string_metric 0.1.3 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b2a2bb553d840612ee0f9bb1eb36f561e55e5d50
-  data.tar.gz: be1ba43ed05865948bf84e51b6eb9b85fabc170f
+  metadata.gz: 821e68238da86612c3b0f498d66186195ccf2c19
+  data.tar.gz: 895062d2498b783460c56efea880f4349cfd8a82
 SHA512:
-  metadata.gz: f449e8a3cd1303c9ee92bd71fe57bde264b03c201c85bfbe538353da26fbc3c91ad2d5394691535e6d2e59a025f391a5408b479182ba8fca2683aeada3aa2ff5
-  data.tar.gz: 52da6e7fd9076dd34c6b2b6325fdbea4e61a0dc37c38594ac76d3700adb61f10de960648a9b23dc8a9d730f5ea6bbcbab244a260371b3202be7fde2e680b4149
+  metadata.gz: d3ea26bc39ffe8311523d43fb07cb86f713374ad152cd170b7257edecebff740bf5e1756f0af56d1bc313aaa30cf8cd17725aadb6e5316be35d06c83731840da
+  data.tar.gz: 4e9d498807aac322012033b4782b2401e24399c7ba78e52b10cfdee4461fa9ba64b3e44aea38118f340a1f4f7af834f080c3cdbbfc5c86a297fcf8de9afac0b3

data/README.md

@@ -48,7 +48,7 @@ __Options__
   penalty. Can be `Fixum` or `Float`.
 
 * `:strategy`: The desired strategy for Levenshtein distance. Supported
-  strategies are `:recursive`, `:two_matrix_rows`, `:two_matrix_rows_v2`,
+  strategies are `:recursive`, `:two_matrix_rows`, `:two_matrix_rows_v2`, `:two_matrix_rows_ext`,
   `:full_matrix` and `:experiment`. The default strategy is
   `:two_matrix_rows_v2` for MRI and `:two_matrix_rows` for other platforms
   One should not depend on `:experiment` strategy.
@@ -101,11 +101,12 @@ __Levenshtein__
 
 Implementation                                   | User      | Real
 -------------------------------------------------|-----------|-----------
-Levenshtein::IterativeWithFullMatrix             | 2.260000  | 2.265873
-Levenshtein::IterativeWithTwoMatrixRows          | 1.970000  | 1.971205
-Levenshtein::Experiment                          | 1.680000  | 1.684419
-Levenshtein::IterativeWithTwoMatrixRowsOptimized | 1.270000  | 1.269643
-Text::Levenshtein (from gem text)                | 2.180000  | 2.186258
+Levenshtein::IterativeWithFullMatrix             | 2.320000  | 2.343141
+Levenshtein::IterativeWithTwoMatrixRows          | 2.020000  | 2.044638
+Levenshtein::Experiment                          | 1.750000  | 1.779868
+Levenshtein::IterativeWithTwoMatrixRowsOptimized | 1.320000  | 1.343095
+Levenshtein::IterativeWithTwoMatrixRowsExt       | 0.220000  | 0.228965
+Text::Levenshtein (from gem text)                | 2.240000  | 2.308803
 
 _Currently the set of fixtures is very small - ruby 2.1.0 is used_
 

data/Rakefile

@@ -12,7 +12,14 @@ if RUBY_ENGINE == "ruby"
 
   Rake::ExtensionTask.new do |ext|
     ext.name = "trie_radix_tree_ext"
-    ext.ext_dir = "ext/#{dir}"
+    ext.ext_dir = "ext/#{dir}/trie_radix_tree"
+    ext.lib_dir = "lib/#{dir}"
+    ext.gem_spec = spec
+  end
+
+  Rake::ExtensionTask.new do |ext|
+    ext.name = "iterative_with_two_matrix_rows_ext"
+    ext.ext_dir = "ext/#{dir}/iterative_with_two_matrix_rows"
     ext.lib_dir = "lib/#{dir}"
     ext.gem_spec = spec
   end
@@ -24,4 +31,3 @@ if RUBY_ENGINE == "ruby"
 else
   task default: :spec
 end
-

data/ext/string_metric/levenshtein/iterative_with_two_matrix_rows/extconf.rb

@@ -0,0 +1,3 @@
+require 'mkmf'
+
+create_makefile("string_metric/levenshtein/iterative_with_two_matrix_rows_ext")

data/ext/string_metric/levenshtein/iterative_with_two_matrix_rows/iterative_with_two_matrix_rows_ext.c

