memory usage improvements

- add console app for testinng allocations - reduce allocations in various places - some refactoring and cleanup - bump version
vforteli · Feb 24, 2024 · c51b7ed · c51b7ed
1 parent d88cac4
commit c51b7ed
Show file tree

Hide file tree

Showing 17 changed files with 450 additions and 1,660 deletions.
diff --git a/ConsoleTester/.config/dotnet-tools.json b/ConsoleTester/.config/dotnet-tools.json
@@ -0,0 +1,12 @@
+{
+  "version": 1,
+  "isRoot": true,
+  "tools": {
+    "dotnet-counters": {
+      "version": "8.0.452401",
+      "commands": [
+        "dotnet-counters"
+      ]
+    }
+  }
+}
diff --git a/ConsoleTester/ConsoleTester.csproj b/ConsoleTester/ConsoleTester.csproj
@@ -0,0 +1,13 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net6.0</TargetFramework>
+    <ImplicitUsings>enable</ImplicitUsings>
+    <Nullable>enable</Nullable>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <ProjectReference Include="..\FuzzySearchNet\FuzzySearchNet.csproj" />
+  </ItemGroup>
+</Project>
diff --git a/ConsoleTester/Program.cs b/ConsoleTester/Program.cs
@@ -0,0 +1,29 @@
+using System.Text;
+using FuzzySearchNet;
+
+// Console.WriteLine("Press enter to begin");
+// Console.ReadLine();
+
+//const string term = "foo";
+const string term2 = "fooo--foo-----fo";
+const string text = "foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--";
+
+var count = 100000;
+
+var stream = new MemoryStream(Encoding.UTF8.GetBytes(text));
+
+Console.WriteLine($"Running fuzzy search with count {count}");
+for (int i = 0; i < count; i++)
+{
+
+    await foreach (var _ in FuzzySearch.FindLevenshteinAsync(term2, stream, new FuzzySearchOptions(3), leaveOpen: true))
+    {
+
+    }
+    stream.Position = 0;
+}
+
+// for (int i = 0; i < count; i++)
+// {
+//     _ = FuzzySearch.FindLevenshtein(term2, text, new FuzzySearchOptions(3)).ToList();
+// }
diff --git a/FuzzySearchNet.Benchmark/BenchmarkFuzzySearch.cs b/FuzzySearchNet.Benchmark/BenchmarkFuzzySearch.cs
@@ -1,4 +1,5 @@
-using BenchmarkDotNet.Attributes;
+using System.Text;
+using BenchmarkDotNet.Attributes;
 
 namespace FuzzySearchNet.Benchmark;
 
@@ -8,20 +9,35 @@ public class BenchmarkFuzzySearch
     private const string term2 = "fooo--foo-----fo";
     private const string text = "foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--foo-----fo--foo-f--fooo--";
 
-    //[Benchmark]
-    //public void SubstitutionOnlyBufferingShort() => FuzzySearch.FindSubstitutionsOnlyBuffering(term, text, 1);
+    public static readonly MemoryStream textStream = new MemoryStream(Encoding.UTF8.GetBytes(text));
 
-    //[Benchmark]
-    //public void SubstitutionOnlyBufferingLong() => FuzzySearch.FindSubstitutionsOnlyBuffering(term2, text, 1);
+    [Benchmark]
+    public void SubstitutionOnlyShort() => FuzzySearch.FindSubstitutionsOnly(term, text, 1);
+
+    [Benchmark]
+    public void SubstitutionOnlyLong() => FuzzySearch.FindSubstitutionsOnly(term2, text, 1);
 
+    [Benchmark]
+    public void SubstitutionOnlyShor_3_distance() => FuzzySearch.FindSubstitutionsOnly(term, text, 3);
+
+    [Benchmark]
+    public void SubstitutionOnlyLong_3_distance() => FuzzySearch.FindSubstitutionsOnly(term2, text, 3);
 
-    //[Benchmark]
-    //public void SubstitutionOnlyBufferingLong3distance() => FuzzySearch.FindSubstitutionsOnlyBuffering(term2, text, 3);
 
 
     [Benchmark]
     public void LevenshteinLong()
     {
         _ = FuzzySearch.FindLevenshtein(term2, text, new FuzzySearchOptions(3)).ToList();
     }
+
+    [Benchmark]
+    public async Task LevenshteinLongAsync()
+    {
+        var stream = new MemoryStream(Encoding.UTF8.GetBytes(text));    // this is not ideal, but the effect of doing this here is basically within the stddev of the benchmark
+        await foreach (var _ in FuzzySearch.FindLevenshteinAsync(term2, stream, new FuzzySearchOptions(3)))
+        {
+
+        }
+    }
 }
diff --git a/FuzzySearchNet.Benchmark/FuzzySearchNet.Benchmark.csproj b/FuzzySearchNet.Benchmark/FuzzySearchNet.Benchmark.csproj
@@ -13,7 +13,7 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="BenchmarkDotNet" Version="0.13.1" />
+    <PackageReference Include="BenchmarkDotNet" Version="0.13.10" />
   </ItemGroup>
 
   <ItemGroup>

diff --git a/FuzzySearchNet.Benchmark/Program.cs b/FuzzySearchNet.Benchmark/Program.cs
@@ -1,71 +1,4 @@
 using BenchmarkDotNet.Running;
 using FuzzySearchNet.Benchmark;
 
-var summary = BenchmarkRunner.Run<BenchmarkFuzzySearch>();
-
-
-
-/*
- *
-
-
-
-|           Method |     Mean |     Error |    StdDev |
-|----------------- |---------:|----------:|----------:|
-| SubstitutionOnly | 4.564 us | 0.0905 us | 0.2551 us |
-
-
-
-
-
-
-|           Method |     Mean |    Error |   StdDev |
-|----------------- |---------:|---------:|---------:|
-| SubstitutionOnly | 18.98 us | 0.373 us | 0.510 us |
-
-
-
-
-
-
-
-
-|                                 Method |      Mean |     Error |    StdDev |
-|--------------------------------------- |----------:|----------:|----------:|
-|         SubstitutionOnlyBufferingShort |  3.704 us | 0.0740 us | 0.0909 us |
-|          SubstitutionOnlyBufferingLong |  2.818 us | 0.0292 us | 0.0244 us |
-| SubstitutionOnlyBufferingLong3distance |  4.396 us | 0.0501 us | 0.0492 us |
-
-
-
-|                                 Method |     Mean |     Error |    StdDev |
-|--------------------------------------- |---------:|----------:|----------:|
-|         SubstitutionOnlyBufferingShort | 2.628 us | 0.0389 us | 0.0570 us |
-|          SubstitutionOnlyBufferingLong | 2.031 us | 0.0404 us | 0.0580 us |
-| SubstitutionOnlyBufferingLong3distance | 2.808 us | 0.0548 us | 0.0988 us |
-
-
-|                                 Method |     Mean |     Error |    StdDev |   Median |
-|--------------------------------------- |---------:|----------:|----------:|---------:|
-|         SubstitutionOnlyBufferingShort | 2.434 us | 0.0264 us | 0.0220 us | 2.438 us |
-|          SubstitutionOnlyBufferingLong | 1.843 us | 0.0366 us | 0.0740 us | 1.810 us |
-| SubstitutionOnlyBufferingLong3distance | 2.471 us | 0.0494 us | 0.1105 us | 2.468 us |
-
-
-
-|                                 Method |     Mean |     Error |    StdDev |   Median |
-|--------------------------------------- |---------:|----------:|----------:|---------:|
-|         SubstitutionOnlyBufferingShort | 2.582 us | 0.0515 us | 0.1233 us | 2.543 us |
-|          SubstitutionOnlyBufferingLong | 2.053 us | 0.0407 us | 0.0743 us | 2.019 us |
-| SubstitutionOnlyBufferingLong3distance | 2.819 us | 0.0559 us | 0.1285 us | 2.782 us |
-
-|                                 Method |     Mean |     Error |    StdDev |   Median |
-|--------------------------------------- |---------:|----------:|----------:|---------:|
-|         SubstitutionOnlyBufferingShort | 2.690 us | 0.0536 us | 0.1539 us | 2.720 us |
-|          SubstitutionOnlyBufferingLong | 2.042 us | 0.0405 us | 0.1074 us | 2.001 us |
-| SubstitutionOnlyBufferingLong3distance | 2.824 us | 0.0563 us | 0.1412 us | 2.754 us |
-
-
-
-
-*/
+_ = BenchmarkRunner.Run<BenchmarkFuzzySearch>();