Add compiled_sql as a template fields on ExecutionMode.WATCHER when using run_results.json

[pankajastro]

The PR #2068 exposes dbt SQL files as templated fields when end-users have installed dbt and Cosmos/Airflow in the same Python virtual environment. The current PR implements the same feature, but for the case when dbt and Cosmos/Airflow are not being installed in the same Python virtual environment, and Cosmos uses the run_results.json to update the status of downstream tasks.

closes: #1968
closes: https://github.com/astronomer/oss-integrations-private/issues/243

[Copilot]

Pull Request Overview

This PR enhances the DbtConsumerWatcherSensor to extract and handle compiled_sql from run results when using the fallback execution mode (when dbt events are not available).

Key changes:

• Modified _get_status_from_run_results to extract compiled_code and call _override_rtif when present
• Added context parameter to _get_status_from_run_results method
• Removed unused Sequence import
• Added test coverage for the new functionality

Reviewed Changes

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

File	Description
cosmos/operators/watcher.py	Updated _get_status_from_run_results to extract compiled_code and trigger RTIF override; removed unused Sequence import; fixed template_fields type annotation
tests/operators/test_watcher.py	Added test to verify compiled_sql extraction and _override_rtif call when processing run_results

---

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

[tatiana]

@pankajastro just double-checking to make sure I understood correctly: the run results already included the compiled SQL?

Could you attach a screenshot of the templated SQL rendered when Cosmos and dbt are not installed in the same Python virtualenv?

Also, could you please fix the failing unit tests? Happy for the PR to be merged after these two changes.

[pankajastro]

@pankajastro just double-checking to make sure I understood correctly: the run results already included the compiled SQL?

Could you attach a screenshot of the templated SQL rendered when Cosmos and dbt are not installed in the same Python virtualenv?

Also, could you please fix the failing unit tests? Happy for the PR to be merged after these two changes.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.77%. Comparing base (98d62bb) to head (76d1597).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2070   +/-   ##
=======================================
  Coverage   97.77%   97.77%           
=======================================
  Files          91       91           
  Lines        5845     5848    +3     
=======================================
+ Hits         5715     5718    +3     
  Misses        130      130           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.