git+https://github.com/google/differential-privacy.git#subdirectory=python/dp_accounting",
     "jax[cpu]",
     "more-itertools",
     "networkx",