@@ -0,0 +1,110 @@
+#include <ruby.h>
+
+#define MIN2(a, b) (((a) < (b)) ? (a) : (b))
+#define MIN3(a, b, c) (MIN2(MIN2((a), (b)), (c)))
+
+#define MALLOC_W(ptr, size) do { \
+    ptr = malloc(size);          \
+    if (!ptr)                    \
+        rb_memerror();           \
+} while (0)
+
+static long min(long *array, long len)
+{
+    long i, min = array[0];
+
+    for (i = 1; i < len; i++)
+        if (array[i] < min)
+            min = array[i];
+
+    return min;
+}
+
+VALUE distance_ext(VALUE self, VALUE v_from, VALUE v_to, VALUE v_from_len, VALUE v_to_len,
+                   VALUE v_max_distance, VALUE v_insertion_cost, VALUE v_deletion_cost,
+                   VALUE v_substitution_cost);
+
+void
+Init_iterative_with_two_matrix_rows_ext(void)
+{
+    VALUE StringMetric;
+    VALUE Levenshtein;
+    VALUE IterativeWithTwoMatrixRowsExt;
+
+    StringMetric = rb_define_module("StringMetric");
+    Levenshtein = rb_define_module_under(StringMetric, "Levenshtein");
+
+    IterativeWithTwoMatrixRowsExt =
+        rb_define_class_under(Levenshtein, "IterativeWithTwoMatrixRowsExt", rb_cObject);
+
+    rb_define_singleton_method(IterativeWithTwoMatrixRowsExt, "distance_ext", distance_ext, 8);
+}
+
+VALUE
+distance_ext(VALUE self, VALUE v_from, VALUE v_to, VALUE v_from_len, VALUE v_to_len,
+             VALUE v_max_distance, VALUE v_insertion_cost, VALUE v_deletion_cost,
+             VALUE v_substitution_cost)
+{
+    /* Ruby to C datatype conversions & variables declaration-initialization */
+    long max_distance =      NUM2LONG(v_max_distance);
+    long insertion_cost =    NUM2LONG(v_insertion_cost);
+    long deletion_cost =     NUM2LONG(v_deletion_cost);
+    long substitution_cost = NUM2LONG(v_substitution_cost);
+
+    long from_len = NUM2LONG(v_from_len);
+    long to_len =   NUM2LONG(v_to_len);
+
+    long v0_len = (from_len + 1) * sizeof(long);
+
+    int *from, *to;
+    long *v0, current, i, j, sub_cell, ins_cell, cost, x = 0;
+
+    /* Use malloc for these arrays in order to avoid stack overflow from big input strings */
+    MALLOC_W(from, from_len * sizeof(int));
+    MALLOC_W(to,   to_len   * sizeof(int));
+    MALLOC_W(v0,   v0_len);
+
+    /* Fill ´from´ & ´to´ C arrays with values from the corresponding ruby arrays */
+    for (i = 0; i < from_len; i++)
+        from[i] = NUM2INT(rb_ary_entry(v_from, i));
+
+    for (i = 0; i < to_len; i++)
+        to[i] = NUM2INT(rb_ary_entry(v_to, i));
+
+    /* Beggining of the algorithm */
+    for (i = 0; i <= from_len; i++)
+        v0[i] = i;
+
+    for (i = 0; i < to_len; i++)
+    {
+        current = x = i + 1;
+        sub_cell = v0[0];
+
+        for (j = 0; j < from_len; j++)
+        {
+            cost = (from[j] == to[i]) ? 0 : substitution_cost;
+
+            ins_cell = v0[j + 1];
+
+            x = MIN3(current + deletion_cost, ins_cell + insertion_cost, sub_cell + cost);
+
+            v0[j] = current;
+            current = x;
+            sub_cell = ins_cell;
+        }
+
+        v0[from_len] = x;
+        if (max_distance && min(v0, v0_len) > max_distance)
+            break;
+    }
+
+    /* Clean up allocated memory */
+    free(from);
+    free(to);
+    free(v0);
+
+    if (max_distance && x > max_distance)
+      return LONG2NUM(max_distance);
+    else
+      return LONG2NUM(x);
+}

data/lib/string_metric/levenshtein.rb

@@ -8,6 +8,7 @@ require_relative "levenshtein/recursive"
 require_relative "levenshtein/trie_node"
 require_relative "levenshtein/trie_radix_tree"
 require_relative "levenshtein/trie_radix_tree_ext"
+require_relative "levenshtein/iterative_with_two_matrix_rows_ext" if RUBY_ENGINE == "ruby"
 
 module StringMetric
   # Levenshtein Distance implementation
@@ -23,6 +24,10 @@ module StringMetric
       two_matrix_rows_v2: IterativeWithTwoMatrixRowsOptimized
     }
 
+    if RUBY_ENGINE == "ruby"
+      STRATEGIES[:two_matrix_rows_ext] = IterativeWithTwoMatrixRowsExt
+    end
+
     # Levenshtein Distance of two strings
     #
     # @param from [String] the first string
@@ -53,7 +58,7 @@ module StringMetric
     # Currently the default strategy is set to IterativeWithTwoMatrixRows
     def default_strategy
       if RUBY_ENGINE == "ruby"
-        pick_strategy(:two_matrix_rows_v2)
+        pick_strategy(:two_matrix_rows_ext)
       else
         pick_strategy(:two_matrix_rows)
       end

data/lib/string_metric/levenshtein/iterative_with_two_matrix_rows_ext.rb

@@ -0,0 +1,33 @@
+# coding: utf-8
+
+require_relative "iterative_with_two_matrix_rows_ext.so"
+
+module StringMetric
+  module Levenshtein
+    class IterativeWithTwoMatrixRowsExt
+      def self.distance(from, to, options = {})
+        max_distance      = options[:max_distance]
+        insertion_cost    = options[:insertion_cost]    || 1
+        deletion_cost     = options[:deletion_cost]     || 1
+        substitution_cost = options[:substitution_cost] || 1
+
+        from_len = from.length
+        to_len =   to.length
+
+        if max_distance && (to_len - from_len).abs >= max_distance
+            return max_distance
+        end
+
+        return 0 if from == to
+        return to_len if from_len.zero?
+        return from_len if to_len.zero?
+
+        from = from.codepoints.to_a
+        to = to.codepoints.to_a
+
+        distance_ext(from, to, from_len, to_len, max_distance || 0, insertion_cost,
+                     deletion_cost, substitution_cost)
+      end
+    end
+  end
+end

data/lib/string_metric/levenshtein/trie_radix_tree.rb

@@ -10,7 +10,7 @@ module StringMetric
         @deletion_cost     = options[:deletion_cost]     || 1
         @substitution_cost = options[:substitution_cost] || 1
 
-        results = []
+        results = {}
         word = from.codepoints
         currentRow = (0..word.length).to_a
 
@@ -35,7 +35,7 @@ module StringMetric
         end
 
         if currentRow.last <= @max_distance && !node.word.nil?
-          results << [node.word, currentRow.last]
+          results[node.word] = currentRow.last
         end
 
         if currentRow.min <= @max_distance

data/lib/string_metric/version.rb

@@ -1,3 +1,3 @@
 module StringMetric
-  VERSION = "0.1.3"
+  VERSION = "0.1.4"
 end

data/spec/lib/levenshtein/iterative_with_two_matrix_rows_ext_spec.rb

@@ -0,0 +1,9 @@
+# coding: utf-8
+
+if RUBY_ENGINE == "ruby"
+  require "spec_helper"
+
+  describe StringMetric::Levenshtein::IterativeWithTwoMatrixRowsExt do
+    it_behaves_like "Levenshtein Distance", { strategy: :two_matrix_rows_ext }
+  end
+end

data/spec/support/levenshtein.rb

@@ -34,6 +34,7 @@ shared_examples "Levenshtein Distance" do |options|
           expect(described_class.distance("gambol", "gumbo", max_distance: 1)).to eq 1
           expect(described_class.distance("kitten", "", max_distance: 2)).to eq 2
           expect(described_class.distance("", "kitten", max_distance: 3)).to eq 3
+          expect(described_class.distance("a", "reallylongwordfortest", max_distance: 5)).to eq 5
         end
       end
       context "and normal distance is less than max_distance" do
@@ -51,6 +52,7 @@ shared_examples "Levenshtein Distance" do |options|
           expect(described_class.distance("", "cat", max_distance: 4)).to eq 3
           expect(described_class.distance("cat", "", max_distance: 5)).to eq 3
           expect(described_class.distance("", "", max_distance: 2)).to eq 0
+          expect(described_class.distance("a", "reallylongwordfortest", max_distance: 25)).to eq 20
         end
       end
       context "and normal distance is same as max_distance" do
@@ -66,6 +68,7 @@ shared_examples "Levenshtein Distance" do |options|
           expect(described_class.distance("", "cat", max_distance: 3)).to eq 3
           expect(described_class.distance("cat", "", max_distance: 3)).to eq 3
           expect(described_class.distance("", "", max_distance: 0)).to eq 0
+          expect(described_class.distance("a", "reallylongwordfortest", max_distance: 20)).to eq 20
         end
       end
     end

