llm based pk detector added #543
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mwojtyczka
requested changes
Aug 25, 2025
There was a problem hiding this comment.
Thanks for the PR. We should clarify the scope. The primary purpose of PK detection is to use it in compare_datasets check, for cases where user don't know pk keys for comparison. There should be a way to call this as a standalone method as well. Profiler seems to be a good place. So a new method that can be called from the profiler should be added, e.g. detect_primary_keys_with_llm. If we want to generate uniqueness check from the profiler, then it should suggest existing is_unique check func. Yes, we can add this as as another profile, and use it for rules generation.
mwojtyczka
reviewed
Aug 25, 2025
mwojtyczka
reviewed
Aug 25, 2025
mwojtyczka
reviewed
Aug 25, 2025
mwojtyczka
reviewed
Aug 25, 2025
mwojtyczka
reviewed
Aug 25, 2025
mwojtyczka
reviewed
Aug 25, 2025
mwojtyczka
reviewed
Aug 25, 2025
There was a problem hiding this comment.
Pull Request Overview
This PR adds LLM-based primary key detection capabilities to the DQX data quality framework. The functionality is completely optional and only activates when explicitly requested by users.
Key changes:
- Implements intelligent primary key detection using Large Language Models via DSPy and Databricks Model Serving
- Adds comprehensive configuration options for LLM-based detection with graceful fallback when dependencies are unavailable
- Integrates seamlessly with existing profiling workflow while maintaining backward compatibility
Reviewed Changes
Copilot reviewed 10 out of 11 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
src/databricks/labs/dqx/llm/pk_identifier.py |
Core LLM detection engine with table metadata analysis and duplicate validation |
src/databricks/labs/dqx/profiler/profiler.py |
Enhanced profiler with LLM detection methods and lazy import handling |
src/databricks/labs/dqx/profiler/generator.py |
Added primary key rule generation with LLM-specific metadata |
src/databricks/labs/dqx/profiler/runner.py |
Updated runner to support table-based profiling with PK detection |
src/databricks/labs/dqx/config.py |
Added LLM configuration fields to ProfilerConfig |
src/databricks/labs/dqx/check_funcs.py |
Implemented is_primary_key validation function |
tests/unit/test_llm_based_pk_identifier.py |
Comprehensive unit tests with graceful dependency handling |
tests/integration/test_pk_detection_integration.py |
End-to-end integration tests for the complete workflow |
src/databricks/labs/dqx/llm/demo.py |
Usage demonstration showing optional LLM activation |
src/databricks/labs/dqx/llm/README.md |
Detailed documentation with examples and best practices |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
|
❌ 335/336 passed, 1 flaky, 1 failed, 2 skipped, 3h29m10s total ❌ test_e2e_workflow: pyspark.errors.exceptions.connect.SparkConnectGrpcException: () BAD_REQUEST: session_id is no longer usable. Generate a new session_id by detaching and reattaching the compute and then try again [sessionId=44e61503-95e0-4676-ad4b-5325ef53b894, reason=INACTIVITY_TIMEOUT]. (requestId=282dff53-84a6-475d-9b06-15de584c23af) (10m45.881s)Flaky tests:
Running from acceptance #2424 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Changes
LLM based Pk detector
Linked issues
#484
Resolves #..
Tests