[
https://issues.apache.org/jira/browse/BEAM-11064?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17214216#comment-17214216
]
Brian Hulette commented on BEAM-11064:
--------------------------------------
It's distinct in that it tells us about the batched element type, which could
be a DataFrame or an arrow RecordBatch or Table.
I think what you're suggesting is we should always use Rows/RowType as the
element type. In the pandas/arrow case, row instances (if needed) would always
be views into an underlying RecordBatch or DataFrame. Is that right?
> Consider adding a pandas typehint
> ---------------------------------
>
> Key: BEAM-11064
> URL: https://issues.apache.org/jira/browse/BEAM-11064
> Project: Beam
> Issue Type: Improvement
> Components: sdk-py-core
> Reporter: Brian Hulette
> Assignee: Brian Hulette
> Priority: P2
>
> The typehint should be parameterized by a proxy object. This way we could
> retrieve the proxy for use in {{to_dataframe}}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
