WIP

ctolkien · Apr 30, 2020 · ef34172 · ef34172
1 parent 874b638
commit ef34172
Show file tree

Hide file tree

Showing 9 changed files with 215 additions and 9 deletions.
diff --git a/Slugify.sln b/Slugify.sln
@@ -10,6 +10,8 @@ Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Slugify.Core", "src\Slugify
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Slugify.Core.Tests", "tests\Slugify.Core.Tests\Slugify.Core.Tests.csproj", "{A3717955-9FC4-41EB-B389-1A610D64B3B6}"
 EndProject
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "Slugify.Core.Benchmarks", "tests\Slugify.Core.Benchmarks\Slugify.Core.Benchmarks.csproj", "{A859F823-7992-4820-905C-640F3440BAF2}"
+EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution
 		Debug|Any CPU = Debug|Any CPU
@@ -44,13 +46,26 @@ Global
 		{A3717955-9FC4-41EB-B389-1A610D64B3B6}.Release|x64.Build.0 = Release|Any CPU
 		{A3717955-9FC4-41EB-B389-1A610D64B3B6}.Release|x86.ActiveCfg = Release|Any CPU
 		{A3717955-9FC4-41EB-B389-1A610D64B3B6}.Release|x86.Build.0 = Release|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Debug|x64.ActiveCfg = Debug|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Debug|x64.Build.0 = Debug|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Debug|x86.ActiveCfg = Debug|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Debug|x86.Build.0 = Debug|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Release|Any CPU.Build.0 = Release|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Release|x64.ActiveCfg = Release|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Release|x64.Build.0 = Release|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Release|x86.ActiveCfg = Release|Any CPU
+		{A859F823-7992-4820-905C-640F3440BAF2}.Release|x86.Build.0 = Release|Any CPU
 	EndGlobalSection
 	GlobalSection(SolutionProperties) = preSolution
 		HideSolutionNode = FALSE
 	EndGlobalSection
 	GlobalSection(NestedProjects) = preSolution
 		{CC5D3071-57A3-442C-8ACC-E543DA0A8E20} = {321AAEAA-F675-4B15-80B5-3FF2E6A15602}
 		{A3717955-9FC4-41EB-B389-1A610D64B3B6} = {28DEE6D8-053D-4364-9DD5-F3744E58CF54}
+		{A859F823-7992-4820-905C-640F3440BAF2} = {28DEE6D8-053D-4364-9DD5-F3744E58CF54}
 	EndGlobalSection
 	GlobalSection(ExtensibilityGlobals) = postSolution
 		SolutionGuid = {189B8864-9AE7-4F2D-BFAC-5368F94F26E2}

diff --git a/src/Slugify.Core/SlugHelper.cs b/src/Slugify.Core/SlugHelper.cs
@@ -3,6 +3,7 @@
 using System.Globalization;
 using System.Text;
 using System.Text.RegularExpressions;
+using System.Xml;
 
 namespace Slugify
 {
@@ -45,6 +46,72 @@ public string GenerateSlug(string inputString)
             return inputString;
         }
 
+        public string GenerateSlug2(string inputString)
+        {
+            if (_config.TrimWhitespace)
+            {
+                inputString = inputString.Trim();
+            }
+
+            if (_config.ForceLowerCase)
+            {
+                inputString = inputString.ToLower();
+            }
+
+            inputString = CleanWhiteSpace(inputString, _config.CollapseWhiteSpace);
+            inputString = ApplyReplacements(inputString, _config.StringReplacements);
+            inputString = RemoveDiacritics(inputString);
+            inputString = DeleteCharacters(inputString, _config.DeniedCharactersRegex);
+
+            if (_config.CollapseDashes)
+            {
+                inputString = Regex.Replace(inputString, "--+", "-");
+            }
+
+            return inputString;
+
+            string CleanWhiteSpace(string str, bool collapse)
+            {
+                return Regex.Replace(str, collapse ? @"\s+" : @"\s", " ");
+            }
+
+            string RemoveDiacritics(string str)
+            {
+                var stFormD = str.Normalize(NormalizationForm.FormD);
+                var sb = new StringBuilder(stFormD.Length);
+
+
+                for (var ich = 0; ich < stFormD.Length; ich++)
+                {
+                    var uc = CharUnicodeInfo.GetUnicodeCategory(stFormD[ich]);
+                    if (uc != UnicodeCategory.NonSpacingMark)
+                    {
+                        sb.Append(stFormD[ich]);
+                    }
+                }
+
+                return sb.ToString().Normalize(NormalizationForm.FormC);
+            }
+
+
+            string ApplyReplacements(string str, Dictionary<string, string> replacements)
+            {
+                foreach (var replacement in replacements)
+                {
+                    str = str.Replace(replacement.Key, replacement.Value);
+                }
+
+                return str;
+            }
+
+            string DeleteCharacters(string str, string regex)
+            {
+                return Regex.Replace(str, regex, "");
+            }
+
+        }
+
+
         protected string CleanWhiteSpace(string str, bool collapse)
         {
             return Regex.Replace(str, collapse ? @"\s+" : @"\s", " ");
@@ -90,15 +157,11 @@ protected string DeleteCharacters(string str, string regex)
         /// </summary>
         public class Config
         {
-            public Config()
-            {
-                StringReplacements = new Dictionary<string, string>
+            public Dictionary<string, string> StringReplacements { get; set; } = new Dictionary<string, string>
                 {
                     { " ", "-" }
                 };
-            }
-
-            public Dictionary<string, string> StringReplacements { get; set; }
+
             public bool ForceLowerCase { get; set; } = true;
             public bool CollapseWhiteSpace { get; set; } = true;
             public string DeniedCharactersRegex { get; set; } = @"[^a-zA-Z0-9\-\._]";

