polars-df 0.20.0-x86_64-darwin → 0.21.1-x86_64-darwin

data/lib/polars/list_name_space.rb

@@ -66,7 +66,7 @@ module Polars
     #
     # @example
     #   s = Polars::Series.new([[1, 2, 3], [5]])
-    #   s.list.lengths
+    #   s.list.len
     #   # =>
     #   # shape: (2,)
     #   # Series: '' [u32]
@@ -74,9 +74,10 @@ module Polars
     #   #         3
     #   #         1
     #   # ]
-    def lengths
+    def len
       super
     end
+    alias_method :lengths, :len
 
     # Drop all null values in the list.
     #
@@ -123,7 +124,7 @@ module Polars
     #   # shape: (2,)
     #   # Series: 'values' [list[i64]]
     #   # [
-    #   #         [2, 1]
+    #   #         [2, 3]
     #   #         [5]
     #   # ]
     def sample(n: nil, fraction: nil, with_replacement: false, shuffle: false, seed: nil)
@@ -202,6 +203,60 @@ module Polars
       super
     end
 
+    # Compute the median value of the arrays in the list.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("values", [[-1, 0, 1], [1, 10]])
+    #   s.list.median
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'values' [f64]
+    #   # [
+    #   #         0.0
+    #   #         5.5
+    #   # ]
+    def median
+      super
+    end
+
+    # Compute the std value of the arrays in the list.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("values", [[-1, 0, 1], [1, 10]])
+    #   s.list.std
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'values' [f64]
+    #   # [
+    #   #         1.0
+    #   #         6.363961
+    #   # ]
+    def std(ddof: 1)
+      super
+    end
+
+    # Compute the var value of the arrays in the list.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("values", [[-1, 0, 1], [1, 10]])
+    #   s.list.var
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'values' [f64]
+    #   # [
+    #   #         1.0
+    #   #         40.5
+    #   # ]
+    def var(ddof: 1)
+      super
+    end
+
     # Sort the arrays in the list.
     #
     # @return [Series]
@@ -254,7 +309,7 @@ module Polars
     #
     # @example
     #   s = Polars::Series.new("a", [[1, 1, 2], [2, 3, 3]])
-    #   s.list.unique()
+    #   s.list.unique
     #   # =>
     #   # shape: (2,)
     #   # Series: 'a' [list[i64]]
@@ -266,6 +321,24 @@ module Polars
       super
     end
 
+    # Count the number of unique values in every sub-lists.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("a", [[1, 1, 2], [2, 3, 4]])
+    #   s.list.n_unique
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'a' [u32]
+    #   # [
+    #   #         2
+    #   #         3
+    #   # ]
+    def n_unique
+      super
+    end
+
     # Concat the arrays in a Series dtype List in linear time.
     #
     # @param other [Object]
@@ -292,7 +365,7 @@ module Polars
     #
     # So index `0` would return the first item of every sublist
     # and index `-1` would return the last item of every sublist
-    # if an index is out of bounds, it will return a `None`.
+    # if an index is out of bounds, it will return a `nil`.
     #
     # @param index [Integer]
     #   Index to return per sublist
@@ -318,6 +391,63 @@ module Polars
       super
     end
 
+    # Take sublists by multiple indices.
+    #
+    # The indices may be defined in a single column, or by sublists in another
+    # column of dtype `List`.
+    #
+    # @param indices [Object]
+    #   Indices to return per sublist
+    # @param null_on_oob [Boolean]
+    #   Behavior if an index is out of bounds:
+    #   True -> set as null
+    #   False -> raise an error
+    #   Note that defaulting to raising an error is much cheaper
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("a", [[3, 2, 1], [], [1, 2]])
+    #   s.list.gather([0, 2], null_on_oob: true)
+    #   # =>
+    #   # shape: (3,)
+    #   # Series: 'a' [list[i64]]
+    #   # [
+    #   #         [3, 1]
+    #   #         [null, null]
+    #   #         [1, null]
+    #   # ]
+    def gather(
+      indices,
+      null_on_oob: false
+    )
+      super
+    end
+
+    # Take every n-th value start from offset in sublists.
+    #
+    # @param n [Integer]
+    #   Gather every n-th element.
+    # @param offset [Integer]
+    #   Starting index.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("a", [[1, 2, 3], [], [6, 7, 8, 9]])
+    #   s.list.gather_every(2, 1)
+    #   # =>
+    #   # shape: (3,)
+    #   # Series: 'a' [list[i64]]
+    #   # [
+    #   #         [2]
+    #   #         []
+    #   #         [7, 9]
+    #   # ]
+    def gather_every(n, offset = 0)
+      super
+    end
+
     # Get the value by index in the sublists.
     #
     # @return [Series]
@@ -554,6 +684,73 @@ module Polars
       super
     end
 
