[R] Add a schema method for arrow_dplyr_query

[asfimport]

we have implicit_schema() which can generate the final schema for a query, though it's not exported. Maybe we "just" export that for people to be able to get what the schema of the resulting query will be.

Alternatively we could make a schema (S3) method that would return the (implicit) schema with schema(query_obj). Though this might be overloading schema since that is not how we retrieve schemas elsewhere (e.g. schema(arrow_table(mtcars)) does not currently work).

One use case: duckdb/duckdb#3299

Reporter: Jonathan Keane / @jonkeane

_{Note: This issue was originally created as ARROW-16030. Please see the migration documentation for further details.}

[asfimport]

Dewey Dunnington / @paleolimbot:
Just some thoughts! It relates to ARROW-15168 in that it would be another S3 generic that produces an arrow object.

I think there are three things going on with a schema...construction (schema(...), coercion (as_arrow_schema(...) maybe), and extraction (???). Lots of libraries lump bits of that together...(e.g., sf::st_crs(...) does both all three!). Substrait will need extraction, too, so it should definitely exist. Maybe extract_schema(x)? Should it also infer the schema from a data.frame or database connection (like type(x)?).

[github-actions]

This issue has been marked as stale because it has had no activity in the past 365 days. Please remove the stale label or comment below, or this issue will be closed in 14 days. If this improvement is still desired but has no current owner, please add the 'Status: needs champion' label.