[Entropy] + separate files in plugin

GViewCore/include/GView.hpp

@@ -805,7 +805,7 @@ namespace Regex
 namespace Entropy
 {
     CORE_EXPORT double ShannonEntropy(const BufferView& buffer);
-    CORE_EXPORT double RenyiEntropy(const BufferView& buffer);
+    CORE_EXPORT double RenyiEntropy(const BufferView& buffer, double alpha);
 } // namespace Entropy
 
 /*

GViewCore/src/Entropy/Entropy.cpp

@@ -1,6 +1,7 @@
 #include "Internal.hpp"
 
 #include <math.h>
+#include <array>
 
 constexpr uint32 MAX_NUMBER_OF_BYTES = 256;
 
@@ -15,6 +16,17 @@ void SetFrequencies(const BufferView& buffer, std::array<char, MAX_NUMBER_OF_BYT
     }
 }
 
+/*
+    In physics, the word entropy has important physical implications as the amount of "disorder" of a system.
+    In mathematics, a more abstract definition is used.
+
+    The (Shannon) entropy of a variable X is defined as
+    H(X) congruent - sum_x P(x) log_2[P(x)]
+    bits, where P(x) is the probability that X is in the state x, and P log_2 P is defined as 0 if P = 0.
+
+    The joint entropy of variables X_1, ..., X_n is then defined by
+    H(X_1, ..., X_n) congruent - sum_(x_1) ... sum_(x_n) P(x_1, ..., x_n) log_2[P(x_1, ..., x_n)].
+*/
 double ShannonEntropy_private(const BufferView& buffer, std::array<char, MAX_NUMBER_OF_BYTES>& frequency)
 {
     double entropy = 0.0;
@@ -26,7 +38,7 @@ double ShannonEntropy_private(const BufferView& buffer, std::array<char, MAX_NUM
         entropy -= probability * log2(probability);
     }
 
-    return entropy; // max log2(n) = 8
+    return entropy; // max log2(n) = 8 (the entire sum)
 }
 
 double ShannonEntropy(const BufferView& buffer)
@@ -36,6 +48,14 @@ double ShannonEntropy(const BufferView& buffer)
     return ShannonEntropy_private(buffer, frequency);
 }
 
+/*
+    Rényi entropy is defined as:
+    H_α(p_1, p_2, ..., p_n) = 1/(1 - α) ln( sum_(i = 1)^n p_i^α), where α>0, α!=1.
+    As α->1, H_α(p_1, p_2, ..., p_n) converges to H(p_1, p_2, ..., p_n), which is Shannon's measure of entropy.
+    Rényi's measure satisfies
+    H_α(p_1, p_2, ..., p_n)<=H_α'(p_1, p_2, ..., p_n)
+    for α<=α'.
+*/
 double RenyiEntropy(const BufferView& buffer, double alpha)
 {
     std::array<char, MAX_NUMBER_OF_BYTES> frequency{};
@@ -47,13 +67,14 @@ double RenyiEntropy(const BufferView& buffer, double alpha)
 
     double sum = 0.0;
     for (auto f : frequency) {
-        double probability = static_cast<double>(f) / buffer.GetLength();
-        if (probability > 0) {
+        if (f > 0) {
+            const double probability = static_cast<double>(f) / buffer.GetLength();
             sum += pow(probability, alpha);
         }
     }
 
     // Convert to bits if using log base e
+    // return std::max(((1.0 / (1.0 - alpha)) * log(sum)) / log(2), 0.0);
     return ((1.0 / (1.0 - alpha)) * log(sum)) / log(2);
 }
 } // namespace GView::Entropy

GenericPlugins/EntropyVisualizer/src/CMakeLists.txt

@@ -1 +1 @@
-target_sources(EntropyVisualizer PRIVATE EntropyVisualizer.cpp)
+target_sources(EntropyVisualizer PRIVATE Plugin.cpp EntropyVisualizer.cpp)