Refactor analysis engine

Author: Genbox

Docs/Overview.drawio Docs/.$Architecture - Numeric.drawio.bkpDocs/Overview.drawio renamed to Docs/.$Architecture - Numeric.drawio.bkp

@@ -1,3 +1,5 @@
+using Genbox.FastData.Internal;
+
 namespace Genbox.FastData.Benchmarks.Benchmarks;
 
 [MemoryDiagnoser]
@@ -20,22 +22,22 @@ public void HashSetDedup()
     {
         int[] keys = new int[_intKeys.Length];
         Array.Copy(_intKeys, keys, _intKeys.Length);
-        FastDataGenerator.DeduplicateWithHashSet(keys, Array.Empty<int>(), false, EqualityComparer<int>.Default, out _);
+        Deduplication.DeduplicateWithHashSet(keys, Array.Empty<int>(), false, EqualityComparer<int>.Default, out _);
     }
 
     [Benchmark]
     public void SortDedup()
     {
         int[] keys = new int[_intKeys.Length];
         Array.Copy(_intKeys, keys, _intKeys.Length);
-        FastDataGenerator.DeduplicateWithSort(keys, Array.Empty<int>(), false, EqualityComparer<int>.Default, Comparer<int>.Default, out _);
+        Deduplication.DeduplicateWithSort(keys, Array.Empty<int>(), false, EqualityComparer<int>.Default, Comparer<int>.Default, out _);
     }
 
     [Benchmark]
     public void SortPreserveDedup()
     {
         int[] keys = new int[_intKeys.Length];
         Array.Copy(_intKeys, keys, _intKeys.Length);
-        FastDataGenerator.DeduplicateWithSortPreserveInputOrder(keys, Array.Empty<int>(), false, EqualityComparer<int>.Default, Comparer<int>.Default, out _);
+        Deduplication.DeduplicateWithSortPreserveInputOrder(keys, Array.Empty<int>(), false, EqualityComparer<int>.Default, Comparer<int>.Default, out _);
     }
 }

Docs/Architecture - Numeric.drawio

@@ -1,3 +1,4 @@
+using Genbox.FastData.Config.Analysis;
 using Genbox.FastData.Enums;
 using Genbox.FastData.Internal.Analysis;
 using Genbox.FastData.Internal.Analysis.Analyzers;

Docs/Architecture - String.drawio

@@ -3,12 +3,11 @@
 using BenchmarkDotNet.Jobs;
 using BenchmarkDotNet.Running;
 using BenchmarkDotNet.Validators;
-using Genbox.FastData.Extensions;
+using Genbox.FastData.Config;
+using Genbox.FastData.Enums;
 using Genbox.FastData.Generator.CSharp;
 using Genbox.FastData.Internal.Structures;
-using Genbox.FastData.InternalShared;
 using Genbox.FastData.InternalShared.Helpers;
-using Genbox.FastData.InternalShared.TestClasses;
 
 namespace Genbox.FastData.Benchmarks;
 
@@ -47,7 +46,7 @@ public static void GenerateFiles()
         // This code is used to generate the files in the Generated folder.
 
         int[] sizes = [1, 5, 10, 50, 100, 500, 1000];
-        Type[] structures = [typeof(ArrayStructure<,>), typeof(ConditionalStructure<,>), typeof(BinarySearchStructure<,>), typeof(BinarySearchStructure<,>), typeof(HashTableStructure<,>)];
+        Type[] structures = [typeof(ArrayStructure<,>), typeof(ConditionalStructure<,>), typeof(BinarySearchStructure<,>), typeof(HashTableStructure<,>)];
 
         foreach (Type type in structures)
         {
@@ -61,8 +60,12 @@ private static void DoStructure(Type type, int size)
         Random rng = new Random(42);
         string[] data = Enumerable.Range(0, size).Select(_ => TestHelper.GenerateRandomString(rng, rng.Next(5, 10))).ToArray();
 
-        TestVector<string> vector = new TestVector<string>(type, data, []);
-        string source = TestGenerator.Generate(CSharpCodeGenerator.Create(new CSharpCodeGeneratorConfig("fastdata")), vector);
-        File.WriteAllText("Gen-" + type.GetCleanName() + "-" + size + ".cs", source);
+        CSharpCodeGenerator generator = CSharpCodeGenerator.Create(new CSharpCodeGeneratorConfig("fastdata"));
+
+        StringDataConfig config = new StringDataConfig();
+        config.StructureTypeOverride = type;
+
+        string source = FastDataGenerator.Generate(data, config, generator);
+        File.WriteAllText("Gen-" + type + "-" + size + ".cs", source);
     }
 }

Src/FastData.Benchmarks/Benchmarks/DeduplicationBenchmarks.cs

@@ -0,0 +1,21 @@
+namespace Genbox.FastData.Cli;
+
+public enum StructureType : byte
+{
+    Auto = 0,
+
+    Array,
+    BinarySearch,
+    BinarySearchInterpolation,
+    BitSet,
+    BloomFilter,
+    Conditional,
+    EliasFano,
+    HashTableCompact,
+    HashTablePerfect,
+    HashTable,
+    KeyLength,
+    Range,
+    RrrBitVector,
+    SingleValue
+}

Src/FastData.Benchmarks/Benchmarks/GPerfAnalyzerBenchmarks.cs

@@ -1,3 +1,4 @@
+using Genbox.FastData.Config;
 using Genbox.FastData.Generator.CSharp;
 
 namespace Genbox.FastData.Examples;
@@ -6,7 +7,7 @@ internal static class Program
 {
     private static void Main()
     {
-        FastDataConfig config = new FastDataConfig();
+        StringDataConfig config = new StringDataConfig();
         CSharpCodeGenerator generator = CSharpCodeGenerator.Create(new CSharpCodeGeneratorConfig("Dogs"));
 
         string source = FastDataGenerator.Generate(["Labrador", "German Shepherd", "Golden Retriever"], config, generator);

Src/FastData.Benchmarks/Program.cs

@@ -1,5 +1,6 @@
 using Genbox.FastData.Generator.Extensions;
 using Genbox.FastData.InternalShared.Harness;
+using Genbox.FastData.InternalShared.Harness.Enums;
 using Genbox.FastData.InternalShared.Helpers;
 using Genbox.FastData.InternalShared.TestClasses;
 using static System.Linq.Enumerable;

Src/FastData.Cli/StructureType.cs

@@ -3,6 +3,7 @@
 using Genbox.FastData.Generator.Framework;
 using Genbox.FastData.Generators.Abstracts;
 using Genbox.FastData.InternalShared.Harness;
+using Genbox.FastData.InternalShared.Harness.Enums;
 
 namespace Genbox.FastData.Generator.CPlusPlus.TestHarness;