polars-df 0.20.0-x86_64-darwin → 0.21.0-x86_64-darwin

data/lib/polars/series.rb

@@ -563,7 +563,7 @@ module Polars
     #   # => false
     def any?(ignore_nulls: true, &block)
       if block_given?
-        apply(skip_nulls: ignore_nulls, &block).any?
+        apply(return_dtype: Boolean, skip_nulls: ignore_nulls, &block).any?
       else
         _s.any(ignore_nulls)
       end
@@ -587,7 +587,7 @@ module Polars
     #   # => true
     def all?(ignore_nulls: true, &block)
       if block_given?
-        apply(skip_nulls: ignore_nulls, &block).all?
+        apply(return_dtype: Boolean, skip_nulls: ignore_nulls, &block).all?
       else
         _s.all(ignore_nulls)
       end
@@ -611,7 +611,7 @@ module Polars
     #   # => true
     def none?(&block)
       if block_given?
-        apply(&block).none?
+        apply(return_dtype: Boolean, &block).none?
       else
         to_frame.select(Polars.col(name).is_not.all).to_series[0]
       end
@@ -1004,6 +1004,13 @@ module Polars
 
     # Get dummy variables.
     #
+    # @param separator [String]
+    #   Separator/delimiter used when generating column names.
+    # @param drop_first [Boolean]
+    #   Remove the first category from the variable being encoded.
+    # @param drop_nulls [Boolean]
+    #   If there are `nil` values in the series, a `null` column is not generated.
+    #
     # @return [DataFrame]
     #
     # @example
@@ -1020,8 +1027,8 @@ module Polars
     #   # │ 0   ┆ 1   ┆ 0   │
     #   # │ 0   ┆ 0   ┆ 1   │
     #   # └─────┴─────┴─────┘
-    def to_dummies(separator: "_", drop_first: false)
-      Utils.wrap_df(_s.to_dummies(separator, drop_first))
+    def to_dummies(separator: "_", drop_first: false, drop_nulls: false)
+      Utils.wrap_df(_s.to_dummies(separator, drop_first, drop_nulls))
     end
 
     # Bin continuous values into discrete categories.
@@ -1234,6 +1241,14 @@ module Polars
     #
     # @param sort [Boolean]
     #   Ensure the output is sorted from most values to least.
+    # @param parallel [Boolean]
+    #   Execute the computation in parallel.
+    # @param name [String]
+    #   Give the resulting count column a specific name; if `normalize` is
+    #   true this defaults to "proportion", otherwise defaults to "count".
+    # @param normalize [Boolean]
+    #   If true, the count is returned as the relative frequency of unique
+    #   values normalized to 1.0.
     #
     # @return [DataFrame]
     #
@@ -1722,6 +1737,10 @@ module Polars
     #
     # @param reverse [Boolean]
     #   Reverse sort.
+    # @param nulls_last [Boolean]
+    #   Place null values last instead of first.
+    # @param multithreaded [Boolean]
+    #   Sort using multiple threads.
     # @param in_place [Boolean]
     #   Sort in place.
     #
@@ -2154,7 +2173,7 @@ module Polars
     # @return [Series]
     #
     # @example
-    #   s = Polars::Series.new("a", [1, 2, 3])
+    #   s = Polars::Series.new("a", [[1, 2, 3]])
     #   s2 = Polars::Series.new("b", [2, 4, nil])
     #   s2.is_in(s)
     #   # =>
@@ -3132,7 +3151,7 @@ module Polars
     #
     # @example
     #   s = Polars::Series.new("a", [1, 2, 3])
-    #   s.map_elements { |x| x + 10 }
+    #   s.map_elements(return_dtype: Polars::Int64) { |x| x + 10 }
     #   # =>
     #   # shape: (3,)
     #   # Series: 'a' [i64]
@@ -3147,7 +3166,7 @@ module Polars
       else
         pl_return_dtype = Utils.rb_type_to_dtype(return_dtype)
       end
-      Utils.wrap_s(_s.apply_lambda(func, pl_return_dtype, skip_nulls))
+      Utils.wrap_s(_s.map_elements(func, pl_return_dtype, skip_nulls))
     end
     alias_method :map, :map_elements
     alias_method :apply, :map_elements
@@ -3421,6 +3440,8 @@ module Polars
     #   a result. If None, it will be set equal to window size.
     # @param center [Boolean]
     #   Set the labels at the center of the window
+    # @param ddof [Integer]
+    #   "Delta Degrees of Freedom": The divisor for a length N window is N - ddof
     #
     # @return [Series]
     #
@@ -3464,6 +3485,8 @@ module Polars
     #   a result. If None, it will be set equal to window size.
     # @param center [Boolean]
     #   Set the labels at the center of the window
+    # @param ddof [Integer]
+    #   "Delta Degrees of Freedom": The divisor for a length N window is N - ddof
     #
     # @return [Series]
     #
@@ -3560,10 +3583,10 @@ module Polars
     #   # [
     #   #         null
     #   #         null
-    #   #         1.0
     #   #         2.0
     #   #         3.0
     #   #         4.0
+    #   #         6.0
     #   # ]
     #
     # @example
@@ -3644,7 +3667,7 @@ module Polars
     #   # Series: 'a' [i64]
     #   # [
     #   #         5
-    #   #         3
+    #   #         2
     #   # ]
     def sample(
       n: nil,
@@ -4256,8 +4279,8 @@ module Polars
     #   # Series: 'a' [i64]
     #   # [
     #   #         2
-    #   #         1
     #   #         3
+    #   #         1
     #   # ]
     def shuffle(seed: nil)
       super

