Skip to content

feat(showcase-ecommerce): add Redshift and BigQuery datapacks - #209

Open
askumar27 wants to merge 1 commit into
feat/showcase-ecommerce-add-databricksfrom
feat/showcase-ecommerce-add-redshift-bigquery
Open

askumar27 wants to merge 1 commit into
feat/showcase-ecommerce-add-databricksfrom
feat/showcase-ecommerce-add-redshift-bigquery

Conversation

@askumar27

Copy link
Copy Markdown

What

Adds two standalone datapacks that mirror the Databricks showcase-ecommerce datapack with Amazon Redshift and Google BigQuery as the warehouse platform.

datapacks/showcase-ecommerce-redshift/
datapacks/showcase-ecommerce-bigquery/

Each is self-contained (01-definitions05-context + index.json) and can be loaded independently or alongside the Databricks / Snowflake packs — no URN collisions.

How it was derived

Generated from showcase-ecommerce-databricks with a platform transform:

Dimension Redshift BigQuery
Platform URN dataPlatform:redshift dataPlatform:bigquery
Container subTypes Database / Schema Project / Dataset
Container hashes regenerated (disjoint) regenerated (disjoint)
SQL dialect near-identity (ARRAY_JOINLISTAGG, DATE_FORMATTO_CHAR, TO_DATECAST AS DATE) full rewrite (TIMESTAMPDATETIME, DATETIME_ADD/SUB/DIFF/TRUNC, TO_DATEDATE, DATE_FORMATFORMAT_DATETIME, ARRAY_JOINSTRING_AGG, PERCENTILE_CONT … WITHIN GROUPAPPROX_QUANTILES, backtick-quoted tables)
  • Dataset names kept identical (order_entry_db.order_entry.orders) — valid 3-part for all platforms.
  • Databricks-only metadata stripped (Unity storage_location, spark.* / delta.* configs, unity_schema).
  • Full governance parity: tags, glossary terms, ownership, domains, structured properties, lineage, siblings, schema fields, 312 queries with queryUsageFeatures.
  • Agent-context docs repointed from Snowflake prose/URNs to the target warehouse.

Validation

  • SQL dialect — all 312/312 translated statements validated against live engines: BigQuery dry-run (full semantic check against real acryl-staging tables) and Redshift EXPLAIN. Zero dialect failures.
  • Ingestion — both datapacks ingested into a local DataHub via datahub ingest mcps (index order): zero failed MCPs.
  • Resolution — verified via GraphQL/SDK: 14 datasets, 4 containers (correct subTypes), 312 queries per platform; downstream lineage (order_details ← 11 upstreams) and query→subject references all resolve.

🤖 Generated with Claude Code

Add `showcase-ecommerce-redshift` and `showcase-ecommerce-bigquery` datapacks
that mirror the Databricks showcase-ecommerce datapack with Redshift / BigQuery
as the warehouse platform.

- 15 warehouse datasets each (order_entry_db.order_entry + analytics schemas)
- Full governance parity: tags, glossary terms, ownership, domains, structured
  properties, editable descriptions, lineage, siblings, schema fields
- 312 queries translated from Databricks SQL to each target dialect:
  - Redshift: ARRAY_JOIN -> LISTAGG, DATE_FORMAT -> TO_CHAR, TO_DATE -> CAST AS DATE
    (DATEADD / DATEDIFF / DATE_TRUNC / PERCENTILE_CONT ... WITHIN GROUP kept as-is)
  - BigQuery: TIMESTAMP -> DATETIME, DATEADD -> DATETIME_ADD/SUB, DATEDIFF ->
    DATETIME_DIFF, DATE_TRUNC arg reorder, TO_DATE -> DATE, DATE_FORMAT ->
    FORMAT_DATETIME, ARRAY_JOIN -> STRING_AGG, PERCENTILE_CONT WITHIN GROUP ->
    APPROX_QUANTILES, backtick-quoted table refs
- Container subTypes per platform: Redshift Database/Schema, BigQuery Project/Dataset
- Databricks-only metadata stripped (Unity storage_location, spark.* / delta.* configs)
- Container hashes regenerated per platform: no URN collisions when co-loaded
  alongside the Databricks (or each other's) datapack
- Agent-context docs repointed from Snowflake prose/URNs to the target warehouse
- All cross-platform entities included (dbt, Looker, PowerBI, Tableau, Spark, etc.)

Validated: all 312 queries pass dialect validation against live BigQuery (dry-run)
and Redshift (EXPLAIN); both datapacks ingest into DataHub with zero failed MCPs,
and datasets / containers / queries / lineage all resolve.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@vercel

vercel Bot commented Jun 30, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment
Project Deployment Actions Updated (UTC)
datahub-docs-archive Ignored Ignored Jun 30, 2026 7:47pm

Request Review

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant