namespace QuanTAlib.Tests; public class HanmaTests { [Fact] public void Hanma_Constructor_ValidatesInput() { var ex1 = Assert.Throws(() => new Hanma(0)); Assert.Equal("period", ex1.ParamName); var ex2 = Assert.Throws(() => new Hanma(-1)); Assert.Equal("period", ex2.ParamName); var hanma = new Hanma(10); Assert.NotNull(hanma); } [Fact] public void Hanma_Calc_ReturnsValue() { // Note: Hanning window has edge weight = 0, so first value with count=1 // gets zero weight. We need at least 2 values for non-zero result. var hanma = new Hanma(10); hanma.Update(new TValue(DateTime.UtcNow, 100)); TValue result = hanma.Update(new TValue(DateTime.UtcNow, 100)); Assert.True(result.Value > 0); } [Fact] public void Hanma_IsHot_BecomesTrueWhenBufferFull() { var hanma = new Hanma(5); Assert.False(hanma.IsHot); for (int i = 0; i < 4; i++) { hanma.Update(new TValue(DateTime.UtcNow, 100)); Assert.False(hanma.IsHot); } hanma.Update(new TValue(DateTime.UtcNow, 100)); Assert.True(hanma.IsHot); } [Fact] public void Hanma_StreamingMatchesBatch() { var hanmaStreaming = new Hanma(10); var hanmaBatch = new Hanma(10); var gbm = new GBM(startPrice: 100.0, mu: 0.05, sigma: 0.2, seed: 42); var series = new TSeries(); for (int i = 0; i < 100; i++) { var bar = gbm.Next(isNew: true); series.Add(new TValue(bar.Time, bar.Close)); } // Streaming var streamingResults = new TSeries(); Assert.True(series.Count > 0); foreach (var item in series) { streamingResults.Add(hanmaStreaming.Update(item)); } // Batch var batchResults = hanmaBatch.Update(series); Assert.Equal(streamingResults.Count, batchResults.Count); for (int i = 0; i < batchResults.Count; i++) { Assert.Equal(streamingResults[i].Value, batchResults[i].Value, 1e-9); } } [Fact] public void Hanma_StaticCalculate_MatchesInstance() { var series = new TSeries(); var gbm = new GBM(startPrice: 100.0, mu: 0.05, sigma: 0.2, seed: 42); for (int i = 0; i < 100; i++) { var bar = gbm.Next(isNew: true); series.Add(bar.Time, bar.Close); } var instanceResults = new Hanma(10).Update(series); var staticResults = Hanma.Batch(series, 10); for (int i = 0; i < instanceResults.Count; i++) { Assert.Equal(instanceResults[i].Value, staticResults[i].Value, 1e-9); } } [Fact] public void Hanma_SpanCalculate_MatchesSeries() { var series = new TSeries(); var gbm = new GBM(startPrice: 100.0, mu: 0.05, sigma: 0.2, seed: 42); for (int i = 0; i < 100; i++) { var bar = gbm.Next(isNew: true); series.Add(bar.Time, bar.Close); } var seriesResults = Hanma.Batch(series, 10); double[] input = series.Values.ToArray(); double[] output = new double[input.Length]; Hanma.Batch(input.AsSpan(), output.AsSpan(), 10); for (int i = 0; i < input.Length; i++) { Assert.Equal(seriesResults[i].Value, output[i], 1e-9); } } [Fact] public void Hanma_Update_IsNewFalse_CorrectsValue() { // Note: Hanning window has edge weight = 0, so the newest value (position period-1) // contributes 0 to the weighted average when buffer is full. We need to test // correction on a value that has non-zero weight, so we add one more value // after the correction target to shift it away from the edge. var hanma = new Hanma(10); var gbm = new GBM(startPrice: 100.0, mu: 0.05, sigma: 0.2, seed: 42); // Feed initial data for (int i = 0; i < 20; i++) { var bar = gbm.Next(isNew: true); hanma.Update(new TValue(bar.Time, bar.Close), isNew: true); } // Add a value that we'll correct var targetBar = gbm.Next(isNew: true); hanma.Update(new TValue(targetBar.Time, targetBar.Close), isNew: true); // Add one more value so the target bar moves to position period-2 (which has non-zero weight) var nextBar = gbm.Next(isNew: true); hanma.Update(new TValue(nextBar.Time, nextBar.Close), isNew: true); double valueAfterCommit = hanma.Last.Value; // Now correct the previous bar (targetBar) with a different value using 2 corrections: // First rollback the last bar, then update targetBar with different value, then re-add nextBar // This simulates