Skip to content

Commit 4760bac

Browse files
committed
comment changes
moved comments and added new comments
1 parent 75df46f commit 4760bac

File tree

2 files changed

+5
-3
lines changed

2 files changed

+5
-3
lines changed

metadata-ingestion/src/datahub/ingestion/source/sql/sql_common.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -354,6 +354,8 @@ def __init__(self, config: SQLCommonConfig, ctx: PipelineContext, platform: str)
354354

355355
def _add_default_options(self, sql_config: SQLCommonConfig) -> None:
356356
"""Add default SQLAlchemy options. Can be overridden by subclasses to add additional defaults."""
357+
# Extra default SQLAlchemy option for better connection pooling and threading.
358+
# https://docs.sqlalchemy.org/en/14/core/pooling.html#sqlalchemy.pool.QueuePool.params.max_overflow
357359
if sql_config.is_profiling_enabled():
358360
sql_config.options.setdefault(
359361
"max_overflow", sql_config.profiling.max_workers
@@ -526,8 +528,6 @@ def get_workunits_internal(self) -> Iterable[Union[MetadataWorkUnit, SqlWorkUnit
526528
# Known issue with sqlalchemy https://stackoverflow.com/questions/60804288/pycharm-duplicated-log-for-sqlalchemy-echo-true
527529
sqlalchemy_log._add_default_handler = lambda x: None # type: ignore
528530

529-
# Extra default SQLAlchemy option for better connection pooling and threading.
530-
# https://docs.sqlalchemy.org/en/14/core/pooling.html#sqlalchemy.pool.QueuePool.params.max_overflow
531531
self._add_default_options(sql_config)
532532

533533
for inspector in self.get_inspectors():

metadata-ingestion/src/datahub/ingestion/source/sql/teradata.py

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -681,8 +681,10 @@ def __init__(self, config: TeradataConfig, ctx: PipelineContext):
681681

682682
def _add_default_options(self, sql_config: SQLCommonConfig) -> None:
683683
"""Add Teradata-specific default options"""
684-
# Teradata does not support max_overflow, instead we use QueuePool when profiling
684+
super()._add_default_options(sql_config)
685685
if sql_config.is_profiling_enabled():
686+
# By default, Teradata uses SingletonThreadPool, which is not supported by sqlalchemy
687+
# QueuePool used for parallel connections when profiling is enabled
686688
sql_config.options.setdefault("poolclass", QueuePool)
687689

688690
@classmethod

0 commit comments

Comments
 (0)