Skip to content

Speed up clearing downstream tasks on large Dags #71203

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

Open
ColtenOuO wants to merge 2 commits into
base: main
Choose a base branch
from
+9 −7
Open

Speed up clearing downstream tasks on large Dags #71203

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
726f3c9
Speed up clearing downstream tasks on large Dags
ColtenOuO Aug 5, 2026
888be9a
Track partial_subset matches as a single id set, not id set + object …
ColtenOuO Aug 11, 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
16 changes: 9 additions & 7 deletions airflow-core/src/airflow/serialization/definitions/dag.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -307,16 +307,17 @@ def is_task(obj) -> TypeIs[SerializedOperator]:
dag = copy.deepcopy(self, memo)

if isinstance(task_ids, str):
matched_tasks = [t for t in self.tasks if task_ids in t.task_id]
matched_task_ids = {t.task_id for t in self.tasks if task_ids in t.task_id}
else:
matched_tasks = [t for t in self.tasks if t.task_id in task_ids]
matched_task_ids = {t.task_id for t in self.tasks if t.task_id in task_ids}

also_include_ids: set[str] = set()
for t in matched_tasks:
for tid in matched_task_ids:
t = self.task_dict[tid]
if include_downstream:
for rel in t.get_flat_relatives(upstream=False, depth=depth):
also_include_ids.add(rel.task_id)
if rel not in matched_tasks: # if it's in there, we're already processing it
if rel.task_id not in matched_task_ids: # if it's in there, we're already processing it
# need to include setups and teardowns for tasks that are in multiple
# non-collinear setup/teardown paths
if not rel.is_setup and not rel.is_teardown:
Expand All @@ -334,7 +335,7 @@ def is_task(obj) -> TypeIs[SerializedOperator]:
also_include: list[SerializedOperator] = [self.task_dict[x] for x in also_include_ids]
direct_upstreams: list[SerializedOperator] = []
if include_direct_upstream:
for t in itertools.chain(matched_tasks, also_include):
for t in itertools.chain((self.task_dict[x] for x in matched_task_ids), also_include):
direct_upstreams.extend(u for u in t.upstream_list if is_task(u))

# Make sure to not recursively deepcopy the dag or task_group while copying the task.
Expand All @@ -344,8 +345,9 @@ def _deepcopy_task(t) -> SerializedOperator:
return copy.deepcopy(t, memo)

# Compiling the unique list of tasks that made the cut
if exclude_original:
matched_tasks = []
matched_tasks: list[SerializedOperator] = (
[] if exclude_original else [self.task_dict[x] for x in matched_task_ids]
)
dag.task_dict = {
t.task_id: _deepcopy_task(t)
for t in itertools.chain(matched_tasks, also_include, direct_upstreams)
Expand Down