Fix various small things in Python API (#1849)

py-polars/polars/cfg.py

@@ -5,7 +5,9 @@
 
 
 class Config:
-    "Configure polars"
+    """
+    Configure polars
+    """
 
     @classmethod
     def set_utf8_tables(cls) -> "Type[Config]":

py-polars/polars/internals/expr.py

@@ -488,7 +488,7 @@ def is_not(self) -> "Expr":
         ├╌╌╌╌╌╌╌┼╌╌╌╌╌╌┤
         │ false ┆ null │
         ╰───────┴──────╯
-        >>> df.select(col("a").is_not())
+        >>> df.select(pl.col("a").is_not())
         shape: (3, 1)
         ╭───────╮
         │ a     │
@@ -576,7 +576,7 @@ def drop_nulls(self) -> "Expr":
         """
         Syntactic sugar for:
 
-        >>> col("foo").filter(col("foo").is_not_null())
+        >>> pl.col("foo").filter(pl.col("foo").is_not_null())
         """
         return self.filter(self.is_not_null())
 
@@ -667,7 +667,7 @@ def cast(self, dtype: Type[Any], strict: bool = True) -> "Expr":
 
         Parameters
         ----------
-        data_type
+        dtype
             DataType to cast to
         strict
             Throw an error if a cast could not be done for instance due to an overflow
@@ -945,8 +945,8 @@ def over(self, expr: Union[str, "Expr", tp.List[Union["Expr", str]]]) -> "Expr":
         >>>})
         >>> (df.lazy()
         >>>    .select([
-        >>>       col("groups")
-        >>>       sum("values").over("groups"))
+        >>>       pl.col("groups")
+        >>>       sum("values").over("groups")
         >>>   ]).collect())
             ╭────────┬────────╮
             │ groups ┆ values │
@@ -1021,7 +1021,14 @@ def filter(self, predicate: "Expr") -> "Expr":
         return wrap_expr(self._pyexpr.filter(predicate._pyexpr))
 
     def where(self, predicate: "Expr") -> "Expr":
-        "alias for filter"
+        """
+        Alias for filter
+
+        Parameters
+        ----------
+        predicate
+            Boolean expression.
+        """
         return self.filter(predicate)
 
     def map(
@@ -1042,6 +1049,8 @@ def map(
             Lambda/ function to apply.
         return_dtype
             Dtype of the output Series.
+        agg_list
+
         """
         if return_dtype == str:
             return_dtype = Utf8
@@ -1084,7 +1093,7 @@ def apply(
 
         >>> df = pl.DataFrame({"a": [1,  2,  1,  1],
                    "b": ["a", "b", "c", "c"]})
-        >>> (df
+        >>> df
          .lazy()
          .groupby("b")
          .agg([col("a").apply(lambda x: x.sum())])
@@ -1272,7 +1281,7 @@ def inspect(self, fmt: str = "{}") -> "Expr":  # type: ignore
         """
         Prints the value that this expression evaluates to and passes on the value.
 
-        >>> df.select(col("foo").cumsum().inspect("value is: {}").alias("bar"))
+        >>> df.select(pl.col("foo").cumsum().inspect("value is: {}").alias("bar"))
         """
 
         def inspect(s: "pli.Series") -> "pli.Series":
@@ -1386,7 +1395,7 @@ def rolling_mean(
         >>>     }
         >>> )
         >>> df.select([
-        >>>     col("A").rolling_mean(window_size=2)
+        >>>     pl.col("A").rolling_mean(window_size=2)
         >>> ])
         shape: (6, 1)
         ┌──────┐
@@ -1542,7 +1551,7 @@ def rolling_apply(
         >>>     }
         >>> )
         >>> df.select([
-        >>>     col("A").rolling_apply(3, lambda s: s.std())
+        >>>     pl.col("A").rolling_apply(3, lambda s: s.std())
         >>> ])
         shape: (5, 1)
         ┌────────────────────┐
@@ -1752,7 +1761,7 @@ def str_concat(self, delimiter: str = "-") -> "Expr":  # type: ignore
 
