fix: handle small max_templated_field_length safely by Jeevankumar-s · Pull Request #59882 · apache/airflow

Jeevankumar-s · 2025-12-29T12:27:57Z

Fixes incorrect truncation when [core] max_templated_field_length is set to very small values.

Previously, negative slicing could result in empty or malformed output.
This change ensures truncation always respects the configured maximum length.

Related: #59877

boring-cyborg · 2025-12-29T12:28:00Z

Congratulations on your first Pull Request and welcome to the Apache Airflow community! If you have any issues or are unsure about any anything please check our Contributors' Guide (https://github.com/apache/airflow/blob/main/contributing-docs/README.rst)
Here are some useful points:

Pay attention to the quality of your code (ruff, mypy and type annotations). Our prek-hooks will help you with that.
In case of a new feature add useful documentation (in docstrings or in docs/ directory). Adding a new operator? Check this short guide Consider adding an example DAG that shows how users should use it.
Consider using Breeze environment for testing locally, it's a heavy docker but it ships with a working Airflow and a lot of integrations.
Be patient and persistent. It might take some time to get a review or get the final approval from Committers.
Please follow ASF Code of Conduct for all communication including (but not limited to) comments on Pull Requests, Mailing list and Slack.
Be sure to read the Airflow Coding style.
Always keep your Pull Requests rebased, otherwise your build might fail due to changes not related to your commits.
Apache Airflow is a community-driven project and together we are making it better 🚀.
In case of doubts contact the developers at:
Mailing List: dev@airflow.apache.org
Slack: https://s.apache.org/airflow-slack

potiuk · 2025-12-29T20:47:32Z

LGTM. @amoghrajesh ?

amoghrajesh

This change also needs to go in https://github.com/astronomer/airflow/blob/main/task-sdk/src/airflow/sdk/execution_time/task_runner.py#L830-L901 which is where things happen at runtime for a task

amoghrajesh · 2025-12-30T06:32:58Z

airflow-core/src/airflow/serialization/helpers.py

+def _truncate_rendered_value(rendered: str, max_length: int) -> str:
+    if max_length <= 0:
+        return ""
+
+    prefix = "Truncated. You can change this behaviour in [core]max_templated_field_length. "
+    suffix = "..."
+
+    if max_length <= len(prefix):
+        return rendered[:max_length]
+
+    available = max_length - len(prefix) - len(suffix)
+    if available <= 0:
+        return rendered[:max_length]
+
+    return f"{prefix}{rendered[:available]!r}{suffix}"


While this is good, from a user perspective, if max_length is set to 1, it would just display rendered[0]. I think from a user perspective, we should prioritise communication to user over "correctness". Or the user wont know what happened.

Always better to present problem + solution vs misleading partial result ig

Thanks for pointing this out.
I agree that showing just a single character can be confusing from a user perspective.
I’ll update the logic so that, even for very small max_templated_field_length values, we prioritize clearly communicating that truncation has occurred rather than returning a misleading partial result.
I’ll also apply the same change in the runtime path (task_runner.py) so the behavior stays consistent.
I’ll push an update shortly.

amoghrajesh

I have some concerns with this implementation, check review comments for details.

amoghrajesh · 2025-12-31T05:58:30Z

airflow-core/src/airflow/serialization/helpers.py

+def _truncate_rendered_value(rendered: str, max_length: int) -> str:
+    if max_length <= 0:
+        return ""
+
+    prefix = "Truncated. You can change this behaviour in [core]max_templated_field_length. "
+    suffix = "..."
+
+    if max_length <= len(suffix):
+        return suffix[:max_length]
+
+    if max_length <= len(prefix) + len(suffix):
+        return (prefix + suffix)[:max_length]
+
+    available = max_length - len(prefix) - len(suffix)
+    return f"{prefix}{rendered[:available]}{suffix}"
+
+
+
+def _safe_truncate_rendered_value(rendered: Any, max_length: int) -> str:
+    return _truncate_rendered_value(str(rendered), max_length)


This still doesn't prioritise message first. Take these test cases for example:

[(1, 'test', 'Minimum value'), (3, 'test', 'At ellipsis length'), (5, 'test', 'Very small'), (10, 'password123', 'Small'), (20, 'secret_value', 'Small with content'), (50, 'This is a test string', 'Medium'), (83, 'Hello World', 'At prefix+suffix boundary v1'), (84, 'Hello World', 'Just above boundary v1'), (86, 'Hello World', 'At overhead boundary v2'), (90, 'short', 'Normal case - short string'), (100, 'This is a longer string', 'Normal case'), (150, 'xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx', 'Large max_length, long string'), (100, 'None', "String 'None'"), (100, 'True', "String 'True'"), (100, "{'key': 'value'}", 'Dict-like string'), (100, "test's", 'String with apostrophe'), (90, '"quoted"', 'String with quotes')]

