Phillip Cloud
Phillip Cloud
This can be done with `_`: ``` In [18]: import ibis In [19]: from ibis import _ In [20]: t = ibis.table(dict(a="int", b="string")) In [21]: t.limit(100).aggregate(sum_a=_.a.sum()) Out[21]: r0 := UnboundTable:...
@icexelloss Thanks for chiming in. Is there any way we could get some information about the things that are breaking? If there are no `BREAKING CHANGE` commits between your current...
@icexelloss It sounds like you are tracking `master` or working off of a commit that isn't part of a release. That has the potential to cause frequent breakage that no...
@jreback @icexelloss Please review. The API is now preserved, with the original test verifying that the UDF is executed once **passing**. We'd like to merge this PR this week.
@icexelloss I get what you're saying, but I'm a bit stuck on this bit: > the table expr that the current expr is build on top of To me, the...
I wonder if there's a less disruptive way to execute this renaming? Maybe a short guide from `a -> b` in the contributing docs?
I'm going to close this out for now. I'd like to address this at some point, but keeping an issue open for it is just noise.
xref: #1767
@goodwanghan This is an interesting problem. Pandas handling of `NaN` here is consistent with the way in which `NaN`s are doubles, but it's not consistent with pandas using `NaN` as...
I think this would be good to get into 3.x and help us move away from pandas as required dependency.