         Examples
         >>> df = pl.DataFrame({"foo": [1, None, 2]})
-        >>> df = df.select(col("foo").str_concat("-"))
+        >>> df = df.select(pl.col("foo").str_concat("-"))
         shape: (1, 1)
         ┌──────────┐
         │ foo      │

py-polars/polars/internals/frame.py

@@ -424,6 +424,8 @@ def read_csv(
             - str -> all values encountered equal to this string will be null
             - tp.List[str] -> A null value per column.
             - Dict[str, str] -> A dictionary that maps column name to a null value string.
+        parse_dates
+            Whether to attempt to parse dates or not
 
         Returns
         -------
@@ -1156,7 +1158,7 @@ def __setitem__(self, key: Union[str, int, Tuple[Any, Any]], value: Any) -> None
             elif isinstance(col_selection, str):
                 self.replace(col_selection, s)
         else:
-            return NotImplemented
+            raise NotImplementedError
 
     def __len__(self) -> int:
         return self.height
@@ -1552,7 +1554,7 @@ def sort(
         **Sort by multiple columns.**
         For multiple columns we can also use expression syntax.
 
-        >>> df.sort([col("foo"), col("bar") ** 2], reverse=[True, False])
+        >>> df.sort([pl.col("foo"), pl.col("bar") ** 2], reverse=[True, False])
 
         """
         if type(by) is list or isinstance(by, pli.Expr):
@@ -2201,12 +2203,12 @@ def with_column(self, column: Union["pli.Series", "pli.Expr"]) -> "DataFrame":
 
     def with_column_renamed(self, existing_name: str, new_name: str) -> "DataFrame":
         """
-        Return a new DataFrame with the column added or replaced.
+        Return a new DataFrame with the column renamed.
 
         Parameters
         ----------
-        column
-            Series, where the name of the Series refers to the column in the DataFrame.
+        existing_name
+        new_name
         """
         return (
             self.lazy()

py-polars/polars/internals/lazy_frame.py

@@ -204,9 +204,9 @@ def inspect(self, fmt: str = "{}") -> "LazyFrame":  # type: ignore
         """
         Prints the value that this node in the computation graph evaluates to and passes on the value.
 
-        >>> (df.select(col("foo").cumsum().alias("bar"))
+        >>> (df.select(pl.col("foo").cumsum().alias("bar"))
         >>>    .inspect()  # print the node before the filter
-        >>>    .filter(col("bar") == col("foo")))
+        >>>    .filter(pl.col("bar") == pl.col("foo")))
         """
 
         def inspect(s: pli.DataFrame) -> pli.DataFrame:

py-polars/polars/internals/lazy_functions.py

@@ -592,7 +592,7 @@ def map(
 
     Parameters
     ----------
-    columns
+    exprs
         Input Series to f
     f
         Function to apply over the input
@@ -627,7 +627,7 @@ def apply(
 
     Parameters
     ----------
-    columns
+    exprs
         Input Series to f
     f
         Function to apply over the input

py-polars/polars/internals/series.py

@@ -678,14 +678,14 @@ def describe(self) -> "pli.DataFrame":
         if self.len() == 0:
             raise ValueError("Series must contain at least one value")
         elif self.is_numeric():
-            self = self.cast(Float64)
+            s = self.cast(Float64)
             stats = {
-                "min": self.min(),
-                "max": self.max(),
-                "null_count": self.null_count(),
-                "mean": self.mean(),
-                "std": self.std(),
-                "count": self.len(),
+                "min": s.min(),
+                "max": s.max(),
+                "null_count": s.null_count(),
+                "mean": s.mean(),
+                "std": s.std(),
+                "count": s.len(),
             }
         elif self.is_boolean():
             stats = {

py-polars/polars/io.py

@@ -406,6 +406,8 @@ def scan_csv(
     ----------
     file
         Path to a file.
+    infer_schema_length
+        The number of rows Polars will read to try to determine the schema.
     has_headers
         If the CSV file has headers or not.
     ignore_errors