daru_lite 0.3.0 → 0.3.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2ecfb59a965f8f05a20a331298549d00db4116e8a8f897dc5fdeb4c4b18dc98e
-  data.tar.gz: fefeb73d29528353d75c291817932146ff283c3160182f755e275e1561882ee1
+  metadata.gz: f72d838a7ae6b15ca8d5d4d2942f733ee731815a432dec1b76e51296cd7d5fe7
+  data.tar.gz: 5653ee135b2fcf671eb28ad8af54cb11e1d8d6187bdfdef284d425b23d0da290
 SHA512:
-  metadata.gz: 668c346ce88a77d215a322cd09adf8e9abe07db7d14233800e154119015d059fab6f64c5b6daccdf15844d61fbdde27d6ea99594aea4f683edba328222bedd58
-  data.tar.gz: 1c6d4e68271787f6abc543806e6409cdbcd139dbcab69d2dc55f2f4c4b7a18a389e0b1ad813ec958670c48322da35903c306f8156ab9f08c9b22e03cae9210f1
+  metadata.gz: b3780b06dcb392bd866af2424ea0a19ddf263cb125b13092c3d6674ab8d6d1c576bbf6b78e6bfb794cc21aac73ac1f1af4faacb0905196c6cfb5a5190be56455
+  data.tar.gz: 4e39865f597bf9ecb63270c9f6df963ad4e8bdf6fecda6de0a305fb81b5c81b58787e28267a7e84644194a4bd63442e98fb56f642c03ef327e08b888210637ca

data/lib/daru_lite/dataframe.rb

@@ -89,6 +89,7 @@ module DaruLite
         data = Hash.new do |h, col|
           h[col] = row_index.map { |r| [r, nil] }.to_h
         end
+        validate_no_duplicate_pairs(rows, columns)
         columns.zip(rows, values).each { |c, r, v| data[c][r] = v }
 
         # FIXME: in fact, WITHOUT this line you'll obtain more "right"
@@ -101,6 +102,13 @@ module DaruLite
 
       private
 
+      def validate_no_duplicate_pairs(rows, columns)
+        seen = Set.new
+        columns.to_a.zip(rows.to_a).each do |pair|
+          raise IndexError, "Duplicate (column, row) pair: #{pair.inspect}" unless seen.add?(pair)
+        end
+      end
+
       def guess_order(source)
         case source.first
         when Vector # assume that all are Vectors

data/lib/daru_lite/index/categorical_index.rb

@@ -65,6 +65,22 @@ module DaruLite
       positions.size == 1 ? positions.first : positions.sort
     end
 
+    # Returns the position(s) of the given category/categories or position(s).
+    # Mirrors Index#[] but resolves against the categorical structure
+    # (@cat_hash / @array) instead of @relation_hash, which CategoricalIndex
+    # never populates.
+    # @param keys [Array<object>] categories or positions to look up
+    # @return [Integer, Array<Integer>, nil] position(s), or nil if absent
+    # @example
+    #   idx = DaruLite::CategoricalIndex.new [:a, :b, :a]
+    #   idx[:b]   # => 1
+    #   idx[:z]   # => nil
+    def [](*keys)
+      pos(*keys)
+    rescue IndexError
+      nil
+    end
+
     # Returns index value from position
     # @param pos [Integer] the position to look for
     # @return [object] category corresponding to position

data/lib/daru_lite/version.rb

@@ -1,3 +1,3 @@
 module DaruLite
-  VERSION = '0.3.0'.freeze
+  VERSION = '0.3.2'.freeze
 end

data/spec/dataframe_spec.rb

@@ -886,6 +886,23 @@ describe DaruLite::DataFrame do
       expect(df).to eq(DaruLite::DataFrame.new({b: [11,12,14,15], a: [1,2,4,5],
       c: [11,22,44,55]}, order: [:a, :b, :c], index: [:one, :two, :four, :five]))
     end
+
+    context "when the data frame has a CategoricalIndex" do
+      let(:categorical_df) do
+        frame = DaruLite::DataFrame.new(
+          { a: [1, 2, 3], b: [11, 22, 33] },
+          order: [:a, :b],
+          index: [:base, :one, :two]
+        )
+        frame.index = DaruLite::CategoricalIndex.new([:base, :one, :two])
+        frame
+      end
+
+      it "deletes the specified row without raising" do
+        expect { categorical_df.delete_row :base }.not_to raise_error
+        expect(categorical_df.index.to_a).to eq [:one, :two]
+      end
+    end
   end
 
   context "#rename_vectors!" do
@@ -1087,6 +1104,16 @@ describe DaruLite::DataFrame do
       expect(df['x'].to_a).to eq([1, 2])
       expect(df['y'].to_a).to eq([3, 4])
     end
+
+    it 'raises IndexError when given duplicate (row, column) pairs' do
+      rows    = DaruLite::Vector.new(%w[a a b])
+      columns = DaruLite::Vector.new(%w[x x y])
+      values  = DaruLite::Vector.new([1, 2, 3])
+
+      expect {
+        described_class.crosstab_by_assignation(rows, columns, values)
+      }.to raise_error(IndexError, /Duplicate/)
+    end
   end
 
   context '#inspect' do

data/spec/index/categorical_index_spec.rb

@@ -68,6 +68,38 @@ describe DaruLite::CategoricalIndex do
     end
   end
 
+  describe "#[]" do
+    context "when the category occurs once" do
+      subject { index[:b] }
+
+      it { is_expected.to eq 1 }
+    end
+
+    context "when the category occurs multiple times" do
+      subject { index[:a] }
+
+      it { is_expected.to eq [0, 2, 3] }
+    end
+
+    context "when given a positional index" do
+      subject { index[0] }
+
+      it { is_expected.to eq 0 }
+    end
+
+    context "when given multiple categories" do
+      subject { index[:a, :c] }
+
+      it { is_expected.to eq [0, 2, 3, 4] }
+    end
+
+    context "when the category is absent" do
+      subject { index[:z] }
+
+      it { is_expected.to be_nil }
+    end
+  end
+
   context "#subset" do
     let(:idx) { described_class.new [:a, 1, :a, 1, :c] }
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: daru_lite
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.3.2
 platform: ruby
 authors:
 - Thomas Naude-Filonnière
@@ -9,10 +9,9 @@ authors:
 - Julie Thomas
 - Amar Slaoua
 - Mourtada Belhantri
-autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-04-14 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activerecord
@@ -317,7 +316,6 @@ description: |
   and can be used with many others like mixed_models, gnuplotrb and iruby.
 
   Daru Lite is a fork of Daru that aims to focus on data manipulation and stability.
-email:
 executables: []
 extensions: []
 extra_rdoc_files: []
@@ -538,7 +536,6 @@ homepage: https://github.com/pollandroll/daru
 licenses:
 - BSD-2-Clause
 metadata: {}
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -553,8 +550,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.10
-signing_key:
+rubygems_version: 4.0.9
 specification_version: 4
 summary: Data Analysis in RUby, stripped down
 test_files: