diff options
Diffstat (limited to 'core/src')
75 files changed, 4114 insertions, 0 deletions
diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/CaliperBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/CaliperBenchmark.scala new file mode 100755 index 0000000..55a6238 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/CaliperBenchmark.scala @@ -0,0 +1,7 @@ +package com.rockymadden.stringmetric + +import com.google.caliper.SimpleBenchmark + +trait CaliperBenchmark extends SimpleBenchmark { + def run(reps: Int)(code: => Unit) = (0 until reps).foreach(i => code) +}
\ No newline at end of file diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/CaliperRunner.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/CaliperRunner.scala new file mode 100755 index 0000000..4474a8d --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/CaliperRunner.scala @@ -0,0 +1,7 @@ +package com.rockymadden.stringmetric + +import com.google.caliper.{Benchmark, Runner} + +abstract class CaliperRunner(private[this] val suite: java.lang.Class[_ <: Benchmark]) { + def main(args: Array[String]): Unit = Runner.main(suite, args) +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithmBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithmBenchmark.scala new file mode 100755 index 0000000..fe5c80d --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithmBenchmark.scala @@ -0,0 +1,26 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class MetaphoneAlgorithmBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string: String = _ + var charArray: Array[Char] = _ + + override protected def setUp() { + string = Random.alphanumeric.filter(_ > '9').take(length).mkString + charArray = string.toCharArray + } + + def timeComputeWithCharArray(reps: Int) = run(reps) { + MetaphoneAlgorithm.compute(charArray) + } + + def timeComputeWithString(reps: Int) = run(reps) { + MetaphoneAlgorithm.compute(string) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetricBenchmark.scala new file mode 100755 index 0000000..c9b59fc --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetricBenchmark.scala @@ -0,0 +1,49 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.annotation.tailrec +import scala.util.Random + +final class MetaphoneMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.filter(_ > '9').take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + MetaphoneMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + MetaphoneMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + MetaphoneMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + MetaphoneMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithmBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithmBenchmark.scala new file mode 100755 index 0000000..ac04cd7 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithmBenchmark.scala @@ -0,0 +1,26 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class NysiisAlgorithmBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string: String = _ + var charArray: Array[Char] = _ + + override protected def setUp() { + string = Random.alphanumeric.filter(_ > '9').take(length).mkString + charArray = string.toCharArray + } + + def timeComputeWithCharArray(reps: Int) = run(reps) { + NysiisAlgorithm.compute(charArray) + } + + def timeComputeWithString(reps: Int) = run(reps) { + NysiisAlgorithm.compute(string) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/NysiisMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/NysiisMetricBenchmark.scala new file mode 100755 index 0000000..f975d29 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/NysiisMetricBenchmark.scala @@ -0,0 +1,49 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.annotation.tailrec +import scala.util.Random + +final class NysiisMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.filter(_ > '9').take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + NysiisMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + NysiisMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + NysiisMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + NysiisMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithmBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithmBenchmark.scala new file mode 100755 index 0000000..4141b37 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithmBenchmark.scala @@ -0,0 +1,26 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class RefinedNysiisAlgorithmBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string: String = _ + var charArray: Array[Char] = _ + + override protected def setUp() { + string = Random.alphanumeric.filter(_ > '9').take(length).mkString + charArray = string.toCharArray + } + + def timeComputeWithCharArray(reps: Int) = run(reps) { + RefinedNysiisAlgorithm.compute(charArray) + } + + def timeComputeWithString(reps: Int) = run(reps) { + RefinedNysiisAlgorithm.compute(string) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetricBenchmark.scala new file mode 100755 index 0000000..d927f18 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetricBenchmark.scala @@ -0,0 +1,49 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.annotation.tailrec +import scala.util.Random + +final class RefinedNysiisMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.filter(_ > '9').take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + RefinedNysiisMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + RefinedNysiisMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + RefinedNysiisMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + RefinedNysiisMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithmBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithmBenchmark.scala new file mode 100755 index 0000000..ec8d53c --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithmBenchmark.scala @@ -0,0 +1,26 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class RefinedSoundexAlgorithmBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string: String = _ + var charArray: Array[Char] = _ + + override protected def setUp() { + string = Random.alphanumeric.filter(_ > '9').take(length).mkString + charArray = string.toCharArray + } + + def timeComputeWithCharArray(reps: Int) = run(reps) { + RefinedSoundexAlgorithm.compute(charArray) + } + + def timeComputeWithString(reps: Int) = run(reps) { + RefinedSoundexAlgorithm.compute(string) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetricBenchmark.scala new file mode 100755 index 0000000..6c52e8d --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetricBenchmark.scala @@ -0,0 +1,49 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.annotation.tailrec +import scala.util.Random + +final class RefinedSoundexMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.filter(_ > '9').take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + RefinedSoundexMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + RefinedSoundexMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + RefinedSoundexMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + RefinedSoundexMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithmBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithmBenchmark.scala new file mode 100755 index 0000000..9dc67de --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithmBenchmark.scala @@ -0,0 +1,26 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class SoundexAlgorithmBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string: String = _ + var charArray: Array[Char] = _ + + override protected def setUp() { + string = Random.alphanumeric.filter(_ > '9').take(length).mkString + charArray = string.toCharArray + } + + def timeComputeWithCharArray(reps: Int) = run(reps) { + SoundexAlgorithm.compute(charArray) + } + + def timeComputeWithString(reps: Int) = run(reps) { + SoundexAlgorithm.compute(string) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/SoundexMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/SoundexMetricBenchmark.scala new file mode 100755 index 0000000..7707019 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/phonetic/SoundexMetricBenchmark.scala @@ -0,0 +1,49 @@ +package com.rockymadden.stringmetric.phonetic + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.annotation.tailrec +import scala.util.Random + +final class SoundexMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.filter(_ > '9').take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + SoundexMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + SoundexMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + SoundexMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + SoundexMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricBenchmark.scala new file mode 100755 index 0000000..5df27fb --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class DiceSorensenMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + DiceSorensenMetric(2).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/HammingMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/HammingMetricBenchmark.scala new file mode 100755 index 0000000..3d04074 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/HammingMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class HammingMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + HammingMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + HammingMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + HammingMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + HammingMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaccardMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaccardMetricBenchmark.scala new file mode 100755 index 0000000..ddeef7a --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaccardMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class JaccardMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + JaccardMetric(2).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + JaccardMetric(2).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + JaccardMetric(2).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + JaccardMetric(2).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroMetricBenchmark.scala new file mode 100755 index 0000000..47baa6d --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class JaroMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + JaroMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + JaroMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + JaroMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + JaroMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricBenchmark.scala new file mode 100755 index 0000000..c11a01c --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class JaroWinklerMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + JaroWinklerMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + JaroWinklerMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + JaroWinklerMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + JaroWinklerMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricBenchmark.scala new file mode 100755 index 0000000..a8460e2 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class LevenshteinMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + LevenshteinMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + LevenshteinMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + LevenshteinMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + LevenshteinMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/NGramMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/NGramMetricBenchmark.scala new file mode 100755 index 0000000..8ba1bb0 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/NGramMetricBenchmark.scala @@ -0,0 +1,51 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class NGramMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + @Param(Array("2", "3")) + var n: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + NGramMetric(n).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + NGramMetric(n).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + NGramMetric(n).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + NGramMetric(n).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/OverlapMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/OverlapMetricBenchmark.scala new file mode 100755 index 0000000..89207f2 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/OverlapMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class OverlapMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + OverlapMetric(2).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + OverlapMetric(2).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + OverlapMetric(2).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + OverlapMetric(2).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricBenchmark.scala new file mode 100755 index 0000000..86196ab --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class RatcliffObershelpMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + RatcliffObershelpMetric.compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricBenchmark.scala new file mode 100755 index 0000000..837ce01 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricBenchmark.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class WeightedLevenshteinMetricBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + var string1: String = _ + var charArray1: Array[Char] = _ + var string2: String = _ + var charArray2: Array[Char] = _ + + override protected def setUp() { + @annotation.tailrec + def random(l: Int, ps: String = null): String = + if (l == 0) "" + else { + val s = Random.alphanumeric.take(l).mkString + + if (ps == null || s != ps) s + else random(l, ps) + } + + string1 = random(length) + string2 = random(length, string1) + charArray1 = string1.toCharArray + charArray2 = string2.toCharArray + } + + def timeCompareWithDifferentCharArrays(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(charArray1, charArray2) + } + + def timeCompareWithDifferentStrings(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(string1, string2) + } + + def timeCompareWithIdenticalCharArrays(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(charArray1, charArray1) + } + + def timeCompareWithIdenticalStrings(reps: Int) = run(reps) { + WeightedLevenshteinMetric(1, 1, 1).compare(string1, string1) + } +} diff --git a/core/src/benchmark/scala/com/rockymadden/stringmetric/tokenize/NGramTokenizerBenchmark.scala b/core/src/benchmark/scala/com/rockymadden/stringmetric/tokenize/NGramTokenizerBenchmark.scala new file mode 100755 index 0000000..d66bf88 --- /dev/null +++ b/core/src/benchmark/scala/com/rockymadden/stringmetric/tokenize/NGramTokenizerBenchmark.scala @@ -0,0 +1,29 @@ +package com.rockymadden.stringmetric.tokenize + +import com.google.caliper.Param +import com.rockymadden.stringmetric.{CaliperBenchmark, CaliperRunner} +import scala.util.Random + +final class NGramTokenizerBenchmark extends CaliperBenchmark { + @Param(Array("0", "1", "2", "4", "8", "16")) + var length: Int = _ + + @Param(Array("2", "3")) + var n: Int = _ + + var string: String = _ + var charArray: Array[Char] = _ + + override protected def setUp() { + string = Random.alphanumeric.take(length).mkString + charArray = string.toCharArray + } + + def timeComputeWithCharArray(reps: Int) = run(reps) { + NGramTokenizer(n).tokenize(charArray) + } + + def timeComputeWithString(reps: Int) = run(reps) { + NGramTokenizer(n).tokenize(string) + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/Algorithm.scala b/core/src/main/scala/com/rockymadden/stringmetric/Algorithm.scala new file mode 100755 index 0000000..84f136d --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/Algorithm.scala @@ -0,0 +1,46 @@ +package com.rockymadden.stringmetric + +object Algorithm { + import Transform.StringTransform + + + trait Algorithm[A] { + def compute(a: A): Option[A] + } + + + trait StringAlgorithm extends Algorithm[Array[Char]] { + def compute(a: String): Option[String] + } + + + object StringAlgorithm { + final val Metaphone = phonetic.MetaphoneAlgorithm + final val Nysiis = phonetic.NysiisAlgorithm + final val RefinedNysiis = phonetic.RefinedNysiisAlgorithm + final val RefinedSoundex = phonetic.RefinedSoundexAlgorithm + final val Soundex = phonetic.SoundexAlgorithm + + def computeWithMetaphone(a: Array[Char]) = Metaphone.compute(a) + + def computeWithNysiis(a: Array[Char]) = Nysiis.compute(a) + + def computeWithRefinedNysiis(a: Array[Char]) = RefinedNysiis.compute(a) + + def computeWithRefinedSoundex(a: Array[Char]) = RefinedSoundex.compute(a) + + def computeWithSoundex(a: Array[Char]) = Soundex.compute(a) + } + + + final class StringAlgorithmDecorator(val sa: StringAlgorithm) { + val withTransform: (StringTransform => StringAlgorithm) = (st) => new StringAlgorithm { + private[this] val self: StringAlgorithm = sa + private[this] val transform: StringTransform = st + + override def compute(a: Array[Char]): Option[Array[Char]] = self.compute(transform(a)) + + override def compute(a: String): Option[String] = self.compute(transform(a.toCharArray)).map(_.mkString) + } + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/Alphabet.scala b/core/src/main/scala/com/rockymadden/stringmetric/Alphabet.scala new file mode 100755 index 0000000..5e666d2 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/Alphabet.scala @@ -0,0 +1,42 @@ +package com.rockymadden.stringmetric + +import scala.collection.immutable.Set + +object Alphabet { + sealed abstract class AlphabetSet(val chars: Set[Char]) { + def isSuperset(a: Char): Boolean = chars.contains(a) + + def isSuperset(a: Array[Char]): Boolean = a.length > 0 && a.takeWhile(chars.contains).length == a.length + + def isSuperset(a: String): Boolean = isSuperset(a.toCharArray) + } + + + case object LowercaseConsonant extends AlphabetSet( + Set('b', 'c', 'd', 'f', 'g', 'h', 'j', 'k', 'l', 'm', 'n', 'p', 'q', 'r', 's', 't', 'v', 'w', 'x' ,'z') + ) + + case object UppercaseConsonant extends AlphabetSet( + Set('B', 'C', 'D', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'V', 'W', 'X' ,'Z') + ) + + case object Consonant extends AlphabetSet(LowercaseConsonant.chars ++ UppercaseConsonant.chars) + + case object LowercaseVowel extends AlphabetSet(Set('a', 'e', 'i', 'o', 'u')) + + case object UppercaseVowel extends AlphabetSet(Set('A', 'E', 'I', 'O', 'U')) + + case object Vowel extends AlphabetSet(LowercaseVowel.chars ++ UppercaseVowel.chars) + + case object LowercaseY extends AlphabetSet(Set('y')) + + case object UppercaseY extends AlphabetSet(Set('Y')) + + case object Y extends AlphabetSet(LowercaseY.chars ++ UppercaseY.chars) + + case object LowercaseAlpha extends AlphabetSet(LowercaseConsonant.chars ++ LowercaseVowel.chars ++ LowercaseY.chars) + + case object UppercaseAlpha extends AlphabetSet(UppercaseConsonant.chars ++ UppercaseVowel.chars ++ UppercaseY.chars) + + case object Alpha extends AlphabetSet(LowercaseAlpha.chars ++ UppercaseAlpha.chars) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/Metric.scala b/core/src/main/scala/com/rockymadden/stringmetric/Metric.scala new file mode 100755 index 0000000..f45dd14 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/Metric.scala @@ -0,0 +1,77 @@ +package com.rockymadden.stringmetric + +object Metric { + import Transform.StringTransform + + + trait Metric[A, B] { + def compare(a: A, b: A): Option[B] + } + + + trait StringMetric[A] extends Metric[Array[Char], A] { + def compare(a: String, b: String): Option[A] + } + + + object StringMetric { + final val DiceSorensen = similarity.DiceSorensenMetric + final val Hamming = similarity.HammingMetric + final val Jaccard = similarity.JaccardMetric + final val Jaro = similarity.JaroMetric + final val JaroWinkler = similarity.JaroWinklerMetric + final val Levenshtein = similarity.LevenshteinMetric + final val Metaphone = phonetic.MetaphoneMetric + final val NGram = similarity.NGramMetric + final val Nysiis = phonetic.NysiisMetric + final val Overlap = similarity.OverlapMetric + final val RefinedNysiis = phonetic.RefinedNysiisMetric + final val RefinedSoundex = phonetic.RefinedSoundexMetric + final val Soundex = phonetic.SoundexMetric + final val WeightedLevenshtein = similarity.WeightedLevenshteinMetric + + def compareWithDiceSorensen(n: Int)(a: Array[Char], b: Array[Char]) = DiceSorensen(n).compare(a, b) + + def compareWithHamming(a: Array[Char], b: Array[Char]) = Hamming.compare(a, b) + + def compareWithJaccard(n: Int)(a: Array[Char], b: Array[Char]) = Jaccard(n).compare(a, b) + + def compareWithJaro(a: Array[Char], b: Array[Char]) = Jaro.compare(a, b) + + def compareWithJaroWinkler(a: Array[Char], b: Array[Char]) = JaroWinkler.compare(a, b) + + def compareWithLevenshtein(a: Array[Char], b: Array[Char]) = Levenshtein.compare(a, b) + + def compareWithMetaphone(a: Array[Char], b: Array[Char]) = Metaphone.compare(a, b) + + def compareWithNGram(n: Int)(a: Array[Char], b: Array[Char]) = NGram(n).compare(a, b) + + def compareWithNysiis(a: Array[Char], b: Array[Char]) = Nysiis.compare(a, b) + + def compareWithOverlap(n: Int)(a: Array[Char], b: Array[Char]) = Overlap(n).compare(a, b) + + def compareWithRefinedNysiis(a: Array[Char], b: Array[Char]) = RefinedNysiis.compare(a, b) + + def compareWithRefinedSoundex(a: Array[Char], b: Array[Char]) = RefinedSoundex.compare(a, b) + + def compareWithSoundex(a: Array[Char], b: Array[Char]) = Soundex.compare(a, b) + + def compareWithWeightedLevenshtein(delete: BigDecimal, insert: BigDecimal, substitute: BigDecimal) + (a: Array[Char], b: Array[Char]) = + + WeightedLevenshtein(delete, insert, substitute).compare(a, b) + } + + final class StringMetricDecorator[A](val sm: StringMetric[A]) { + val withTransform: (StringTransform => StringMetric[A]) = (st) => new StringMetric[A] { + private[this] val self: StringMetric[A] = sm + private[this] val transform: StringTransform = st + + override def compare(a: Array[Char], b: Array[Char]): Option[A] = + self.compare(transform(a), transform(b)) + + override def compare(a: String, b: String): Option[A] = + self.compare(transform(a.toCharArray), transform(b.toCharArray)) + } + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/Tokenize.scala b/core/src/main/scala/com/rockymadden/stringmetric/Tokenize.scala new file mode 100755 index 0000000..a011c96 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/Tokenize.scala @@ -0,0 +1,33 @@ +package com.rockymadden.stringmetric + +object Tokenize { + sealed trait Tokenizer[A] { + def tokenize(a: A): Option[Array[A]] + } + + + sealed trait StringTokenizer extends Tokenizer[Array[Char]] { + def tokenize(a: String): Option[Array[String]] + } + + + object StringTokenizer { + val NGram = NGramTokenizer + + def tokenizeWithNGram(n: Int)(charArray: Array[Char]) = NGram(n).tokenize(charArray) + } + + + final case class NGramTokenizer(n: Int) extends StringTokenizer { + override def tokenize(a: Array[Char]): Option[Array[Array[Char]]] = + if (n <= 0 || a.length < n) None + else Some(sequence(a, Array.empty[Array[Char]], n)) + + override def tokenize(a: String): Option[Array[String]] = tokenize(a.toCharArray).map(_.map(_.mkString)) + + @annotation.tailrec + private val sequence: ((Array[Char], Array[Array[Char]], Int) => Array[Array[Char]]) = (i, o, n) => + if (i.length <= n) o :+ i + else sequence(i.tail, o :+ i.take(n), n) + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/Transform.scala b/core/src/main/scala/com/rockymadden/stringmetric/Transform.scala new file mode 100644 index 0000000..c2cdace --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/Transform.scala @@ -0,0 +1,71 @@ +package com.rockymadden.stringmetric + +object Transform { + import scala.collection.immutable.NumericRange + + + type Transform[A] = (A => A) + type StringTransform = Transform[Array[Char]] + + + object StringTransform { + private final val Ascii = NumericRange(0x00, 0x7F, 1) + private final val ExtendedAscii = NumericRange(0x00, 0x7F, 1) + private final val Latin = NumericRange(0x00, 0x24F, 1) + private final val LowerCase = NumericRange(0x61, 0x7A, 1) + private final val Numbers = NumericRange(0x30, 0x39, 1) + private final val UpperCase = NumericRange(0x41, 0x5A, 1) + + private final val filter: ((Array[Char], (Char => Boolean)) => String) = (ca, f) => + ca.filter(c => f(c)).mkString + + private final val filterNot: ((Array[Char], (Char => Boolean)) => String) = (ca, f) => + ca.filterNot(c => f(c)).mkString + + val filterAlpha: StringTransform = (ca) => filter(ca, c => { + val ci = c.toInt + LowerCase.contains(ci) || UpperCase.contains(ci) + }) + + val filterNotAlpha: StringTransform = (ca) => filterNot(ca, c => { + val ci = c.toInt + LowerCase.contains(ci) || UpperCase.contains(ci) + }) + + val filterAlphaNumeric: StringTransform = (ca) => filter(ca, c => { + val ci = c.toInt + LowerCase.contains(ci) || UpperCase.contains(ci) || Numbers.contains(ci) + }) + + val filterNotAlphaNumeric: StringTransform = (ca) => filterNot(ca, c => { + val ci = c.toInt + LowerCase.contains(ci) || UpperCase.contains(ci) || Numbers.contains(ci) + }) + + val filterAscii: StringTransform = (ca) => filter(ca, c => Ascii.contains(c.toInt)) + + val filterNotAscii: StringTransform = (ca) => filterNot(ca, c => Ascii.contains(c.toInt)) + + val filterExtendedAscii: StringTransform = (ca) => filter(ca, c => ExtendedAscii.contains(c.toInt)) + + val filterNotExtendedAscii: StringTransform = (ca) => filterNot(ca, c => ExtendedAscii.contains(c.toInt)) + + val filterLatin: StringTransform = (ca) => filter(ca, c => Latin.contains(c.toInt)) + + val filterNotLatin: StringTransform = (ca) => filterNot(ca, c => Latin.contains(c.toInt)) + + val filterLowerCase: StringTransform = (ca) => filter(ca, c => LowerCase.contains(c.toInt)) + + val filterNotLowerCase: StringTransform = (ca) => filterNot(ca, c => LowerCase.contains(c.toInt)) + + val filterNumeric: StringTransform = (ca) => filter(ca, c => Numbers.contains(c.toInt)) + + val filterNotNumeric: StringTransform = (ca) => filterNot(ca, c => Numbers.contains(c.toInt)) + + val filterUpperCase: StringTransform = (ca) => filter(ca, c => UpperCase.contains(c.toInt)) + + val filterNotUpperCase: StringTransform = (ca) => filterNot(ca, c => UpperCase.contains(c.toInt)) + + val ignoreAlphaCase: StringTransform = (ca) => ca.map(c => if (c >= 65 && c <= 90) (c + 32).toChar else c) + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/package.scala b/core/src/main/scala/com/rockymadden/stringmetric/package.scala new file mode 100755 index 0000000..e5bc19d --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/package.scala @@ -0,0 +1,17 @@ +package com.rockymadden + +package object stringmetric { + import scala.language.implicitConversions + import Algorithm._ + import Metric._ + + type CompareTuple[T] = (Array[T], Array[T]) + type MatchTuple[T] = (Array[T], Array[T]) + + implicit def stringToCharArray(s: String): Array[Char] = + s.toCharArray + implicit def stringAlgorithmToDecoratedStringAlgorithm(sa: StringAlgorithm): StringAlgorithmDecorator = + new StringAlgorithmDecorator(sa) + implicit def stringMetricToDecoratedStringMetric[A](sa: StringMetric[A]): StringMetricDecorator[A] = + new StringMetricDecorator[A](sa) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithm.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithm.scala new file mode 100755 index 0000000..3abe7cc --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithm.scala @@ -0,0 +1,105 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Algorithm.StringAlgorithm + +case object MetaphoneAlgorithm extends StringAlgorithm { + import com.rockymadden.stringmetric.Alphabet.{Alpha, LowercaseVowel} + + override def compute(a: Array[Char]): Option[Array[Char]] = + if (a.length == 0 || !(Alpha isSuperset a.head)) None + else { + val th = (transcodeHead andThen deduplicate)(a.map(_.toLower)) + val t = transcode(Array.empty[Char], th.head, th.tail, Array.empty[Char]) + + if (t.length == 0) None else Some(t) // Single Y or W would have 0 length. + } + + override def compute(a: String): Option[String] = compute(a.toCharArray).map(_.mkString) + + private val deduplicate: (Array[Char] => Array[Char]) = (ca) => + if (ca.length <= 1) ca + else ca.sliding(2).withFilter(a => a(0) == 'c' || a(0) != a(1)).map(_(0)).toArray[Char] :+ ca.last + + @annotation.tailrec + private val transcode: ((Array[Char], Char, Array[Char], Array[Char]) => Array[Char]) = (l, c, r, o) => + if (c == '\0' && r.length == 0) o + else { + def shift(d: Int, ca: Array[Char]) = { + val sca = r.splitAt(d - 1) + + ( + if (sca._1.length > 0) (l :+ c) ++ sca._1 else l :+ c, + if (sca._2.length > 0) sca._2.head else '\0', + if (sca._2.length > 1) sca._2.tail else Array.empty[Char], + ca + ) + } + + val t = { + (c: @annotation.switch) match { + case 'a' | 'e' | 'i' | 'o' | 'u' => if (l.length == 0) shift(1, o:+ c) else shift(1, o) + case 'f' | 'j' | 'l' | 'm' | 'n' | 'r' => shift(1, o :+ c) + case 'b' => if (l.length >= 1 && l.last == 'm' && r.length == 0) shift(1, o) else shift(1, o :+ 'b') + case 'c' => + if (r.length >= 1 && r.head == 'h' && l.length >= 1 && l.last == 's') shift(1, o :+ 'k') + else if (r.length >= 2 && r.head == 'i' && r(1) == 'a') shift(3, o :+ 'x') + else if ((r.length >= 1 && r.head == 'h') + || (l.length >= 1 && r.length >= 1 && l.last == 's' && r.head == 'h')) shift(2, o :+ 'x') + else if (l.length >= 1 && r.length >= 1 && l.last == 's' + && (r.head == 'i' || r.head == 'e' || r.head == 'y')) shift(1, o) + else if (r.length >= 1 && (r.head == 'i' || r.head == 'e' || r.head == 'y')) shift(1, o :+ 's') + else shift(1, o :+ 'k') + case 'd' => + if (r.length >= 2 && r.head == 'g' + && (r(1) == 'e' || r(1) == 'y' || r(1) == 'i')) shift(1, o :+ 'j') + else shift(1, o :+ 't') + case 'g' => + if ((r.length > 1 && r.head == 'h') + || (r.length == 1 && r.head == 'n') + || (r.length == 3 && r.head == 'n' && r(1) == 'e' && r(2) == 'd')) shift(1, o) + else if (r.length >= 1 && (r.head == 'i' || r.head == 'e' || r.head == 'y')) shift(2, o :+ 'j') + else shift(1, o :+ 'k') + case 'h' => + if ((l.length >= 1 && (LowercaseVowel isSuperset l.last) && (r.length == 0 || !(LowercaseVowel isSuperset r.head))) + || (l.length >= 2 && l.last == 'h' + && (l(l.length - 2) == 'c' || l(l.length - 2) == 's' || l(l.length - 2) == 'p' + || l(l.length - 2) == 't' || l(l.length - 2) == 'g'))) shift(1, o) + else shift(1, o :+ 'h') + case 'k' => if (l.length >= 1 && l.last == 'c') shift(1, o) else shift(1, o :+ 'k') + case 'p' => if (r.length >= 1 && r.head == 'h') shift(2, o :+ 'f') else shift(1, o :+ 'p') + case 'q' => shift(1, o :+ 'k') + case 's' => + if (r.length >= 2 && r.head == 'i' && (r(1) == 'o' || r(1) == 'a')) shift(3, o :+ 'x') + else if (r.length >= 1 && r.head == 'h') shift(2, o :+ 'x') + else shift(1, o :+ 's') + case 't' => + if (r.length >= 2 && r.head == 'i' && (r(1) == 'a' || r(1) == 'o')) shift(3, o :+ 'x') + else if (r.length >= 1 && r.head == 'h') shift(2, o :+ '0') + else if (r.length >= 2 && r.head == 'c' && r(1) == 'h') shift(1, o) + else shift(1, o :+ 't') + case 'v' => shift(1, o :+ 'f') + case 'w' | 'y' => if (r.length == 0 || !(LowercaseVowel isSuperset r.head)) shift(1, o) else shift(1, o :+ c) + case 'x' => shift(1, (o :+ 'k') :+ 's') + case 'z' => shift(1, o :+ 's') + case _ => shift(1, o) + } + } + + transcode(t._1, t._2, t._3, t._4) + } + + private val transcodeHead: (Array[Char] => Array[Char]) = (ca) => + (ca.length: @annotation.switch) match { + case 0 => ca + case 1 => if (ca.head == 'x') Array('s') else ca + case _ => + (ca.head: @annotation.switch) match { + case 'a' if ca(1) == 'e' => ca.tail + case 'g' | 'k' | 'p' if ca(1) == 'n' => ca.tail + case 'w' if ca(1) == 'r' => ca.tail + case 'w' if ca(1) == 'h' => 'w' +: ca.drop(2) + case 'x' => 's' +: ca.tail + case _ => ca + } + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetric.scala new file mode 100755 index 0000000..d06f774 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetric.scala @@ -0,0 +1,15 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object MetaphoneMetric extends StringMetric[Boolean] { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compare(a: Array[Char], b: Array[Char]): Option[Boolean] = + if (a.length == 0 || !(Alpha isSuperset a.head) || b.length == 0 || !(Alpha isSuperset b.head)) None + else MetaphoneAlgorithm.compute(a).filter(_.length > 0).flatMap { mp1 => + MetaphoneAlgorithm.compute(b).filter(_.length > 0).map(mp1.sameElements(_)) + } + + override def compare(a: String, b: String): Option[Boolean] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithm.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithm.scala new file mode 100755 index 0000000..3e46675 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithm.scala @@ -0,0 +1,115 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Algorithm.StringAlgorithm + +case object NysiisAlgorithm extends StringAlgorithm { + import com.rockymadden.stringmetric.Alphabet.{Alpha, LowercaseVowel} + + override def compute(a: Array[Char]): Option[Array[Char]] = + if (a.length == 0 || !(Alpha isSuperset a.head)) None + else { + val tr = transcodeRight(a.map(_.toLower)) + val tl = transcodeLeft(tr._1) + val t = + if (tl._2.length == 0) tl._1 ++ tr._2 + else tl._1 ++ transcodeCenter( + Array.empty[Char], + tl._2.head, + if (tl._2.length > 1) tl._2.tail else Array.empty[Char], + Array.empty[Char] + ) ++ tr._2 + + if (t.length == 1) Some(t) + else Some(t.head +: (cleanLast andThen cleanTerminal andThen deduplicate)(t.tail)) + } + + override def compute(a: String): Option[String] = compute(a.toCharArray).map(_.mkString) + + private val cleanLast: (Array[Char] => Array[Char]) = (ca) => + if (ca.length == 0) ca + else if(ca.last == 'a' || ca.last == 's') + ca.dropRight(ca.reverseIterator.takeWhile(c => c == 'a' || c == 's').length) + else ca + + private val cleanTerminal: (Array[Char] => Array[Char]) = (ca) => + if (ca.length >= 2 && ca.last == 'y' && ca(ca.length - 2) == 'a') ca.dropRight(2) :+ 'y' + else ca + + private val deduplicate: (Array[Char] => Array[Char]) = (ca) => + if (ca.length <= 1) ca + else ca.sliding(2).withFilter(a => a(0) != a(1)).map(_(0)).toArray[Char] :+ ca.last + + @annotation.tailrec + private val transcodeCenter: ((Array[Char], Char, Array[Char], Array[Char]) => Array[Char]) = (l, c, r, o) => + if (c == '\0' && r.length == 0) o + else { + def shift(d: Int, ca: Array[Char]) = { + val sca = r.splitAt(d - 1) + + ( + if (sca._1.length > 0) (l :+ c) ++ sca._1 else l :+ c, + if (sca._2.length > 0) sca._2.head else '\0', + if (sca._2.length > 1) sca._2.tail else Array.empty[Char], + ca + ) + } + + val t = { + (c: @annotation.switch) match { + case 'a' | 'i' | 'o' | 'u' => shift(1, o :+ 'a') + case 'b' | 'c' | 'd' | 'f' | 'g' | 'j' | 'l' | 'n' | 'r' | 't' | 'v' | 'x' | 'y' => shift(1, o :+ c) + case 'e' => + if (r.length >= 1 && r.head == 'v') shift(2, o ++ Array('a', 'f')) + else shift(1, o :+ 'a') + case 'h' => + if (l.length >= 1 && (!(LowercaseVowel isSuperset l.last) || (r.length >= 1 && !