+    # Returns a column with a separate row for every list element.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("a", [[1, 2, 3], [4, 5, 6]])
+    #   s.list.explode
+    #   # =>
+    #   # shape: (6,)
+    #   # Series: 'a' [i64]
+    #   # [
+    #   #         1
+    #   #         2
+    #   #         3
+    #   #         4
+    #   #         5
+    #   #         6
+    #   # ]
+    def explode
+      super
+    end
+
+    # Count how often the value produced by `element` occurs.
+    #
+    # @param element [Object]
+    #   An expression that produces a single value
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("a", [[0], [1], [1, 2, 3, 2], [1, 2, 1], [4, 4]])
+    #   s.list.count_matches(1)
+    #   # =>
+    #   # shape: (5,)
+    #   # Series: 'a' [u32]
+    #   # [
+    #   #         0
+    #   #         1
+    #   #         1
+    #   #         2
+    #   #         0
+    #   # ]
+    def count_matches(element)
+      super
+    end
+
+    # Convert a List column into an Array column with the same inner data type.
+    #
+    # @param width [Integer]
+    #   Width of the resulting Array column.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new([[1, 2], [3, 4]], dtype: Polars::List.new(Polars::Int8))
+    #   s.list.to_array(2)
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: '' [array[i8, 2]]
+    #   # [
+    #   #         [1, 2]
+    #   #         [3, 4]
+    #   # ]
+    def to_array(width)
+      super
+    end
+
     # Convert the series of type `List` to a series of type `Struct`.
     #
     # @param n_field_strategy ["first_non_null", "max_width"]
@@ -608,5 +805,124 @@ module Polars
     def eval(expr)
       super
     end
+
+    # Filter elements in each list by a boolean expression, returning a new Series of lists.
+    #
+    # @param predicate [Object]
+    #   A boolean expression evaluated on each list element.
+    #   Use `Polars.element` to refer to the current element.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new("a", [[1, 4], [8, 5], [3, 2]])
+    #   s.list.filter(Polars.element % 2 == 0)
+    #   # =>
+    #   # shape: (3,)
+    #   # Series: 'a' [list[i64]]
+    #   # [
+    #   #         [4]
+    #   #         [8]
+    #   #         [2]
+    #   # ]
+    def filter(predicate)
+      super
+    end
+
+    # Compute the SET UNION between the elements in this list and the elements of `other`.
+    #
+    # @param other [Object]
+    #   Right hand side of the set operation.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   a = Polars::Series.new([[1, 2, 3], [], [nil, 3], [5, 6, 7]])
+    #   b = Polars::Series.new([[2, 3, 4], [3], [3, 4, nil], [6, 8]])
+    #   a.list.set_union(b)
+    #   # =>
+    #   # shape: (4,)
+    #   # Series: '' [list[i64]]
+    #   # [
+    #   #         [1, 2, … 4]
+    #   #         [3]
+    #   #         [null, 3, 4]
+    #   #         [5, 6, … 8]
+    #   # ]
+    def set_union(other)
+      super
+    end
+
+    # Compute the SET DIFFERENCE between the elements in this list and the elements of `other`.
+    #
+    # @param other [Object]
+    #   Right hand side of the set operation.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   a = Polars::Series.new([[1, 2, 3], [], [nil, 3], [5, 6, 7]])
+    #   b = Polars::Series.new([[2, 3, 4], [3], [3, 4, nil], [6, 8]])
+    #   a.list.set_difference(b)
+    #   # =>
+    #   # shape: (4,)
+    #   # Series: '' [list[i64]]
+    #   # [
+    #   #         [1]
+    #   #         []
+    #   #         []
+    #   #         [5, 7]
+    #   # ]
+    def set_difference(other)
+      super
+    end
+
+    # Compute the SET INTERSECTION between the elements in this list and the elements of `other`.
+    #
+    # @param other [Object]
+    #   Right hand side of the set operation.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   a = Polars::Series.new([[1, 2, 3], [], [nil, 3], [5, 6, 7]])
+    #   b = Polars::Series.new([[2, 3, 4], [3], [3, 4, nil], [6, 8]])
+    #   a.list.set_intersection(b)
+    #   # =>
+    #   # shape: (4,)
+    #   # Series: '' [list[i64]]
+    #   # [
+    #   #         [2, 3]
+    #   #         []
+    #   #         [null, 3]
+    #   #         [6]
+    #   # ]
+    def set_intersection(other)
+      super
+    end
+
+    # Compute the SET SYMMETRIC DIFFERENCE between the elements in this list and the elements of `other`.
+    #
+    # @param other [Object]
+    #   Right hand side of the set operation.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   a = Polars::Series.new([[1, 2, 3], [], [nil, 3], [5, 6, 7]])
+    #   b = Polars::Series.new([[2, 3, 4], [3], [3, 4, nil], [6, 8]])
+    #   a.list.set_symmetric_difference(b)
+    #   # =>
+    #   # shape: (4,)
+    #   # Series: '' [list[i64]]
+    #   # [
+    #   #         [1, 4]
+    #   #         [3]
+    #   #         [4]
+    #   #         [5, 7, 8]
+    #   # ]
+    def set_symmetric_difference(other)
+      super
+    end
   end
 end

