Re:Querying Parquet with Precision Using DuckDB
by scottmcdot on Friday June 25, 2021 @05:32PM
(#27636530)
Which GUI do you use now? For me I have to use Impala via HUE, only after I have created an external table from the parquet file. I really don't like web based GUIs, especially when my employer has a crappy server running the GUI rendering.
[ Parent
| on HN ]
Re:Querying Parquet with Precision Using DuckDB
by wenc on Friday June 25, 2021 @05:41PM
(#27636615)
I don't because I haven't found a good one for Parquet. I would really like something like SSMS for SQL Server.
I would poke at local Parquet files either through a local instance of Spark (or DataBricks when I was using that for remote data folders) or use Pandas' read_parquet from Jupyter.
Re:Querying Parquet with Precision Using DuckDB
by scottmcdot on Friday June 25, 2021 @07:50PM
(#27637640)
Same, I would really love a local-running tool like SSMS or Teradata SQL Assistant. CDSW has been good to use with Python (interacting with spark.sql) as a somewhat IDE however it's also web-based. The best thing I have gotten to work is Python within VS Code (running locally) tunnelling to a CDSW project, where I again can run spark.sql. However, it still prints tables out in ASCII way instead of the nice way of SSMS or Teradata SQL Assistant.