Replies: 1 comment 1 reply
-
As context, pre universal-SQL support, this is how all Evidence queries worked (ran against data warehouse). The problem this had, among others, is that you could only connect to a single database, and there was no support for client side interactivity. We plan to add support for querying external data sources in the future, though this will likely be via DuckDB rather than using the data warehouse directly. In the meantime, you can do aggregation in the source layer if you need to, and just query the aggregates. |
Beta Was this translation helpful? Give feedback.
1 reply
# for free
to join this conversation on GitHub.
Already have an account?
# to comment
-
If I understand correctly, Evidence will always extract data to a parquet file before querying it?
In my evidence page, I have multiple sales related queries.
For this mySQL source, I have to define a source query (e.g. select * from db.sales), so that I can use other queries in my page (e.g. select max(saleAmount) from sales).
So for each refresh, all full export is made from the source db, while I only need the summarized KPI which would have been performed more efficiently on the source db itself instead of locally on the parquet files?
Is there a way to choose whether the query is exectud on the db or on the parquet file?
Beta Was this translation helpful? Give feedback.
All reactions