data/lib/polars/string_expr.rb

@@ -63,6 +63,13 @@ module Polars
     #   in the target string.
     # @param cache [Boolean]
     #   Use a cache of unique, converted datetimes to apply the conversion.
+    # @param ambiguous ['raise', 'earliest', 'latest', 'null']
+    #   Determine how to deal with ambiguous datetimes:
+    #
+    #   - `'raise'` (default): raise
+    #   - `'earliest'`: use the earliest datetime
+    #   - `'latest'`: use the latest datetime
+    #   - `'null'`: set to null
     #
     # @return [Expr]
     #
@@ -145,6 +152,8 @@ module Polars
     # @param exact [Boolean]
     #   - If true, require an exact format match.
     #   - If false, allow the format to match anywhere in the target string.
+    # @param cache [Boolean]
+    #   Use a cache of unique, converted dates to apply the datetime conversion.
     # @param utc [Boolean]
     #   Parse timezone aware datetimes as UTC. This may be useful if you have data
     #   with mixed offsets.
@@ -590,6 +599,7 @@ module Polars
     #   # │ null         ┆ null         │
     #   # └──────────────┴──────────────┘
     def pad_start(length, fill_char = " ")
+      length = Utils.parse_into_expression(length)
       Utils.wrap_expr(_rbexpr.str_pad_start(length, fill_char))
     end
     alias_method :rjust, :pad_start
@@ -620,6 +630,7 @@ module Polars
     #   # │ null         ┆ null         │
     #   # └──────────────┴──────────────┘
     def pad_end(length, fill_char = " ")
+      length = Utils.parse_into_expression(length)
       Utils.wrap_expr(_rbexpr.str_pad_end(length, fill_char))
     end
     alias_method :ljust, :pad_end
@@ -664,6 +675,9 @@ module Polars
     #   A valid regex pattern.
     # @param literal [Boolean]
     #   Treat pattern as a literal string.
+    # @param strict [Boolean]
+    #   Raise an error if the underlying pattern is not a valid regex,
+    #   otherwise mask out with a null value.
     #
     # @return [Expr]
     #
@@ -780,6 +794,9 @@ module Polars
     # @param dtype [Object]
     #   The dtype to cast the extracted value to. If nil, the dtype will be
     #   inferred from the JSON value.
+    # @param infer_schema_length [Integer]
+    #   The maximum number of rows to scan for schema inference.
+    #   If set to `nil`, the full data may be scanned *(this is slow)*.
     #
     # @return [Expr]
     #
@@ -1036,6 +1053,8 @@ module Polars
     #
     # @param pattern [String]
     #   A valid regex pattern
+    # @param literal [Boolean]
+    #   Treat `pattern` as a literal string, not as a regular expression.
     #
     # @return [Expr]
     #
@@ -1177,6 +1196,8 @@ module Polars
     #   Replacement string.
     # @param literal [Boolean]
     #   Treat pattern as a literal string.
+    # @param n [Integer]
+    #   Number of matches to replace.
     #
     # @return [Expr]
     #
@@ -1328,9 +1349,9 @@ module Polars
     #   # │ cafe ┆ 51966  │
     #   # │ null ┆ null   │
     #   # └──────┴────────┘
-    def to_integer(base: 10, strict: true)
+    def to_integer(base: 10, dtype: Int64, strict: true)
       base = Utils.parse_into_expression(base, str_as_lit: false)
-      Utils.wrap_expr(_rbexpr.str_to_integer(base, strict))
+      Utils.wrap_expr(_rbexpr.str_to_integer(base, dtype, strict))
     end
 
