feat: Add {Expr,Series}.any_value
#3315
Conversation
FBruzzesi
changed the title
{Expr, Series}.any_value{Expr,Series}.any_value
|
thanks! some initial thoughts:
|
|
Thanks @MarcoGorelli
Wouldn't using
Sure I will add such option 🙏🏼 |
Sure, but we don't guarantee it - an in, it happens to return the first value, but it's not guaranteed (and the name Polars does something similar in joins - in the eager case it may happen to return rows in the same order when it's more efficient to do so, but it's not guaranteed (and it's documented that it's not guaranteed) |
| # !NOTE: ibis arbitrary returns a random non-null value | ||
| # See: https://ibis-project.org/reference/expression-generic.html#ibis.expr.types.generic.Column.arbitrary | ||
| expr = cast("ir.Column", expr) | ||
| return expr.arbitrary() if ignore_nulls else expr.first(include_null=True) |
There was a problem hiding this comment.
expr.first(include_null=not ignore_nulls) is not behaving as expected 🫠
| if kwargs.get("ignore_nulls"): | ||
| msg = ( | ||
| "`Expr.any_value(ignore_nulls=True)` is not supported in a `over` " | ||
| "context for pandas-like backend." | ||
| ) | ||
| raise NotImplementedError(msg) |
There was a problem hiding this comment.
This is an issue if a group has all null values
| if last_node.kwargs.get("ignore_nulls"): | ||
| msg = ( | ||
| "`Expr.any_value(ignore_nulls=True)` is not supported in a `group_by` " | ||
| "context for pandas-like backend" | ||
| ) | ||
| raise NotImplementedError(msg) |
There was a problem hiding this comment.
This is an issue if a group has all null values
|
thanks for updating, will try to review next week |
|
@MarcoGorelli just to make sure we align before going on a tangent. We briefly touched the possibility of explicitly marking this feature as unstable as it's outside of polars API. Next steps would be:
Let me know what your thoughts are :) Update: Ok I pushed a proposal in e8d44f6 |
What type of PR is this? (check all applicable)
Related issues
Expr.any_value? #3298Checklist