mlund

FROM ubuntu:22.04 as llvm_mos_base
RUN apt-get update
RUN DEBIAN_FRONTEND=noninteractive apt-get install --no-install-recommends -y \
    git curl ca-certificates ninja-build cmake build-essential  \
    libstdc++-10-dev libxml2-dev libssl-dev pkg-config python3-pip \
    swig python3-dev libedit-dev libncurses5-dev  liblzma-dev

FROM llvm_mos_base as build

WORKDIR /tmp

mlund

# Copyright (c) Jupyter Development Team.
# Distributed under the terms of the Modified BSD License.
#
# What is installed?

# - https://jupyter-docker-stacks.readthedocs.io/en/latest/using/selecting.html
# - GCC 11, CMake, OpenMPI, parmed, openmm, nglview, mdtraj
# - Faunus main branch incl. manual and examples
#
ARG OWNER=jupyter

mlund

#include <vector>
#include <iostream>
#include <type_traits>
#include <Eigen/Eigen>

/**
 * @brief Eigen::Map facade to access data members in STL vectors/arrays
 *
 * No data is copied and modifications of the Eigen object
 * modifies the original container and vice versa.

mlund

Ubuntu terminal setup for macOS

This will setup a Ubuntu-like terminal in MacOS X.

Installation

cat bash_profile >> ~/.bash_profile
open Ubuntu.terminal

mlund

#!/usr/bin/env python

# Copyright (c) 2015-2023 Mikael Lund
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions
# are met:
#
# 1. Redistributions of source code must retain the above copyright
#    notice, this list of conditions and the following disclaimer.

mlund

private:
enum baseparticle_ : Index {RX=0,RY,RZ,ID,Q,LAST};
enum multipoleparticle_ : Index {MUX=LAST,MUY,MUZ};

public:
// Particle charge
EIGEN_STRONG_INLINE Scalar charge() const { return (*this)[Q]; }
EIGEN_STRONG_INLINE Scalar& charge() { return (*this)[Q]; }

// Particle id

mlund

#include <iostream>
#include <functional>
#include <map>
#include <cmath>
#include <memory>

/* ordered pair */
template<class T>
struct opair : public std::pair<T,T> {
  typedef std::pair<T,T> base;

mlund

Cell list with periodic boundaries

This maps a vector of 3d positions and map each index into an evenly spaced lattice of grid points (cells). The neighbors of a given index in the position vector can be used to quickly find all other index in the vicinity.

More inspiration: http://cacs.usc.edu/education/cs596/01-1LinkedListCell.pdf

mlund

#!/bin/bash
curl -LH "Accept: application/x-bibtex;q=1" http://dx.doi.org/$1
echo

mlund

#include <random>
#include <iostream>
template<typename T, typename Tengine=std::mt19937> 
class RandOne {
  std::uniform_real_distribution<T> dist_;
  Tengine eng_;
  public:
  /* constructor w. non-deterministic seed */
  RandOne() : dist_(0,1) {
    std::random_device rd;