#openspending

pudo: any plans for joins in dataset?

I really liked dataset and was a bit sad I couldn't really use it

except to run a query

haha

I don't know

so I want to avoid it becoming a proper ORM

it's a tricky thing to o

yeah

but I think it's going to get virtual tables at some point

so table[

'col1', 'col2']

and then why not table.meet(other_table, say_hi='other_table.id')

I'm thinking something like extend

and then you can run queries on these constructs

yes exactly

guys, quick Q: is it fair to say that OS is planning to support a cubes API in the medium-term? Or is the default right now that it'll only be flat-file?

pudo: I think it's fair to say that it will be supported (although I'm inclined to say "probably supported") we're focussing on gettting flat files now and then provide a static aggregation process based on those, but the idea is that the flat files don't only have to be processed by static aggregation but can also be fed into something with a cubes API

urgh time to hit the server: https://twitter.com/OpenDataRU/status/614061346...

I don't like error reports that are limited to 140 characters