I write a lot of SQL, both to support distributed systems, as well as for BI/analytics. I find most database tools to be bloated and annoying, so I built Sidequery, an exploratory data analysis tool. This demo version runs 100% in your browser via DuckDB WASM and can query remote or local CSV or Parquet, either via drag and drop or attaching an entire directory of files. Millions of rows are no problem if you're on a quick machine.
PS: Make sure to select the explore button on bottom right after selecting some rows. If you click a particular "dimension" value, all of the other charts will then reflect your filter.
PPS: There is also a desktop version in the works with native database connectivity. A feature I'm proud of is federated query capability, allowing joining tables across database connections— even between Postgres & MySQL (and toss a CSV backed table in there for good measure).
Protips:
- Connect a folder of .csv or .parquet files and they'll all be registered as tables
- There is very little chart configuration required and it's easy for both SQL enjoyers and non alike to explore data
- Coordinated timeseries hover makes it simple to compare metrics at the same time period
- Any DuckDB aggregation function is supported as a custom expression when adding a metric