(LowercaseVowel isSuperset r.head)))) + shift(1, o) + else shift(1, o :+ c) + case 'k' => if (r.length >= 1 && r.head == 'n') shift(2, o :+ 'n') else shift(1, o :+ 'c') + case 'm' => shift(1, o :+ 'n') + case 'p' => if (r.length >= 1 && r.head == 'h') shift(2, o :+ 'f') else shift(1, o :+ c) + case 'q' => shift(1, o :+ 'g') + case 's' => + if (r.length >= 2 && r.head == 'c' && r(1) == 'h') shift(3, o :+ c) + else shift(1, o :+ c) + case 'w' => + if (l.length >= 1 && (LowercaseVowel isSuperset l.last)) shift(1, o) + else shift(1, o :+ c) + case 'z' => shift(1, o :+ 's') + case _ => shift(1, o) + } + } + + transcodeCenter(t._1, t._2, t._3, t._4) + } + + private val transcodeLeft: (Array[Char] => (Array[Char], Array[Char])) = (ca) => + if (ca.length == 0) (Array.empty[Char], ca) + else { + lazy val tr2 = ca.takeRight(ca.length - 2) + lazy val tr3 = ca.takeRight(ca.length - 3) + + (ca.head: @annotation.switch) match { + case 'k' if ca.length >= 2 && ca(1) == 'n' => (Array('n', 'n'), tr2) + case 'k' => (Array('c'), ca.tail) + case 'm' if ca.length >= 3 && (ca(1) == 'a' && ca(2) == 'c') => (Array('m', 'c'), tr3) + case 'p' if ca.length >= 2 && (ca(1) == 'h' || ca(1) == 'f') => (Array('f', 'f'), tr2) + case 's' if ca.length >= 3 && (ca(1) == 'c' && ca(2) == 'h') => (Array('s', 's'), tr3) + case _ => (Array(ca.head), ca.tail) + } + } + + private val transcodeRight: (Array[Char] => (Array[Char], Array[Char])) = (ca) => + if (ca.length >= 2) { + val lc = ca(ca.length - 1) + val lcm1 = ca(ca.length - 2) + lazy val t2 = ca.take(ca.length - 2) + + (lc: @annotation.switch) match { + case 'd' if lcm1 == 'n' || lcm1 == 'r' => (t2, Array('d')) + case 'e' if lcm1 == 'e' || lcm1 == 'i' => (t2, Array('y')) + case 't' if lcm1 == 'd' || lcm1 == 'n' || lcm1 == 'r' => (t2, Array('d')) + case _ => (ca, Array.empty[Char]) + } + } else (ca, Array.empty[Char]) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/NysiisMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/NysiisMetric.scala new file mode 100755 index 0000000..c9a0914 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/NysiisMetric.scala @@ -0,0 +1,24 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object NysiisMetric extends StringMetric[Boolean] { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compare(a: Array[Char], b: Array[Char]): Option[Boolean] = { + val unequal: ((Char, Char) => Boolean) = (c1, c2) => { + val lc1 = c1.toLower + val lc2 = c2.toLower + + (if (lc1 == 'k') 'c' else lc1) != (if (lc2 == 'k') 'c' else lc2) + } + + if (a.length == 0 || !(Alpha isSuperset a.head) || b.length == 0 || !(Alpha isSuperset b.head)) None + else if (unequal(a.head, b.head)) Some(false) + else NysiisAlgorithm.compute(a).filter(_.length > 0).flatMap { ny1 => + NysiisAlgorithm.compute(b).filter(_.length > 0).map(ny1.sameElements(_)) + } + } + + override def compare(a: String, b: String): Option[Boolean] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithm.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithm.scala new file mode 100755 index 0000000..9976847 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithm.scala @@ -0,0 +1,121 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Algorithm.StringAlgorithm + +case object RefinedNysiisAlgorithm extends StringAlgorithm { + import com.rockymadden.stringmetric.Alphabet.{Alpha, LowercaseVowel} + + override def compute(a: Array[Char]): Option[Array[Char]] = + if (a.length == 0 || !(Alpha isSuperset a.head)) None + else { + val lca = a.map(_.toLower) + val tlh = (transcodeHead andThen transcodeLast)(lca.head +: cleanLast(lca.tail, Set('s', 'z'))) + val t = transcode(Array.empty[Char], tlh.head, tlh.tail, Array.empty[Char]) + + if (t.length == 1) Some(t) + else Some(deduplicate( + t.head +: (cleanLast.tupled andThen cleanTerminal)(t.tail, Set('a')) + )) + } + + override def compute(string: String): Option[String] = compute(string.toCharArray).map(_.mkString) + + private val cleanLast: ((Array[Char], Set[Char]) => Array[Char]) = (ca, s) => + if (ca.length == 0) ca + else if(s.contains(ca.last)) ca.dropRight(ca.reverseIterator.takeWhile(c => s.contains(c)).length) + else ca + + private val cleanTerminal: (Array[Char] => Array[Char]) = (ca) => + if (ca.length >= 2 && ca.last == 'y' && ca(ca.length - 2) == 'a') ca.dropRight(2) :+ 'y' + else ca + + private val deduplicate: (Array[Char] => Array[Char]) = (ca) => + if (ca.length <= 1) ca + else ca.sliding(2).withFilter(a => a(0) != a(1)).map(a => a(0)).toArray[Char] :+ ca.last + + @annotation.tailrec + private val transcode: ((Array[Char], Char, Array[Char], Array[Char]) => Array[Char]) = (l, c, r, o) => + if (c == '\0' && r.length == 0) o + else { + def shift(d: Int, ca: Array[Char]) = { + val sca = r.splitAt(d - 1) + + ( + if (sca._1.length > 0) (l :+ c) ++ sca._1 else l :+ c, + if (sca._2.length > 0) sca._2.head else '\0', + if (sca._2.length > 1) sca._2.tail else Array.empty[Char], + ca + ) + } + + val t = { + (c: @annotation.switch) match { + case 'a' | 'i' | 'o' | 'u' => + if (l.length == 0) shift(1, o :+ c) + else shift(1, o :+ 'a') + case 'b' | 'c' | 'f' | 'j' | 'l' | 'n' | 'r' | 't' | 'v' | 'x' => shift(1, o :+ c) + case 'd' => + if (r.length >= 1 && r.head == 'g') shift(2, o :+ 'g') else shift(1, o :+ c) + case 'e' => + if (l.length == 0) shift(1, o :+ c) + else if (r.length >= 1 && r.head == 'v') shift(2, o ++ Array('a', 'f')) + else shift(1, o :+ 'a') + case 'g' => + if (r.length >= 2 && r.head == 'h' && r(1) == 't') shift(3, o ++ Array('g', 't')) + else shift(1, o :+ c) + case 'h' => + if (l.length == 0) shift(1, o :+ c) + else if (!(LowercaseVowel isSuperset l.last) || (r.length >= 1 && !(LowercaseVowel isSuperset r.head))) + shift(1, o) + else shift(1, o :+ c) + case 'k' => if (r.length >= 1 && r.head == 'n') shift(2, o :+ 'n') else shift(1, o :+ 'c') + case 'm' => if (l.length == 0) shift(1, o :+ c) else shift(1, o :+ 'n') + case 'p' => if (r.length >= 1 && r.head == 'h') shift(2, o :+ 'f') else shift(1, o :+ c) + case 'q' => if (l.length == 0) shift(1, o :+ c) else shift(1, o :+ 'g') + case 's' => + if (r.length >= 2 && r.head == 'c' && r(1) == 'h') shift(3, o :+ c) + else if (r.length >= 1 && r.head == 'h') shift(2, o :+ c) + else shift(1, o :+ c) + case 'w' => + if (l.length >= 1 && (LowercaseVowel isSuperset l.last)) shift(1, o) + else if (r.length >= 1 && r.head == 'r') shift(2, o :+ 'r') + else shift(1, o :+ c) + case 'y' => + if (l.length >= 1 && r.length >= 2 && r.head == 'w') shift(2, o :+ 'a') + else if (r.length >= 1 && r.head == 'w') shift(2, o :+ c) + else if (l.length >= 1 && r.length >= 1) shift(1, o :+ 'a') + else shift(1, o :+ c) + case 'z' => if (l.length == 0) shift(1, o :+ c) else shift(1, o :+ 's') + case _ => shift(1, o) + } + } + + transcode(t._1, t._2, t._3, t._4) + } + + private val transcodeHead: (Array[Char] => Array[Char]) = (ca) => + if (ca.length == 0) ca + else + (ca.head: @annotation.switch) match { + case 'm' if ca.length >= 3 && ca(1) == 'a' && ca(2) == 'c' => + Array('m', 'c') ++ ca.takeRight(ca.length - 3) + case 'p' if ca.length >= 2 && ca(1) == 'f' =>'f' +: ca.takeRight(ca.length - 2) + case _ => ca + } + + private val transcodeLast: (Array[Char] => Array[Char]) = (ca) => + if (ca.length >= 2) { + val lc = ca(ca.length - 1) + val lcm1 = ca(ca.length - 2) + lazy val t2 = ca.take(ca.length - 2) + + (lc: @annotation.switch) match { + case 'd' if lcm1 == 'n' || lcm1 == 'r' => t2 :+ 'd' + case 'e' if lcm1 == 'e' || lcm1 == 'i' || lcm1 =='y' => t2 :+ 'y' + case 't' if lcm1 == 'd' || lcm1 == 'n' || lcm1 == 'r' => t2 :+ 'd' + case 'x' if lcm1 == 'e' => t2 ++ Array('e', 'c') + case 'x' if lcm1 == 'i' => t2 ++ Array('i', 'c') + case _ => ca + } + } else ca +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetric.scala new file mode 100755 index 0000000..488f261 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetric.scala @@ -0,0 +1,24 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object RefinedNysiisMetric extends StringMetric[Boolean] { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compare(a: Array[Char], b: Array[Char]): Option[Boolean] = { + val unequal = (c1: Char, c2: Char) => { + val lc1 = c1.toLower + val lc2 = c2.toLower + + (if (lc1 == 'k') 'c' else lc1) != (if (lc2 == 'k') 'c' else lc2) + } + + if (a.length == 0 || !(Alpha isSuperset a.head) || b.length == 0 || !(Alpha isSuperset b.head)) None + else if (unequal(a.head, b.head)) Some(false) + else RefinedNysiisAlgorithm.compute(a).filter(_.length > 0).flatMap { rny1 => + RefinedNysiisAlgorithm.compute(b).filter(_.length > 0).map(rny1.sameElements(_)) + } + } + + override def compare(a: String, b: String): Option[Boolean] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithm.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithm.scala new file mode 100755 index 0000000..e8f3af6 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithm.scala @@ -0,0 +1,59 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Algorithm.StringAlgorithm + +case object RefinedSoundexAlgorithm extends StringAlgorithm { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compute(a: Array[Char]): Option[Array[Char]] = + if (a.length == 0 || !(Alpha isSuperset a.head)) None + else Some(transcode(a, Array(a.head.toLower))) + + override def compute(a: String): Option[String] = compute(a.toCharArray).map(_.mkString) + + @annotation.tailrec + private val transcode: ((Array[Char], Array[Char]) => Array[Char]) = (i, o) => + if (i.length == 0) o + else { + val c = i.head.toLower + val m2 = (mc: Char) => (mc: @annotation.switch) match { + case 'a' | 'e' | 'h' | 'i' | 'o' | 'u' | 'w' | 'y' => '0' + case 'b' | 'p' => '1' + case 'f' | 'v' => '2' + case 'c' | 'k' | 's' => '3' + case 'g' | 'j' => '4' + case 'q' | 'x' | 'z' => '5' + case 'd' | 't' => '6' + case 'l' => '7' + case 'm' | 'n' => '8' + case 'r' => '9' + case _ => '\0' + } + val m1 = (mc: Char, pc: Char) => (mc: @annotation.switch) match { + case 'a' | 'e' | 'h' | 'i' | 'o' | 'u' | 'w' | 'y' if pc != '0' => '0' + case 'b' | 'p' if pc != '1' => '1' + case 'f' | 'v' if pc != '2' => '2' + case 'c' | 'k' | 's' if pc != '3' => '3' + case 'g' | 'j' if pc != '4' => '4' + case 'q' | 'x' | 'z' if pc != '5' => '5' + case 'd' | 't' if pc != '6' => '6' + case 'l' if pc != '7' => '7' + case 'm' | 'n' if pc != '8' => '8' + case 'r' if pc != '9' => '9' + case _ => '\0' + } + val a = + // Code twice. + if (o.length == 1) m2(c) + // Code once. + else m1( + c, + (o.last: @annotation.switch) match { + case '0' | '1' | '2' | '3' | '4' | '5' | '6' | '7' | '8' | '9' => o.last + case _ => m2(o.last) + } + ) + + transcode(i.tail, if (a != '\0') o :+ a else o) + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetric.scala new file mode 100755 index 0000000..289fe29 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetric.scala @@ -0,0 +1,16 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object RefinedSoundexMetric extends StringMetric[Boolean] { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compare(a: Array[Char], b: Array[Char]): Option[Boolean] = + if (a.length == 0 || !(Alpha isSuperset a.head) || b.length == 0 || !(Alpha isSuperset b.head)) None + else if (a.head.toLower != b.head.toLower) Some(false) + else RefinedSoundexAlgorithm.compute(a).filter(_.length > 0).flatMap { rse1 => + RefinedSoundexAlgorithm.compute(b).filter(_.length > 0).map(rse1.sameElements(_)) + } + + override def compare(a: String, b: String): Option[Boolean] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithm.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithm.scala new file mode 100755 index 0000000..b211908 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithm.scala @@ -0,0 +1,57 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Algorithm.StringAlgorithm + +case object SoundexAlgorithm extends StringAlgorithm { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compute(a: Array[Char]): Option[Array[Char]] = + if (a.length == 0 || !(Alpha isSuperset a.head)) None + else { + val fc = a.head.toLower + + Some(transcode(a.tail, fc, Array(fc)).padTo(4, '0')) + } + + override def compute(string: String): Option[String] = compute(string.toCharArray).map(_.mkString) + + @annotation.tailrec + private val transcode: ((Array[Char], Char, Array[Char]) => Array[Char]) = (i, pc, o) => + if (i.length == 0) o + else { + val c = i.head.toLower + val m2 = (mc: Char) => (mc: @annotation.switch) match { + case 'b' | 'f' | 'p' | 'v' => '1' + case 'c' | 'g' | 'j' | 'k' | 'q' | 's' | 'x' | 'z' => '2' + case 'd' | 't' => '3' + case 'l' => '4' + case 'm' | 'n' => '5' + case 'r' => '6' + case _ => '\0' + } + val m1 = (mc: Char, pc: Char) => (mc: @annotation.switch) match { + case 'b' | 'f' | 'p' | 'v' if pc != '1' => '1' + case 'c' | 'g' | 'j' | 'k' | 'q' | 's' | 'x' | 'z' if pc != '2' => '2' + case 'd' | 't' if pc != '3' => '3' + case 'l' if pc != '4' => '4' + case 'm' | 'n' if pc != '5' => '5' + case 'r' if pc != '6' => '6' + case _ => '\0' + } + val a = pc match { + // Code twice. + case 'a' | 'e' | 'i' | 'o' | 'u' | 'y' => m2(c) + // Code once. + case _ => m1( + c, + (o.last: @annotation.switch) match { + case '1' | '2' | '3' | '4' | '5' | '6' => o.last + case _ => m2(o.last) + } + ) + } + + if (o.length == 3 && a != '\0') o :+ a + else transcode(i.tail, c, if (a != '\0') o :+ a else o) + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/phonetic/SoundexMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/SoundexMetric.scala new file mode 100755 index 0000000..eca32db --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/phonetic/SoundexMetric.scala @@ -0,0 +1,16 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object SoundexMetric extends StringMetric[Boolean] { + import com.rockymadden.stringmetric.Alphabet.Alpha + + override def compare(a: Array[Char], b: Array[Char]): Option[Boolean] = + if (a.length == 0 || !(Alpha isSuperset a.head) || b.length == 0 || !(Alpha isSuperset b.head)) None + else if (a.head.toLower != b.head.toLower) Some(false) + else SoundexAlgorithm.compute(a).filter(_.length > 0).flatMap { se1 => + SoundexAlgorithm.compute(b).filter(_.length > 0).map(se1.sameElements(_)) + } + + final override def compare(a: String, b: String): Option[Boolean] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetric.scala new file mode 100755 index 0000000..0ad3915 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetric.scala @@ -0,0 +1,27 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +/** + * An implementation of the Dice/Sorensen metric. This implementation differs in that n-gram size is required. + * Traditionally, the algorithm uses bigrams. + */ +final case class DiceSorensenMetric(n: Int) extends StringMetric[Double] { + import com.rockymadden.stringmetric.Tokenize.NGramTokenizer + import com.rockymadden.stringmetric.MatchTuple + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (n <= 0 || a.length < n || b.length < n) None // Because length is less than n, it is not possible to compare. + else if (a.sameElements(b)) Some(1d) + else NGramTokenizer(n).tokenize(a).flatMap { ca1bg => + NGramTokenizer(n).tokenize(b).map { ca2bg => + val ms = scoreMatches(ca1bg.map(_.mkString), ca2bg.map(_.mkString)) + + (2d * ms) / (ca1bg.length + ca2bg.length) + } + } + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) + + private val scoreMatches: (MatchTuple[String] => Int) = (mt) => mt._1.intersect(mt._2).length +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/HammingMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/HammingMetric.scala new file mode 100755 index 0000000..4a90f32 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/HammingMetric.scala @@ -0,0 +1,18 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object HammingMetric extends StringMetric[Int] { + import com.rockymadden.stringmetric.CompareTuple + + override def compare(a: Array[Char], b: Array[Char]): Option[Int] = + if (a.length == 0 || b.length == 0 || a.length != b.length) None + else if (a.sameElements(b)) Some(0) + else Some(hamming(a, b)) + + override def compare(a: String, b: String): Option[Int] = compare(a.toCharArray, b.toCharArray) + + private val hamming: (CompareTuple[Char] => Int) = (ct) => + if (ct._1.length == 0) 0 + else ct._1.zip(ct._2).count(t => t._1 != t._2) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaccardMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaccardMetric.scala new file mode 100755 index 0000000..6ec5db4 --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaccardMetric.scala @@ -0,0 +1,20 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +final case class JaccardMetric(n: Int) extends StringMetric[Double] { + import com.rockymadden.stringmetric.Tokenize.NGramTokenizer + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (n <= 0 || a.length < n || b.length < n) None // Because length is less than n, it is not possible to compare. + else if (a.sameElements(b)) Some(1d) + else NGramTokenizer(n).tokenize(a).flatMap { ca1bg => + NGramTokenizer(n).tokenize(b).map { ca2bg => + val i = (ca1bg.map(_.mkString) intersect ca2bg.map(_.mkString)).length + + i.toDouble / (ca1bg.length + ca2bg.length - i) + } + } + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaroMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaroMetric.scala new file mode 100755 index 0000000..575d67a --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaroMetric.scala @@ -0,0 +1,66 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric +import scala.Some + +/** + * An implementation of the Jaro metric. One differing detail in this implementation is that if a character is matched + * in string2, it cannot be matched upon again. This results in a more penalized distance in these scenarios. + */ +case object JaroMetric extends StringMetric[Double] { + import com.rockymadden.stringmetric.{CompareTuple, MatchTuple} + import scala.collection.mutable.{ArrayBuffer, HashSet} + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (a.length == 0 || b.length == 0) None + else if (a.sameElements(b)) Some(1d) + else { + val mt = `match`(a, b) + val ms = scoreMatches(mt._1, mt._2) + + if (ms == 0) Some(0d) + else { + val ts = scoreTranspositions(mt._1, mt._2) + + Some(((ms.toDouble / a.length) + (ms.toDouble / b.length) + ((ms.toDouble - ts) / ms)) / 3) + } + } + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) + + private val `match`: (CompareTuple[Char] => MatchTuple[Char]) = (ct) => { + lazy val window = math.abs((math.max(ct._1.length, ct._2.length) / 2d).floor.toInt - 1) + val one = ArrayBuffer.empty[Int] + val two = HashSet.empty[Int] + var i = 0 + var bi = false + + while (i < ct._1.length && !bi) { + val start = if (i - window <= 0) 0 else i - window + val end = if (i + window >= ct._2.length - 1) ct._2.length - 1 else i + window + + if (start > ct._2.length - 1) bi = !bi + else { + var ii = start + var bii = false + + while (ii <= end && !bii) { + if (!two.contains(ii) && ct._1(i) == ct._2(ii)) { + one += i + two += ii + bii = !bii + } else ii += 1 + } + + i += 1 + } + } + + (one.toArray.map(ct._1(_)), two.toArray.sortWith(_ < _).map(ct._2(_))) + } + + private val scoreMatches: (MatchTuple[Char] => Int) = (mt) => mt._1.length + + private val scoreTranspositions: (MatchTuple[Char] => Int) = (mt) => + (mt._1.zip(mt._2).count(t => t._1 != t._2) / 2d).floor.toInt +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetric.scala new file mode 100755 index 0000000..e83f73f --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetric.scala @@ -0,0 +1,23 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +/** + * An implementation of the Jaro-Winkler metric. One differing detail in this implementation is that if a character is + * matched in string2, it cannot be matched upon again. This results in a more penalized distance in these scenarios + * (e.g. comparing henka and henkan distance is 0.9666 versus the typical 0.9722). + */ +case object JaroWinklerMetric extends StringMetric[Double] { + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + JaroMetric.compare(a, b).map { + case 0d => 0d + case 1d => 1d + case jaro => { + val prefix = a.zip(b).takeWhile(t => t._1 == t._2) + + jaro + ((if (prefix.length <= 4) prefix.length else 4) * 0.1d * (1 - jaro)) + } + } + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetric.scala new file mode 100755 index 0000000..fb90cdc --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetric.scala @@ -0,0 +1,40 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object LevenshteinMetric extends StringMetric[Int] { + import com.rockymadden.stringmetric.CompareTuple + + override def compare(a: Array[Char], b: Array[Char]): Option[Int] = + if (a.length == 0 || b.length == 0) None + else if (a.sameElements(b)) Some(0) + else Some(levenshtein(a, b)) + + override def compare(a: String, b: String): Option[Int] = compare(a.toCharArray, b.toCharArray) + + private val levenshtein: (CompareTuple[Char] => Int) = (ct) => { + val m = Array.fill[Int](ct._1.length + 1, ct._2.length + 1)(-1) + + def distance(t: (Int, Int)): Int = t match { + case (r, 0) => r + case (0, c) => c + case (r, c) if m(r)(c) != -1 => m(r)(c) + case (r, c) => { + val min = + if (ct._1(r - 1) == ct._2(c - 1)) distance(r - 1, c - 1) + else math.min( + math.min( + distance(r - 1, c) + 1, // Delete (left). + distance(r, c - 1) + 1 // Insert (up). + ), + distance(r - 1, c - 1) + 1 // Substitute (left-up). + ) + + m(r)(c) = min + min + } + } + + distance(ct._1.length, ct._2.length) + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/NGramMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/NGramMetric.scala new file mode 100755 index 0000000..8c194ce --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/NGramMetric.scala @@ -0,0 +1,24 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +final case class NGramMetric(n: Int) extends StringMetric[Double] { + import com.rockymadden.stringmetric.MatchTuple + import com.rockymadden.stringmetric.Tokenize.NGramTokenizer + import scala.math + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (n <= 0 || a.length < n || b.length < n) None // Because length is less than n, it is not possible to compare. + else if (a.sameElements(b)) Some(1d) + else NGramTokenizer(n).tokenize(a).flatMap { ca1bg => + NGramTokenizer(n).tokenize(b).map { ca2bg => + val ms = scoreMatches((ca1bg.map(_.mkString), ca2bg.map(_.mkString))) + + ms.toDouble / math.max(ca1bg.length, ca2bg.length) + } + } + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) + + private val scoreMatches: (MatchTuple[String] => Int) = (mt) => mt._1.intersect(mt._2).length +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/OverlapMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/OverlapMetric.scala new file mode 100755 index 0000000..8f0418b --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/OverlapMetric.scala @@ -0,0 +1,24 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +final case class OverlapMetric(n: Int) extends StringMetric[Double] { + import com.rockymadden.stringmetric.MatchTuple + import com.rockymadden.stringmetric.Tokenize.NGramTokenizer + import scala.math + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (n <= 0 || a.length < n || b.length < n) None // Because length is less than n, it is not possible to compare. + else if (a.sameElements(b)) Some(1d) + else NGramTokenizer(n).tokenize(a).flatMap { ca1bg => + NGramTokenizer(n).tokenize(b).map { ca2bg => + val ms = scoreMatches(ca1bg.map(_.mkString), ca2bg.map(_.mkString)) + + ms.toDouble / math.min(ca1bg.length, ca2bg.length) + } + } + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) + + private val scoreMatches: (MatchTuple[String] => Int) = (mt) => mt._1.intersect(mt._2).length +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetric.scala new file mode 100755 index 0000000..fa113bc --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetric.scala @@ -0,0 +1,43 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +case object RatcliffObershelpMetric extends StringMetric[Double] { + import com.rockymadden.stringmetric.CompareTuple + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (a.length == 0 || b.length == 0) None + else if (a.sameElements(b)) Some(1d) + else Some(2d * commonSequences(a, b).foldLeft(0)(_ + _.length) / (a.length + b.length)) + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) + + private def longestCommonSubsequence(ct: CompareTuple[Char]) = { + val m = Array.ofDim[Int](ct._1.length + 1, ct._2.length + 1) + var lrc = (0, 0, 0) // Length, row, column. + + for (r <- 0 to ct._1.length - 1; c <- 0 to ct._2.length - 1) { + if (ct._1(r) == ct._2(c)) { + val l = m(r)(c) + 1 + m(r + 1)(c + 1) = l + if (l > lrc._1) lrc = (l, r + 1, c + 1) + } + } + + lrc + } + + private val commonSequences: (CompareTuple[Char] => Array[Array[Char]]) = (ct) => { + val lcs = longestCommonSubsequence(ct) + + if (lcs._1 == 0) Array.empty + else { + val sct1 = (ct._1.take(lcs._2 - lcs._1), ct._1.takeRight(ct._1.length - lcs._2)) + val sct2 = (ct._2.take(lcs._3 - lcs._1), ct._2.takeRight(ct._2.length - lcs._3)) + + Array(ct._1.slice(lcs._2 - lcs._1, lcs._2)) ++ + commonSequences(sct1._1, sct2._1) ++ + commonSequences(sct1._2, sct2._2) + } + } +} diff --git a/core/src/main/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetric.scala b/core/src/main/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetric.scala new file mode 100755 index 0000000..ae6f49c --- /dev/null +++ b/core/src/main/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetric.scala @@ -0,0 +1,36 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.Metric.StringMetric + +final case class WeightedLevenshteinMetric(delete: BigDecimal, insert: BigDecimal, substitute: BigDecimal) + extends StringMetric[Double] { + + import com.rockymadden.stringmetric.CompareTuple + + override def compare(a: Array[Char], b: Array[Char]): Option[Double] = + if (a.length == 0 || b.length == 0) None + else if (a.sameElements(b)) Some(0d) + else Some(weightedLevenshtein((a, b), (delete, insert, substitute)).toDouble) + + override def compare(a: String, b: String): Option[Double] = compare(a.toCharArray, b.toCharArray) + + private val weightedLevenshtein: ((CompareTuple[Char], (BigDecimal, BigDecimal, BigDecimal)) => BigDecimal) = + (ct, w) => { + val m = Array.ofDim[BigDecimal](ct._1.length + 1, ct._2.length + 1) + + for (r <- 0 to ct._1.length) m(r)(0) = w._1 * r + for (c <- 0 to ct._2.length) m(0)(c) = w._2 * c + + for (r <- 1 to ct._1.length; c <- 1 to ct._2.length) { + m(r)(c) = + if (ct._1(r - 1) == ct._2(c - 1)) m(r - 1)(c - 1) + else (m(r - 1)(c) + w._1).min( // Delete (left). + (m(r)(c - 1) + w._2).min( // Insert (up). + m(r - 1)(c - 1) + w._3 // Substitute (left-up). + ) + ) + } + + m(ct._1.length)(ct._2.length) + } +} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/AlgorithmSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/AlgorithmSpec.scala new file mode 100644 index 0000000..d727145 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/AlgorithmSpec.scala @@ -0,0 +1,34 @@ +package com.rockymadden.stringmetric + +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class AlgorithmSpec extends ScalaTest { + import phonetic._ + import Algorithm._ + import Transform.StringTransform + + "StringAlgorithm" should provide { + "compute method and companion object pass through" in { + StringAlgorithm.computeWithMetaphone("testone").get should + equal (MetaphoneAlgorithm.compute("testone".toCharArray).get) + StringAlgorithm.computeWithNysiis("testone").get should + equal (NysiisAlgorithm.compute("testone".toCharArray).get) + StringAlgorithm.computeWithRefinedNysiis("testone").get should + equal (RefinedNysiisAlgorithm.compute("testone".toCharArray).get) + StringAlgorithm.computeWithRefinedSoundex("testone").get should + equal (RefinedSoundexAlgorithm.compute("testone".toCharArray).get) + StringAlgorithm.computeWithSoundex("testone").get should + equal (SoundexAlgorithm.compute("testone".toCharArray).get) + } + } + + "StringAlgorithmDecorator" should provide { + "withTransform()" in { + (MetaphoneAlgorithm withTransform StringTransform.filterAlpha).compute("abc123").get should + equal (MetaphoneAlgorithm.compute("abc").get) + } + } +} + diff --git a/core/src/test/scala/com/rockymadden/stringmetric/AlphabetSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/AlphabetSpec.scala new file mode 100755 index 0000000..5a6a1dc --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/AlphabetSpec.scala @@ -0,0 +1,95 @@ +package com.rockymadden.stringmetric + +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class AlphabetSpec extends ScalaTest { "Alphabet" should provide { + import Alphabet.{Alpha, Vowel} + + "an overloaded isSuperset method which accepts Char" when passed { + "non-alphabet argument" should returns { + "false" in { + Alpha isSuperset '0' should be (false) + } + } + "alphabet argument" should returns { + "true" in { + Alpha isSuperset 'a' should be (true) + Alpha isSuperset 'A' should be (true) + } + } + "non-vowel argument" should returns { + "false" in { + Vowel isSuperset 'y' should be (false) + } + } + "vowel argument" should returns { + "true" in { + Vowel isSuperset 'a' should be (true) + Vowel isSuperset 'A' should be (true) + } + } + } + "an overloaded isSuperset method which accepts Array[Char]" when passed { + "empty argument" should returns { + "false" in { + Alpha isSuperset Array.empty[Char] should be (false) + } + } + "non-alphabet argument" should returns { + "false" in { + Alpha isSuperset "hi!".toCharArray should be (false) + Alpha isSuperset "helloworld!".toCharArray should be (false) + } + } + "alphabet argument" should returns { + "true" in { + Alpha isSuperset "hi".toCharArray should be (true) + Alpha isSuperset "helloworld".toCharArray should be (true) + Alpha isSuperset "HI".toCharArray should be (true) + Alpha isSuperset "HELLOWORLD".toCharArray should be (true) + } + } + "non-vowel argument" should returns { + "false" in { + Vowel isSuperset "y".toCharArray should be (false) + } + } + "vowel argument" should returns { + "true" in { + Vowel isSuperset "a".toCharArray should be (true) + Vowel isSuperset "A".toCharArray should be (true) + } + } + } + "an overloaded isSuperset method which accepts String" when passed { + "empty argument" should returns { + "false" in { + Alpha isSuperset "" should be (false) + } + } + "non-alphabet argument" should returns { + "false" in { + Alpha isSuperset "helloworld!" should be (false) + } + } + "alphabet argument" should returns { + "true" in { + Alpha isSuperset "helloworld" should be (true) + Alpha isSuperset "HELLOWORLD" should be (true) + } + } + "non-vowel argument" should returns { + "false" in { + Vowel isSuperset "y" should be (false) + } + } + "vowel argument" should returns { + "true" in { + Vowel isSuperset "a" should be (true) + Vowel isSuperset "A" should be (true) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/MetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/MetricSpec.scala new file mode 100644 index 0000000..3b9021d --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/MetricSpec.scala @@ -0,0 +1,56 @@ +package com.rockymadden.stringmetric + +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class MetricSpec extends ScalaTest { + import phonetic._ + import similarity._ + import Metric._ + import Transform.StringTransform + + "StringMetric standalone object" should provide { + "compare method and companion object pass through" in { + StringMetric.compareWithDiceSorensen(1)("testone", "testtwo").get should + equal (DiceSorensenMetric(1).compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithHamming("testone", "testtwo").get should + equal (HammingMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithJaccard(1)("testone", "testtwo").get should + equal (JaccardMetric(1).compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithJaro("testone", "testtwo").get should + equal (JaroMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithJaroWinkler("testone", "testtwo").get should + equal (JaroWinklerMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithLevenshtein("testone", "testtwo").get should + equal (LevenshteinMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithMetaphone("testone", "testtwo").get should + equal (MetaphoneMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithNGram(1)("testone", "testtwo").get should + equal (NGramMetric(1).compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithNysiis("testone", "testtwo").get should + equal (NysiisMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithOverlap(1)("testone", "testtwo").get should + equal (OverlapMetric(1).compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithRefinedNysiis("testone", "testtwo").get should + equal (RefinedNysiisMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithRefinedSoundex("testone", "testtwo").get should + equal (RefinedSoundexMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithSoundex("testone", "testtwo").get should + equal (SoundexMetric.compare("testone".toCharArray, "testtwo".toCharArray).get) + StringMetric.compareWithWeightedLevenshtein(1, 2, 3)("testone", "testtwo").get should + equal (WeightedLevenshteinMetric(1, 2, 3).compare("testone".toCharArray, "testtwo".toCharArray).get) + } + } + + "StringMetricDecorator" should provide { + "withTransform()" in { + (MetaphoneMetric withTransform StringTransform.filterAlpha).compare("abc123", "abc456").get should + equal (true) + (DiceSorensenMetric(1) withTransform StringTransform.filterAlpha).compare("abc123", "abc456").get should + equal (1.0) + + (MetaphoneMetric withTransform (StringTransform.filterAlpha andThen StringTransform.filterUpperCase)).compare("abc123", "abc456") + } + } +} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/ScalaTest.scala b/core/src/test/scala/com/rockymadden/stringmetric/ScalaTest.scala new file mode 100755 index 0000000..5f4ab62 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/ScalaTest.scala @@ -0,0 +1,18 @@ +package com.rockymadden.stringmetric + +import org.scalatest.