inital

Author: mlobl

.gitignore

@@ -0,0 +1,12 @@
+/doc/
+/lib/
+/bin/
+/.shards/
+.idea/
+*.dwarf
+
+# Libraries don't need dependency lock
+# Dependencies will be locked in application that uses them
+/shard.lock
+
+*.editorconfig

.travis.yml

@@ -0,0 +1 @@
+language: crystal

LICENSE

@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2018 Meyer Lobl
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.

README.md

@@ -0,0 +1,40 @@
+# string-metrics
+
+Ports some popular string algorithms for Crystal:
+* [Levenshtein distance](https://en.wikipedia.org/wiki/Levenshtein_distance)
+* [Hamming distance](https://en.wikipedia.org/wiki/Hamming_distance)
+* [Damerau–Levenshtein distance](https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_distance)
+
+## Installation
+
+Add this to your application's `shard.yml`:
+
+```yaml
+dependencies:
+  string-metrics:
+    github: mlobl/string-metrics
+```
+
+## Usage
+
+```crystal
+require "string-metrics"
+```
+
+TODO: Write usage instructions here
+
+## Development
+
+TODO: Write development instructions here
+
+## Contributing
+
+1. Fork it ( https://github.com/mlobl/string-metrics/fork )
+2. Create your feature branch (git checkout -b my-new-feature)
+3. Commit your changes (git commit -am 'Add some feature')
+4. Push to the branch (git push origin my-new-feature)
+5. Create a new Pull Request
+
+## Contributors
+
+- [mlobl](https://github.com/mlobl) Meyer Lobl - creator, maintainer

benchmark_profiling/levenshtein.cr

@@ -0,0 +1,9 @@
+require "../src/string-metrics"
+
+# A stress test to profile with
+a = "n information theory, linguistics and computer science, the Levenshtein distance is a string metric for measuring the difference between two sequences. Informally, the Levenshtein distance between two words is the minimum number of single-character edits (insertions, deletions or substitutions) required to change one word into the other. It is named after the Soviet mathematician Vladimir Levenshtein, who considered this distance in 1965"
+b = "The Levenshtein distance can also be computed between two longer strings, but the cost to compute it, which is roughly proportional to the product of the two string lengths, makes this impractical. Thus, when used to aid in fuzzy string searching in applications such as record linkage, the compared strings are usually short to help improve speed of comparisons."
+start_time = Time.now()
+result = StringMetrics.levenshtein(a*50, b*50)
+timing = (Time.now() - start_time)
+puts "Distance was #{result} and took #{timing}"

shard.yml

@@ -0,0 +1,9 @@
+name: string-metrics
+version: 0.1.0
+
+authors:
+  - Meyer Lobl <[email protected]>
+
+crystal: 0.24.1
+
+license: MIT

spec/spec_helper.cr

@@ -0,0 +1,2 @@
+require "spec"
+require "../src/string-metrics"

spec/string-metrics_spec.cr

@@ -0,0 +1,60 @@
+require "./spec_helper"
+
+describe "StringMetrics" do
+  it "levenshtein both empty" do
+    StringMetrics.levenshtein("", "").should eq(0)
+  end
+
+  it "levenshtein one empty" do
+    StringMetrics.levenshtein("", "hi").should eq(2)
+  end
+
+  it "levenshtein one missing" do
+    StringMetrics.levenshtein("h", "hi").should eq(1)
+  end
+
+  it "levenshtein one sub" do
+    StringMetrics.levenshtein("ha", "hi").should eq(1)
+  end
+
+  it "levenshtein one deletion" do
+    StringMetrics.levenshtein("hai", "hi").should eq(1)
+  end
+
+  it "levenshtein multiple insertions" do
+    StringMetrics.levenshtein("char", "charity").should eq(3)
+  end
+
+  it "levenshtein multiple insertions and one substitution" do
+    StringMetrics.levenshtein("char", "sharity").should eq(4)
+  end
+
+  it "levenshtein with many operations" do
+    StringMetrics.levenshtein("example", "samples").should eq(3)
+    StringMetrics.levenshtein("sturgeon", "urgently").should eq(6)
+    StringMetrics.levenshtein("levenshtein", "frankenstein").should eq(6)
+    StringMetrics.levenshtein("distance", "difference").should eq(5)
+  end
+
+  it "hamming with differing length strings" do
+    StringMetrics.hamming("hi", "ha").should eq(1)
+  end
+
+  it "hamming with empty strings" do
+    StringMetrics.hamming("", "").should eq(0)
+  end
+
+  it "hamming with differing length strings" do
+    expect_raises ArgumentError, do
+      StringMetrics.hamming("", "check")
+    end
+  end
+
+  it "damerau levenshtein" do
+    StringMetrics.damerau_levenshtein("char", "hcra").should eq(2)
+  end
+
+  it "empty strings damerau" do
+    StringMetrics.damerau_levenshtein("", "").should eq(0)
+  end
+end

