3 Comments

Sounds like dlt as soon as we merge this data contract Pull request. Are you shilling for us without knowing :) ?

- auto deploy - we have CLI 1 line deploy to airflow or git actions. Embeddable in any orchestrator.

- standardised ingestion with auto schema inference, typing, normalisation and evolution

- Validate and generate data contracts for input and output data assets - PR in review.

- shift left: there is no more left than before ingestion, this is it.

- data catalogs and lineages: We can currently register tables in catalogs. Ingestion row and column level lineage is included. Schema is exportable and can be stored somewhere like github for reuse as contract.

https://dlthub.com/docs/intro

thank you!

Expand full comment

I mentioned you in the data lineage section 😉

I didn't know about the latest feature, and it looks really good. I'll take a look and maybe prepare an article in the future for my Substack.

Thanks for the summary! 🙌🏼

Expand full comment

That's how I found the article, really grateful for it! But i mean we are going for those principles and I am glad to see the alignment in thinking!

this is the PR https://github.com/dlt-hub/dlt/pull/594, it contains confiburable evolution between hard contract (row must match), soft contract (data row must contain contract row) or configurable partial or complete evolution

- Adrian

Expand full comment