diff --git a/src/Slugify.Core/Slugify.Core.csproj b/src/Slugify.Core/Slugify.Core.csproj
@@ -34,6 +34,7 @@ With default settings, you will get an hyphenized, lowercase, alphanumeric versi
       <PrivateAssets>all</PrivateAssets>
       <IncludeAssets>runtime; build; native; contentfiles; analyzers; buildtransitive</IncludeAssets>
     </PackageReference>
+    <PackageReference Include="System.Memory" Version="4.5.4" />
   </ItemGroup>
 
 </Project>
diff --git a/...et.Artifacts/results/Slugify.Core.Benchmarks.SlugifyBenchmarks-report-github.md b/...et.Artifacts/results/Slugify.Core.Benchmarks.SlugifyBenchmarks-report-github.md
@@ -0,0 +1,17 @@
+``` ini
+
+BenchmarkDotNet=v0.12.1, OS=Windows 10.0.18363.778 (1909/November2018Update/19H2)
+Intel Core i9-9900K CPU 3.60GHz (Coffee Lake), 1 CPU, 16 logical and 8 physical cores
+.NET Core SDK=3.1.201
+  [Host]     : .NET Core 3.1.3 (CoreCLR 4.700.20.11803, CoreFX 4.700.20.12001), X64 RyuJIT
+  DefaultJob : .NET Core 3.1.3 (CoreCLR 4.700.20.11803, CoreFX 4.700.20.12001), X64 RyuJIT
+
+
+```
+|   Method | N |                 word |       Mean |    Error |   StdDev | Ratio | RatioSD |  Gen 0 | Gen 1 | Gen 2 | Allocated |
+|--------- |-- |--------------------- |-----------:|---------:|---------:|------:|--------:|-------:|------:|------:|----------:|
+| **Baseline** | **1** |                **Hello** |   **660.0 ns** |  **2.01 ns** |  **1.78 ns** |  **1.00** |    **0.00** | **0.0734** |     **-** |     **-** |     **616 B** |
+| Improved | 1 |                Hello |   643.8 ns | 12.74 ns | 11.92 ns |  0.98 |    0.02 | 0.0553 |     - |     - |     464 B |
+|          |   |                      |            |          |          |       |         |        |       |       |           |
+| **Baseline** | **1** | **lorem(...)oot!! [40]** | **4,174.1 ns** | **11.46 ns** | **10.72 ns** |  **1.00** |    **0.00** | **0.4501** |     **-** |     **-** |    **3824 B** |
+| Improved | 1 | lorem(...)oot!! [40] | 3,924.9 ns |  7.03 ns |  5.87 ns |  0.94 |    0.00 | 0.3967 |     - |     - |    3360 B |
diff --git a/...ks/BenchmarkDotNet.Artifacts/results/Slugify.Core.Benchmarks.SlugifyBenchmarks-report.csv b/...ks/BenchmarkDotNet.Artifacts/results/Slugify.Core.Benchmarks.SlugifyBenchmarks-report.csv
@@ -0,0 +1,5 @@
+Method,Job,AnalyzeLaunchVariance,EvaluateOverhead,MaxAbsoluteError,MaxRelativeError,MinInvokeCount,MinIterationTime,OutlierMode,Affinity,EnvironmentVariables,Jit,Platform,PowerPlanMode,Runtime,AllowVeryLargeObjects,Concurrent,CpuGroups,Force,HeapAffinitizeMask,HeapCount,NoAffinitize,RetainVm,Server,Arguments,BuildConfiguration,Clock,EngineFactory,NuGetReferences,Toolchain,IsMutator,InvocationCount,IterationCount,IterationTime,LaunchCount,MaxIterationCount,MaxWarmupIterationCount,MinIterationCount,MinWarmupIterationCount,RunStrategy,UnrollFactor,WarmupCount,N,word,Mean,Error,StdDev,Ratio,RatioSD,Gen 0,Gen 1,Gen 2,Allocated
+Baseline,DefaultJob,False,Default,Default,Default,Default,Default,Default,1111111111111111,Empty,RyuJit,X64,8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c,.NET Core 3.1,False,True,False,True,Default,Default,False,False,False,Default,Default,Default,Default,Default,Default,Default,1,Default,Default,Default,Default,Default,Default,Default,Default,16,Default,1,Hello,660.0 ns,2.01 ns,1.78 ns,1.00,0.00,0.0734,0.0000,0.0000,616 B
+Improved,DefaultJob,False,Default,Default,Default,Default,Default,Default,1111111111111111,Empty,RyuJit,X64,8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c,.NET Core 3.1,False,True,False,True,Default,Default,False,False,False,Default,Default,Default,Default,Default,Default,Default,1,Default,Default,Default,Default,Default,Default,Default,Default,16,Default,1,Hello,643.8 ns,12.74 ns,11.92 ns,0.98,0.02,0.0553,0.0000,0.0000,464 B
+Baseline,DefaultJob,False,Default,Default,Default,Default,Default,Default,1111111111111111,Empty,RyuJit,X64,8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c,.NET Core 3.1,False,True,False,True,Default,Default,False,False,False,Default,Default,Default,Default,Default,Default,Default,1,Default,Default,Default,Default,Default,Default,Default,Default,16,Default,1,lorem(...)oot!! [40],"4,174.1 ns",11.46 ns,10.72 ns,1.00,0.00,0.4501,0.0000,0.0000,3824 B
+Improved,DefaultJob,False,Default,Default,Default,Default,Default,Default,1111111111111111,Empty,RyuJit,X64,8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c,.NET Core 3.1,False,True,False,True,Default,Default,False,False,False,Default,Default,Default,Default,Default,Default,Default,1,Default,Default,Default,Default,Default,Default,Default,Default,16,Default,1,lorem(...)oot!! [40],"3,924.9 ns",7.03 ns,5.87 ns,0.94,0.00,0.3967,0.0000,0.0000,3360 B
diff --git a/...s/BenchmarkDotNet.Artifacts/results/Slugify.Core.Benchmarks.SlugifyBenchmarks-report.html b/...s/BenchmarkDotNet.Artifacts/results/Slugify.Core.Benchmarks.SlugifyBenchmarks-report.html
@@ -0,0 +1,33 @@
+<!DOCTYPE html>
+<html lang='en'>
+<head>
+<meta charset='utf-8' />
+<title>Slugify.Core.Benchmarks.SlugifyBenchmarks-20200430-160656</title>
+
+<style type="text/css">
+	table { border-collapse: collapse; display: block; width: 100%; overflow: auto; }
+	td, th { padding: 6px 13px; border: 1px solid #ddd; text-align: right; }
+	tr { background-color: #fff; border-top: 1px solid #ccc; }
+	tr:nth-child(even) { background: #f8f8f8; }
+</style>
+</head>
+<body>
+<pre><code>
+BenchmarkDotNet=v0.12.1, OS=Windows 10.0.18363.778 (1909/November2018Update/19H2)
+Intel Core i9-9900K CPU 3.60GHz (Coffee Lake), 1 CPU, 16 logical and 8 physical cores
+.NET Core SDK=3.1.201
+  [Host]     : .NET Core 3.1.3 (CoreCLR 4.700.20.11803, CoreFX 4.700.20.12001), X64 RyuJIT
+  DefaultJob : .NET Core 3.1.3 (CoreCLR 4.700.20.11803, CoreFX 4.700.20.12001), X64 RyuJIT
+</code></pre>
+<pre><code></code></pre>
+
+<table>
+<thead><tr><th>Method</th><th>N</th><th>          word</th><th>Mean</th><th>Error</th><th>StdDev</th><th>Ratio</th><th>RatioSD</th><th>Gen 0</th><th>Gen 1</th><th>Gen 2</th><th>Allocated</th>
+</tr>
+</thead><tbody><tr><td>Baseline</td><td>1</td><td>Hello</td><td>660.0 ns</td><td>2.01 ns</td><td>1.78 ns</td><td>1.00</td><td>0.00</td><td>0.0734</td><td>-</td><td>-</td><td>616 B</td>
+</tr><tr><td>Improved</td><td>1</td><td>Hello</td><td>643.8 ns</td><td>12.74 ns</td><td>11.92 ns</td><td>0.98</td><td>0.02</td><td>0.0553</td><td>-</td><td>-</td><td>464 B</td>
+</tr><tr><td>Baseline</td><td>1</td><td>lorem(...)oot!! [40]</td><td>4,174.1 ns</td><td>11.46 ns</td><td>10.72 ns</td><td>1.00</td><td>0.00</td><td>0.4501</td><td>-</td><td>-</td><td>3824 B</td>
+</tr><tr><td>Improved</td><td>1</td><td>lorem(...)oot!! [40]</td><td>3,924.9 ns</td><td>7.03 ns</td><td>5.87 ns</td><td>0.94</td><td>0.00</td><td>0.3967</td><td>-</td><td>-</td><td>3360 B</td>
+</tr></tbody></table>
+</body>
+</html>
diff --git a/tests/Slugify.Core.Benchmarks/Program.cs b/tests/Slugify.Core.Benchmarks/Program.cs
@@ -0,0 +1,55 @@
+using BenchmarkDotNet.Attributes;
+using BenchmarkDotNet.Running;
+using System;
+using System.Collections.Generic;
+using System.Linq;
+using System.Text;
+
+namespace Slugify.Core.Benchmarks
+{
+    internal class Program
+    {
+        private static void Main(string[] args)
+        {
+            var summary = BenchmarkRunner.Run<SlugifyBenchmarks>();
+        }
+    }
+
+    [MemoryDiagnoser]
+    public class SlugifyBenchmarks
+    {
+        private SlugHelper _slugHelper;
+
+        [GlobalSetup]
+        public void GlobalSetup()
+        {
+            _slugHelper = new SlugHelper();
+        }
+
+        [Params(1)]
+        public int N;
+
+        [Params("Hello", "lorem & ipsome Some Other thing!_ woot!!")]
+        public string word;
+
+        [Benchmark(Baseline = true)]
+        public void Baseline()
+        {
+            for (int i = 0; i < N; i++)
+            {
+                _slugHelper.GenerateSlug(word);
+            }
+        }
+
+        [Benchmark]
+        public void Improved()
+        {
+            for (int i = 0; i < N; i++)
+            {
+                _slugHelper.GenerateSlug2(word);
+            }
+        }
+
+
+    }
+}
diff --git a/tests/Slugify.Core.Benchmarks/Slugify.Core.Benchmarks.csproj b/tests/Slugify.Core.Benchmarks/Slugify.Core.Benchmarks.csproj
@@ -0,0 +1,17 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>netcoreapp3.1</TargetFramework>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="BenchmarkDotNet" Version="0.12.1" />
+  </ItemGroup>
+
+  <ItemGroup>
+    <ProjectReference Include="..\..\src\Slugify.Core\Slugify.Core.csproj" />
+  </ItemGroup>
+
+
+</Project>
diff --git a/tests/Slugify.Core.Tests/SlugHelperTests.cs b/tests/Slugify.Core.Tests/SlugHelperTests.cs
@@ -12,7 +12,7 @@ public void TestEmptyConfig()
             var config = new SlugHelper.Config();
             Assert.True(config.ForceLowerCase);
             Assert.True(config.CollapseWhiteSpace);
-            Assert.Equal(1, config.StringReplacements.Count);
+            Assert.Single(config.StringReplacements);
             Assert.NotNull(new Regex(config.DeniedCharactersRegex));
         }
 
@@ -21,7 +21,7 @@ public void TestDefaultConfig()
         {
             var config = new SlugHelper.Config();
 
-            Assert.Equal(1, config.StringReplacements.Count);
+            Assert.Single(config.StringReplacements);
             Assert.Equal("-", config.StringReplacements[" "]);
         }
 
@@ -46,7 +46,7 @@ public void TestLoweCaseEnforcement()
 
             var helper = new SlugHelper();
 
-            Assert.Equal(expected, helper.GenerateSlug(original));
+            Assert.Equal(expected, helper.GenerateSlug2(original));
         }
 
         [Fact]