Curious what this is / where it's live?

[kindly]

@jpmckinney The closest thing to this tool is probably datasette, but based on duckdb. But could also be considered a web based SQL/duckdb client. It allows you to run SQL against a variety of formats such as CSV, XLSX, PARQUET, JSON (loading data with flatterer) sqlite and postgres, using duckdb to query them. The idea is to make it the quickest way to run SQL against pretty much any data you have, hence the name. It was originally intended as a way to publish such data like datasette, being easy to host as a single binary.

I actually use this pretty much every day. If I get a CSV file that I need to view or do basic manipulation on, I normally load it using this instead of say Excel. It allows me to do things like compare the CSV file to data in postgres as it allows you to join between them and then export the CSV at the end. So it is basically now the SQL client and the CSV tool, I use most of the time.

However, even though I use it and think it would be useful to others, I do not have the time to work on it, document it or promote it. This is emphasised by the fact the sql box currenly ONLY has vim keybindings!

There is an MacOS binary of it in the releases, which should work. The CLI goes someting like sqlnow -v db=postgresql://david@localhost/gemlive -t ukraine_solar_approx_20240926.csv mydb.duckdb

-v means make a view of the file/url you supply and -t means it loads the data into the database. For the above case you get a duckdb database but if you do not supply the database file it will use an in memory one.

[jpmckinney]

Thank you! Sounds cool. I'll let you know if me or the team play round with it :)