{BeforeAndAfter, ParallelTestExecution, WordSpec} +import org.scalatest.matchers.ShouldMatchers + +trait ScalaTest extends WordSpec with ShouldMatchers with BeforeAndAfter with ParallelTestExecution { + def allows = afterWord("allow") + + def executes = afterWord("execute") + + def passed = afterWord("passed") + + def provide = afterWord("provide") + + def returns = afterWord("return") + + def throws = afterWord("throw") +} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/TokenizeSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/TokenizeSpec.scala new file mode 100755 index 0000000..cfba0f7 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/TokenizeSpec.scala @@ -0,0 +1,45 @@ +package com.rockymadden.stringmetric + +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class TokenizeSpec extends ScalaTest { "NGramTokenizer" should provide { + import Tokenize._ + + "tokenize method" when passed { + "empty argument" should returns { + "None" in { + NGramTokenizer(1).tokenize("").isDefined should be (false) + } + } + "invalid n argument" should returns { + "None" in { + NGramTokenizer(0).tokenize("").isDefined should be (false) + NGramTokenizer(-1).tokenize("").isDefined should be (false) + } + } + "valid argument" should returns { + "Array[String]" in { + NGramTokenizer(1).tokenize("abcdefghijklmnopqrstuvwxyz").get should equal ( + Array( + "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", + "s", "t", "u", "v", "w", "x", "y", "z" + ) + ) + NGramTokenizer(2).tokenize("abcdefghijklmnopqrstuvwxyz").get should equal ( + Array( + "ab", "bc", "cd", "de", "ef", "fg", "gh", "hi", "ij", "jk", "kl", "lm", "mn", "no", "op", + "pq", "qr", "rs", "st", "tu", "uv", "vw", "wx", "xy", "yz" + ) + ) + NGramTokenizer(3).tokenize("abcdefghijklmnopqrstuvwxyz").get should equal ( + Array( + "abc", "bcd", "cde", "def", "efg", "fgh", "ghi", "hij", "ijk", "jkl", "klm", "lmn", "mno", + "nop", "opq", "pqr", "qrs", "rst", "stu", "tuv", "uvw", "vwx", "wxy", "xyz" + ) + ) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/TransformSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/TransformSpec.scala new file mode 100644 index 0000000..5e79c62 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/TransformSpec.scala @@ -0,0 +1,181 @@ +package com.rockymadden.stringmetric + +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class TransformSpec extends ScalaTest { "StringTransform" should provide { + import Transform._ + + "filterAlpha()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterAlpha( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ("aBc".toCharArray) + } + } + } + "filterNotAlpha()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotAlpha( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ( + ("123" + 0x250.toChar).toCharArray + ) + } + } + } + "filterAlphaNumeric()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterAlphaNumeric( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ("aBc123".toCharArray) + } + } + } + "filterNotAlphaNumeric()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotAlphaNumeric( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ( + ("" + 0x250.toChar).toCharArray + ) + } + } + } + "filterAscii()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterAscii( + ("aBc" + 0x80.toChar).toCharArray + ) should equal ("aBc".toCharArray) + } + } + } + "filterNotAscii()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotAscii( + ("aBc" + 0x100.toChar).toCharArray + ) should equal ( + ("" + 0x100.toChar).toCharArray + ) + } + } + } + "filterExtendedAscii()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterExtendedAscii( + ("aBc" + 0x100.toChar).toCharArray + ) should equal ("aBc".toCharArray) + } + } + } + "filterNotExtendedAscii()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotExtendedAscii( + ("aBc" + 0x250.toChar).toCharArray + ) should equal ( + ("" + 0x250.toChar).toCharArray + ) + } + } + } + "filterLatin()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterLatin( + ("aBc" + 0x250.toChar).toCharArray + ) should equal ("aBc".toCharArray) + } + } + } + "filterNotLatin()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotLatin( + ("aBc" + 0x300.toChar).toCharArray + ) should equal ( + ("" + 0x300.toChar).toCharArray + ) + } + } + } + "filterLowerCase()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterLowerCase( + "aBc123" + 0x250.toChar + ) should equal ("ac".toCharArray) + } + } + } + "filterNotLowerCase()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotLowerCase( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ( + ("B123" + 0x250.toChar).toCharArray + ) + } + } + } + "filterNumeric()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNumeric( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ("123".toCharArray) + } + } + } + "filterNotNumeric()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotNumeric( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ( + ("aBc" + 0x250.toChar).toCharArray + ) + } + } + } + "filterUpperCase()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterUpperCase( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ("B".toCharArray) + } + } + } + "filterNotUpperCase()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.filterNotUpperCase( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ( + ("ac123" + 0x250.toChar).toCharArray + ) + } + } + } + "ignoreAlphaCase()" when passed { + "String" should returns { + "transformed String" in { + StringTransform.ignoreAlphaCase( + ("aBc123" + 0x250.toChar).toCharArray + ) should equal ( + ("abc123" + 0x250.toChar).toCharArray + ) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithmSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithmSpec.scala new file mode 100755 index 0000000..6f56b6d --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/MetaphoneAlgorithmSpec.scala @@ -0,0 +1,211 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class MetaphoneAlgorithmSpec extends ScalaTest { "MetaphoneAlgorithm" should provide { + "compute method" when passed { + "empty argument" should returns { + "None" in { + MetaphoneAlgorithm.compute("").isDefined should be (false) + } + } + "non-phonetic argument" should returns { + "None" in { + MetaphoneAlgorithm.compute("123").isDefined should be (false) + } + } + "phonetic argument" should returns { + "Some" in { + // z + MetaphoneAlgorithm.compute("z").get should equal ("s") + MetaphoneAlgorithm.compute("zz").get should equal ("s") + + // y + MetaphoneAlgorithm.compute("y").isDefined should be (false) + MetaphoneAlgorithm.compute("zy").get should equal ("s") + MetaphoneAlgorithm.compute("zyz").get should equal ("ss") + MetaphoneAlgorithm.compute("zya").get should equal ("sy") + + // x + MetaphoneAlgorithm.compute("x").get should equal ("s") + MetaphoneAlgorithm.compute("zx").get should equal ("sks") + MetaphoneAlgorithm.compute("zxz").get should equal ("skss") + + // w + MetaphoneAlgorithm.compute("w").isDefined should be (false) + MetaphoneAlgorithm.compute("zw").get should equal ("s") + MetaphoneAlgorithm.compute("zwz").get should equal ("ss") + MetaphoneAlgorithm.compute("zwa").get should equal ("sw") + + // v + MetaphoneAlgorithm.compute("v").get should equal ("f") + MetaphoneAlgorithm.compute("zv").get should equal ("sf") + MetaphoneAlgorithm.compute("zvz").get should equal ("sfs") + + // u + MetaphoneAlgorithm.compute("u").get should equal ("u") + MetaphoneAlgorithm.compute("zu").get should equal ("s") + + // t + MetaphoneAlgorithm.compute("t").get should equal ("t") + MetaphoneAlgorithm.compute("ztiaz").get should equal ("sxs") + MetaphoneAlgorithm.compute("ztioz").get should equal ("sxs") + MetaphoneAlgorithm.compute("zthz").get should equal ("s0s") + MetaphoneAlgorithm.compute("ztchz").get should equal ("sxs") + MetaphoneAlgorithm.compute("ztz").get should equal ("sts") + + // s + MetaphoneAlgorithm.compute("s").get should equal ("s") + MetaphoneAlgorithm.compute("zshz").get should equal ("sxs") + MetaphoneAlgorithm.compute("zsioz").get should equal ("sxs") + MetaphoneAlgorithm.compute("zsiaz").get should equal ("sxs") + MetaphoneAlgorithm.compute("zs").get should equal ("ss") + MetaphoneAlgorithm.compute("zsz").get should equal ("sss") + + // r + MetaphoneAlgorithm.compute("r").get should equal ("r") + MetaphoneAlgorithm.compute("zr").get should equal ("sr") + MetaphoneAlgorithm.compute("zrz").get should equal ("srs") + + // q + MetaphoneAlgorithm.compute("q").get should equal ("k") + MetaphoneAlgorithm.compute("zq").get should equal ("sk") + MetaphoneAlgorithm.compute("zqz").get should equal ("sks") + + // p + MetaphoneAlgorithm.compute("p").get should equal ("p") + MetaphoneAlgorithm.compute("zp").get should equal ("sp") + MetaphoneAlgorithm.compute("zph").get should equal ("sf") + MetaphoneAlgorithm.compute("zpz").get should equal ("sps") + + // o + MetaphoneAlgorithm.compute("o").get should equal ("o") + MetaphoneAlgorithm.compute("zo").get should equal ("s") + + // n + MetaphoneAlgorithm.compute("n").get should equal ("n") + MetaphoneAlgorithm.compute("zn").get should equal ("sn") + MetaphoneAlgorithm.compute("znz").get should equal ("sns") + + // m + MetaphoneAlgorithm.compute("m").get should equal ("m") + MetaphoneAlgorithm.compute("zm").get should equal ("sm") + MetaphoneAlgorithm.compute("zmz").get should equal ("sms") + + // l + MetaphoneAlgorithm.compute("l").get should equal ("l") + MetaphoneAlgorithm.compute("zl").get should equal ("sl") + MetaphoneAlgorithm.compute("zlz").get should equal ("sls") + + // k + MetaphoneAlgorithm.compute("k").get should equal ("k") + MetaphoneAlgorithm.compute("zk").get should equal ("sk") + MetaphoneAlgorithm.compute("zck").get should equal ("sk") + + // j + MetaphoneAlgorithm.compute("j").get should equal ("j") + MetaphoneAlgorithm.compute("zj").get should equal ("sj") + MetaphoneAlgorithm.compute("zjz").get should equal ("sjs") + + // i + MetaphoneAlgorithm.compute("i").get should equal ("i") + MetaphoneAlgorithm.compute("zi").get should equal ("s") + + // h + MetaphoneAlgorithm.compute("h").get should equal ("h") // php wrongly says nothing + MetaphoneAlgorithm.compute("zh").get should equal ("sh") // php wrongly says s + MetaphoneAlgorithm.compute("zah").get should equal ("s") + MetaphoneAlgorithm.compute("zchh").get should equal ("sx") + MetaphoneAlgorithm.compute("ha").get should equal ("h") + + // g + MetaphoneAlgorithm.compute("g").get should equal ("k") + MetaphoneAlgorithm.compute("zg").get should equal ("sk") + MetaphoneAlgorithm.compute("zgh").get should equal ("skh") // php wrongly says sf + MetaphoneAlgorithm.compute("zghz").get should equal ("shs") // php wrongly says sfs + MetaphoneAlgorithm.compute("zgha").get should equal ("sh") // php wrongly says sf others wrongly say skh + MetaphoneAlgorithm.compute("zgn").get should equal ("sn") + MetaphoneAlgorithm.compute("zgns").get should equal ("skns") + MetaphoneAlgorithm.compute("zgned").get should equal ("snt") // others wrongly says sknt + MetaphoneAlgorithm.compute("zgneds").get should equal ("sknts") // php wrongly says snts + MetaphoneAlgorithm.compute("zgi").get should equal ("sj") + MetaphoneAlgorithm.compute("zgiz").get should equal ("sjs") + MetaphoneAlgorithm.compute("zge").get should equal ("sj") + MetaphoneAlgorithm.compute("zgez").get should equal ("sjs") + MetaphoneAlgorithm.compute("zgy").get should equal ("sj") + MetaphoneAlgorithm.compute("zgyz").get should equal ("sjs") + MetaphoneAlgorithm.compute("zgz").get should equal ("sks") + + // f + MetaphoneAlgorithm.compute("f").get should equal ("f") + MetaphoneAlgorithm.compute("zf").get should equal ("sf") + MetaphoneAlgorithm.compute("zfz").get should equal ("sfs") + + // e + MetaphoneAlgorithm.compute("e").get should equal ("e") + MetaphoneAlgorithm.compute("ze").get should equal ("s") + + // d + MetaphoneAlgorithm.compute("d").get should equal ("t") + MetaphoneAlgorithm.compute("fudge").get should equal ("fjj") // php wrongly says fj + MetaphoneAlgorithm.compute("dodgy").get should equal ("tjj") // php wrongly says tj others wrongly say tjjy + MetaphoneAlgorithm.compute("dodgi").get should equal ("tjj") // php wrongly says tj + MetaphoneAlgorithm.compute("zd").get should equal ("st") + MetaphoneAlgorithm.compute("zdz").get should equal ("sts") + + // c + MetaphoneAlgorithm.compute("c").get should equal ("k") + MetaphoneAlgorithm.compute("zcia").get should equal ("sx") + MetaphoneAlgorithm.compute("zciaz").get should equal ("sxs") + MetaphoneAlgorithm.compute("zch").get should equal ("sx") + MetaphoneAlgorithm.compute("zchz").get should equal ("sxs") + MetaphoneAlgorithm.compute("zci").get should equal ("ss") + MetaphoneAlgorithm.compute("zciz").get should equal ("sss") + MetaphoneAlgorithm.compute("zce").get should equal ("ss") + MetaphoneAlgorithm.compute("zcez").get should equal ("sss") + MetaphoneAlgorithm.compute("zcy").get should equal ("ss") + MetaphoneAlgorithm.compute("zcyz").get should equal ("sss") + MetaphoneAlgorithm.compute("zsci").get should equal ("ss") + MetaphoneAlgorithm.compute("zsciz").get should equal ("sss") + MetaphoneAlgorithm.compute("zsce").get should equal ("ss") + MetaphoneAlgorithm.compute("zscez").get should equal ("sss") + MetaphoneAlgorithm.compute("zscy").get should equal ("ss") + MetaphoneAlgorithm.compute("zscyz").get should equal ("sss") + MetaphoneAlgorithm.compute("zsch").get should equal ("sskh") // php wrongly says ssx + MetaphoneAlgorithm.compute("zc").get should equal ("sk") + MetaphoneAlgorithm.compute("zcz").get should equal ("sks") + + // b + MetaphoneAlgorithm.compute("b").get should equal ("b") + MetaphoneAlgorithm.compute("zb").get should equal ("sb") + MetaphoneAlgorithm.compute("zbz").get should equal ("sbs") + MetaphoneAlgorithm.compute("zmb").get should equal ("sm") + + // a + MetaphoneAlgorithm.compute("a").get should equal ("a") + MetaphoneAlgorithm.compute("za").get should equal ("s") + + // Miscellaneous. + MetaphoneAlgorithm.compute("dumb").get should equal ("tm") + MetaphoneAlgorithm.compute("smith").get should equal ("sm0") + MetaphoneAlgorithm.compute("school").get should equal ("skhl") // php wrongly says sxl + MetaphoneAlgorithm.compute("merci").get should equal ("mrs") + MetaphoneAlgorithm.compute("cool").get should equal ("kl") + MetaphoneAlgorithm.compute("aebersold").get should equal ("ebrslt") + MetaphoneAlgorithm.compute("gnagy").get should equal ("nj") + MetaphoneAlgorithm.compute("knuth").get should equal ("n0") + MetaphoneAlgorithm.compute("pniewski").get should equal ("nsk") + MetaphoneAlgorithm.compute("wright").get should equal ("rht") // php wrongly says rft + MetaphoneAlgorithm.compute("phone").get should equal ("fn") + MetaphoneAlgorithm.compute("aggregate").get should equal ("akrkt") + MetaphoneAlgorithm.compute("accuracy").get should equal ("akkrs") + MetaphoneAlgorithm.compute("encyclopedia").get should equal ("ensklpt") + MetaphoneAlgorithm.compute("honorificabilitudinitatibus").get should equal ("hnrfkblttnttbs") + MetaphoneAlgorithm.compute("antidisestablishmentarianism").get should equal ("anttsstblxmntrnsm") + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetricSpec.scala new file mode 100755 index 0000000..05401b0 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/MetaphoneMetricSpec.scala @@ -0,0 +1,39 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class MetaphoneMetricSpec extends ScalaTest { "MetaphoneMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + MetaphoneMetric.compare("", "").isDefined should be (false) + MetaphoneMetric.compare("abc", "").isDefined should be (false) + MetaphoneMetric.compare("", "xyz").isDefined should be (false) + } + } + "non-phonetic arguments" should returns { + "None" in { + MetaphoneMetric.compare("123", "123").isDefined should be (false) + MetaphoneMetric.compare("123", "").isDefined should be (false) + MetaphoneMetric.compare("", "123").isDefined should be (false) + } + } + "phonetically similar arguments" should returns { + "Boolean indicating true" in { + MetaphoneMetric.compare("dumb", "dum").get should be (true) + MetaphoneMetric.compare("smith", "smeth").get should be (true) + MetaphoneMetric.compare("merci", "mercy").get should be (true) + } + } + "phonetically dissimilar arguments" should returns { + "Boolean indicating false" in { + MetaphoneMetric.compare("dumb", "gum").get should be (false) + MetaphoneMetric.compare("smith", "kiss").get should be (false) + MetaphoneMetric.compare("merci", "burpy").get should be (false) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithmSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithmSpec.scala new file mode 100755 index 0000000..db8e073 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/NysiisAlgorithmSpec.scala @@ -0,0 +1,189 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class NysiisAlgorithmSpec extends ScalaTest { "NysiisAlgorithm" should provide { + "compute method" when passed { + "empty argument" should returns { + "None" in { + NysiisAlgorithm.compute("").isDefined should be (false) + } + } + "non-phonetic argument" should returns { + "None" in { + NysiisAlgorithm.compute("123").isDefined should be (false) + } + } + "phonetic argument" should returns { + "Some" in { + // a + NysiisAlgorithm.compute("a").get should equal ("a") + NysiisAlgorithm.compute("aa").get should equal ("a") + + // b + NysiisAlgorithm.compute("b").get should equal ("b") + NysiisAlgorithm.compute("bb").get should equal ("bb") + + // c + NysiisAlgorithm.compute("c").get should equal ("c") + NysiisAlgorithm.compute("cc").get should equal ("cc") + + // d + NysiisAlgorithm.compute("d").get should equal ("d") + NysiisAlgorithm.compute("dd").get should equal ("dd") + + // e + NysiisAlgorithm.compute("e").get should equal ("e") + NysiisAlgorithm.compute("ee").get should equal ("y") + + // f + NysiisAlgorithm.compute("f").get should equal ("f") + NysiisAlgorithm.compute("ff").get should equal ("ff") + + // g + NysiisAlgorithm.compute("g").get should equal ("g") + NysiisAlgorithm.compute("gg").get should equal ("gg") + + // h + NysiisAlgorithm.compute("h").get should equal ("h") + NysiisAlgorithm.compute("hh").get should equal ("hh") + + // i + NysiisAlgorithm.compute("i").get should equal ("i") + NysiisAlgorithm.compute("ii").get should equal ("i") + + // j + NysiisAlgorithm.compute("j").get should equal ("j") + NysiisAlgorithm.compute("jj").get should equal ("jj") + + // k + NysiisAlgorithm.compute("k").get should equal ("c") + NysiisAlgorithm.compute("kk").get should equal ("cc") + + // l + NysiisAlgorithm.compute("l").get should equal ("l") + NysiisAlgorithm.compute("ll").get should equal ("ll") + + // m + NysiisAlgorithm.compute("m").get should