[
https://issues.apache.org/jira/browse/SPARK-42032?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17677142#comment-17677142
]
Ruifeng Zheng commented on SPARK-42032:
---------------------------------------
Have an offline discussion with [~beliefer], Spark Connect has the same
behavior as Scala Dataset API, which is different from PySpark in some cases:
Dataset API
{code:java}
scala> spark.createDataFrame(Seq((1, Map("foo" -> -2.0, "bar" ->
2.0)))).show(100, 100)
+---+-------------------------+
| _1| _2|
+---+-------------------------+
| 1|{foo -> -2.0, bar -> 2.0}|
+---+-------------------------+
{code}
PySpark:
{code:java}
In [2]: spark.createDataFrame([(1, {"foo": -2.0, "bar": 2.0})]).show(100, 100)
+---+-------------------------+
| _1| _2|
+---+-------------------------+
| 1|{bar -> 2.0, foo -> -2.0}|
+---+-------------------------+
{code}
this should not be a bug in Connect.
> Map data show in different order
> --------------------------------
>
> Key: SPARK-42032
> URL: https://issues.apache.org/jira/browse/SPARK-42032
> Project: Spark
> Issue Type: Sub-task
> Components: Connect, PySpark
> Affects Versions: 3.4.0
> Reporter: Ruifeng Zheng
> Priority: Major
>
> not sure whether this needs to be fixed:
> {code:java}
> **********************************************************************
> File
> "/Users/ruifeng.zheng/Dev/spark/python/pyspark/sql/connect/functions.py",
> line 1623, in pyspark.sql.connect.functions.transform_keys
> Failed example:
> df.select(transform_keys(
> "data", lambda k, _: upper(k)).alias("data_upper")
> ).show(truncate=False)
> Expected:
> +-------------------------+
> |data_upper |
> +-------------------------+
> |{BAR -> 2.0, FOO -> -2.0}|
> +-------------------------+
> Got:
> +-------------------------+
> |data_upper |
> +-------------------------+
> |{FOO -> -2.0, BAR -> 2.0}|
> +-------------------------+
> <BLANKLINE>
> **********************************************************************
> File
> "/Users/ruifeng.zheng/Dev/spark/python/pyspark/sql/connect/functions.py",
> line 1630, in pyspark.sql.connect.functions.transform_values
> Failed example:
> df.select(transform_values(
> "data", lambda k, v: when(k.isin("IT", "OPS"), v + 10.0).otherwise(v)
> ).alias("new_data")).show(truncate=False)
> Expected:
> +---------------------------------------+
> |new_data |
> +---------------------------------------+
> |{OPS -> 34.0, IT -> 20.0, SALES -> 2.0}|
> +---------------------------------------+
> Got:
> +---------------------------------------+
> |new_data |
> +---------------------------------------+
> |{IT -> 20.0, SALES -> 2.0, OPS -> 34.0}|
> +---------------------------------------+
> <BLANKLINE>
> **********************************************************************
> 1 of 2 in pyspark.sql.connect.functions.transform_keys
> 1 of 2 in pyspark.sql.connect.functions.transform_values
> {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
