Violetta Mishechkina

Hey @zilto good point! I think the reason it was done this way -- so we have an incremental field. But string could also be incremental, so I agree with...

Hi @zilto! I think what we can do -- is to create the variable in str type right away: ``` created_at: dlt.sources.incremental[str] = dlt.sources.incremental( "ts", initial_value=start_dt, end_value=end_dt, allow_external_schedulers=True, ), ```...

Hey, @rob-stover could you add a description?

Also, please add the same info about adapter for qdrant https://deploy-preview-1662--dlt-hub-docs.netlify.app/docs/dlt-ecosystem/destinations/qdrant/#qdrant_adapter and lancedb https://deploy-preview-1662--dlt-hub-docs.netlify.app/docs/dlt-ecosystem/destinations/lancedb/#run-the-pipeline

Regarding using the pandas for xml. People say that Beautiful soap is better. Pandas can only work with shallow xml https://dlthub-community.slack.com/archives/C04DQA7JJN6/p1722953520158219

@willi-mueller this is an important feature, thank you! I have a question: records with None value at the incremental cursor will be re-loaded each time, right? Could maybe some fall-back...

Also the link to the corresponding Slack discussion https://dlthub-community.slack.com/archives/C04DQA7JJN6/p1720787605031039