[
https://issues.apache.org/jira/browse/SPARK-61165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hyukjin Kwon reassigned SPARK-61165:
------------------------------------
Assignee: Hyukjin Kwon
> Keep the metadata of a StructField as its own dict, so that edits of it show
> in the field and its struct, in the Python client as PySpark does
> ----------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: SPARK-61165
> URL: https://issues.apache.org/jira/browse/SPARK-61165
> Project: Spark
> Issue Type: Bug
> Components: Connect
> Affects Versions: connect-rust-4.3.0
> Reporter: Hyukjin Kwon
> Assignee: Hyukjin Kwon
> Priority: Major
> Labels: pull-request-available
> Fix For: connect-rust-4.3.0
>
>
> In-place edits of a StructField's metadata are silently lost in the Python
> client: `f.metadata["comment"] = "x"` changes a temporary dict, so
> `f.metadata` is `{}` again, and the edit is not in `f.jsonValue()`, the
> struct holding the field, or the schema of a DataFrame created with it.
> PySpark's StructField keeps its metadata dict, so the edit persists.
> Repro: `f = StructField("a", IntegerType()); f.metadata["comment"] = "x";
> f.metadata` gives `{}` in the client and `{'comment': 'x'}` in PySpark 4.3.0.
> The fix keeps the dict in the field and reads the metadata from it when the
> field is converted to a Rust type.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
