polars-df 0.7.0 → 0.8.0

data/lib/polars/data_frame.rb

@@ -119,10 +119,10 @@ module Polars
 
       processed_null_values = Utils._process_null_values(null_values)
 
-      if columns.is_a?(String)
+      if columns.is_a?(::String)
         columns = [columns]
       end
-      if file.is_a?(String) && file.include?("*")
+      if file.is_a?(::String) && file.include?("*")
         dtypes_dict = nil
         if !dtype_list.nil?
           dtypes_dict = dtype_list.to_h
@@ -206,11 +206,11 @@ module Polars
       if Utils.pathlike?(source)
         source = Utils.normalise_filepath(source)
       end
-      if columns.is_a?(String)
+      if columns.is_a?(::String)
         columns = [columns]
       end
 
-      if source.is_a?(String) && source.include?("*") && Utils.local_file?(source)
+      if source.is_a?(::String) && source.include?("*") && Utils.local_file?(source)
         scan =
           Polars.scan_parquet(
             source,
@@ -269,11 +269,11 @@ module Polars
       if Utils.pathlike?(file)
         file = Utils.normalise_filepath(file)
       end
-      if columns.is_a?(String)
+      if columns.is_a?(::String)
         columns = [columns]
       end
 
-      if file.is_a?(String) && file.include?("*")
+      if file.is_a?(::String) && file.include?("*")
         raise Todo
       end
 
@@ -411,7 +411,7 @@ module Polars
     #     }
     #   )
     #   df.dtypes
-    #   # => [Polars::Int64, Polars::Float64, Polars::Utf8]
+    #   # => [Polars::Int64, Polars::Float64, Polars::String]
     def dtypes
       _df.dtypes
     end
@@ -429,7 +429,7 @@ module Polars
     #     }
     #   )
     #   df.schema
-    #   # => {"foo"=>Polars::Int64, "bar"=>Polars::Float64, "ham"=>Polars::Utf8}
+    #   # => {"foo"=>Polars::Int64, "bar"=>Polars::Float64, "ham"=>Polars::String}
     def schema
       columns.zip(dtypes).to_h
     end
@@ -589,13 +589,13 @@ module Polars
             return df.slice(row_selection, 1)
           end
           # df[2, "a"]
-          if col_selection.is_a?(String) || col_selection.is_a?(Symbol)
+          if col_selection.is_a?(::String) || col_selection.is_a?(Symbol)
             return self[col_selection][row_selection]
           end
         end
 
         # column selection can be "a" and ["a", "b"]
-        if col_selection.is_a?(String) || col_selection.is_a?(Symbol)
+        if col_selection.is_a?(::String) || col_selection.is_a?(Symbol)
           col_selection = [col_selection]
         end
 
@@ -621,7 +621,7 @@ module Polars
 
         # select single column
         # df["foo"]
-        if item.is_a?(String) || item.is_a?(Symbol)
+        if item.is_a?(::String) || item.is_a?(Symbol)
           return Utils.wrap_s(_df.column(item.to_s))
         end
 
@@ -647,7 +647,7 @@ module Polars
 
         if item.is_a?(Series)
           dtype = item.dtype
-          if dtype == Utf8
+          if dtype == String
             return _from_rbdf(_df.select(item))
           elsif dtype == UInt32
             return _from_rbdf(_df.take_with_series(item._s))
@@ -698,7 +698,7 @@ module Polars
         s[row_selection] = value
 
         if col_selection.is_a?(Integer)
-          replace_at_idx(col_selection, s)
+          replace_column(col_selection, s)
         elsif Utils.strlike?(col_selection)
           replace(col_selection, s)
         end
@@ -1222,7 +1222,7 @@ module Polars
     # @example
     #   df = Polars::DataFrame.new({"foo" => [1, 2, 3], "bar" => [4, 5, 6]})
     #   s = Polars::Series.new("baz", [97, 98, 99])
-    #   df.insert_at_idx(1, s)
+    #   df.insert_column(1, s)
     #   # =>
     #   # shape: (3, 3)
     #   # ┌─────┬─────┬─────┐
@@ -1244,7 +1244,7 @@ module Polars
     #     }
     #   )
     #   s = Polars::Series.new("d", [-2.5, 15, 20.5, 0])
-    #   df.insert_at_idx(3, s)
+    #   df.insert_column(3, s)
     #   # =>
     #   # shape: (4, 4)
     #   # ┌─────┬──────┬───────┬──────┐
@@ -1257,13 +1257,14 @@ module Polars
     #   # │ 3   ┆ 10.0 ┆ false ┆ 20.5 │
     #   # │ 4   ┆ 13.0 ┆ true  ┆ 0.0  │
     #   # └─────┴──────┴───────┴──────┘
-    def insert_at_idx(index, series)
+    def insert_column(index, series)
       if index < 0
         index = columns.length + index
       end
-      _df.insert_at_idx(index, series._s)
+      _df.insert_column(index, series._s)
       self
     end
+    alias_method :insert_at_idx, :insert_column
 
