Robert Yi
Robert Yi
As-is, codecov is treating the rust and python libraries separately, meaning code coverage is calculated only from the latest file that is uploaded to codecov, rather than merging the two...
Currently, our Jinja engine (`wh run`) combines the templating and execution of the provided query into one step. For pipelining purposes, it could be useful to enable `wh render` that...
The rust side of things is embarrassingly uncovered by tests. :) This will be a standing issue, until we get to the point where our test coverage matches Python.
@cantzakas created the SQL query necessary to pull metadata in (https://github.com/dataframehq/whale/pull/140) -- we just have to make the greenplum extractor scaffolding. This should just follow the exact same shape as...
We currently allow users to write `--!wh-run` in either SQL files or ````sql` blocks in Markdown files, and have this dynamically replaced with results of the surrounding query queries in-line...
We currently allow users to write `--!wh-run` in either SQL files or ````sql` blocks in Markdown files, and have this dynamically replaced with results of the surrounding query queries in-line...
E.g. if one's full snowflake address is `aeo123.eu-central.snowflakecomputing.com` the account that needs to be specified in `wh connections` is `aeo123.eu-central`
Currently, we restrict the "metrics" feature to queries that can only return single values. We should enable multiple values to be returned and viewed in the rendered markdown metrics payload....
Given our [recent discussion](https://talk-whale.slack.com/archives/C01BLT4KPNE/p1605694343088000) in slack, ideally we should adjust the format in which Bigquery saves tags to the `TableMetadata` model to have a consistent format between both Glue and...