#ifndef SPINLOCKQUEUE_HPP
#define SPINLOCKQUEUE_HPP 1
#include <cstddef>
#include <algorithm>
#include <atomic>
#include <thread>
#include <mutex>
#include <condition_variable>
#include <utility>
#include <cassert>
#include <vector>
#include <iostream>
#include <boost/date_time/posix_time/posix_time.hpp>
using namespace boost::posix_time;

template <typename T>
class SpinLockQueue {
private:
  struct Node {
    Node( T *val ) : value(val), next(0) { }
    Node( Node && rhs)
    {
    *this = std::move(rhs);
    }
    Node( Node const &) = delete;
    Node& operator=(Node const &) = delete;
    Node& operator=(Node && rhs) {
    if (this != &rhs)
    {
        value = rhs.value;
        rhs.value = 0;
        next = rhs.next;
        rhs.next = 0;
    }

    return *this;
    }

    ~Node() {
    delete value;
    }
    T* value;
    Node* next;
  } __attribute__((aligned(64)));



  std::atomic<std::size_t> mWaiters;
  std::mutex mMutex;
  std::condition_variable mCondition;


  Node* first __attribute__((aligned(16)));

  // shared among consumers
  atomic<long> consumerLock;
  char pad[ 64 - sizeof(consumerLock) - sizeof(first)];
  // for one producer at a time
  Node* last __attribute__((aligned(32)));

  // shared among producers
  atomic<long> producerLock __attribute__((aligned(32)));


  atomic<Node*> nodeList __attribute__((aligned(64)));


  Node* getNewNode(T const & t) {
    Node* n = nodeList;
    if (n)
    {
        while (n && !nodeList.compare_exchange_strong(n,n->next))
        {
            n = nodeList;
        }

        if (n)
        {
            n->value = new T(t);
            n->next = 0;
            return n;
        } else {
            return new Node(new T(t));
        }

    } else {
        return new Node(new T(t));
    }
  }

  Node* getNewNode(T* t) {
    Node * n = nodeList;
    if (n)
    {
        Node* next = n->next;
        while (!nodeList.compare_exchange_strong(n,next))
        {
            n = nodeList;
            if (n)
            {
                next = n->next;
            }
        }

        if (n)
        {
            n->value = t;
            n->next = 0;
            return n;
        }
        else
        {
            return new Node(n);
        }
    }
    else
    {
        return new Node(n);
    }
  }

  void addNodeToList(Node* node)
  {
        Node *n(0);
                do {
                        n = nodeList;
                        node->next = n;
                } while (!nodeList.compare_exchange_strong(n,node));

  }

public:
  SpinLockQueue() : mWaiters(0) {
    nodeList = new Node(0);
    first  = last = new Node(0);           // add dummy separator

    producerLock = false;
    consumerLock = false;
  }
  ~SpinLockQueue() {
    Node* tmp;
    while( (tmp = first) != 0 ) {   // release the list
      first = tmp->next;
      delete tmp;
    }

    while ( (tmp = nodeList) != 0) {
    nodeList = tmp->next;
    delete tmp;
    }
  }


  void push( const T& t ) {
    Node* newlast = getNewNode(t);
    assert(newlast);
    assert(newlast->value);
    long oldvalue = false;
    while (!producerLock.compare_exchange_strong(oldvalue,true)) {}
    assert(producerLock);
// own producer lock

    last->next = newlast;
    last = newlast;

    assert(producerLock.exchange(false));

    if (mWaiters)
    {
    mCondition.notify_one();
    }
  }

  bool try_pop( T& result ) {
    // No nodes point to front
    long oldvalue = false;
    while (!consumerLock.compare_exchange_strong(oldvalue,true)) { continue; }
    assert(consumerLock.load());

    Node * theFirst = first;
    Node * next = theFirst->next;
    while (next)
    {
    if (next->value == 0)
    {

        first = next;
                theFirst->value = 0;
        addNodeToList(theFirst);
        theFirst = first;
        next = theFirst->next;
        continue;
    }
    else
    {
        T* val = next->value;
        next->value = 0;
        first = next;
        theFirst->value = 0;
        addNodeToList(theFirst);
        long exchangedVal = consumerLock.exchange(false);
        assert(exchangedVal);
        assert(val);
        result = std::move(*val);
        delete val;

    }

    return true;
    }

    assert(consumerLock.exchange(false));
    return false;
  }


 void pop(T& result) {
    ++mWaiters;

    std::unique_lock<std::mutex> lk(mMutex);
    while (!try_pop(result))
    {
        mCondition.wait(lk);
    }

    --mWaiters;
    }

};

#if 1

static std::atomic<int> pushed, popped;

void process(SpinLockQueue<std::vector<std::string> >* queue)
{

        for (int i = 0; i < 100000; ++i)
        {
                queue->push(std::vector<std::string>());
        ++pushed;
        }

    std::vector<std::string> a;
#pragma omp parallel private(a)
    while (queue->try_pop(a))
    {
        ++popped;
    }
#pragma omp  barrier
}



int main(int argc, char * argv[])
{
    SpinLockQueue<std::vector<std::string> > *queue = new SpinLockQueue<std::vector<std::string> >;
    pushed = 0;
    popped = 0;

    ptime begin(microsec_clock::local_time());
    process(queue);
    ptime end(microsec_clock::local_time());

    std::cout << "Numbers: pushes: " << pushed.load() << " pops: " << popped.load() << std::endl;
    std::cout << "Time taken: " << (end-begin) << std::endl;
    delete queue;
    return 0;
}


#endif

#endif