polars-df 0.15.0 → 0.26.0

data/lib/polars/series_plot.rb

@@ -0,0 +1,72 @@
+module Polars
+  # Series.plot namespace.
+  class SeriesPlot
+    # @private
+    def initialize(s)
+      require "vega"
+
+      name = s.name || "value"
+      @df = s.to_frame(name)
+      @series_name = name
+    end
+
+    # Draw histogram.
+    #
+    # @return [Vega::LiteChart]
+    def hist
+      encoding = {
+        x: {field: @series_name, bin: true},
+        y: {aggregate: "count"}
+      }
+
+      Vega.lite
+        .data(@df.rows(named: true))
+        .mark(type: "bar", tooltip: true)
+        .encoding(encoding)
+        .config(axis: {labelFontSize: 12})
+    end
+
+    # Draw kernel density estimate plot.
+    #
+    # @return [Vega::LiteChart]
+    def kde
+      if @series_name == "density"
+        msg = "cannot use `plot.kde` when Series name is `'density'`"
+        raise ArgumentError, msg
+      end
+
+      encoding = {
+        x: {field: @series_name, type: "quantitative"},
+        y: {field: "density", type: "quantitative"}
+      }
+
+      Vega.lite
+        .data(@df.rows(named: true))
+        .transform(density: @series_name, as: [@series_name, "density"])
+        .mark(type: "area", tooltip: true)
+        .encoding(encoding)
+        .config(axis: {labelFontSize: 12})
+    end
+
+    # Draw line plot.
+    #
+    # @return [Vega::LiteChart]
+    def line
+      if @series_name == "index"
+        msg = "cannot call `plot.line` when Series name is 'index'"
+        raise ArgumentError, msg
+      end
+
+      encoding = {
+        x: {field: "index", type: "quantitative"},
+        y: {field: @series_name, type: "quantitative"}
+      }
+
+      Vega.lite
+        .data(@df.with_row_index.rows(named: true))
+        .mark(type: "line", tooltip: true)
+        .encoding(encoding)
+        .config(axis: {labelFontSize: 12})
+    end
+  end
+end

data/lib/polars/slice.rb

@@ -12,7 +12,7 @@ module Polars
 
       # check for fast-paths / single-operation calls
       if @slice_length == 0
-        @obj.cleared
+        @obj.clear
       elsif @is_unbounded && [-1, 1].include?(@stride)
         @stride < 0 ? @obj.reverse : @obj.clone
       elsif @start >= 0 && @stop >= 0 && @stride == 1

data/lib/polars/sql_context.rb

@@ -5,11 +5,17 @@ module Polars
     attr_accessor :_ctxt, :_eager_execution
 
     # Initialize a new `SQLContext`.
-    def initialize(frames = nil, eager_execution: false, **named_frames)
+    def initialize(frames = nil, register_globals: false, eager: false, **named_frames)
+      Utils.issue_unstable_warning(
+        "`SQLContext` is considered **unstable**, although it is close to being considered stable."
+      )
       self._ctxt = RbSQLContext.new
-      self._eager_execution = eager_execution
+      self._eager_execution = eager
 
       frames = (frames || {}).to_h
+      if register_globals
+        raise Todo
+      end
 
       if frames.any? || named_frames.any?
         register_many(frames, **named_frames)
@@ -37,7 +43,8 @@ module Polars
     #       ["Pulp Fiction", 1994, 8_000_000, 107_930_000, 8.9],
     #       ["The Shawshank Redemption", 1994, 25_000_000, 28_341_469, 9.3],
     #     ],
-    #     schema: ["title", "release_year", "budget", "gross", "imdb_score"]
+    #     schema: ["title", "release_year", "budget", "gross", "imdb_score"],
+    #     orient: "row"
     #   )
     #   ctx = Polars::SQLContext.new(films: df)
     #   ctx.execute(
@@ -66,11 +73,11 @@ module Polars
     #   ctx.execute(
     #     "
     #     SELECT
-    #         MAX(release_year / 10) * 10 AS decade,
+    #         MAX(release_year // 10) * 10 AS decade,
     #         SUM(gross) AS total_gross,
     #         COUNT(title) AS n_films,
     #     FROM films
-    #     GROUP BY (release_year / 10) -- decade
+    #     GROUP BY (release_year // 10) -- decade
     #     ORDER BY total_gross DESC
     #     ",
     #     eager: true
@@ -129,7 +136,7 @@ module Polars
     #   Named eager/lazy frames, provided as kwargs.
     #
     # @return [SQLContext]
-    def register_many(frames, **named_frames)
+    def register_many(frames = nil, **named_frames)
       frames = (frames || {}).to_h
       frames = frames.merge(named_frames)
       frames.each do |name, frame|