data/string_metric.gemspec

@@ -25,7 +25,8 @@ Gem::Specification.new do |spec|
 
   if RUBY_ENGINE == "ruby"
     spec.add_development_dependency "rake-compiler", "~> 0.9.2"
-    spec.extensions << "ext/#{spec.name}/levenshtein/extconf.rb"
+    spec.extensions << "ext/#{spec.name}/levenshtein/trie_radix_tree/extconf.rb"
+    spec.extensions << "ext/#{spec.name}/levenshtein/iterative_with_two_matrix_rows/extconf.rb"
 
     if RUBY_VERSION > "1.9.3"
       spec.add_development_dependency "pry-byebug", "~> 1.2.1"

metadata

@@ -1,97 +1,97 @@
 --- !ruby/object:Gem::Specification
 name: string_metric
 version: !ruby/object:Gem::Version
-  version: 0.1.3
+  version: 0.1.4
 platform: ruby
 authors:
 - Giorgos Tsiftsis
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-05-23 00:00:00.000000000 Z
+date: 2015-05-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.5'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.5'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 10.1.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 10.1.1
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 2.14.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 2.14.1
 - !ruby/object:Gem::Dependency
   name: text
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.2.3
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.2.3
 - !ruby/object:Gem::Dependency
   name: rake-compiler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 0.9.2
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 0.9.2
 - !ruby/object:Gem::Dependency
   name: pry-byebug
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.2.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: 1.2.1
 description: A simple library with String Metric algorithms
@@ -99,25 +99,29 @@ email:
 - giorgos.tsiftsis@skroutz.gr
 executables: []
 extensions:
-- ext/string_metric/levenshtein/extconf.rb
+- ext/string_metric/levenshtein/trie_radix_tree/extconf.rb
+- ext/string_metric/levenshtein/iterative_with_two_matrix_rows/extconf.rb
 extra_rdoc_files: []
 files:
-- ".gitignore"
-- ".rspec"
-- ".travis.yml"
+- .gitignore
+- .rspec
+- .travis.yml
 - Gemfile
 - LICENSE.txt
 - README.md
 - Rakefile
 - benchmarks/dictionary.rb
 - benchmarks/levenshtein.rb
-- ext/string_metric/levenshtein/extconf.rb
-- ext/string_metric/levenshtein/trie_radix_tree_ext.c
+- ext/string_metric/levenshtein/iterative_with_two_matrix_rows/extconf.rb
+- ext/string_metric/levenshtein/iterative_with_two_matrix_rows/iterative_with_two_matrix_rows_ext.c
+- ext/string_metric/levenshtein/trie_radix_tree/extconf.rb
+- ext/string_metric/levenshtein/trie_radix_tree/trie_radix_tree_ext.c
 - lib/string_metric.rb
 - lib/string_metric/levenshtein.rb
 - lib/string_metric/levenshtein/experiment.rb
 - lib/string_metric/levenshtein/iterative_with_full_matrix.rb
 - lib/string_metric/levenshtein/iterative_with_two_matrix_rows.rb
+- lib/string_metric/levenshtein/iterative_with_two_matrix_rows_ext.rb
 - lib/string_metric/levenshtein/iterative_with_two_matrix_rows_optimized.rb
 - lib/string_metric/levenshtein/recursive.rb
 - lib/string_metric/levenshtein/trie_node.rb
@@ -128,6 +132,7 @@ files:
 - spec/fixtures/levenshtein.csv
 - spec/lib/levenshtein/experiment_spec.rb
 - spec/lib/levenshtein/iterative_with_full_matric_spec.rb
+- spec/lib/levenshtein/iterative_with_two_matrix_rows_ext_spec.rb
 - spec/lib/levenshtein/iterative_with_two_matrix_rows_optimized_spec.rb
 - spec/lib/levenshtein/iterative_with_two_matrix_rows_spec.rb
 - spec/lib/levenshtein/recursive_spec.rb
@@ -145,17 +150,17 @@ require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.6
+rubygems_version: 2.1.11
 signing_key: 
 specification_version: 4
 summary: A simple library with String Metric algorithms
@@ -164,6 +169,7 @@ test_files:
 - spec/fixtures/levenshtein.csv
 - spec/lib/levenshtein/experiment_spec.rb
 - spec/lib/levenshtein/iterative_with_full_matric_spec.rb
+- spec/lib/levenshtein/iterative_with_two_matrix_rows_ext_spec.rb
 - spec/lib/levenshtein/iterative_with_two_matrix_rows_optimized_spec.rb
 - spec/lib/levenshtein/iterative_with_two_matrix_rows_spec.rb
 - spec/lib/levenshtein/recursive_spec.rb