fix(tableau): skip ingesting filtered project assets

[mihai103]

Issue Description:

Problem: The Tableau asset ingestion process is designed to filter and ingest allowed projects first. Subsequent layer-by-layer asset ingestion incorrectly includes published data sources from projects that should be filtered out if these data sources are upstream of embedded data sources in allowed projects. This results in unauthorized assets being ingested without correct 'browsePaths', as their parent projects are not ingested.

Expected: The ingestion should strictly honor project filters and not ingest any upstream assets from disallowed projects. Assets only in allowed projects should have their 'browsePaths' created and ingested.

Checklist

• The PR conforms to DataHub's Contributing Guideline (particularly Commit Message Format)
• Links to related issues (if applicable)
• Tests for the changes have been added/updated (if applicable)
• Docs related to the changes have been added/updated (if applicable). If a new feature has been added a Usage Guide has been added for the same.
• For any breaking change/potential downtime/deprecation/big changes an entry has been made in Updating DataHub

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Files with missing lines	Coverage Δ
...on/src/datahub/ingestion/source/tableau/tableau.py	89.02% <100.00%> (ø)

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5ba7767...a2c9852. Read the comment docs.

[sgomezvillamor]

Thanks for raising the issue and the proposal. Overall, it looks good.

As a general comment, wondering if it would be possible to do the filtering upper in the call stack? Without a heavy refactor, of course. Ideally, and not sure if we are very consistent with that, emit methods should be exclusively responsible for generating the aspects and build the MetadataWorkUnit, not doing the filtering; filtering should happens before.

[mihai103]

Thanks for raising the issue and the proposal. Overall, it looks good.

As a general comment, wondering if it would be possible to do the filtering upper in the call stack? Without a heavy refactor, of course. Ideally, and not sure if we are very consistent with that, emit methods should be exclusively responsible for generating the aspects and build the MetadataWorkUnit, not doing the filtering; filtering should happens before.

I agree, but due to the way everything works, by traversing everything from downstream to upstream and the fact that even though published datasource is an upstream for embedded datasources, they still share all the upstreams, I can't think of a way to do it without a heavy refactoring.

[sgomezvillamor]

.pre-commit-config.yaml
datahub-web-react/src/app/ingest/source/IngestionSourceList.tsx
metadata-service/iceberg-catalog/build.gradle
metadata-service/iceberg-catalog/pyproject.toml
metadata-service/iceberg-catalog/requirements.txt
metadata-service/iceberg-catalog/src/integrationTest/integration_test.py

Those files may have been accidentally pushed. Maybe some issue during the merge?

Once they are removed from the PR, this is ready to be merged.

[sgomezvillamor]

LGTM
Thanks for the contrib!