     # Filter the rows in the DataFrame based on a predicate expression.
     #
@@ -1367,7 +1368,7 @@ module Polars
           ]
         )._df
       )
-      summary.insert_at_idx(
+      summary.insert_column(
         0,
         Polars::Series.new(
           "describe",
@@ -1388,11 +1389,12 @@ module Polars
     #   df = Polars::DataFrame.new(
     #     {"foo" => [1, 2, 3], "bar" => [6, 7, 8], "ham" => ["a", "b", "c"]}
     #   )
-    #   df.find_idx_by_name("ham")
+    #   df.get_column_index("ham")
     #   # => 2
-    def find_idx_by_name(name)
-      _df.find_idx_by_name(name)
+    def get_column_index(name)
+      _df.get_column_index(name)
     end
+    alias_method :find_idx_by_name, :get_column_index
 
     # Replace a column at an index location.
     #
@@ -1412,7 +1414,7 @@ module Polars
     #     }
     #   )
     #   s = Polars::Series.new("apple", [10, 20, 30])
-    #   df.replace_at_idx(0, s)
+    #   df.replace_column(0, s)
     #   # =>
     #   # shape: (3, 3)
     #   # ┌───────┬─────┬─────┐
@@ -1424,13 +1426,14 @@ module Polars
     #   # │ 20    ┆ 7   ┆ b   │
     #   # │ 30    ┆ 8   ┆ c   │
     #   # └───────┴─────┴─────┘
-    def replace_at_idx(index, series)
+    def replace_column(index, series)
       if index < 0
         index = columns.length + index
       end
-      _df.replace_at_idx(index, series._s)
+      _df.replace_column(index, series._s)
       self
     end
+    alias_method :replace_at_idx, :replace_column
 
     # Sort the DataFrame by column.
     #
@@ -1524,13 +1527,14 @@ module Polars
     #       "ham" => ["c", "b", "a"]
     #     }
     #   )
-    #   df1.frame_equal(df1)
+    #   df1.equals(df1)
     #   # => true
-    #   df1.frame_equal(df2)
+    #   df1.equals(df2)
     #   # => false
-    def frame_equal(other, null_equal: true)
-      _df.frame_equal(other._df, null_equal)
+    def equals(other, null_equal: true)
+      _df.equals(other._df, null_equal)
     end
+    alias_method :frame_equal, :equals
 
     # Replace a column by a new Series.
     #
@@ -1716,7 +1720,7 @@ module Polars
     #   # │ 3   ┆ 8   ┆ c   │
     #   # └─────┴─────┴─────┘
     def drop_nulls(subset: nil)
-      if subset.is_a?(String)
+      if subset.is_a?(::String)
         subset = [subset]
       end
       _from_rbdf(_df.drop_nulls(subset))
@@ -2267,7 +2271,7 @@ module Polars
       if by.nil?
         by = []
       end
-      if by.is_a?(String)
+      if by.is_a?(::String)
         by = [by]
       end
       if offset.nil?
@@ -2461,17 +2465,17 @@ module Polars
     # @example
     #   df.join(other_df, on: "ham", how: "outer")
     #   # =>
-    #   # shape: (4, 4)
-    #   # ┌──────┬──────┬─────┬───────┐
-    #   # │ foo  ┆ bar  ┆ ham ┆ apple │
-    #   # │ ---  ┆ ---  ┆ --- ┆ ---   │
-    #   # │ i64  ┆ f64  ┆ str ┆ str   │
-    #   # ╞══════╪══════╪═════╪═══════╡
-    #   # │ 1    ┆ 6.0  ┆ a   ┆ x     │
-    #   # │ 2    ┆ 7.0  ┆ b   ┆ y     │
-    #   # │ null ┆ null ┆ d   ┆ z     │
-    #   # │ 3    ┆ 8.0  ┆ c   ┆ null  │
-    #   # └──────┴──────┴─────┴───────┘
+    #   # shape: (4, 5)
+    #   # ┌──────┬──────┬──────┬───────┬───────────┐
+    #   # │ foo  ┆ bar  ┆ ham  ┆ apple ┆ ham_right │
+    #   # │ ---  ┆ ---  ┆ ---  ┆ ---   ┆ ---       │
+    #   # │ i64  ┆ f64  ┆ str  ┆ str   ┆ str       │
+    #   # ╞══════╪══════╪══════╪═══════╪═══════════╡
+    #   # │ 1    ┆ 6.0  ┆ a    ┆ x     ┆ a         │
+    #   # │ 2    ┆ 7.0  ┆ b    ┆ y     ┆ b         │
+    #   # │ null ┆ null ┆ null ┆ z     ┆ d         │
+    #   # │ 3    ┆ 8.0  ┆ c    ┆ null  ┆ null      │
+    #   # └──────┴──────┴──────┴───────┴───────────┘
     #
     # @example
     #   df.join(other_df, on: "ham", how: "left")
@@ -3111,17 +3115,17 @@ module Polars
       sort_columns: false,
       separator: "_"
     )
-      if values.is_a?(String)
+      if values.is_a?(::String)
         values = [values]
       end
-      if index.is_a?(String)
+      if index.is_a?(::String)
         index = [index]
       end
-      if columns.is_a?(String)
+      if columns.is_a?(::String)
         columns = [columns]
       end
 
