zero323 edited a comment on pull request #32431:
URL: https://github.com/apache/spark/pull/32431#issuecomment-1029850928
Since we are revisiting it have a counter-proposal ‒ instead exposing new
API, let's just improve UX of what we already have. For example, if we tweak
`select` to support keyword arguments like this:
```patch
diff --git a/python/pyspark/sql/dataframe.py
b/python/pyspark/sql/dataframe.py
index ee68865c98..00a7a4543e 100644
--- a/python/pyspark/sql/dataframe.py
+++ b/python/pyspark/sql/dataframe.py
@@ -1941,14 +1941,18 @@ class DataFrame(PandasMapOpsMixin,
PandasConversionMixin):
return Column(jc)
@overload
- def select(self, *cols: "ColumnOrName") -> "DataFrame":
+ def select(self, *cols: "ColumnOrName", **acols: Column) -> "DataFrame":
...
@overload
- def select(self, __cols: Union[List[Column], List[str]]) -> "DataFrame":
+ def select(self, __cols: Union[List[Column], List[str]], **acols:
Column) -> "DataFrame":
...
- def select(self, *cols: "ColumnOrName") -> "DataFrame": # type:
ignore[misc]
+ def select( # type: ignore[misc]
+ self,
+ *cols: "ColumnOrName",
+ **namedCols: Column,
+ ) -> "DataFrame":
"""Projects a set of expressions and returns a new
:class:`DataFrame`.
.. versionadded:: 1.3.0
@@ -1959,6 +1963,8 @@ class DataFrame(PandasMapOpsMixin,
PandasConversionMixin):
column names (string) or expressions (:class:`Column`).
If one of the column names is '*', that column is expanded to
include all columns
in the current :class:`DataFrame`.
+ namedCols : :class:`Column`
+ exprasssions select under given alaias.
Examples
--------
@@ -1968,8 +1974,17 @@ class DataFrame(PandasMapOpsMixin,
PandasConversionMixin):
[Row(name='Alice', age=2), Row(name='Bob', age=5)]
>>> df.select(df.name, (df.age + 10).alias('age')).collect()
[Row(name='Alice', age=12), Row(name='Bob', age=15)]
- """
- jdf = self._jdf.select(self._jcols(*cols))
+ >>> df.select(
+ ... "age",
+ ... name_lower=lower("name"),
+ ... name_upper=upper("name"),
+ ... age_plus_one=col("age") + 1,
+ ... ).limit(1).collect()
+ [Row(age=2, name_lower='alice', name_upper='ALICE', age_plus_one=3)]
+ """
+ if len(cols) == 1 and isinstance(cols[0], (list, tuple)):
+ cols = cols[0] # type: ignore[assignment]
+ jdf = self._jdf.select(self._jcols(*cols, *[c.alias(a) for a, c in
namedCols.items()]))
return DataFrame(jdf, self.sql_ctx)
@overload
```
will be able to use it like this:
```python
df = spark.range(10).select(rand(42).alias("id"))
df.select(
"id", plus_one=col("id") + 1, times_two=col("id") * 2, log_id=log("id")
).show()
```
making it similar to `dplyr::mutate`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