data/lib/polars/string_cache.rb

@@ -1,30 +1,11 @@
 module Polars
-  # Class for enabling and disabling the global string cache.
+  # Does nothing.
   #
-  # @example Construct two Series using the same global string cache.
-  #   s1 = nil
-  #   s2 = nil
-  #   Polars::StringCache.new do
-  #     s1 = Polars::Series.new("color", ["red", "green", "red"], dtype: Polars::Categorical)
-  #     s2 = Polars::Series.new("color", ["blue", "red", "green"], dtype: Polars::Categorical)
-  #   end
-  #
-  # @example As both Series are constructed under the same global string cache, they can be concatenated.
-  #   Polars.concat([s1, s2])
-  #   # =>
-  #   # shape: (6,)
-  #   # Series: 'color' [cat]
-  #   # [
-  #   #         "red"
-  #   #         "green"
-  #   #         "red"
-  #   #         "blue"
-  #   #         "red"
-  #   #         "green"
-  #   # ]
+  # @deprecated
+  #   The string cache was used to maintain the mapping for the Categorical
+  #   dtype, this is now done through `Polars::Categories`.
   class StringCache
     def initialize(&block)
-      RbStringCacheHolder.hold(&block)
     end
   end
 
@@ -33,69 +14,35 @@ module Polars
   end
 
   module Functions
-    # Enable the global string cache.
+    # Does nothing.
     #
-    # `Categorical` columns created under the same global string cache have
-    # the same underlying physical value when string values are equal. This allows the
-    # columns to be concatenated or used in a join operation, for example.
+    # @deprecated
+    #   The string cache was used to maintain the mapping for the Categorical
+    #   dtype, this is now done through `Polars::Categories`.
     #
     # @return [nil]
-    #
-    # @example Construct two Series using the same global string cache.
-    #   Polars.enable_string_cache
-    #   s1 = Polars::Series.new("color", ["red", "green", "red"], dtype: Polars::Categorical)
-    #   s2 = Polars::Series.new("color", ["blue", "red", "green"], dtype: Polars::Categorical)
-    #   Polars.disable_string_cache
-    #
-    # @example As both Series are constructed under the same global string cache, they can be concatenated.
-    #   Polars.concat([s1, s2])
-    #   # =>
-    #   # shape: (6,)
-    #   # Series: 'color' [cat]
-    #   # [
-    #   #         "red"
-    #   #         "green"
-    #   #         "red"
-    #   #         "blue"
-    #   #         "red"
-    #   #         "green"
-    #   # ]
     def enable_string_cache
-      Plr.enable_string_cache
     end
 
-    # Disable and clear the global string cache.
+    # Does nothing.
     #
-    # @return [nil]
+    # @deprecated
+    #     The string cache was used to maintain the mapping for the Categorical
+    #     dtype, this is now done through `Polars::Categories`.
     #
-    # @example Construct two Series using the same global string cache.
-    #   Polars.enable_string_cache
-    #   s1 = Polars::Series.new("color", ["red", "green", "red"], dtype: Polars::Categorical)
-    #   s2 = Polars::Series.new("color", ["blue", "red", "green"], dtype: Polars::Categorical)
-    #   Polars.disable_string_cache
-    #
-    # @example As both Series are constructed under the same global string cache, they can be concatenated.
-    #   Polars.concat([s1, s2])
-    #   # =>
-    #   # shape: (6,)
-    #   # Series: 'color' [cat]
-    #   # [
-    #   #         "red"
-    #   #         "green"
-    #   #         "red"
-    #   #         "blue"
-    #   #         "red"
-    #   #         "green"
-    #   # ]
+    # @return [nil]
     def disable_string_cache
-      Plr.disable_string_cache
     end
 
-    # Check whether the global string cache is enabled.
+    # Always returns true.
+    #
+    # @deprecated
+    #   The string cache was used to maintain the mapping for the Categorical
+    #   dtype, this is now done through `Polars::Categories`.
     #
     # @return [Boolean]
     def using_string_cache
-      Plr.using_string_cache
+      true
     end
   end
 end