-      if aggregate_fn.is_a?(String)
+      if aggregate_fn.is_a?(::String)
         case aggregate_fn
         when "first"
           aggregate_expr = Polars.element.first._rbexpr
@@ -3206,10 +3210,10 @@ module Polars
     #   # │ z   ┆ c        ┆ 6     │
     #   # └─────┴──────────┴───────┘
     def melt(id_vars: nil, value_vars: nil, variable_name: nil, value_name: nil)
-      if value_vars.is_a?(String)
+      if value_vars.is_a?(::String)
         value_vars = [value_vars]
       end
-      if id_vars.is_a?(String)
+      if id_vars.is_a?(::String)
         id_vars = [id_vars]
       end
       if value_vars.nil?
@@ -3423,7 +3427,7 @@ module Polars
     #   # │ C   ┆ 2   ┆ l   │
     #   # └─────┴─────┴─────┘}
     def partition_by(groups, maintain_order: true, include_key: true, as_dict: false)
-      if groups.is_a?(String)
+      if groups.is_a?(::String)
         groups = [groups]
       elsif !groups.is_a?(::Array)
         groups = Array(groups)
@@ -3774,7 +3778,7 @@ module Polars
     #   # └─────┴─────┴─────┘
     def max(axis: 0)
       if axis == 0
-        _from_rbdf(_df.max)
+        lazy.max.collect(_eager: true)
       elsif axis == 1
         Utils.wrap_s(_df.max_horizontal)
       else
@@ -3806,7 +3810,7 @@ module Polars
     #   # └─────┴─────┴─────┘
     def min(axis: 0)
       if axis == 0
-        _from_rbdf(_df.min)
+        lazy.min.collect(_eager: true)
       elsif axis == 1
         Utils.wrap_s(_df.min_horizontal)
       else
@@ -3855,7 +3859,7 @@ module Polars
     def sum(axis: 0, null_strategy: "ignore")
       case axis
       when 0
-        _from_rbdf(_df.sum)
+        lazy.sum.collect(_eager: true)
       when 1
         Utils.wrap_s(_df.sum_horizontal(null_strategy))
       else
@@ -3893,7 +3897,7 @@ module Polars
     def mean(axis: 0, null_strategy: "ignore")
       case axis
       when 0
-        _from_rbdf(_df.mean)
+        lazy.mean.collect(_eager: true)
       when 1
         Utils.wrap_s(_df.mean_horizontal(null_strategy))
       else
@@ -3939,7 +3943,7 @@ module Polars
     #   # │ 0.816497 ┆ 0.816497 ┆ null │
     #   # └──────────┴──────────┴──────┘
     def std(ddof: 1)
-      _from_rbdf(_df.std(ddof))
+      lazy.std(ddof: ddof).collect(_eager: true)
     end
 
     # Aggregate the columns of this DataFrame to their variance value.
@@ -3980,7 +3984,7 @@ module Polars
     #   # │ 0.666667 ┆ 0.666667 ┆ null │
     #   # └──────────┴──────────┴──────┘
     def var(ddof: 1)
-      _from_rbdf(_df.var(ddof))
+      lazy.var(ddof: ddof).collect(_eager: true)
     end
 
