Connector Improvement: Remove HTML tags while ingesting
Connectors such as Braze expose the html body of email campaigns, which is ingested by FiveTran, however they do not provide a human-readable version with html tags stripped out. This means that the end user has to create a bespoke tag stripping pipeline. These ad-hoc solutions can be error prone and expose security holes for customers if whatever html parser used isn't correctly designed. A more elegant solution would be a FiveTran method to strip out html from a field during ingestion, which could be broadly applicable to different data sources that include raw html code.
-
Official comment
Hi Barry,
Luke from the Product team here! Thanks for submitting this feature request. This isn't something that we plan to support.
I understand how this could be valuable. However, Fivetran focuses on an Extract, Load, Transform (ELT) method rather than traditional ETL. This request is for a transformation before loading, which would be ETL. In this blog post, you can read more about why Fivetran believes ELT is the better approach. To summarize: we avoid in-flight transformations to reduce failure rates, improve performance, and decrease complexity. We enable customers to do transformations like this based on their use cases in their destinations.
Let me know if you have any questions.
Cheers,
Luke
Please sign in to leave a comment.
Comments
1 comment