equal ("m") + NysiisAlgorithm.compute("mm").get should equal ("mn") + + // n + NysiisAlgorithm.compute("n").get should equal ("n") + NysiisAlgorithm.compute("nn").get should equal ("nn") + + // o + NysiisAlgorithm.compute("o").get should equal ("o") + NysiisAlgorithm.compute("oo").get should equal ("o") + + // p + NysiisAlgorithm.compute("p").get should equal ("p") + NysiisAlgorithm.compute("pp").get should equal ("pp") + + // q + NysiisAlgorithm.compute("q").get should equal ("q") + NysiisAlgorithm.compute("qq").get should equal ("qg") + + // r + NysiisAlgorithm.compute("r").get should equal ("r") + NysiisAlgorithm.compute("rr").get should equal ("rr") + + // s + NysiisAlgorithm.compute("s").get should equal ("s") + NysiisAlgorithm.compute("ss").get should equal ("s") + + // t + NysiisAlgorithm.compute("t").get should equal ("t") + NysiisAlgorithm.compute("tt").get should equal ("tt") + + // u + NysiisAlgorithm.compute("u").get should equal ("u") + NysiisAlgorithm.compute("uu").get should equal ("u") + + // v + NysiisAlgorithm.compute("v").get should equal ("v") + NysiisAlgorithm.compute("vv").get should equal ("vv") + + // w + NysiisAlgorithm.compute("w").get should equal ("w") + NysiisAlgorithm.compute("ww").get should equal ("ww") + + // x + NysiisAlgorithm.compute("x").get should equal ("x") + NysiisAlgorithm.compute("xx").get should equal ("xx") + + // y + NysiisAlgorithm.compute("y").get should equal ("y") + NysiisAlgorithm.compute("yy").get should equal ("yy") + + // z + NysiisAlgorithm.compute("z").get should equal ("z") + NysiisAlgorithm.compute("zz").get should equal ("z") + + // Head cases. + NysiisAlgorithm.compute("mac").get should equal ("mc") + NysiisAlgorithm.compute("kn").get should equal ("nn") + NysiisAlgorithm.compute("k").get should equal ("c") + NysiisAlgorithm.compute("ph").get should equal ("ff") + NysiisAlgorithm.compute("pf").get should equal ("ff") + NysiisAlgorithm.compute("sch").get should equal ("s") // dropby wrongly says ss + + // Last cases. + NysiisAlgorithm.compute("ee").get should equal ("y") + NysiisAlgorithm.compute("ie").get should equal ("y") + NysiisAlgorithm.compute("dt").get should equal ("d") + NysiisAlgorithm.compute("rt").get should equal ("d") + NysiisAlgorithm.compute("rd").get should equal ("d") + NysiisAlgorithm.compute("nt").get should equal ("d") + NysiisAlgorithm.compute("nd").get should equal ("d") + + // Core cases. + NysiisAlgorithm.compute("eev").get should equal ("eaf") + NysiisAlgorithm.compute("zev").get should equal ("zaf") + NysiisAlgorithm.compute("kkn").get should equal ("cn") + NysiisAlgorithm.compute("sschn").get should equal ("ssn") + NysiisAlgorithm.compute("pph").get should equal ("pf") + + // Miscellaneous. + NysiisAlgorithm.compute("macdonald").get should equal ("mcdanald") + NysiisAlgorithm.compute("phone").get should equal ("ffan") + NysiisAlgorithm.compute("aggregate").get should equal ("agragat") + NysiisAlgorithm.compute("accuracy").get should equal ("acaracy") + NysiisAlgorithm.compute("encyclopedia").get should equal ("encyclapad") + NysiisAlgorithm.compute("honorificabilitudinitatibus").get should equal ("hanarafacabalatadanatatab") + NysiisAlgorithm.compute("antidisestablishmentarianism").get should equal ("antadasastablasnantaranasn") + + // Dropby. + NysiisAlgorithm.compute("macintosh").get should equal ("mcant") + NysiisAlgorithm.compute("knuth").get should equal ("nnat") + NysiisAlgorithm.compute("koehn").get should equal ("can") // dropby wrongly says c + NysiisAlgorithm.compute("phillipson").get should equal ("ffalapsan") + NysiisAlgorithm.compute("pfeister").get should equal ("ffastar") + NysiisAlgorithm.compute("schoenhoeft").get should equal ("ssanaft") + NysiisAlgorithm.compute("mckee").get should equal ("mcy") + NysiisAlgorithm.compute("heitschmedt").get should equal ("hatsnad") + NysiisAlgorithm.compute("bart").get should equal ("bad") + NysiisAlgorithm.compute("hurd").get should equal ("had") + NysiisAlgorithm.compute("hunt").get should equal ("had") + NysiisAlgorithm.compute("westerlund").get should equal ("wastarlad") + NysiisAlgorithm.compute("casstevens").get should equal ("castafan") + NysiisAlgorithm.compute("vasquez").get should equal ("vasg") + NysiisAlgorithm.compute("frazier").get should equal ("frasar") + NysiisAlgorithm.compute("bowman").get should equal ("banan") + NysiisAlgorithm.compute("mcknight").get should equal ("mcnagt") + NysiisAlgorithm.compute("rickert").get should equal ("racad") + NysiisAlgorithm.compute("deutsch").get should equal ("dat") // dropby wrongly says dats + NysiisAlgorithm.compute("westphal").get should equal ("wastfal") + NysiisAlgorithm.compute("shriver").get should equal ("shravar") + NysiisAlgorithm.compute("kuhl").get should equal ("cal") // dropby wrongly says c + NysiisAlgorithm.compute("rawson").get should equal ("rasan") + NysiisAlgorithm.compute("jiles").get should equal ("jal") + NysiisAlgorithm.compute("carraway").get should equal ("caray") + NysiisAlgorithm.compute("yamada").get should equal ("yanad") + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/NysiisMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/NysiisMetricSpec.scala new file mode 100755 index 0000000..7e52101 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/NysiisMetricSpec.scala @@ -0,0 +1,37 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class NysiisMetricSpec extends ScalaTest { + "NysiisMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + NysiisMetric.compare("", "").isDefined should be (false) + NysiisMetric.compare("abc", "").isDefined should be (false) + NysiisMetric.compare("", "xyz").isDefined should be (false) + } + } + "non-phonetic arguments" should returns { + "None" in { + NysiisMetric.compare("123", "123").isDefined should be (false) + NysiisMetric.compare("123", "").isDefined should be (false) + NysiisMetric.compare("", "123").isDefined should be (false) + } + } + "phonetically similar arguments" should returns { + "Boolean indicating true" in { + NysiisMetric.compare("ham", "hum").get should be (true) + } + } + "phonetically dissimilar arguments" should returns { + "Boolean indicating false" in { + NysiisMetric.compare("dumb", "gum").get should be (false) + } + } + } + } +} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithmSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithmSpec.scala new file mode 100755 index 0000000..9602f16 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisAlgorithmSpec.scala @@ -0,0 +1,206 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class RefinedNysiisAlgorithmSpec extends ScalaTest { "RefinedNysiisAlgorithm" should provide { + "compute method" when passed { + "empty argument" should returns { + "None" in { + RefinedNysiisAlgorithm.compute("").isDefined should be (false) + } + } + "non-phonetic argument" should returns { + "None" in { + RefinedNysiisAlgorithm.compute("123").isDefined should be (false) + } + } + "phonetic argument" should returns { + "Some" in { + // a + RefinedNysiisAlgorithm.compute("a").get should equal ("a") + RefinedNysiisAlgorithm.compute("aa").get should equal ("a") + + // b + RefinedNysiisAlgorithm.compute("b").get should equal ("b") + RefinedNysiisAlgorithm.compute("bb").get should equal ("b") + + // c + RefinedNysiisAlgorithm.compute("c").get should equal ("c") + RefinedNysiisAlgorithm.compute("cc").get should equal ("c") + + // d + RefinedNysiisAlgorithm.compute("d").get should equal ("d") + RefinedNysiisAlgorithm.compute("dd").get should equal ("d") + + // e + RefinedNysiisAlgorithm.compute("e").get should equal ("e") + RefinedNysiisAlgorithm.compute("ee").get should equal ("y") + + // f + RefinedNysiisAlgorithm.compute("f").get should equal ("f") + RefinedNysiisAlgorithm.compute("ff").get should equal ("f") + + // g + RefinedNysiisAlgorithm.compute("g").get should equal ("g") + RefinedNysiisAlgorithm.compute("gg").get should equal ("g") + + // h + RefinedNysiisAlgorithm.compute("h").get should equal ("h") + RefinedNysiisAlgorithm.compute("hh").get should equal ("h") + + // i + RefinedNysiisAlgorithm.compute("i").get should equal ("i") + RefinedNysiisAlgorithm.compute("ii").get should equal ("i") + + // j + RefinedNysiisAlgorithm.compute("j").get should equal ("j") + RefinedNysiisAlgorithm.compute("jj").get should equal ("j") + + // k + RefinedNysiisAlgorithm.compute("k").get should equal ("c") + RefinedNysiisAlgorithm.compute("kk").get should equal ("c") + + // l + RefinedNysiisAlgorithm.compute("l").get should equal ("l") + RefinedNysiisAlgorithm.compute("ll").get should equal ("l") + + // m + RefinedNysiisAlgorithm.compute("m").get should equal ("m") + RefinedNysiisAlgorithm.compute("mm").get should equal ("mn") + + // n + RefinedNysiisAlgorithm.compute("n").get should equal ("n") + RefinedNysiisAlgorithm.compute("nn").get should equal ("n") + + // o + RefinedNysiisAlgorithm.compute("o").get should equal ("o") + RefinedNysiisAlgorithm.compute("oo").get should equal ("o") + + // p + RefinedNysiisAlgorithm.compute("p").get should equal ("p") + RefinedNysiisAlgorithm.compute("pp").get should equal ("p") + + // q + RefinedNysiisAlgorithm.compute("q").get should equal ("q") + RefinedNysiisAlgorithm.compute("qq").get should equal ("qg") + + // r + RefinedNysiisAlgorithm.compute("r").get should equal ("r") + RefinedNysiisAlgorithm.compute("rr").get should equal ("r") + + // s + RefinedNysiisAlgorithm.compute("s").get should equal ("s") + RefinedNysiisAlgorithm.compute("ss").get should equal ("s") + + // t + RefinedNysiisAlgorithm.compute("t").get should equal ("t") + RefinedNysiisAlgorithm.compute("tt").get should equal ("t") + + // u + RefinedNysiisAlgorithm.compute("u").get should equal ("u") + RefinedNysiisAlgorithm.compute("uu").get should equal ("u") + + // v + RefinedNysiisAlgorithm.compute("v").get should equal ("v") + RefinedNysiisAlgorithm.compute("vv").get should equal ("v") + + // w + RefinedNysiisAlgorithm.compute("w").get should equal ("w") + RefinedNysiisAlgorithm.compute("ww").get should equal ("w") + + // x + RefinedNysiisAlgorithm.compute("x").get should equal ("x") + RefinedNysiisAlgorithm.compute("xx").get should equal ("x") + + // y + RefinedNysiisAlgorithm.compute("y").get should equal ("y") + RefinedNysiisAlgorithm.compute("yy").get should equal ("y") + RefinedNysiisAlgorithm.compute("ybyb").get should equal ("ybab") + + // z + RefinedNysiisAlgorithm.compute("z").get should equal ("z") + RefinedNysiisAlgorithm.compute("zz").get should equal ("z") + + // Head cases. + RefinedNysiisAlgorithm.compute("mac").get should equal ("mc") + RefinedNysiisAlgorithm.compute("pf").get should equal ("f") + + // Last cases. + RefinedNysiisAlgorithm.compute("ix").get should equal ("ic") + RefinedNysiisAlgorithm.compute("ex").get should equal ("ec") + RefinedNysiisAlgorithm.compute("ye").get should equal ("y") + RefinedNysiisAlgorithm.compute("ee").get should equal ("y") + RefinedNysiisAlgorithm.compute("ie").get should equal ("y") + RefinedNysiisAlgorithm.compute("dt").get should equal ("d") + RefinedNysiisAlgorithm.compute("rt").get should equal ("d") + RefinedNysiisAlgorithm.compute("rd").get should equal ("d") + RefinedNysiisAlgorithm.compute("nt").get should equal ("d") + RefinedNysiisAlgorithm.compute("nd").get should equal ("d") + + // Core cases. + RefinedNysiisAlgorithm.compute("bevb").get should equal ("bafb") + RefinedNysiisAlgorithm.compute("bghtb").get should equal ("bgtb") + RefinedNysiisAlgorithm.compute("bdgb").get should equal ("bgb") + RefinedNysiisAlgorithm.compute("bphb").get should equal ("bfb") + RefinedNysiisAlgorithm.compute("bknb").get should equal ("bnb") + RefinedNysiisAlgorithm.compute("bshb").get should equal ("bsb") + RefinedNysiisAlgorithm.compute("bschb").get should equal ("bsb") + RefinedNysiisAlgorithm.compute("bywb").get should equal ("bab") + RefinedNysiisAlgorithm.compute("byw").get should equal ("by") + RefinedNysiisAlgorithm.compute("ywb").get should equal ("yb") + RefinedNysiisAlgorithm.compute("bwrb").get should equal ("brb") + + // Transcode cases. + RefinedNysiisAlgorithm.compute("bay").get should equal ("by") + + // Miscellaneous. + RefinedNysiisAlgorithm.compute("macdonald").get should equal ("mcdanald") + RefinedNysiisAlgorithm.compute("phone").get should equal ("fan") + RefinedNysiisAlgorithm.compute("aggregate").get should equal ("agragat") + RefinedNysiisAlgorithm.compute("accuracy").get should equal ("acaracy") + RefinedNysiisAlgorithm.compute("encyclopedia").get should equal ("encaclapad") + RefinedNysiisAlgorithm.compute("honorificabilitudinitatibus").get should equal ("hanarafacabalatadanatatab") + RefinedNysiisAlgorithm.compute("antidisestablishmentarianism").get should equal ("antadasastablasnantaranasn") + + // Dropby. + RefinedNysiisAlgorithm.compute("edwards").get should equal ("edwad") + RefinedNysiisAlgorithm.compute("parez").get should equal ("par") + RefinedNysiisAlgorithm.compute("macintosh").get should equal ("mcantas") + RefinedNysiisAlgorithm.compute("phillipson").get should equal ("falapsan") + RefinedNysiisAlgorithm.compute("haddix").get should equal ("hadac") + RefinedNysiisAlgorithm.compute("essex").get should equal ("esac") + RefinedNysiisAlgorithm.compute("moye").get should equal ("my") + RefinedNysiisAlgorithm.compute("mckee").get should equal ("mcy") + RefinedNysiisAlgorithm.compute("mackie").get should equal ("mcy") + RefinedNysiisAlgorithm.compute("heitschmidt").get should equal ("hatsnad") + RefinedNysiisAlgorithm.compute("bart").get should equal ("bad") + RefinedNysiisAlgorithm.compute("hurd").get should equal ("had") + RefinedNysiisAlgorithm.compute("hunt").get should equal ("had") + RefinedNysiisAlgorithm.compute("westerlund").get should equal ("wastarlad") + RefinedNysiisAlgorithm.compute("evers").get should equal ("evar") + RefinedNysiisAlgorithm.compute("devito").get should equal ("dafat") + RefinedNysiisAlgorithm.compute("rawson").get should equal ("rasan") + RefinedNysiisAlgorithm.compute("shoulders").get should equal ("saldar") + RefinedNysiisAlgorithm.compute("leighton").get should equal ("lagtan") + RefinedNysiisAlgorithm.compute("wooldridge").get should equal ("waldrag") + RefinedNysiisAlgorithm.compute("oliphant").get should equal ("olafad") + RefinedNysiisAlgorithm.compute("hatchett").get should equal ("hatcat") + RefinedNysiisAlgorithm.compute("mcknight").get should equal ("mcnagt") + RefinedNysiisAlgorithm.compute("rickert").get should equal ("racad") + RefinedNysiisAlgorithm.compute("bowman").get should equal ("banan") + RefinedNysiisAlgorithm.compute("vasquez").get should equal ("vasg") + RefinedNysiisAlgorithm.compute("bashaw").get should equal ("bas") + RefinedNysiisAlgorithm.compute("schoenhoeft").get should equal ("sanaft") // dropby wrongly says scanaft + RefinedNysiisAlgorithm.compute("heywood").get should equal ("had") + RefinedNysiisAlgorithm.compute("hayman").get should equal ("hanan") + RefinedNysiisAlgorithm.compute("seawright").get should equal ("saragt") + RefinedNysiisAlgorithm.compute("kratzer").get should equal ("cratsar") + RefinedNysiisAlgorithm.compute("canaday").get should equal ("canady") + RefinedNysiisAlgorithm.compute("crepeau").get should equal ("crap") + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetricSpec.scala new file mode 100755 index 0000000..8c6d9c1 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedNysiisMetricSpec.scala @@ -0,0 +1,35 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class RefinedNysiisMetricSpec extends ScalaTest { "RefinedNysiisMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + RefinedNysiisMetric.compare("", "").isDefined should be (false) + RefinedNysiisMetric.compare("abc", "").isDefined should be (false) + RefinedNysiisMetric.compare("", "xyz").isDefined should be (false) + } + } + "non-phonetic arguments" should returns { + "None" in { + RefinedNysiisMetric.compare("123", "123").isDefined should be (false) + RefinedNysiisMetric.compare("123", "").isDefined should be (false) + RefinedNysiisMetric.compare("", "123").isDefined should be (false) + } + } + "phonetically similar arguments" should returns { + "Boolean indicating true" in { + RefinedNysiisMetric.compare("ham", "hum").get should be (true) + } + } + "phonetically dissimilar arguments" should returns { + "Boolean indicating false" in { + RefinedNysiisMetric.compare("dumb", "gum").get should be (false) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithmSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithmSpec.scala new file mode 100755 index 0000000..ca39da8 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexAlgorithmSpec.scala @@ -0,0 +1,160 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class RefinedSoundexAlgorithmSpec extends ScalaTest { "RefinedSoundexAlgorithm" should provide { + "compute method" when passed { + "empty argument" should returns { + "None" in { + RefinedSoundexAlgorithm.compute("").isDefined should be (false) + } + } + "non-phonetic argument" should returns { + "None" in { + RefinedSoundexAlgorithm.compute("123").isDefined should be (false) + } + } + "phonetic argument" should returns { + "Some" in { + // a + RefinedSoundexAlgorithm.compute("a").get should equal ("a0") + RefinedSoundexAlgorithm.compute("aa").get should equal ("a0") + + // b + RefinedSoundexAlgorithm.compute("b").get should equal ("b1") + RefinedSoundexAlgorithm.compute("bb").get should equal ("b1") + + // c + RefinedSoundexAlgorithm.compute("c").get should equal ("c3") + RefinedSoundexAlgorithm.compute("cc").get should equal ("c3") + + // d + RefinedSoundexAlgorithm.compute("d").get should equal ("d6") + RefinedSoundexAlgorithm.compute("dd").get should equal ("d6") + + // e + RefinedSoundexAlgorithm.compute("e").get should equal ("e0") + RefinedSoundexAlgorithm.compute("ee").get should equal ("e0") + + // f + RefinedSoundexAlgorithm.compute("f").get should equal ("f2") + RefinedSoundexAlgorithm.compute("ff").get should equal ("f2") + + // g + RefinedSoundexAlgorithm.compute("g").get should equal ("g4") + RefinedSoundexAlgorithm.compute("gg").get should equal ("g4") + + // h + RefinedSoundexAlgorithm.compute("h").get should equal ("h0") + RefinedSoundexAlgorithm.compute("hh").get should equal ("h0") + + // i + RefinedSoundexAlgorithm.compute("i").get should equal ("i0") + RefinedSoundexAlgorithm.compute("ii").get should equal ("i0") + + // j + RefinedSoundexAlgorithm.compute("j").get should equal ("j4") + RefinedSoundexAlgorithm.compute("jj").get should equal ("j4") + + // k + RefinedSoundexAlgorithm.compute("k").get should equal ("k3") + RefinedSoundexAlgorithm.compute("kk").get should equal ("k3") + + // l + RefinedSoundexAlgorithm.compute("l").get should equal ("l7") + RefinedSoundexAlgorithm.compute("ll").get should equal ("l7") + + // m + RefinedSoundexAlgorithm.compute("m").get should equal ("m8") + RefinedSoundexAlgorithm.compute("mm").get should equal ("m8") + + // n + RefinedSoundexAlgorithm.compute("n").get should equal ("n8") + RefinedSoundexAlgorithm.compute("nn").get should equal ("n8") + + // o + RefinedSoundexAlgorithm.compute("o").get should equal ("o0") + RefinedSoundexAlgorithm.compute("oo").get should equal ("o0") + + // p + RefinedSoundexAlgorithm.compute("p").get should equal ("p1") + RefinedSoundexAlgorithm.compute("pp").get should equal ("p1") + + // q + RefinedSoundexAlgorithm.compute("q").get should equal ("q5") + RefinedSoundexAlgorithm.compute("qq").get should equal ("q5") + + // r + RefinedSoundexAlgorithm.compute("r").get should equal ("r9") + RefinedSoundexAlgorithm.compute("rr").get should equal ("r9") + + // s + RefinedSoundexAlgorithm.compute("s").get should equal ("s3") + RefinedSoundexAlgorithm.compute("ss").get should equal ("s3") + + // t + RefinedSoundexAlgorithm.compute("t").get should equal ("t6") + RefinedSoundexAlgorithm.compute("tt").get should equal ("t6") + + // u + RefinedSoundexAlgorithm.compute("u").get should equal ("u0") + RefinedSoundexAlgorithm.compute("uu").get should equal ("u0") + + // v + RefinedSoundexAlgorithm.compute("v").get should equal ("v2") + RefinedSoundexAlgorithm.compute("vv").get should equal ("v2") + + // w + RefinedSoundexAlgorithm.compute("w").get should equal ("w0") + RefinedSoundexAlgorithm.compute("ww").get should equal ("w0") + + // x + RefinedSoundexAlgorithm.compute("x").get should equal ("x5") + RefinedSoundexAlgorithm.compute("xx").get should equal ("x5") + + // y + RefinedSoundexAlgorithm.compute("y").get should equal ("y0") + RefinedSoundexAlgorithm.compute("yy").get should equal ("y0") + + // z + RefinedSoundexAlgorithm.compute("z").get should equal ("z5") + RefinedSoundexAlgorithm.compute("zz").get should equal ("z5") + + // Starting with letter then numbers. + RefinedSoundexAlgorithm.compute("x123456").get should equal ("x5") + RefinedSoundexAlgorithm.compute("a123456").get should equal ("a0") + RefinedSoundexAlgorithm.compute("f123456").get should equal ("f2") + + // Miscellaneous. + RefinedSoundexAlgorithm.compute("braz").get should equal ("b1905") + RefinedSoundexAlgorithm.compute("broz").get should equal ("b1905") + RefinedSoundexAlgorithm.compute("caren").get should equal ("c30908") + RefinedSoundexAlgorithm.compute("carren").get should equal ("c30908") + RefinedSoundexAlgorithm.compute("coram").get should equal ("c30908") + RefinedSoundexAlgorithm.compute("corran").get should equal ("c30908") + RefinedSoundexAlgorithm.compute("curreen").get should equal ("c30908") + RefinedSoundexAlgorithm.compute("curwen").get should equal ("c30908") + RefinedSoundexAlgorithm.compute("hairs").get should equal ("h093") + RefinedSoundexAlgorithm.compute("hark").get should equal ("h093") + RefinedSoundexAlgorithm.compute("hars").get should equal ("h093") + RefinedSoundexAlgorithm.compute("hayers").get should equal ("h093") + RefinedSoundexAlgorithm.compute("heers").get should equal ("h093") + RefinedSoundexAlgorithm.compute("hiers").get should equal ("h093") + RefinedSoundexAlgorithm.compute("lambard").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lambart").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lambert").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lambird").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lampaert").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lampart").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lamport").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("limbert").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("lombard").get should equal ("l7081096") + RefinedSoundexAlgorithm.compute("nolton").get should equal ("n807608") + RefinedSoundexAlgorithm.compute("noulton").get should equal ("n807608") + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetricSpec.scala new file mode 100755 index 0000000..84f547a --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/RefinedSoundexMetricSpec.scala @@ -0,0 +1,35 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class RefinedSoundexMetricSpec extends ScalaTest { "RefinedSoundexMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + RefinedSoundexMetric.compare("", "").isDefined should be (false) + RefinedSoundexMetric.compare("abc", "").isDefined should be (false) + RefinedSoundexMetric.compare("", "xyz").isDefined should be (false) + } + } + "non-phonetic arguments" should returns { + "None" in { + RefinedSoundexMetric.compare("123", "123").isDefined should be (false) + RefinedSoundexMetric.compare("123", "").isDefined should be (false) + RefinedSoundexMetric.compare("", "123").isDefined should be (false) + } + } + "phonetically similar arguments" should returns { + "Boolean indicating true" in { + RefinedSoundexMetric.compare("robert", "rupert").get should be (true) + } + } + "phonetically dissimilar arguments" should returns { + "Boolean indicating false" in { + RefinedSoundexMetric.compare("robert", "rubin").get should be (false) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithmSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithmSpec.scala new file mode 100755 index 0000000..157a24b --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/SoundexAlgorithmSpec.scala @@ -0,0 +1,159 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class SoundexAlgorithmSpec extends ScalaTest { "SoundexAlgorithm" should provide { + "compute method" when passed { + "empty argument" should returns { + "None" in { + SoundexAlgorithm.compute("").isDefined should be (false) + } + } + "non-phonetic argument" should returns { + "None" in { + SoundexAlgorithm.compute("123").isDefined should be (false) + } + } + "phonetic argument" should returns { + "Some" in { + // a + SoundexAlgorithm.compute("a").get should equal ("a000") + SoundexAlgorithm.compute("aa").get should equal ("a000") + + // b + SoundexAlgorithm.compute("b").get should equal ("b000") + SoundexAlgorithm.compute("bb").get should equal ("b000") + + // c + SoundexAlgorithm.compute("c").get should equal ("c000") + SoundexAlgorithm.compute("cc").get should equal ("c000") + + // d + SoundexAlgorithm.compute("d").get should equal ("d000") + SoundexAlgorithm.compute("dd").get should equal ("d000") + + // e + SoundexAlgorithm.compute("e").get should equal ("e000") + SoundexAlgorithm.compute("ee").get should equal ("e000") + + // f + SoundexAlgorithm.compute("f").get should equal ("f000") + SoundexAlgorithm.compute("ff").get should equal ("f000") + + // g + SoundexAlgorithm.compute("g").get should equal ("g000") + SoundexAlgorithm.compute("gg").get should equal ("g000") + + // h + SoundexAlgorithm.compute("h").get should equal ("h000") + SoundexAlgorithm.compute("hh").get should equal ("h000") + + // i + SoundexAlgorithm.compute("i").get should equal ("i000") + SoundexAlgorithm.compute("ii").get should equal ("i000") + + // j + SoundexAlgorithm.compute("j").get should equal ("j000") + SoundexAlgorithm.compute("jj").get should equal ("j000") + + // k + SoundexAlgorithm.compute("k").get should equal ("k000") + SoundexAlgorithm.compute("kk").get should equal ("k000") + + // l + SoundexAlgorithm.compute("l").get should equal ("l000") + SoundexAlgorithm.compute("ll").get should equal ("l000") + + // m + SoundexAlgorithm.compute("m").get should equal ("m000") + SoundexAlgorithm.compute("mm").get should equal ("m000") + + // n + SoundexAlgorithm.compute("n").get should equal ("n000") + SoundexAlgorithm.compute("nn").get should equal ("n000") + + // o + SoundexAlgorithm.compute("o").get should equal ("o000") + SoundexAlgorithm.compute("oo").get should equal ("o000") + + // p + SoundexAlgorithm.compute("p").get should equal ("p000") + SoundexAlgorithm.compute("pp").get should equal ("p000") + + // q + SoundexAlgorithm.compute("q").get should equal ("q000") + SoundexAlgorithm.compute("qq").get should equal ("q000") + + // r + SoundexAlgorithm.compute("r").get should equal ("r000") + SoundexAlgorithm.compute("rr").get should equal ("r000") + + // s + SoundexAlgorithm.compute("s").get should equal ("s000") + SoundexAlgorithm.compute("ss").get should equal ("s000") + + // t + SoundexAlgorithm.compute("t").get should equal ("t000") + SoundexAlgorithm.compute("tt").get should equal ("t000") + + // u + SoundexAlgorithm.compute("u").get should equal ("u000") + SoundexAlgorithm.compute("uu").get should equal ("u000") + + // v + SoundexAlgorithm.compute("v").get should equal ("v000") + SoundexAlgorithm.compute("vv").get should equal ("v000") + + // w + SoundexAlgorithm.compute("w").get should equal ("w000") + SoundexAlgorithm.compute("ww").get should equal ("w000") + + // x + SoundexAlgorithm.compute("x").get should equal ("x000") + SoundexAlgorithm.compute("xx").get should equal ("x000") + + // y + SoundexAlgorithm.compute("y").get should equal ("y000") + SoundexAlgorithm.compute("yy").get should equal ("y000") + + // z + SoundexAlgorithm.compute("z").get should equal ("z000") + SoundexAlgorithm.compute("zz").get should equal ("z000") + + // Starting with letter then numbers. + SoundexAlgorithm.compute("x123456").get should equal ("x000") + SoundexAlgorithm.compute("a123456").get should equal ("a000") + SoundexAlgorithm.compute("f123456").get should equal ("f000") + + // Miscellaneous. + SoundexAlgorithm.compute("abc").get should equal ("a120") + SoundexAlgorithm.compute("xyz").get should equal ("x200") + SoundexAlgorithm.compute("robert").get should equal ("r163") + SoundexAlgorithm.compute("rupert").get should equal ("r163") + SoundexAlgorithm.compute("rubin").get should equal ("r150") + SoundexAlgorithm.compute("ashcraft").get should equal ("a261") + SoundexAlgorithm.compute("tymczak").get should equal ("t522") + SoundexAlgorithm.compute("pfister").get should equal ("p236") + SoundexAlgorithm.compute("euler").get should equal ("e460") + SoundexAlgorithm.compute("gauss").get should equal ("g200") + SoundexAlgorithm.compute("hilbert").get should equal ("h416") + SoundexAlgorithm.compute("knuth").get should equal ("k530") + SoundexAlgorithm.compute("lloyd").get should equal ("l300") + SoundexAlgorithm.compute("lukasiewicz").get should equal ("l222") + SoundexAlgorithm.compute("ashcroft").get should equal ("a261") + SoundexAlgorithm.compute("tymczak").get should equal ("t522") + SoundexAlgorithm.compute("pfister").get should equal ("p236") + SoundexAlgorithm.compute("ellery").get should equal ("e460") + SoundexAlgorithm.compute("ghosh").get should equal ("g200") + SoundexAlgorithm.compute("heilbronn").get should equal ("h416") + SoundexAlgorithm.compute("kant").get should equal ("k530") + SoundexAlgorithm.compute("ladd").get should equal ("l300") + SoundexAlgorithm.compute("lissajous").get should equal ("l222") + SoundexAlgorithm.compute("fusedale").get should equal ("f234") + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/phonetic/SoundexMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/SoundexMetricSpec.scala new file mode 100755 index 0000000..b903add --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/phonetic/SoundexMetricSpec.scala @@ -0,0 +1,35 @@ +package com.rockymadden.stringmetric.phonetic + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class SoundexMetricSpec extends ScalaTest { "SoundexMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + SoundexMetric.compare("", "").isDefined should be (false) + SoundexMetric.compare("abc", "").isDefined should be (false) + SoundexMetric.compare("", "xyz").isDefined should be (false) + } + } + "non-phonetic arguments" should returns { + "None" in { + SoundexMetric.compare("123", "123").isDefined should be (false) + SoundexMetric.compare("123", "").isDefined should be (false) + SoundexMetric.compare("", "123").isDefined should be (false) + } + } + "phonetically similar arguments" should returns { + "Boolean indicating true" in { + SoundexMetric.compare("robert", "rupert").get should be (true) + } + } + "phonetically dissimilar arguments" should returns { + "Boolean indicating false" in { + SoundexMetric.compare("robert", "rubin").get should be (false) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricSpec.scala new file mode 100755 index 0000000..b7a3b58 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/DiceSorensenMetricSpec.scala @@ -0,0 +1,60 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class DiceSorensenMetricSpec extends ScalaTest { "DiceSorensenMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + DiceSorensenMetric(1).compare("", "").isDefined should be (false) + DiceSorensenMetric(1).compare("abc", "").isDefined should be (false) + DiceSorensenMetric(1).compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "1" in { + DiceSorensenMetric(1).compare("abc", "abc").get should be (1) + DiceSorensenMetric(2).compare("abc", "abc").get should be (1) + DiceSorensenMetric(2).compare("abc", "abc").get should be (1) + } + } + "unequal arguments" should returns { + "0" in { + DiceSorensenMetric(1).compare("abc", "xyz").get should be (0) + DiceSorensenMetric(2).compare("abc", "xyz").get should be (0) + DiceSorensenMetric(3).compare("abc", "xyz").get should be (0) + } + } + "invalid arguments" should returns { + "None" in { + DiceSorensenMetric(2).compare("n", "naght").isDefined should be (false) + DiceSorensenMetric(2).compare("night", "n").isDefined should be (false) + DiceSorensenMetric(3).compare("ni", "naght").isDefined should be (false) + DiceSorensenMetric(3).compare("night", "na").isDefined should be (false) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + DiceSorensenMetric(1).compare("night", "nacht").get should be (0.6) + DiceSorensenMetric(1).compare("night", "naght").get should be (0.8) + DiceSorensenMetric(1).compare("context", "contact").get should be (0.7142857142857143) + + DiceSorensenMetric(2).compare("night", "nacht").get should be (0.25) + DiceSorensenMetric(2).compare("night", "naght").get should be (0.5) + DiceSorensenMetric(2).compare("context", "contact").get should be (0.5) + DiceSorensenMetric(2).compare("contextcontext", "contact").get should be (0.3157894736842105) + DiceSorensenMetric(2).compare("context", "contactcontact").get should be (0.3157894736842105) + DiceSorensenMetric(2).compare("ht", "nacht").get should be (0.4) + DiceSorensenMetric(2).compare("xp", "nacht").get should be (0) + DiceSorensenMetric(2).compare("ht", "hththt").get should be (0.3333333333333333) + + DiceSorensenMetric(3).compare("night", "nacht").get should be (0) + DiceSorensenMetric(3).compare("night", "naght").get should be (0.3333333333333333) + DiceSorensenMetric(3).compare("context", "contact").get should be (0.4) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/HammingMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/HammingMetricSpec.scala new file mode 100755 index 0000000..abb56db --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/HammingMetricSpec.scala @@ -0,0 +1,37 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class HammingMetricSpec extends ScalaTest { "HammingMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + HammingMetric.compare("", "").isDefined should be (false) + HammingMetric.compare("abc", "").isDefined should be (false) + HammingMetric.compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "0" in { + HammingMetric.compare("abc", "abc").get should be (0) + HammingMetric.compare("123", "123").get should be (0) + } + } + "unequal arguments" should returns { + "Int indicating distance" in { + HammingMetric.compare("abc", "xyz").get should be (3) + HammingMetric.compare("123", "456").get should be (3) + } + } + "valid arguments" should returns { + "Int indicating distance" in { + HammingMetric.compare("toned", "roses").get should be (3) + HammingMetric.compare("1011101", "1001001").get should be (2) + HammingMetric.compare("2173896", "2233796").get should be (3) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaccardMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaccardMetricSpec.scala new file mode 100755 index 0000000..4c04193 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaccardMetricSpec.scala @@ -0,0 +1,60 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class JaccardMetricSpec extends ScalaTest { "JaccardMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + JaccardMetric(1).compare("", "").isDefined should be (false) + JaccardMetric(1).compare("abc", "").isDefined should be (false) + JaccardMetric(1).compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "1" in { + JaccardMetric(1).compare("abc", "abc").get should be (1) + JaccardMetric(2).compare("abc", "abc").get should be (1) + JaccardMetric(3).compare("abc", "abc").get should be (1) + } + } + "unequal arguments" should returns { + "0" in { + JaccardMetric(1).compare("abc", "xyz").get should be (0) + JaccardMetric(2).compare("abc", "xyz").get should be (0) + JaccardMetric(3).compare("abc", "xyz").get should be (0) + } + } + "invalid arguments" should returns { + "None" in { + JaccardMetric(2).compare("n", "naght").isDefined should be (false) + JaccardMetric(2).compare("night", "n").isDefined should be (false) + JaccardMetric(3).compare("ni", "naght").isDefined should be (false) + JaccardMetric(3).compare("night", "na").isDefined should be (false) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + JaccardMetric(1).compare("night", "nacht").get should be (0.42857142857142855) + JaccardMetric(1).compare("night", "naght").get should be (0.6666666666666666) + JaccardMetric(1).compare("context", "contact").get should be (0.5555555555555556) + + JaccardMetric(2).compare("night", "nacht").get should be (0.14285714285714285) + JaccardMetric(2).compare("night", "naght").get should be (0.3333333333333333) + JaccardMetric(2).compare("context", "contact").get should be (0.3333333333333333) + JaccardMetric(2).compare("contextcontext", "contact").get should be (0.1875) + JaccardMetric(2).compare("context", "contactcontact").get should be (0.1875) + JaccardMetric(2).compare("ht", "nacht").get should be (0.25) + JaccardMetric(2).compare("xp", "nacht").get should be (0) + JaccardMetric(2).compare("ht", "hththt").get should be (0.2) + + JaccardMetric(3).compare("night", "nacht").get should be (0) + JaccardMetric(3).compare("night", "naght").get should be (0.2) + JaccardMetric(3).compare("context", "contact").get should be (0.25) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaroMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaroMetricSpec.scala new file mode 100755 index 0000000..26ae38e --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaroMetricSpec.scala @@ -0,0 +1,51 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class JaroMetricSpec extends ScalaTest { "JaroMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + JaroMetric.compare("", "").isDefined should be (false) + JaroMetric.compare("abc", "").isDefined should be (false) + JaroMetric.compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "1" in { + JaroMetric.compare("a", "a").get should be (1) + JaroMetric.compare("abc", "abc").get should be (1) + JaroMetric.compare("123", "123").get should be (1) + } + } + "unequal arguments" should returns { + "0" in { + JaroMetric.compare("abc", "xyz").get should be (0) + JaroMetric.compare("123", "456").get should be (0) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + JaroMetric.compare("aa", "a").get should be (0.8333333333333334) + JaroMetric.compare("a", "aa").get should be (0.8333333333333334) + JaroMetric.compare("veryveryverylong", "v").get should be (0.6875) + JaroMetric.compare("v", "veryveryverylong").get should be (0.6875) + JaroMetric.compare("martha", "marhta").get should be (0.9444444444444445) + JaroMetric.compare("dwayne", "duane").get should be (0.8222222222222223) + JaroMetric.compare("dixon", "dicksonx").get should be (0.7666666666666666) + JaroMetric.compare("abcvwxyz", "cabvwxyz").get should be (0.9583333333333334) + JaroMetric.compare("jones", "johnson").get should be (0.7904761904761904) + JaroMetric.compare("henka", "henkan").get should be (0.9444444444444445) + JaroMetric.compare("fvie", "ten").get should be (0) + + JaroMetric.compare("zac ephron", "zac efron").get should be > + JaroMetric.compare("zac ephron", "kai ephron").get + JaroMetric.compare("brittney spears", "britney spears").get should be > + JaroMetric.compare("brittney spears", "brittney startzman").get + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricSpec.scala new file mode 100755 index 0000000..d645456 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/JaroWinklerMetricSpec.scala @@ -0,0 +1,51 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class JaroWinklerMetricSpec extends ScalaTest { "JaroWinklerMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + JaroWinklerMetric.compare("", "").isDefined should be (false) + JaroWinklerMetric.compare("abc", "").isDefined should be (false) + JaroWinklerMetric.compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "1" in { + JaroWinklerMetric.compare("a", "a").get should be (1) + JaroWinklerMetric.compare("abc", "abc").get should be (1) + JaroWinklerMetric.compare("123", "123").get should be (1) + } + } + "unequal arguments" should returns { + "0" in { + JaroWinklerMetric.compare("abc", "xyz").get should be (0) + JaroWinklerMetric.compare("123", "456").get should be (0) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + JaroWinklerMetric.compare("aa", "a").get should be (0.8500000000000001) + JaroWinklerMetric.compare("a", "aa").get should be (0.8500000000000001) + JaroWinklerMetric.compare("veryveryverylong", "v").get should be (0.71875) + JaroWinklerMetric.compare("v", "veryveryverylong").get should be (0.71875) + JaroWinklerMetric.compare("martha", "marhta").get should be (0.9611111111111111) + JaroWinklerMetric.compare("dwayne", "duane").get should be (0.8400000000000001) + JaroWinklerMetric.compare("dixon", "dicksonx").get should be (0.8133333333333332) + JaroWinklerMetric.compare("abcvwxyz", "cabvwxyz").get should be (0.9583333333333334) + JaroWinklerMetric.compare("jones", "johnson").get should be (0.8323809523809523) + JaroWinklerMetric.compare("henka", "henkan").get should be (0.9666666666666667) + JaroWinklerMetric.compare("fvie", "ten").get should be (0) + + JaroWinklerMetric.compare("zac ephron", "zac efron").get should be > + JaroWinklerMetric.compare("zac ephron", "kai ephron").get + JaroWinklerMetric.compare("brittney spears", "britney spears").get should be > + JaroWinklerMetric.compare("brittney spears", "brittney startzman").get + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricSpec.scala new file mode 100755 index 0000000..8e1538b --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/LevenshteinMetricSpec.scala @@ -0,0 +1,50 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class LevenshteinMetricSpec extends ScalaTest { "LevenshteinMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + LevenshteinMetric.compare("", "").isDefined should be (false) + LevenshteinMetric.compare("abc", "").isDefined should be (false) + LevenshteinMetric.compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "0" in { + LevenshteinMetric.compare("abc", "abc").get should be (0) + LevenshteinMetric.compare("123", "123").get should be (0) + } + } + "unequal arguments" should returns { + "Int indicating distance" in { + LevenshteinMetric.compare("abc", "xyz").get should be (3) + LevenshteinMetric.compare("123", "456").get should be (3) + } + } + "valid arguments" should returns { + "Int indicating distance" in { + LevenshteinMetric.compare("abc", "a").get should be (2) + LevenshteinMetric.compare("a", "abc").get should be (2) + LevenshteinMetric.compare("abc", "c").get should be (2) + LevenshteinMetric.compare("c", "abc").get should be (2) + LevenshteinMetric.compare("sitting", "kitten").get should be (3) + LevenshteinMetric.compare("kitten", "sitting").get should be (3) + LevenshteinMetric.compare("cake", "drake").get should be (2) + LevenshteinMetric.compare("drake", "cake").get should be (2) + LevenshteinMetric.compare("saturday", "sunday").get should be (3) + LevenshteinMetric.compare("sunday", "saturday").get should be (3) + LevenshteinMetric.compare("book", "back").get should be (2) + LevenshteinMetric.compare("dog", "fog").get should be (1) + LevenshteinMetric.compare("foq", "fog").get should be (1) + LevenshteinMetric.compare("fvg", "fog").get should be (1) + LevenshteinMetric.compare("encyclopedia", "encyclopediaz").get should be (1) + LevenshteinMetric.compare("encyclopediz", "encyclopediaz").get should be (1) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/NGramMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/NGramMetricSpec.scala new file mode 100755 index 0000000..5e1abb8 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/NGramMetricSpec.scala @@ -0,0 +1,60 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class NGramMetricSpec extends ScalaTest { "NGramMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + NGramMetric(1).compare("", "").isDefined should be (false) + NGramMetric(1).compare("abc", "").isDefined should be (false) + NGramMetric(1).compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "1" in { + NGramMetric(1).compare("abc", "abc").get should be (1) + NGramMetric(2).compare("abc", "abc").get should be (1) + NGramMetric(3).compare("abc", "abc").get should be (1) + } + } + "unequal arguments" should returns { + "0" in { + NGramMetric(1).compare("abc", "xyz").get should be (0) + NGramMetric(2).compare("abc", "xyz").get should be (0) + NGramMetric(3).compare("abc", "xyz").get should be (0) + } + } + "invalid arguments" should returns { + "None" in { + NGramMetric(2).compare("n", "naght").isDefined should be (false) + NGramMetric(2).compare("night", "n").isDefined should be (false) + NGramMetric(3).compare("ni", "naght").isDefined should be (false) + NGramMetric(3).compare("night", "na").isDefined should be (false) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + NGramMetric(1).compare("night", "nacht").get should be (0.6) + NGramMetric(1).compare("night", "naght").get should be (0.8) + NGramMetric(1).compare("context", "contact").get should be (0.7142857142857143) + + NGramMetric(2).compare("night", "nacht").get should be (0.25) + NGramMetric(2).compare("night", "naght").get should be (0.5) + NGramMetric(2).compare("context", "contact").get should be (0.5) + NGramMetric(2).compare("contextcontext", "contact").get should be (0.23076923076923078) + NGramMetric(2).compare("context", "contactcontact").get should be (0.23076923076923078) + NGramMetric(2).compare("ht", "nacht").get should be (0.25) + NGramMetric(2).compare("xp", "nacht").get should be (0) + NGramMetric(2).compare("ht", "hththt").get should be (0.2) + + NGramMetric(3).compare("night", "nacht").get should be (0) + NGramMetric(3).compare("night", "naght").get should be (0.3333333333333333) + NGramMetric(3).compare("context", "contact").get should be (0.4) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/OverlapMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/OverlapMetricSpec.scala new file mode 100755 index 0000000..ec1e7b2 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/OverlapMetricSpec.scala @@ -0,0 +1,62 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class OverlapMetricSpec extends ScalaTest { "OverlapMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + OverlapMetric(1).compare("", "").isDefined should be (false) + OverlapMetric(1).compare("abc", "").isDefined should be (false) + OverlapMetric(1).compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "1" in { + OverlapMetric(1).compare("abc", "abc").get should be (1) + OverlapMetric(2).compare("abc", "abc").get should be (1) + OverlapMetric(3).compare("abc", "abc").get should be (1) + } + } + "unequal arguments" should returns { + "0" in { + OverlapMetric(1).compare("abc", "xyz").get should be (0) + OverlapMetric(2).compare("abc", "xyz").get should be (0) + OverlapMetric(3).compare("abc", "xyz").get should be (0) + } + } + "invalid arguments" should returns { + "None" in { + OverlapMetric(2).compare("n", "naght").isDefined should be (false) + OverlapMetric(2).compare("night", "n").isDefined should be (false) + OverlapMetric(3).compare("ni", "naght").isDefined should be (false) + OverlapMetric(3).compare("night", "na").isDefined should be (false) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + OverlapMetric(1).compare("bob", "bobman").get should be (1) + OverlapMetric(1).compare("bob", "manbobman").get should be (1) + OverlapMetric(1).compare("night", "nacht").get should be (0.6) + OverlapMetric(1).compare("night", "naght").get should be (0.8) + OverlapMetric(1).compare("context", "contact").get should be (0.7142857142857143) + + OverlapMetric(2).compare("night", "nacht").get should be (0.25) + OverlapMetric(2).compare("night", "naght").get should be (0.5) + OverlapMetric(2).compare("context", "contact").get should be (0.5) + OverlapMetric(2).compare("contextcontext", "contact").get should be (0.5) + OverlapMetric(2).compare("context", "contactcontact").get should be (0.5) + OverlapMetric(2).compare("ht", "nacht").get should be (1) + OverlapMetric(2).compare("xp", "nacht").get should be (0) + OverlapMetric(2).compare("ht", "hththt").get should be (1) + + OverlapMetric(3).compare("night", "nacht").get should be (0) + OverlapMetric(3).compare("night", "naght").get should be (0.3333333333333333) + OverlapMetric(3).compare("context", "contact").get should be (0.4) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricSpec.scala new file mode 100755 index 0000000..30bbf49 --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/RatcliffObershelpMetricSpec.scala @@ -0,0 +1,40 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class RatcliffObershelpMetricSpec extends ScalaTest { "RatcliffObershelpMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + RatcliffObershelpMetric.compare("", "").isDefined should be (false) + RatcliffObershelpMetric.compare("abc", "").isDefined should be (false) + RatcliffObershelpMetric.compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "0" in { + RatcliffObershelpMetric.compare("abc", "abc").get should be (1) + RatcliffObershelpMetric.compare("123", "123").get should be (1) + } + } + "unequal arguments" should returns { + "Double indicating distance" in { + RatcliffObershelpMetric.compare("abc", "xyz").get should be (0) + RatcliffObershelpMetric.compare("123", "456").get should be (0) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + RatcliffObershelpMetric.compare("aleksander", "alexandre").get should be (0.7368421052631579) + RatcliffObershelpMetric.compare("alexandre", "aleksander").get should be (0.7368421052631579) + RatcliffObershelpMetric.compare("pennsylvania", "pencilvaneya").get should be (0.6666666666666666) + RatcliffObershelpMetric.compare("pencilvaneya", "pennsylvania").get should be (0.6666666666666666) + RatcliffObershelpMetric.compare("abcefglmn", "abefglmo").get should be (0.8235294117647058) + RatcliffObershelpMetric.compare("abefglmo", "abcefglmn").get should be (0.8235294117647058) + } + } + } +}} diff --git a/core/src/test/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricSpec.scala b/core/src/test/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricSpec.scala new file mode 100755 index 0000000..070458a --- /dev/null +++ b/core/src/test/scala/com/rockymadden/stringmetric/similarity/WeightedLevenshteinMetricSpec.scala @@ -0,0 +1,48 @@ +package com.rockymadden.stringmetric.similarity + +import com.rockymadden.stringmetric.ScalaTest +import org.junit.runner.RunWith +import org.scalatest.junit.JUnitRunner + +@RunWith(classOf[JUnitRunner]) +final class WeightedLevenshteinMetricSpec extends ScalaTest { "WeightedLevenshteinMetric" should provide { + "compare method" when passed { + "empty arguments" should returns { + "None" in { + WeightedLevenshteinMetric(10, 0.1, 1).compare("", "").isDefined should be (false) + WeightedLevenshteinMetric(10, 0.1, 1).compare("abc", "").isDefined should be (false) + WeightedLevenshteinMetric(10, 0.1, 1).compare("", "xyz").isDefined should be (false) + } + } + "equal arguments" should returns { + "0" in { + WeightedLevenshteinMetric(10, 0.1, 1).compare("abc", "abc").get should be (0) + WeightedLevenshteinMetric(10, 0.1, 1).compare("123", "123").get should be (0) + } + } + "unequal arguments" should returns { + "Double indicating distance" in { + WeightedLevenshteinMetric(10, 0.1, 1).compare("abc", "xyz").get should be (3) + WeightedLevenshteinMetric(10, 0.1, 1).compare("123", "456").get should be (3) + } + } + "valid arguments" should returns { + "Double indicating distance" in { + WeightedLevenshteinMetric(10, 0.1, 1).compare("az", "z").get should be (10) + WeightedLevenshteinMetric(10, 0.1, 1).compare("z", "az").get should be (0.1) + WeightedLevenshteinMetric(10, 0.1, 1).compare("a", "z").get should be (1) + WeightedLevenshteinMetric(10, 0.1, 1).compare("z", "a").get should be (1) + WeightedLevenshteinMetric(10, 0.1, 1).compare("ab", "yz").get should be (2) + WeightedLevenshteinMetric(10, 0.1, 1).compare("yz", "ab").get should be (2) + WeightedLevenshteinMetric(10, 0.1, 1).compare("0", "0123456789").get should be (0.9) + WeightedLevenshteinMetric(10, 0.1, 1).compare("0123456789", "0").get should be (90) + WeightedLevenshteinMetric(10, 0.1, 1).compare("book", "back").get should be (2) + WeightedLevenshteinMetric(10, 0.1, 1).compare("back", "book").get should be (2) + WeightedLevenshteinMetric(10, 0.1, 1).compare("hosp", "hospital").get should be (0.4) + WeightedLevenshteinMetric(10, 0.1, 1).compare("hospital", "hosp").get should be (40) + WeightedLevenshteinMetric(10, 0.1, 1).compare("clmbs blvd", "columbus boulevard").get should be (0.8) + WeightedLevenshteinMetric(10, 0.1, 1).compare("columbus boulevard", "clmbs blvd").get should be (80) + } + } + } +}} |