     # Aggregate the columns of this DataFrame to their median value.
@@ -4006,7 +4010,7 @@ module Polars
     #   # │ 2.0 ┆ 7.0 ┆ null │
     #   # └─────┴─────┴──────┘
     def median
-      _from_rbdf(_df.median)
+      lazy.median.collect(_eager: true)
     end
 
     # Aggregate the columns of this DataFrame to their product values.
@@ -4063,7 +4067,7 @@ module Polars
     #   # │ 2.0 ┆ 7.0 ┆ null │
     #   # └─────┴─────┴──────┘
     def quantile(quantile, interpolation: "nearest")
-      _from_rbdf(_df.quantile(quantile, interpolation))
+      lazy.quantile(quantile, interpolation: interpolation).collect(_eager: true)
     end
 
     # Get one hot encoded dummy variables.
@@ -4094,7 +4098,7 @@ module Polars
     #   # │ 0     ┆ 1     ┆ 0     ┆ 1     ┆ 0     ┆ 1     │
     #   # └───────┴───────┴───────┴───────┴───────┴───────┘
     def to_dummies(columns: nil, separator: "_", drop_first: false)
-      if columns.is_a?(String)
+      if columns.is_a?(::String)
         columns = [columns]
       end
       _from_rbdf(_df.to_dummies(columns, separator, drop_first))
@@ -4603,8 +4607,8 @@ module Polars
     #   # │ 1   ┆ 5   │
     #   # │ 3   ┆ 7   │
     #   # └─────┴─────┘
-    def gather_every(n)
-      select(Utils.col("*").gather_every(n))
+    def gather_every(n, offset = 0)
+      select(Utils.col("*").gather_every(n, offset))
     end
     alias_method :take_every, :gather_every
 
@@ -4754,7 +4758,7 @@ module Polars
     #   # │ bar    ┆ 2   ┆ b   ┆ null ┆ [3]       ┆ womp  │
     #   # └────────┴─────┴─────┴──────┴───────────┴───────┘
     def unnest(names)
-      if names.is_a?(String)
+      if names.is_a?(::String)
         names = [names]
       end
       _from_rbdf(_df.unnest(names))
@@ -4867,10 +4871,10 @@ module Polars
             if val.is_a?(Hash) && dtype != Struct
               updated_data[name] = DataFrame.new(val).to_struct(name)
             elsif !Utils.arrlen(val).nil?
-              updated_data[name] = Series.new(String.new(name), val, dtype: dtype)
-            elsif val.nil? || [Integer, Float, TrueClass, FalseClass, String, ::Date, ::DateTime, ::Time].any? { |cls| val.is_a?(cls) }
+              updated_data[name] = Series.new(::String.new(name), val, dtype: dtype)
+            elsif val.nil? || [Integer, Float, TrueClass, FalseClass, ::String, ::Date, ::DateTime, ::Time].any? { |cls| val.is_a?(cls) }
               dtype = Polars::Float64 if val.nil? && dtype.nil?
-              updated_data[name] = Series.new(String.new(name), [val], dtype: dtype).extend_constant(val, array_len - 1)
+              updated_data[name] = Series.new(::String.new(name), [val], dtype: dtype).extend_constant(val, array_len - 1)
             else
               raise Todo
             end
@@ -4927,7 +4931,7 @@ module Polars
       end
       column_names =
         (schema || []).map.with_index do |col, i|
-          if col.is_a?(String)
+          if col.is_a?(::String)
             col || "column_#{i}"
           else
             col[0]
@@ -4940,7 +4944,7 @@ module Polars
       lookup = column_names.zip(lookup_names || []).to_h
 
       column_dtypes =
-        (schema || []).select { |col| !col.is_a?(String) && col[1] }.to_h do |col|
+        (schema || []).select { |col| !col.is_a?(::String) && col[1] }.to_h do |col|
           [lookup[col[0]] || col[0], col[1]]
         end
 

data/lib/polars/data_types.rb

@@ -120,9 +120,12 @@ module Polars
   end
 
   # UTF-8 encoded string type.
-  class Utf8 < DataType
+  class String < DataType
   end
 
+  # Allow Utf8 as an alias for String
+  Utf8 = String
+
   # Binary type.
   class Binary < DataType
   end

