(airflow) branch main updated: Add dynamic task mapping no-op example (#67022)

[ch...@apache.org]

This is an automated email from the ASF dual-hosted git repository.

choo121600 pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/airflow.git


The following commit(s) were added to refs/heads/main by this push:
     new 1f5b99948a7 Add dynamic task mapping no-op example (#67022)
1f5b99948a7 is described below

commit 1f5b99948a7afe8411bda8f201e725e515543269
Author: 백형준 <13...@users.noreply.github.com>
AuthorDate: Thu May 28 09:55:34 2026 +0900

    Add dynamic task mapping no-op example (#67022)
---
 .../dynamic-task-mapping.rst                       | 30 ++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/airflow-core/docs/authoring-and-scheduling/dynamic-task-mapping.rst b/airflow-core/docs/authoring-and-scheduling/dynamic-task-mapping.rst
index 1a5366cc4c3..d9dfec21fe8 100644
--- a/airflow-core/docs/authoring-and-scheduling/dynamic-task-mapping.rst
+++ b/airflow-core/docs/authoring-and-scheduling/dynamic-task-mapping.rst
@@ -676,3 +676,33 @@ Automatically skipping zero-length maps
 =======================================
 
 If the input is empty (zero length), no new tasks will be created and the mapped task will be marked as ``SKIPPED``.
+
+This can be useful when a Dag discovers work to do at runtime, but sometimes there is no work for that run.
+For example, a scan-and-repair Dag can return an empty list when it does not find anything to repair.
+In that case, the mapped task is skipped, and a downstream summary task can still treat the run as a successful no-op if it uses a trigger rule that allows skipped upstream tasks.
+
+.. code-block:: python
+
+    from airflow.sdk import TriggerRule, task
+
+
+    @task
+    def find_work_items():
+        # Return an empty list when no files, records, or partitions need repair.
+        return []
+
+
+    @task
+    def repair(item): ...
+
+
+    @task(trigger_rule=TriggerRule.NONE_FAILED)
+    def summarize(repaired_items):
+        if not repaired_items:
+            print("No work found; nothing to repair.")
+            return
+        print(f"Repaired {len(repaired_items)} item(s).")
+
+
+    repaired_items = repair.expand(item=find_work_items())
+    summarize(repaired_items)