     # Parse integers with base radix from strings.
@@ -1437,7 +1458,7 @@ module Polars
     #     Polars.col("lyrics")
     #     .str.replace_many(
     #       ["me", "you", "they"],
-    #       ""
+    #       [""]
     #     )
     #     .alias("removes_pronouns")
     #   )

data/lib/polars/string_name_space.rb

@@ -63,6 +63,13 @@ module Polars
     #   in the target string.
     # @param cache [Boolean]
     #   Use a cache of unique, converted datetimes to apply the conversion.
+    # @param ambiguous ['raise', 'earliest', 'latest', 'null']
+    #   Determine how to deal with ambiguous datetimes:
+    #
+    #   - `'raise'` (default): raise
+    #   - `'earliest'`: use the earliest datetime
+    #   - `'latest'`: use the latest datetime
+    #   - `'null'`: set to null
     #
     # @return [Series]
     #
@@ -229,6 +236,10 @@ module Polars
     #
     # @param delimiter [String]
     #   The delimiter to insert between consecutive string values.
+    # @param ignore_nulls [Boolean]
+    #   Ignore null values (default).
+    #   If set to `False`, null values will be propagated. This means that
+    #   if the column contains any null values, the output is null.
     #
     # @return [Series]
     #

data/lib/polars/utils/parse.rb

@@ -33,6 +33,46 @@ module Polars
       exprs
     end
 
+    def self.parse_into_selector(i, strict: true)
+      if i.is_a?(::String)
+        cs = Selectors
+
+        cs.by_name([i], require_all: strict)
+      elsif i.is_a?(Selector)
+        i
+      elsif i.is_a?(Expr)
+        i.meta.as_selector
+      else
+        msg = "cannot turn #{i.inspect} into selector"
+        raise TypeError, msg
+      end
+    end
+
+    def self.parse_list_into_selector(inputs, strict: true)
+      if inputs.is_a?(::Array)
+        cs = Selectors
+
+        columns = inputs.select { |i| i.is_a?(::String) }
+        selector = cs.by_name(columns, require_all: strict)
+
+        if columns.length == inputs.length
+          return selector
+        end
+
+        # A bit cleaner
+        if columns.length == 0
+          selector = cs.empty
+        end
+
+        inputs.each do |i|
+          selector |= parse_into_selector(i, strict: strict)
+        end
+        selector
+      else
+        parse_into_selector(inputs, strict: strict)
+      end
+    end
+
     def self._parse_positional_inputs(inputs, structify: false)
       inputs_iter = _parse_inputs_as_iterable(inputs)
       inputs_iter.map { |e| parse_into_expression(e, structify: structify) }