This is what your function returns:

for max_length, rendered, description in test_cases: v1_result = truncate_v1(rendered, max_length) print(v1_result) . ... Trunc Truncated. Truncated. You can c Truncated. You can change this behaviour in [core] Truncated. You can change this behaviour in [core]max_templated_field_length. He... Truncated. You can change this behaviour in [core]max_templated_field_length. Hel... Truncated. You can change this behaviour in [core]max_templated_field_length. Hello... Truncated. You can change this behaviour in [core]max_templated_field_length. short... Truncated. You can change this behaviour in [core]max_templated_field_length. This is a longer st... Truncated. You can change this behaviour in [core]max_templated_field_length. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx... Truncated. You can change this behaviour in [core]max_templated_field_length. None... Truncated. You can change this behaviour in [core]max_templated_field_length. True... Truncated. You can change this behaviour in [core]max_templated_field_length. {'key': 'value'}... Truncated. You can change this behaviour in [core]max_templated_field_length. test's... Truncated. You can change this behaviour in [core]max_templated_field_length. "quoted"...

This is what I would expect:

Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. ... Truncated. You can change this behaviour in [core]max_templated_field_length. 'He'... Truncated. You can change this behaviour in [core]max_templated_field_length. 'short'... Truncated. You can change this behaviour in [core]max_templated_field_length. 'This is a longer'... Truncated. You can change this behaviour in [core]max_templated_field_length. 'xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'... Truncated. You can change this behaviour in [core]max_templated_field_length. 'None'... Truncated. You can change this behaviour in [core]max_templated_field_length. 'True'... Truncated. You can change this behaviour in [core]max_templated_field_length. "{'key': 'value'}"... Truncated. You can change this behaviour in [core]max_templated_field_length. "test's"... Truncated. You can change this behaviour in [core]max_templated_field_length. '"quote'...

Thanks for the examples. I agree and have updated the logic to always prioritise the truncation message and avoid partial prefix or value output. The results now match the expected outputs you shared.

amoghrajesh · 2026-01-05T05:54:39Z

@Jeevankumar-s lot of test failures, could you adhere to them before I can look?

Jeevankumar-s · 2026-01-05T06:00:53Z

@Jeevankumar-s lot of test failures, could you adhere to them before I can look?

Stopped working on this PR since #59999 is actively addressing the issue.

fix: handle small max_templated_field_length safely

c342bb0

Jeevankumar-s requested review from ashb and bolkedebruin as code owners December 29, 2025 12:27

boring-cyborg bot added the area:serialization label Dec 29, 2025

henry3260 mentioned this pull request Dec 29, 2025

Incorrect rendered template truncation for small max_templated_fields_length config #59877

Closed

2 tasks

potiuk approved these changes Dec 29, 2025

View reviewed changes

fix(mypy): ensure truncate helper receives str

5565800

amoghrajesh reviewed Dec 30, 2025

View reviewed changes

fix: align rendered field truncation in serialization and runtime

2a00aea

Jeevankumar-s requested a review from kaxil as a code owner December 30, 2025 07:05

amoghrajesh requested changes Dec 31, 2025

View reviewed changes

Prioritise truncation message over partial rendered values

71ce7ac

Ajay9704 mentioned this pull request Dec 31, 2025

Fix small max templated field length #59999

Closed

Jeevankumar-s marked this pull request as draft January 1, 2026 10:34

Jeevankumar-s requested a review from amoghrajesh January 2, 2026 06:39

Jeevankumar-s closed this Jan 5, 2026

Jeevankumar-s deleted the fix-small-max-templated-field-length branch January 22, 2026 09:08

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix: handle small max_templated_field_length safely#59882

fix: handle small max_templated_field_length safely#59882
Jeevankumar-s wants to merge 4 commits intoapache:mainfrom
Jeevankumar-s:fix-small-max-templated-field-length

Jeevankumar-s commented Dec 29, 2025

Uh oh!

boring-cyborg bot commented Dec 29, 2025

Uh oh!

potiuk commented Dec 29, 2025

Uh oh!

amoghrajesh left a comment

Uh oh!

amoghrajesh Dec 30, 2025

Uh oh!

Jeevankumar-s Dec 30, 2025

Uh oh!

amoghrajesh left a comment

Uh oh!

amoghrajesh Dec 31, 2025

Uh oh!

Jeevankumar-s Dec 31, 2025

Uh oh!

amoghrajesh commented Jan 5, 2026

Uh oh!

Jeevankumar-s commented Jan 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

Jeevankumar-s commented Dec 29, 2025

Uh oh!

boring-cyborg bot commented Dec 29, 2025

Uh oh!

potiuk commented Dec 29, 2025

Uh oh!

amoghrajesh left a comment

Choose a reason for hiding this comment

Uh oh!

amoghrajesh Dec 30, 2025

Choose a reason for hiding this comment

Uh oh!

Jeevankumar-s Dec 30, 2025

Choose a reason for hiding this comment

Uh oh!

amoghrajesh left a comment

Choose a reason for hiding this comment

Uh oh!

amoghrajesh Dec 31, 2025

Choose a reason for hiding this comment

Uh oh!

Jeevankumar-s Dec 31, 2025

Choose a reason for hiding this comment

Uh oh!

amoghrajesh commented Jan 5, 2026

Uh oh!

Jeevankumar-s commented Jan 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants