mistercrunch commented on issue #3302: Create a PandasDatasource URL: https://github.com/apache/incubator-superset/issues/3302#issuecomment-323603943 I'm guessing your REST API is backend by some sort of database, couldn't you just access that? Alternatively, if you don't own that database and want to run analytical workloads off of it, shouldn't you just write a data pipeline to bring that data over into a local datastore? How would the owner of that service feel when front-loading the cache in the morning? You'd be potentially hitting that REST API pretty hard. The issue with the Pandas approach is you can't "pushdown" the filtering and/or aggregations, meaning the web server has to do all that heavy lifting. It's true that things would get cached, but any slice-dicing beyond the cache would fall off a perf cliff. There's no reason why it wouldn't work for small datasets though, though I doubt is would be very useful to others. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
With regards, Apache Git Services
