AyCode.Core/AyCode.Core.Tests/TestModels/BenchmarkTestDataProvider.cs

using AyCode.Core.Serializers.Binaries;
using System.Collections;
using System.Reflection;
using System.Runtime.CompilerServices;

namespace AyCode.Core.Tests.TestModels;

/// <summary>
/// Charset suffix presets for the per-property string augmentation in
/// <c>BenchmarkTestDataProvider.ToLongString</c>. The benchmark applies the configured suffix
/// to every short (≤ <c>FixStrMaxLength</c>) string property across the test data graph (via reflection
/// in <c>BenchmarkTestDataProvider.EnsureAllStringsBypassFixStr</c>), producing long-string
/// benchmark payloads with a controlled UTF-8 content profile.
///
/// Switch by assigning to <see cref="BenchmarkTestDataProvider.LongStringSuffix"/> from the interactive
/// Settings → Charset submenu (or programmatically). The active charset is recorded in the .LLM
/// markdown output header so per-charset bench files are self-documenting.
/// </summary>
public static class CharsetSuffixes
{
    /// <summary>Empty suffix — short Hungarian baseline strings (e.g. "SharedTag_All_True") stay short, hitting
    /// the FixStr fast-path. Stress-test for FixStr / short-string code paths. Note: the baseline
    /// property values remain Hungarian; only the suffix is empty. Despite the "FixAscii" name, this
    /// option does NOT change baseline values to ASCII — it suppresses the suffix that would otherwise
    /// push every property past the FixStr boundary.</summary>
    public const string Latin1FixAscii = "";

    /// <summary>Short Latin1 mixed (Hungarian, ~24 char) — typical European i18n payload, short
    /// multi-byte runs. Below the 32-char FixStr boundary on the suffix alone, but combined with
    /// baseline values pushes every property past it.</summary>
    public const string Latin1Short = " árvíztűrő tükörfúrógép";

    /// <summary>Long Latin1 mixed (~47 char) — exceeds the 32-char FixStr boundary on the suffix alone,
    /// exercising the StringSmall+ tier path with Latin1 mixed content (Hungarian accented letters).</summary>
    public const string Latin1Long = " árvíztűrő tükörfúrógép a magyar betűzés tesztje";

    /// <summary>CJK BMP (Chinese / Japanese / Korean Basic Multilingual Plane) — long homogeneous
    /// 3-byte UTF-8 runs. Primary win region for V4N2 Phase 3 SIMD multi-byte transcoder work.</summary>
    public const string CjkBmp = " 你好世界 こんにちは 안녕하세요";

    /// <summary>Cyrillic (Russian / Ukrainian / etc.) — long homogeneous 2-byte runs, different shape
    /// than Hungarian mixed (where 2-byte chars are short interspersed runs).</summary>
    public const string Cyrillic = " Привет мир дорогой друг";

    /// <summary>Mixed full-spectrum (Hungarian + CJK + Cyrillic + emoji surrogate pairs) — multi-tier
    /// coverage in one payload. Stresses surrogate-pair handling in the UTF-8 transcoder.</summary>
    public const string Mixed = " árvíz 你好 Привет 😀";
}

public static class BenchmarkTestDataProvider
{
    private const int FixStrMaxLength = 31;

    /// <summary>
    /// Active long-string suffix appended to short string properties during benchmark data construction.
    /// Defaults to <see cref="CharsetSuffixes.Latin1Long"/> (~47-char Latin1 mixed) — backward-compatible
    /// in spirit with the prior fixed default (Latin1 mixed family, ~32 char). Switch from
    /// <see cref="CharsetSuffixes"/> to measure other UTF-8 content profiles.
    /// </summary>
    public static string LongStringSuffix = CharsetSuffixes.Latin1Long;

    private sealed class ReferenceComparer : IEqualityComparer<object>
    {
        public static readonly ReferenceComparer Instance = new();
        public new bool Equals(object? x, object? y) => ReferenceEquals(x, y);
        public int GetHashCode(object obj) => RuntimeHelpers.GetHashCode(obj);
    }

    public static List<TestDataSet> CreateTestDataSets(bool resetId = true)
    {
        return new List<TestDataSet>
        {
            CreateSmallTestData(resetId),
            CreateMediumTestData(resetId),
            CreateLargeTestData(resetId),
            CreateRepeatedStringsTestData(resetId),
            CreateDeepNestedTestData(resetId)
        };
    }

    private static TestDataSet CreateSmallTestData(bool resetId = true)
    {
        if (resetId) TestDataFactory.ResetIdCounter();

        var sharedTag = TestDataFactory.CreateTag("SharedTag_All_True");
        var sharedUser = TestDataFactory.CreateUser("shareduser");

        var order = TestDataFactory.CreateOrder(
            itemCount: 2,
            palletsPerItem: 2,
            measurementsPerPallet: 2,
            pointsPerMeasurement: 2,
            sharedTag: sharedTag,
            sharedUser: sharedUser);

        EnsureAllStringsBypassFixStr(order);

        ClearDeepLevelRefs(order);

        return new TestDataSet("Small (2x2x2x2)", order, iidRefPercent: 20);
    }

    private static TestDataSet CreateMediumTestData(bool resetId = true)
    {
        if (resetId) TestDataFactory.ResetIdCounter();

        var sharedTag = TestDataFactory.CreateTag("SharedTag_All_True");
        var sharedUser = TestDataFactory.CreateUser("shareduser");
        var sharedMeta = TestDataFactory.CreateMetadata("shared", withChild: true);

        var sharedPreferences = new UserPreferences_All_True
        {
            Theme = "dark",
            Language = "hungarian",
            NotificationsEnabled = true,
            EmailDigestFrequency = "weekly"
        };
        sharedUser.Preferences = sharedPreferences;

        var order = TestDataFactory.CreateOrder(
            itemCount: 3,
            palletsPerItem: 3,
            measurementsPerPallet: 3,
            pointsPerMeasurement: 4,
            sharedTag: sharedTag,
            sharedUser: sharedUser,
            sharedMetadata: sharedMeta,
            sharedPreferences: sharedPreferences);

        EnsureAllStringsBypassFixStr(order);

        ClearDeepLevelRefs(order);

        return new TestDataSet("Medium (3x3x3x4)", order, iidRefPercent: 20);
    }

    private static TestDataSet CreateLargeTestData(bool resetId = true)
    {
        if (resetId) TestDataFactory.ResetIdCounter();

        var sharedTag = TestDataFactory.CreateTag("SharedTag_All_True");
        var sharedUser = TestDataFactory.CreateUser("shareduser");

        var sharedPreferences = new UserPreferences_All_True
        {
            Theme = "light",
            Language = "german",
            NotificationsEnabled = false,
            EmailDigestFrequency = "daily"
        };
        sharedUser.Preferences = sharedPreferences;

        var order = TestDataFactory.CreateOrder(
            itemCount: 5,
            palletsPerItem: 5,
            measurementsPerPallet: 5,
            pointsPerMeasurement: 10,
            sharedTag: sharedTag,
            sharedUser: sharedUser,
            sharedPreferences: sharedPreferences);

        EnsureAllStringsBypassFixStr(order);

        ClearDeepLevelRefs(order);

        return new TestDataSet("Large (5x5x5x10)", order, iidRefPercent: 20);
    }

    private static TestDataSet CreateRepeatedStringsTestData(bool resetId = true)
    {
        if (resetId) TestDataFactory.ResetIdCounter();

        var sharedTag = TestDataFactory.CreateTag("RepeatedTag");
        var sharedUser = TestDataFactory.CreateUser("repeateduser");

        var sharedPreferences = new UserPreferences_All_True
        {
            Theme = "dark",
            Language = "hungarian",
            NotificationsEnabled = true,
            EmailDigestFrequency = "weekly"
        };
        sharedUser.Preferences = sharedPreferences;

        var order = TestDataFactory.CreateOrder(
            itemCount: 10,
            palletsPerItem: 2,
            measurementsPerPallet: 2,
            pointsPerMeasurement: 2,
            sharedTag: sharedTag,
            sharedUser: sharedUser,
            sharedPreferences: sharedPreferences);

        // Repeated string fields — ProductName on items + PalletCode on pallets. Both are common
        // across the hierarchy, exercising string-interning deduplication on the Default preset
        // (which has UseStringInterning = All). Targeting ~20% repeated-string share overall.
        // Baselines are short ASCII (≤ FixStrMaxLength) so EnsureAllStringsBypassFixStr appends the
        // active CharsetSuffix — the resulting payload's UTF-8 content profile is governed entirely
        // by the selected charset (not contaminated by hard-coded Hungarian baseline values).
        foreach (var item in order.Items)
        {
            item.Status = TestStatus.Processing;
            item.ProductName = "ProductName";

            foreach (var pallet in item.Pallets)
            {
                pallet.PalletCode = "PalletCode";
            }
        }

        EnsureAllStringsBypassFixStr(order);

        ClearDeepLevelRefs(order);

        return new TestDataSet("Repeated Strings (10 items)", order, iidRefPercent: 20);
    }

    private static TestDataSet CreateDeepNestedTestData(bool resetId = true)
    {
        if (resetId) TestDataFactory.ResetIdCounter();

        var sharedTag = TestDataFactory.CreateTag("DeepTag");
        var sharedUser = TestDataFactory.CreateUser("deepuser");
        var sharedCategory = TestDataFactory.CreateCategory("DeepCategory");

        var sharedPreferences = new UserPreferences_All_True
        {
            Theme = "light",
            Language = "french",
            NotificationsEnabled = false,
            EmailDigestFrequency = "monthly"
        };
        sharedUser.Preferences = sharedPreferences;

        var order = TestDataFactory.CreateOrder(
            itemCount: 2,
            palletsPerItem: 4,
            measurementsPerPallet: 4,
            pointsPerMeasurement: 8,
            sharedTag: sharedTag,
            sharedUser: sharedUser,
            sharedPreferences: sharedPreferences,
            sharedCategory: sharedCategory);

        EnsureAllStringsBypassFixStr(order);

        ClearDeepLevelRefs(order);

        return new TestDataSet("Deep Nested (2x4x4x8)", order, iidRefPercent: 20);
    }

    private static void ClearDeepLevelRefs(TestOrder_All_True order)
    {
        // Keep shared IId refs at the pallet level (Tag + Inspector) — these contribute the bulk of
        // the ~20% IId-ref share that the test data targets. Only Category is cleared at this level
        // (one-of-three clears keep the share moderate). The deeper measurement / point levels are
        // cleared entirely so deep-tree ref noise does not skew the share upward beyond ~20%.
        foreach (var item in order.Items)
        {
            foreach (var pallet in item.Pallets)
            {
                // pallet.Tag = null;       // KEEP for ~20% IId-ref share (was cleared)
                // pallet.Inspector = null; // KEEP for ~20% IId-ref share (was cleared)
                pallet.Category = null;

                foreach (var measurement in pallet.Measurements)
                {
                    measurement.Tag = null;
                    measurement.Operator = null;

                    foreach (var point in measurement.Points)
                    {
                        point.Tag = null;
                        point.Verifier = null;
                    }
                }
            }
        }
    }

    private static void EnsureAllStringsBypassFixStr(object? root)
    {
        if (root == null) return;

        var visited = new HashSet<object>(ReferenceComparer.Instance);
        var stack = new Stack<object>();
        stack.Push(root);

        while (stack.Count > 0)
        {
            var current = stack.Pop();
            if (!visited.Add(current)) continue;

            if (current is IEnumerable enumerable && current is not string)
            {
                foreach (var item in enumerable)
                {
                    if (item != null)
                        stack.Push(item);
                }
                continue;
            }

            var type = current.GetType();
            foreach (var property in type.GetProperties(BindingFlags.Instance | BindingFlags.Public))
            {
                if (!property.CanRead) continue;

                if (property.PropertyType == typeof(string))
                {
                    if (!property.CanWrite) continue;

                    var value = (string?)property.GetValue(current);
                    property.SetValue(current, ToLongString(value));
                    continue;
                }

                if (property.PropertyType.IsValueType || property.PropertyType.IsEnum)
                    continue;

                var child = property.GetValue(current);
                if (child != null)
                    stack.Push(child);
            }
        }
    }

    [MethodImpl(MethodImplOptions.AggressiveInlining)]
    private static string ToLongString(string? value)
    {
        if (string.IsNullOrEmpty(value))
            return "Benchmark_String_Value" + LongStringSuffix;

        if (value.Length > FixStrMaxLength)
            return value;

        return value + LongStringSuffix;
    }
}

public class TestDataSet<TOrder>
{
    public string Name { get; }
    public TOrder Order { get; }

    /// <summary>
    /// Percentage of IId shared references in the data (0-100).
    /// Higher values mean more deduplication benefit for Default mode.
    /// </summary>
    public int IIdRefPercent { get; }

    public TestDataSet(string name, TOrder order, int iidRefPercent = 0)
    {
        Name = name;
        Order = order;
        IIdRefPercent = iidRefPercent;
    }

    public string TypeName => Order.GetType().Name;

    /// <summary>
    /// Gets display name including IId ref percentage if set.
    /// </summary>
    public string DisplayName => IIdRefPercent > 0
        ? $"{Name} [{IIdRefPercent}% IId refs]"
        : Name;
}

public sealed class TestDataSet : TestDataSet<TestOrder_All_True>
{
    public TestDataSet(string name, TestOrder_All_True order, int iidRefPercent = 0)
        : base(name, order, iidRefPercent)
    {
    }
}