data/lib/polars/utils.rb

@@ -107,7 +107,7 @@ module Polars
     end
 
     def self.is_selector(obj)
-      obj.is_a?(Selectors::SelectorProxy)
+      obj.is_a?(Selector)
     end
 
     def self.expand_selector(target, selector, strict: true)
@@ -161,5 +161,9 @@ module Polars
       # escapes _only_ those metachars with meaning to the rust regex crate
       Plr.re_escape(s)
     end
+
+    def self.parse_into_datatype_expr(input)
+      raise Todo
+    end
   end
 end

data/lib/polars/version.rb

@@ -1,4 +1,4 @@
 module Polars
   # @private
-  VERSION = "0.20.0"
+  VERSION = "0.21.0"
 end

data/lib/polars.rb

@@ -19,6 +19,11 @@ require_relative "polars/binary_expr"
 require_relative "polars/binary_name_space"
 require_relative "polars/cat_expr"
 require_relative "polars/cat_name_space"
+require_relative "polars/catalog"
+require_relative "polars/catalog/unity/catalog_info"
+require_relative "polars/catalog/unity/column_info"
+require_relative "polars/catalog/unity/namespace_info"
+require_relative "polars/catalog/unity/table_info"
 require_relative "polars/config"
 require_relative "polars/convert"
 require_relative "polars/plot"
@@ -54,6 +59,7 @@ require_relative "polars/io/ipc"
 require_relative "polars/io/json"
 require_relative "polars/io/ndjson"
 require_relative "polars/io/parquet"
+require_relative "polars/io/scan_options"
 require_relative "polars/lazy_frame"
 require_relative "polars/lazy_group_by"
 require_relative "polars/list_expr"
@@ -61,7 +67,9 @@ require_relative "polars/list_name_space"
 require_relative "polars/meta_expr"
 require_relative "polars/name_expr"
 require_relative "polars/rolling_group_by"
+require_relative "polars/scan_cast_options"
 require_relative "polars/schema"
+require_relative "polars/selector"
 require_relative "polars/selectors"
 require_relative "polars/series"
 require_relative "polars/slice"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: polars-df
 version: !ruby/object:Gem::Version
-  version: 0.20.0
+  version: 0.21.0
 platform: x86_64-darwin
 authors:
 - Andrew Kane
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2025-06-24 00:00:00.000000000 Z
+date: 2025-08-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bigdecimal
@@ -49,6 +49,11 @@ files:
 - lib/polars/binary_name_space.rb
 - lib/polars/cat_expr.rb
 - lib/polars/cat_name_space.rb
+- lib/polars/catalog.rb
+- lib/polars/catalog/unity/catalog_info.rb
+- lib/polars/catalog/unity/column_info.rb
+- lib/polars/catalog/unity/namespace_info.rb
+- lib/polars/catalog/unity/table_info.rb
 - lib/polars/config.rb
 - lib/polars/convert.rb
 - lib/polars/data_frame.rb
@@ -84,6 +89,7 @@ files:
 - lib/polars/io/json.rb
 - lib/polars/io/ndjson.rb
 - lib/polars/io/parquet.rb
+- lib/polars/io/scan_options.rb
 - lib/polars/lazy_frame.rb
 - lib/polars/lazy_group_by.rb
 - lib/polars/list_expr.rb
@@ -92,7 +98,9 @@ files:
 - lib/polars/name_expr.rb
 - lib/polars/plot.rb
 - lib/polars/rolling_group_by.rb
+- lib/polars/scan_cast_options.rb
 - lib/polars/schema.rb
+- lib/polars/selector.rb
 - lib/polars/selectors.rb
 - lib/polars/series.rb
 - lib/polars/slice.rb