data/lib/polars/meta_expr.rb

@@ -97,6 +97,70 @@ module Polars
       _rbexpr.meta_is_regex_projection
     end
 
+    # Indicate if this expression only selects columns (optionally with aliasing).
+    #
+    # This can include bare columns, columns matched by regex or dtype, selectors
+    # and exclude ops, and (optionally) column/expression aliasing.
+    #
+    # @param allow_aliasing [Boolean]
+    #   If false (default), any aliasing is not considered to be column selection.
+    #   Set true to allow for column selection that also includes aliasing.
+    #
+    # @return [Boolean]
+    #
+    # @example
+    #   e = Polars.col("foo")
+    #   e.meta.is_column_selection
+    #   # => true
+    #
+    # @example
+    #   e = Polars.col("foo").alias("bar")
+    #   e.meta.is_column_selection
+    #   # => false
+    #
+    # @example
+    #   e.meta.is_column_selection(allow_aliasing: true)
+    #   # => true
+    #
+    # @example
+    #   e = Polars.col("foo") * Polars.col("bar")
+    #   e.meta.is_column_selection
+    #   # => false
+    #
+    # @example
+    #   e = Polars.cs.starts_with("foo")
+    #   e.meta.is_column_selection
+    #   # => true
+    #
+    # @example
+    #   e = Polars.cs.starts_with("foo").exclude("foo!")
+    #   e.meta.is_column_selection
+    #   # => true
+    def is_column_selection(allow_aliasing: false)
+      _rbexpr.meta_is_column_selection(allow_aliasing)
+    end
+
+    # Indicate if this expression is a literal value (optionally aliased).
+    #
+    # @param allow_aliasing [Boolean]
+    #   If false (default), only a bare literal will match.
+    #   Set true to also allow for aliased literals.
+    #
+    # @return [Boolean]
+    #
+    # @example
+    #   e = Polars.lit(123)
+    #   e.meta.is_literal
+    #   # => true
+    #
+    # @example
+    #   e = Polars.lit(987.654321).alias("foo")
+    #   e.meta.is_literal
+    #   # => false
+    def is_literal(allow_aliasing: false)
+      _rbexpr.meta_is_literal(allow_aliasing)
+    end
+
     # Get the column name that this expression would produce.
     #
     # @return [String]
@@ -171,32 +235,17 @@ module Polars
       Utils.wrap_expr(_rbexpr.meta_undo_aliases)
     end
 
-    # Turn this expression in a selector.
-    #
-    # @return [Expr]
-    def _as_selector
-      Utils.wrap_expr(_rbexpr._meta_as_selector)
-    end
-
-    # Add selectors.
+    # Try to turn this expression in a selector.
     #
-    # @return [Expr]
-    def _selector_add(other)
-      Utils.wrap_expr(_rbexpr._meta_selector_add(other._rbexpr))
-    end
-
-    # Subtract selectors.
+    # Raises if the underlying expressions is not a column or selector.
     #
     # @return [Expr]
-    def _selector_sub(other)
-      Utils.wrap_expr(_rbexpr._meta_selector_sub(other._rbexpr))
-    end
-
-    # & selectors.
     #
-    # @return [Expr]
-    def _selector_and(other)
-      Utils.wrap_expr(_rbexpr._meta_selector_and(other._rbexpr))
+    # @note
+    #   This functionality is considered **unstable**. It may be changed
+    #   at any point without it being considered a breaking change.
+    def as_selector
+      Selector._from_rbselector(_rbexpr.into_selector)
     end
 
     # Format the expression as a tree.

data/lib/polars/name_expr.rb

@@ -194,5 +194,41 @@ module Polars
     def to_uppercase
       Utils.wrap_expr(_rbexpr.name_to_uppercase)
     end
+
+    # Add a prefix to all field names of a struct.
+    #
+    # @note
+    #   This only takes effect for struct columns.
+    #
+    # @param prefix [String]
+    #   Prefix to add to the field name.
+    #
+    # @return [Expr]
+    #
+    # @example
+    #   df = Polars::DataFrame.new({"x" => {"a" => 1, "b" => 2}})
+    #   df.select(Polars.col("x").name.prefix_fields("prefix_")).schema
+    #   # => {"x"=>Polars::Struct({"prefix_a"=>Polars::Int64, "prefix_b"=>Polars::Int64})}
+    def prefix_fields(prefix)
+      Utils.wrap_expr(_rbexpr.name_prefix_fields(prefix))
+    end
+
+    # Add a suffix to all field names of a struct.
+    #
+    # @note
+    #   This only takes effect for struct columns.
+    #
+    # @param suffix [String]
+    #   Suffix to add to the field name.
+    #
+    # @return [Expr]
+    #
+    # @example
+    #   df = Polars::DataFrame.new({"x" => {"a" => 1, "b" => 2}})
+    #   df.select(Polars.col("x").name.suffix_fields("_suffix")).schema
+    #   # => {"x"=>Polars::Struct({"a_suffix"=>Polars::Int64, "b_suffix"=>Polars::Int64})}
+    def suffix_fields(suffix)
+      Utils.wrap_expr(_rbexpr.name_suffix_fields(suffix))
+    end
   end
 end

data/lib/polars/scan_cast_options.rb

@@ -0,0 +1,64 @@
+module Polars
+  # Options for scanning files.
+  class ScanCastOptions
+    # Common configuration for scanning files.
+    #
+    # @note
+    #   This functionality is considered **unstable**. It may be changed
+    #   at any point without it being considered a breaking change.
+    #
+    # @param integer_cast ['upcast', 'forbid']
+    #   Configuration for casting from integer types:
+    #
+    #   * `upcast`: Allow lossless casting to wider integer types.
+    #   * `forbid`: Raises an error if dtypes do not match.
+    #
+    # @param float_cast ['upcast', 'downcast', 'forbid']
+    #   Configuration for casting from float types:
+    #
+    #   * `upcast`: Allow casting to higher precision float types.
+    #   * `downcast`: Allow casting to lower precision float types.
+    #   * `forbid`: Raises an error if dtypes do not match.
+    #
+    # @param datetime_cast ['nanosecond-downcast', 'convert-timezone', 'forbid']
+    #   Configuration for casting from datetime types:
+    #
+    #   * `nanosecond-downcast`: Allow nanosecond precision datetime to be
+    #     downcasted to any lower precision. This has a similar effect to
+    #     PyArrow's `coerce_int96_timestamp_unit`.
+    #   * `convert-timezone`: Allow casting to a different timezone.
+    #   * `forbid`: Raises an error if dtypes do not match.
+    #
+    # @param missing_struct_fields ['insert', 'raise']
+    #   Configuration for behavior when struct fields defined in the schema
+    #   are missing from the data:
+    #
+    #   * `insert`: Inserts the missing fields.
+    #   * `raise`: Raises an error.
+    #
+    # @param extra_struct_fields ['ignore', 'raise']
+    #   Configuration for behavior when extra struct fields outside of the
+    #   defined schema are encountered in the data:
+    #
+    #   * `ignore`: Silently ignores.
+    #   * `raise`: Raises an error.
+    def initialize(
+      integer_cast: "forbid",
+      float_cast: "forbid",
+      datetime_cast: "forbid",
+      missing_struct_fields: "raise",
+      extra_struct_fields: "raise",
+      _internal_call: false
+    )
+      @integer_cast = integer_cast
+      @float_cast = float_cast
+      @datetime_cast = datetime_cast
+      @missing_struct_fields = missing_struct_fields
+      @extra_struct_fields = extra_struct_fields
+    end
+
+    def self.default
+      new(_internal_call: true)
+    end
+  end
+end