data/lib/polars/date_time_expr.rb

@@ -365,7 +365,7 @@ module Polars
     #   # ┌──────┐
     #   # │ date │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 1    │
     #   # │ 2    │
@@ -407,7 +407,7 @@ module Polars
     #   # ┌──────┐
     #   # │ date │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 1    │
     #   # │ 2    │
@@ -449,7 +449,7 @@ module Polars
     #   # ┌──────┐
     #   # │ date │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 1    │
     #   # │ 5    │
@@ -496,7 +496,7 @@ module Polars
     #   # ┌─────────┬──────────────┬─────────────┐
     #   # │ weekday ┆ day_of_month ┆ day_of_year │
     #   # │ ---     ┆ ---          ┆ ---         │
-    #   # │ u32     ┆ u32          ┆ u32         │
+    #   # │ i8      ┆ i8           ┆ i16         │
     #   # ╞═════════╪══════════════╪═════════════╡
     #   # │ 1       ┆ 1            ┆ 1           │
     #   # │ 4       ┆ 4            ┆ 4           │
@@ -544,7 +544,7 @@ module Polars
     #   # ┌─────────┬──────────────┬─────────────┐
     #   # │ weekday ┆ day_of_month ┆ day_of_year │
     #   # │ ---     ┆ ---          ┆ ---         │
-    #   # │ u32     ┆ u32          ┆ u32         │
+    #   # │ i8      ┆ i8           ┆ i16         │
     #   # ╞═════════╪══════════════╪═════════════╡
     #   # │ 1       ┆ 1            ┆ 1           │
     #   # │ 4       ┆ 4            ┆ 4           │
@@ -592,7 +592,7 @@ module Polars
     #   # ┌─────────┬──────────────┬─────────────┐
     #   # │ weekday ┆ day_of_month ┆ day_of_year │
     #   # │ ---     ┆ ---          ┆ ---         │
-    #   # │ u32     ┆ u32          ┆ u32         │
+    #   # │ i8      ┆ i8           ┆ i16         │
     #   # ╞═════════╪══════════════╪═════════════╡
     #   # │ 1       ┆ 1            ┆ 1           │
     #   # │ 4       ┆ 4            ┆ 4           │
@@ -654,7 +654,7 @@ module Polars
     #   # ┌──────┐
     #   # │ date │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 0    │
     #   # │ 12   │
@@ -695,7 +695,7 @@ module Polars
     #   # ┌──────┐
     #   # │ date │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 0    │
     #   # │ 2    │
@@ -744,7 +744,7 @@ module Polars
     #   # ┌──────┐
     #   # │ secs │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 0    │
     #   # │ 3    │
@@ -789,7 +789,7 @@ module Polars
     #   # ┌──────┐
     #   # │ date │
     #   # │ ---  │
-    #   # │ u32  │
+    #   # │ i8   │
     #   # ╞══════╡
     #   # │ 0    │
     #   # │ 2    │

data/lib/polars/date_time_name_space.rb

@@ -220,7 +220,7 @@ module Polars
     #   date.dt.quarter
     #   # =>
     #   # shape: (4,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         1
     #   #         1
@@ -258,7 +258,7 @@ module Polars
     #   date.dt.month
     #   # =>
     #   # shape: (4,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         1
     #   #         2
@@ -296,7 +296,7 @@ module Polars
     #   date.dt.week
     #   # =>
     #   # shape: (4,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         1
     #   #         5
@@ -336,7 +336,7 @@ module Polars
     #   date.dt.weekday
     #   # =>
     #   # shape: (7,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         1
     #   #         2
@@ -378,7 +378,7 @@ module Polars
     #   date.dt.day
     #   # =>
     #   # shape: (5,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         1
     #   #         3
@@ -416,7 +416,7 @@ module Polars
     #   date.dt.ordinal_day
     #   # =>
     #   # shape: (3,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i16]
     #   # [
     #   #         1
     #   #         32
@@ -452,7 +452,7 @@ module Polars
     #   date.dt.hour
     #   # =>
     #   # shape: (4,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         0
     #   #         1
@@ -488,7 +488,7 @@ module Polars
     #   date.dt.minute
     #   # =>
     #   # shape: (3,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         0
     #   #         2
@@ -531,7 +531,7 @@ module Polars
     #   date.dt.second
     #   # =>
     #   # shape: (9,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i8]
     #   # [
     #   #         0
     #   #         0
@@ -593,7 +593,7 @@ module Polars
     #   date.dt.millisecond
     #   # =>
     #   # shape: (9,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i32]
     #   # [
     #   #         0
     #   #         500
@@ -638,7 +638,7 @@ module Polars
     #   date.dt.microsecond
     #   # =>
     #   # shape: (9,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i32]
     #   # [
     #   #         0
     #   #         500000
@@ -683,7 +683,7 @@ module Polars
     #   date.dt.nanosecond
     #   # =>
     #   # shape: (9,)
-    #   # Series: '' [u32]
+    #   # Series: '' [i32]
     #   # [
     #   #         0
     #   #         500000000