src/string-metrics.cr

@@ -0,0 +1,91 @@
+require "./string-metrics/*"
+
+# TODO: Write documentation for `String::Metrics`
+module StringMetrics
+
+  # Gets the min edit distance between two strings.
+  # See https://en.wikipedia.org/wiki/Levenshtein_distance
+  # Ported from https://en.wikibooks.org/wiki/Algorithm_Implementation/Strings/Levenshtein_distance#Python
+  def self.levenshtein(s1 : String, s2 : String) : Int
+    return levenshtein(s2, s1) if s1.size < s2.size
+    return s1.size if s2.size == 0
+
+    previous_row = (0..s2.size).to_a
+    s1_chars = s1.chars
+    s2_chars = s2.chars
+    # it's expensive to continually recreate an array
+    cache = [[0] * (s2.size + 1), [0] * (s2.size + 1)]
+    s1_chars.each_with_index do |c1, i|
+      current_row = cache[i % 2]
+      current_row[0] = i + 1
+      s2_chars.each_with_index do |c2, j|
+        insertions = previous_row.unsafe_at(j + 1) + 1
+        deletions = current_row.unsafe_at(j) + 1
+        substitutions = previous_row.unsafe_at(j)
+        substitutions +=  1 if c1 != c2
+        # could have minned a tuple, but this is a bit faster
+        min = insertions
+        min = deletions if deletions < min
+        min = substitutions if substitutions < min
+        # main bottleneck is here, memory writes to an already allocated address :/
+        current_row[j + 1] = min
+      end
+      previous_row = current_row
+    end
+    previous_row.last
+  end
+
+  def self.hamming(s1 : String, s2 : String) : Int
+    raise ArgumentError.new("input lengths are not equal") if s1.size != s2.size
+    (0...s2.size).sum { |i| (s1[i] != s2[i])? 1 : 0 }
+  end
+
+  # A variation of the Levenshtein distance, this counts transpositions as a single edit.
+  # ```damerau_levenshtein("char", "hcar") == 1``` as opposed to a distance of 2 from levenshtein
+  # on it's own
+  # Ported from https://github.com/jamesturk/jellyfish/blob/master/jellyfish/_jellyfish.py
+  def self.damerau_levenshtein(s1 : String, s2 : String) : Int
+    infinite = s1.size + s2.size
+
+    da = Hash(Char, Int32).new(default_value=0)
+
+    # distance matrix
+    score = (0...s1.size + 2).to_a.map {|i| [0]*(s2.size + 2)}
+    score[0][0] = infinite
+
+    (0...s1.size + 1).each do |i|
+       score[i + 1][0] = infinite
+       score[i + 1][1] = i
+    end
+    (0...s2.size + 1).each do |i|
+      score[0][i + 1] = infinite
+      score[1][i + 1] = i
+    end
+
+    s1_chars = s1.chars
+    s2_chars = s2.chars
+
+    (1..s1.size).each do |i|
+      db = 0
+      (1..s2.size).each do |j|
+        i1 = da[s2_chars[j-1]]
+        j1 = db
+        cost = 1
+        if s1_chars[i - 1] == s2_chars[j - 1]
+          cost = 0
+          db = j
+        end
+
+        score[i + 1][j + 1] = {
+          score[i][j] + cost,
+          score[i+1][j] + 1,
+          score[i][j+1] + 1,
+          score[i1][j1] + (i - i1 - 1) + 1 + (j - j1 - 1)
+        }.min
+      end
+      da[s1_chars[i - 1]] = i
+    end
+    score[s1.size + 1][s2.size + 1]
+  end
+
+end

src/string-metrics/version.cr

@@ -0,0 +1,3 @@
+module StringMetrics
+  VERSION = "0.1.0"
+end