WN 0007 - Nested Computations

[lloydtabb]

Malloy produces nested data. To access nested data today, malloy implicitly unnests while guaranteeing computational accuracy. There are times when we'd like to make computations against nested data without unnesting (count the number of occurrences with a filter) or to transform the nested data or to unnest the data more than once (to find co-occurrence).

https://github.com/malloydata/whatsnext/blob/main/wns/WN-0007/wn-0007.md

[christopherswenson]

I think pretty much all of this sounds good to me, except the exact syntax. I don't like making "return" a special name — that would mean that if you had a table with a column called return, you might run into some problems.

Two possible other syntaxes (I like # 1 better)

1. return aggregate:

aggregate:, group_by:, and calculate: can be prefixed by return to mark them as the return value.

I imagine we'll start adding other prefixes like private later, so this is probably consistent with future features.

Value could be named or unnamed.

dimension: foo is some_join -> {
  return aggregate: count()
}

2. return:

return would accept a dimension or measure, named or unnamed.

Downsides:

• Not clear to a reader whether the return value is an aggregate or scalar
• No way to use calculates as a return, because you can't get at output space values

dimension: foo is some_join -> {
  return: count()
}

[christopherswenson]

It seems like the Nested Sources section should be its own WN?