From af2585c5be539e76e4584d4bf885c5b7fe1bd331 Mon Sep 17 00:00:00 2001
From: Brian Drawert <brian@drawert.net>
Date: Mon, 25 Jul 2022 13:03:55 -0700
Subject: [PATCH] Closes #289

---
 .../NRMConstant/README_NRMConstant.txt        | 16 -----
 .../NRMConstant/demo_NRMConstant_v5.cpp       | 66 -------------------
 2 files changed, 82 deletions(-)
 delete mode 100644 spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/README_NRMConstant.txt
 delete mode 100644 spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/demo_NRMConstant_v5.cpp

diff --git a/spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/README_NRMConstant.txt b/spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/README_NRMConstant.txt
deleted file mode 100644
index 009cf10e..00000000
--- a/spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/README_NRMConstant.txt
+++ /dev/null
@@ -1,16 +0,0 @@
-Sanft added NRMConstant_v4 class (NRMConstant_v4.hpp, .cpp)
-The class is a priority queue that should scale roughly O(1)
-for large discrete stochastic reaction-diffusion models.
-
-demo_NRMConstant.cpp shows how to use it. Compile with:
-g++ demo_NRMConstant.cpp NRMConstant_v4.cpp
-
-Note the key functions:
-NRMConstant_v4 reactionGenerator; // constructor
-reactionGenerator.build(propensities, gen, propensitySum, activeChannels); // set up priority queue from initial propensities
-reactionGenerator.selectReaction(); // get the item (next reaction or diffusion event) that will occur next
-reactionGenerator.update(reactionIndex,propensities[reactionIndex],currentTime,gen); // update event time
-
-Note that the event that fires and all affected propensities should have a corresponding call to update before
-calling selectReaction
-
diff --git a/spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/demo_NRMConstant_v5.cpp b/spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/demo_NRMConstant_v5.cpp
deleted file mode 100644
index ef15692e..00000000
--- a/spatialpy/solvers/c_base/ssa_sdpd-c-simulation-engine/external/NRMConstant/demo_NRMConstant_v5.cpp
+++ /dev/null
@@ -1,66 +0,0 @@
-// compile: g++ *.cpp
-// run: a.out
-#include <iostream>
-#include <random>
-#include "NRMConstant_v5.hpp"
-
-int main() {
-    std::cout << "testing v5...(doesn't use ParticleSystem!)" << "\n";
-
-    /* SET UP RANDOM NUMBER GENERATOR AND "MODEL" */
-    // gen is the random number generator
-    std::mt19937 gen;
-    //    gen.seed(123);
-    std::random_device rd;
-    gen.seed(rd());
-    
-    // create a "MODEL" (just some random propensity values)
-    std::size_t modelsize=10;//total number of propensities
-    std::vector<double> propensities(modelsize); //vector of all propensities  (reaction and diffusion)
-    double propensitySum=0.0;
-    std::size_t activeChannels=0; // initial number of nonzero propensities (typically improves initial performance)
-    
-    
-    //let's fill the propensities with random values using a uniform continuous distribution from 0 to 100 (mean 50)
-    std::uniform_real_distribution<double> uniform100(0.0,100.0);
-    for (int i=0; i<modelsize; i++) {
-        double next_propensity = uniform100(gen); // propensity i would normally be calculated based on state
-        std::cout << "propensity " << i << " = " << next_propensity << "\n";
-        propensities[i]=next_propensity;
-        propensitySum+=next_propensity;
-        activeChannels++;
-    }
-    std::cout << "propensitySum=" << propensitySum << "\n";
-    std::cout << "activeChannels=" << activeChannels << "\n";
-    
-    /* HERE IS THE PRIORITY QUEUE "NRMConstant_v5" */
-    NRMConstant_v5 reactionGenerator;
-    // set up the priority queue:
-    reactionGenerator.build(propensities, gen, propensitySum, activeChannels);
-    
-    /* RUN A "SIMULATION" */
-    // selectReaction() returns a pair: the event firing time and the event index (index of propensity)
-    std::pair<double,int> timeRxnPair;
-    int reactionIndex;
-    double currentTime=0;
-    double endTime=.1;
-    while (currentTime<endTime) {
-        timeRxnPair=reactionGenerator.selectReaction();
-        currentTime=timeRxnPair.first;
-        reactionIndex=timeRxnPair.second;
-        std::cout << "firing reaction " << reactionIndex << " at t = " << currentTime << "\n";
-        // here is where we would update state, recalculate all affected propensities
-        // suppose reaction i affects (itself and) reaction i+1 (would typically be done with a dependency graph)
-        int affectedReaction = (reactionIndex+1) % modelsize;
-        // update reaction that fired
-        propensities[reactionIndex]=uniform100(gen); // propensity would normally be calculated based on state
-        // update affected reaction
-        propensities[affectedReaction]=uniform100(gen); // propensity would normally be calculated based on state
-        // update the priority queue for all changed propensities
-        // update takes the affected reaction index, the new propensity value, current time, and random generator
-        reactionGenerator.update(reactionIndex,propensities[reactionIndex],currentTime,gen);
-        reactionGenerator.update(affectedReaction,propensities[affectedReaction],currentTime,gen);
-    }
-    
-    return 0;
-}