bar correction where we need to re-apply subsequent bars // Alternative approach: test the isNew=false mechanism directly on the LAST bar // even though it has weight=0, we verify the state rollback works correctly _ = hanma.Update(new TValue(nextBar.Time, nextBar.Close + 50.0), isNew: false); // Even though the newest bar has weight=0, the state rollback should still work // and the result may differ due to buffer state restoration // For Hanning, if only the edge changes, result stays same - this is mathematically correct // So we test state restoration instead: hanma.Update(new TValue(nextBar.Time, nextBar.Close), isNew: false); Assert.Equal(valueAfterCommit, hanma.Last.Value, 1e-9); } [Fact] public void Hanma_NaN_Input_UsesLastValidValue() { var hanma = new Hanma(5); hanma.Update(new TValue(DateTime.UtcNow, 100)); hanma.Update(new TValue(DateTime.UtcNow, 110)); var resultAfterNaN = hanma.Update(new TValue(DateTime.UtcNow, double.NaN)); Assert.True(double.IsFinite(resultAfterNaN.Value)); Assert.NotEqual(0, resultAfterNaN.Value); } [Fact] public void Hanma_Reset_ClearsState() { var hanma = new Hanma(10); hanma.Update(new TValue(DateTime.UtcNow, 100)); hanma.Update(new TValue(DateTime.UtcNow, 110)); Assert.True(hanma.Last.Value > 0); hanma.Reset(); Assert.Equal(0, hanma.Last.Value); Assert.False(hanma.IsHot); } [Fact] public void Hanma_FirstValue_ReturnsExpected() { var hanma = new Hanma(10); TValue result = hanma.Update(new TValue(DateTime.UtcNow, 100)); Assert.Equal(100.0, result.Value, 1e-9); } [Fact] public void Hanma_Properties_Accessible() { var hanma = new Hanma(10); Assert.False(hanma.IsHot); Assert.Equal(0, hanma.Last.Value); } [Fact] public void Hanma_Calc_IsNew_AcceptsParameter() { var hanma = new Hanma(10); hanma.Update(new TValue(DateTime.UtcNow, 100), isNew: true); Assert.Equal(100, hanma.Last.Value); } [Fact] public void Hanma_IterativeCorrections_RestoreToOriginalState() { var hanma = new Hanma(10); var gbm = new GBM(startPrice: 100.0, mu: 0.02, sigma: 0.1); // Feed 10 new values TValue tenthInput = default; for (int i = 0; i < 10; i++) { var bar = gbm.Next(isNew: true); tenthInput = new TValue(bar.Time, bar.Close); hanma.Update(tenthInput, isNew: true); } // Remember state after 10 values double valueAfterTen = hanma.Last.Value; // Generate 9 corrections with isNew=false (different values) for (int i = 0; i < 9; i++) { var bar = gbm.Next(isNew: false); hanma.Update(new TValue(bar.Time, bar.Close), isNew: false); } // Feed the remembered 10th input again with isNew=false TValue finalValue = hanma.Update(tenthInput, isNew: false); // Should match the original state after 10 values Assert.Equal(valueAfterTen, finalValue.Value, 1e-9); } [Fact] public void Hanma_Infinity_Input_UsesLastValidValue() { var hanma = new Hanma(10); hanma.Update(new TValue(DateTime.UtcNow, 100)); hanma.Update(new TValue(DateTime.UtcNow, 110)); var resultPosInf = hanma.Update(new TValue(DateTime.UtcNow, double.PositiveInfinity)); Assert.True(double.IsFinite(resultPosInf.Value)); var resultNegInf = hanma.Update(new TValue(DateTime.UtcNow, double.NegativeInfinity)); Assert.True(double.IsFinite(resultNegInf.Value)); } [Fact] public void Hanma_MultipleNaN_ContinuesWithLastValid() { var hanma = new Hanma(10); hanma.Update(new TValue(DateTime.UtcNow, 100)); var r1 = hanma.Update(new TValue(DateTime.UtcNow, double.NaN)); var r2 = hanma.Update(new TValue(DateTime.UtcNow, double.NaN)); Assert.True(double.IsFinite(r1.Value)); Assert.True(double.IsFinite(r2.Value)); } [Fact] public void Hanma_AllModes_ProduceSameResult() { // Arrange const int period = 10; var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.2, seed: 123); var bars = gbm.Fetch(1000, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1)); var series = bars.Close; // 1. Batch Mode var batchSeries = Hanma.Batch(series, period); double expected = batchSeries.Last.Value; // 2. Span Mode var tValues = series.Values.ToArray(); var spanInput = new ReadOnlySpan(tValues); var spanOutput = new double[tValues.Length]; Hanma.Batch(spanInput, spanOutput, period); double spanResult = spanOutput[^1]; // 3. Streaming Mode var streamingInd = new Hanma(period); for (int i = 0; i < series.Count; i++) { streamingInd.Update(series[i]); } double streamingResult = streamingInd.Last.Value; // 4. Eventing Mode var pubSource = new TSeries(); var eventingInd = new Hanma(pubSource, period); for (int i = 0; i < series.Count; i++) { pubSource.Add(series[i]); } double eventingResult = eventingInd.Last.Value; // Assert Assert.Equal(expected, spanResult, 1e-9); Assert.Equal(expected, streamingResult, 1e-9); Assert.Equal(expected, eventingResult, 1e-9); } [Fact] public void Hanma_SpanCalc_ValidatesInput() { double[] source = [1, 2, 3, 4, 5]; double[] output = new double[5]; double[] wrongSizeOutput = new double[3]; Assert.Throws(() => Hanma.Batch(source.AsSpan(), output.AsSpan(), 0)); Assert.Throws(() => Hanma.Batch(source.AsSpan(), wrongSizeOutput.AsSpan(), 3)); } [Fact] public void Hanma_SpanCalc_HandlesNaN() { double[] source = [100, 110, double.NaN, 120, 130]; double[] output = new double[5]; Hanma.Batch(source.AsSpan(), output.AsSpan(), 3); foreach (var val in output) { Assert.True(double.IsFinite(val)); } } [Fact] public void Hanma_HanningWindow_WeightSymmetry() { // Hanning window should be symmetric around center // w[i] = w[period-1-i] for all i int period = 11; // Odd for exact center // Verify weight symmetry by checking equal outputs for symmetric inputs var hanma1 = new Hanma(period); var hanma2 = new Hanma(period); // Feed ascending values to hanma1 double[] ascending = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]; foreach (var v in ascending) { hanma1.Update(new TValue(DateTime.UtcNow, v)); } // Feed descending values to hanma2 double[] descending = [11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1]; foreach (var v in descending) { hanma2.Update(new TValue(DateTime.UtcNow, v)); } // Results should be the same (symmetric weights applied to symmetric data) Assert.Equal(hanma1.Last.Value, hanma2.Last.Value, 1e-9); } [Fact] public void Hanma_KnownValues_ManualCalculation() { // Manual verification with known Hanning weights // period=5: w[i] = 0.5 * (1 - cos(2π*i/4)) // w[0] = 0.5 * (1 - cos(0)) = 0.5 * 0 = 0 // w[1] = 0.5 * (1 - cos(π/2)) = 0.5 * 1 = 0.5 // w[2] = 0.5 * (1 - cos(π)) = 0.5 * 2 = 1.0 // w[3] = 0.5 * (1 - cos(3π/2)) = 0.5 * 1 = 0.5 // w[4] = 0.5 * (1 - cos(2π)) = 0.5 * 0 = 0 int period = 5; var hanma = new Hanma(period); double[] prices = [100, 102, 104, 103, 101]; foreach (var price in prices) { hanma.Update(new TValue(DateTime.UtcNow, price)); } // Calculate expected manually double twoPiOverPm1 = 2.0 * Math.PI / (period - 1); double[] weights = new double[period]; double weightSum = 0; for (int i = 0; i < period; i++) { weights[i] = 0.5 * (1.0 - Math.Cos(twoPiOverPm1 * i)); weightSum += weights[i]; } double expected = 0; for (int i = 0; i < period; i++) { expected += prices[i] * weights[i]; } expected /= weightSum; Assert.Equal(expected, hanma.Last.Value, 1e-9); } [Fact] public void Hanma_PeriodOne_ReturnsInputValue() { var hanma = new Hanma(1); for (int i = 1; i <= 10; i++) { var input = new TValue(DateTime.UtcNow, i * 10.0); var result = hanma.Update(input); Assert.Equal(i * 10.0, result.Value, 1e-9); } } [Fact] public void Hanma_EdgeWeights_AreZero() { // Hanning window uniquely has edge weights = 0 // This means first and last values in window get zero weight int period = 5; var hanma = new Hanma(period); // All same values except edges double[] prices = [999, 100, 100, 100, 888]; foreach (var price in prices) { hanma.Update(new TValue(DateTime.UtcNow, price)); } // Because edge weights are 0, the 999 and 888 should have no effect // Result should be 100.0 (weighted average of middle values only) Assert.Equal(100.0, hanma.Last.Value, 1e-9); } }