polars-df 0.4.0-x86_64-linux → 0.6.0-x86_64-linux

data/README.md

@@ -25,7 +25,7 @@ Polars.read_csv("iris.csv")
   .collect
 ```
 
-You can follow [Polars tutorials](https://pola-rs.github.io/polars-book/user-guide/introduction.html) and convert the code to Ruby in many cases. Feel free to open an issue if you run into problems.
+You can follow [Polars tutorials](https://pola-rs.github.io/polars-book/user-guide/) and convert the code to Ruby in many cases. Feel free to open an issue if you run into problems.
 
 ## Reference
 
@@ -58,9 +58,9 @@ Polars.scan_parquet("file.parquet")
 From Active Record
 
 ```ruby
-Polars.read_sql(User.all)
+Polars.read_database(User.all)
 # or
-Polars.read_sql("SELECT * FROM users")
+Polars.read_database("SELECT * FROM users")
 ```
 
 From JSON
@@ -348,7 +348,7 @@ df.to_numo
 You can specify column types when creating a data frame
 
 ```ruby
-Polars::DataFrame.new(data, columns: {"a" => Polars::Int32, "b" => Polars::Float32})
+Polars::DataFrame.new(data, schema: {"a" => Polars::Int32, "b" => Polars::Float32})
 ```
 
 Supported types are:
@@ -357,8 +357,9 @@ Supported types are:
 - float - `Float64`, `Float32`
 - integer - `Int64`, `Int32`, `Int16`, `Int8`
 - unsigned integer - `UInt64`, `UInt32`, `UInt16`, `UInt8`
-- string - `Utf8`, `Categorical`
+- string - `Utf8`, `Binary`, `Categorical`
 - temporal - `Date`, `Datetime`, `Time`, `Duration`
+- other - `Object`, `List`, `Struct`, `Array` [unreleased]
 
 Get column types
 

data/lib/polars/array_expr.rb

@@ -0,0 +1,84 @@
+module Polars
+  # Namespace for array related expressions.
+  class ArrayExpr
+    # @private
+    attr_accessor :_rbexpr
+
+    # @private
+    def initialize(expr)
+      self._rbexpr = expr._rbexpr
+    end
+
+    # Compute the min values of the sub-arrays.
+    #
+    # @return [Expr]
+    #
+    # @example
+    #   df = Polars::DataFrame.new(
+    #     {"a" => [[1, 2], [4, 3]]},
+    #     schema: {"a" => Polars::Array.new(2, Polars::Int64)}
+    #   )
+    #   df.select(Polars.col("a").arr.min)
+    #   # =>
+    #   # shape: (2, 1)
+    #   # ┌─────┐
+    #   # │ a   │
+    #   # │ --- │
+    #   # │ i64 │
+    #   # ╞═════╡
+    #   # │ 1   │
+    #   # │ 3   │
+    #   # └─────┘
+    def min
+      Utils.wrap_expr(_rbexpr.array_min)
+    end
+
+    # Compute the max values of the sub-arrays.
+    #
+    # @return [Expr]
+    #
+    # @example
+    #   df = Polars::DataFrame.new(
+    #     {"a" => [[1, 2], [4, 3]]},
+    #     schema: {"a" => Polars::Array.new(2, Polars::Int64)}
+    #   )
+    #   df.select(Polars.col("a").arr.max)
+    #   # =>
+    #   # shape: (2, 1)
+    #   # ┌─────┐
+    #   # │ a   │
+    #   # │ --- │
+    #   # │ i64 │
+    #   # ╞═════╡
+    #   # │ 2   │
+    #   # │ 4   │
+    #   # └─────┘
+    def max
+      Utils.wrap_expr(_rbexpr.array_max)
+    end
+
+    # Compute the sum values of the sub-arrays.
+    #
+    # @return [Expr]
+    #
+    # @example
+    #   df = Polars::DataFrame.new(
+    #     {"a" => [[1, 2], [4, 3]]},
+    #     schema: {"a" => Polars::Array.new(2, Polars::Int64)}
+    #   )
+    #   df.select(Polars.col("a").arr.sum)
+    #   # =>
+    #   # shape: (2, 1)
+    #   # ┌─────┐
+    #   # │ a   │
+    #   # │ --- │
+    #   # │ i64 │
+    #   # ╞═════╡
+    #   # │ 3   │
+    #   # │ 7   │
+    #   # └─────┘
+    def sum
+      Utils.wrap_expr(_rbexpr.array_sum)
+    end
+  end
+end

data/lib/polars/array_name_space.rb

@@ -0,0 +1,77 @@
+module Polars
+  # Series.arr namespace.
+  class ArrayNameSpace
+    include ExprDispatch
+
+    self._accessor = "arr"
+
+    # @private
+    def initialize(series)
+      self._s = series._s
+    end
+
+    # Compute the min values of the sub-arrays.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "a", [[1, 2], [4, 3]], dtype: Polars::Array.new(2, Polars::Int64)
+    #   )
+    #   s.arr.min
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'a' [i64]
+    #   # [
+    #   #         1
+    #   #         3
+    #   # ]
+    def min
+      super
+    end
+
+    # Compute the max values of the sub-arrays.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   s = Polars::Series.new(
+    #     "a", [[1, 2], [4, 3]], dtype: Polars::Array.new(2, Polars::Int64)
+    #   )
+    #   s.arr.max
+    #   # =>
+    #   # shape: (2,)
+    #   # Series: 'a' [i64]
+    #   # [
+    #   #         2
+    #   #         4
+    #   # ]
+    def max
+      super
+    end
+
+    # Compute the sum values of the sub-arrays.
+    #
+    # @return [Series]
+    #
+    # @example
+    #   df = Polars::DataFrame.new(
+    #     {"a" => [[1, 2], [4, 3]]},
+    #     schema: {"a" => Polars::Array.new(2, Polars::Int64)}
+    #   )
+    #   df.select(Polars.col("a").arr.sum)
+    #   # =>
+    #   # shape: (2, 1)
+    #   # ┌─────┐
+    #   # │ a   │
+    #   # │ --- │
+    #   # │ i64 │
+    #   # ╞═════╡
+    #   # │ 3   │
+    #   # │ 7   │
+    #   # └─────┘
+    def sum
+      super
+    end
+  end
+end

data/lib/polars/batched_csv_reader.rb

@@ -41,7 +41,7 @@ module Polars
           dtypes.each do|k, v|
             dtype_list << [k, Utils.rb_type_to_dtype(v)]
           end
-        elsif dtypes.is_a?(Array)
+        elsif dtypes.is_a?(::Array)
           dtype_slice = dtypes
         else
           raise ArgumentError, "dtype arg should be list or dict"

data/lib/polars/convert.rb

@@ -26,8 +26,8 @@ module Polars
     #   # │ 1   ┆ 3   │
     #   # │ 2   ┆ 4   │
     #   # └─────┴─────┘
-    def from_hash(data, columns: nil)
-      DataFrame._from_hash(data, columns: columns)
+    def from_hash(data, schema: nil, columns: nil)
+      DataFrame._from_hash(data, schema: schema || columns)
     end
 
     # Construct a DataFrame from a sequence of dictionaries. This operation clones data.