Add DAG importer abstraction layer

[kaxil]

Introduces a pluggable importer architecture for DAG files. This enables future support for non-Python DAG formats (YAML, etc.) while maintaining full backwards compatibility with existing behavior.

This should make it easier for AIP-85 Extendable DAG parsing controls. So not documenting it purposely until then.

---

^ Add meaningful description above
Read the Pull Request Guidelines for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in airflow-core/newsfragments.

[kaxil]

cc @IKholopov

[Copilot]

Pull request overview

This PR introduces a pluggable DAG importer architecture that enables support for multiple DAG file formats (e.g., YAML) while maintaining backwards compatibility with existing Python DAG files. The implementation follows the AIP-85 Extendable DAG parsing controls proposal.

Key Changes:

• Introduces an abstract AbstractDagImporter base class and DagImporterRegistry for managing importers
• Refactors Python DAG import logic into a dedicated PythonDagImporter class
• Modifies DagBag.process_file() to use the new importer registry instead of directly importing modules

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
airflow-core/src/airflow/dag_processing/importers/__init__.py	Exports public API for the new importers module
airflow-core/src/airflow/dag_processing/importers/base.py	Defines abstract importer interface, registry singleton, and result/error data structures
airflow-core/src/airflow/dag_processing/importers/python_importer.py	Implements Python DAG importer by extracting and refactoring logic from DagBag
airflow-core/src/airflow/dag_processing/dagbag.py	Refactored to use the new importer registry pattern instead of direct module loading
airflow-core/tests/unit/dag_processing/importers/__init__.py	Test package initialization file
airflow-core/tests/unit/dag_processing/importers/test_registry.py	Comprehensive tests for the importer registry functionality
airflow-core/tests/unit/dag_processing/test_dagbag.py	Updated mock paths to reflect refactored code structure
airflow-core/.pre-commit-config.yaml	Updated pre-commit exclusion patterns for new importer files

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[ikholopov-omni]

From the point of AIP-85 - LGTM as long as we will have the assumption "at most 1 importer per file extension" well documented and handled.

One deviation from AIP-85 is that filepaths are still handled by DagBag (and list_py_file_paths only lists py/zip paths). Do we have a definitive plan on how we want to handle it? Should it be handled by importer (somewhat makes sense, as it knows what extensions to look for) as was proposed in AIP-85 as well or by something else?

[Copilot]

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 4 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[ikholopov-omni]

Thank you for working on this!