[
https://issues.apache.org/jira/browse/BEAM-12698?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17401884#comment-17401884
]
Yunfeng Zhang commented on BEAM-12698:
--------------------------------------
Thank you very much for taking a look! I tried to create an example using
synthetic data, but it turns out the binary operations do work.
It seems the problem was in my other parts of code that parse and read in json
data. It used code like this
{color:#de350b}beam.Map(json.loads).with_output_types(CustomNamedTupleType){color}
to cast json data into a custom type. But all the fields were parsed as
objects instead of the specified types like int or str. Any suggestions here
would be appreciated!
> Some binary operations on DeferredSeries don't work
> ---------------------------------------------------
>
> Key: BEAM-12698
> URL: https://issues.apache.org/jira/browse/BEAM-12698
> Project: Beam
> Issue Type: Bug
> Components: dsl-dataframe
> Affects Versions: 2.31.0
> Reporter: Yunfeng Zhang
> Priority: P2
>
> I get an error when trying to multiply two columns of a deferred dataframe:
> ib.collect(df.A.mul(df.B))
> {color:#FF0000}TypeError: can't multiply sequence by non-int of type 'str'
> [while running '[35]:
> ComputedExpression[mul_Series_140480471782160]/[ComputedExpression[mul_Series_140480471782160]]:140480472023312/FlatMap(evaluate)/FlatMap(evaluate)']{color}
>
> Similarly, adding a scalar also doesn't work:
> ib.collect(df.A.add(1))
> {color:#FF0000}TypeError: can only concatenate str (not "int") to str [while
> running '[39]:
> ComputedExpression[add_Series_140480422551952]/[ComputedExpression[get_column_Series_140480422552144],
>
> ComputedExpression[add_Series_140480422551952]]:140480472568336/FlatMap(evaluate)/FlatMap(evaluate)']{color}
> {color:#172b4d}But adding two columns work.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
