Pulled the latest version from the mmsp project to the mmsp fork on ACME-Project

.github/ISSUE_TEMPLATE.md

@@ -0,0 +1,13 @@
+### There's a *broken feature* in MMSP
+- Bug appears in {file or program}
+- Encountered running on {CPU} and {accelerator} using {OS} with {relevant software}
+- The expected behavior is 
+- The observed behavior is
+
+### There's a *missing feature* in MMSP
+- The gap is in {file or program}
+- It deserves attention because
+- A possible approach is
+
+### There's a *confusing feature* in MMSP
+- Join the conversation on [gitter](https://gitter.im/mesoscale/mmsp)!

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,7 @@
+The proposed changes are intended to {summary of intent}.
+
+This is achieved by {description of method}.
+
+Fixes #, Addresses #
+
+The modified code may have unintended consequences in {areas of concern}.

CONTRIBUTING.md

@@ -0,0 +1,130 @@
+# Contributing to [MMSP: The Mesoscale Microstructure Simulation Project][_mmsp]
+
+Thank you for spending some time with MMSP. We sincerely appreciate your
+interest, and hope you will consider contributing to the documentation,
+source code, and community at large. This code is open-source, and we really
+appreciate input from the community. There are several ways you can contribute,
+including editing or adding documentation, writing tutorials, submitting bug
+reports, requesting new features, and writing code to incorporate. If you'd like
+to help out, please do!
+
+If you're looking for support or guidance, please visit our [Gitter chat][_gitter]
+to talk with the developers and community members, or send an e-mail to
+<trevor.keller@gmail.com>.
+
+## Ground Rules
+
+- Help us to maintain a considerate and supportive community by reading and
+  enforcing the [Code of Conduct][_conduct]. Be nice to newcomers.
+- Promote clean code by ensuring your code and documentation build with neither
+  warnings nor errors using compiler flags equivalent to [GCC's][_gcc]
+  `-Wall -pedantic`.
+- Engage with the community by creating [issues][_issue] for changes or
+  enhancements you'd like to make. Discuss ideas the open and get feedback.
+- Maximize reusability by keeping code simple. Program in C if possible, and
+  generally follow the [Google C++ Style Guide][_goog]. Avoid creating new
+  classes if possible.
+- Document new functions and classes with [Doxygen][_doxy]-compatible comments
+  in the source code.
+
+## Getting Started
+
+Interested in helping, but unsure where to start? Consider proofreading the PDF
+documentation and reporting or fixing errors! There's bound to be a typo in
+there. Or look through the existing [issues][_issue] for fixes or enhancements
+you're interested in. The number of comments on an issue can indicate its level
+of difficulty, as well as the impact closing it will have.
+
+If you're brand new to open source, welcome! You might benefit from the
+[GitHub Help Pages][_ghhelp] and a [tutorial][_tut].
+
+## Branching Workflow
+
+[MMSP][_mmsp] uses [git][_git] version control with a [branching workflow][_branch],
+summarized below. Following this workflow ensures a pristine `master` branch
+that remains synchronized, and allows collaborators to quickly and easily review
+contributions for content and quality.
+
+The core mantra is ***never commit to `master`***. On [mesoscale/mmsp][_mmsp],
+`master` is the rolling release branch, and code must only be merged into it
+following [continuous integration][_ci] testing, code review, and discussion.
+
+Changes are made on *branches*, using `master` as the trunk. Ideally, a branch
+is created based on an [issue][_issue], which captures a *feature* or *hotfix*
+to MMSP. To paraphrase illustrations of more involved workflows, a simple
+branching model for MMSP looks like the following, with `B` indicating a branch
+named `___`, `M` a commit with message `"___"`, `?` a pull request, and `D`
+deletion of a branch. Note that branches are numbered sequentially, using the
+number of the [issue][_issue] outlining the missing or broken functionality to
+be patched.
+
+```
+            mesoscale/ origin/ origin/
+              master   hotfix  feature  hotfix         feature
+                :
+                |
+                +-----------------------------------------B issue4
+                |                                         |
+                +--------------------------B bug5         M "create function"
+                |                          M "missing ;"  M "update docs"
+                |         ?----------------+              |
+ "merge bug5"   M---------+                |              |
+                |         D                D              |
+                +-----------------------------------------M "merge master"
+                |                 ?-----------------------+
+                |                 |                       M "address comments"
+                |                 M-----------------------+
+ "merge issue4" M-----------------+                       |
+                |                 D                       D
+                : 
+```
+
+## Pull Requests
+
+*Please neither commit directly to the `master` branch, nor push directly to
+mesoscale/mmsp.*
+
+1. Create a fork of [MMSP][_mmsp] on your personal GitHub account.
+2. For obvious changes, such as typos or edits to `.gitignore`, you can edit
+   your fork directly in the browser, then file a [pull request][_pr].
+3. Most changes begin with an [Issue][_issue]. If one does not exist, create it.
+4. On your local machine, use the command line to work on your local fork of
+   [MMSP][_mmsp]. Create a working branch off of `master`. If you're working,
+   for example, on issue #42, *function template documentation*:
+   - [Fork][_ghhelp] mesoscale/mmsp and [clone][_ghhelp] it to your local machine.
+     Your fork is a `git remote` with the alias `origin` by default.
+   - `git pull https://github.com/mesoscale/mmsp.git master` to update your
+     local `master` branch from the main repository.
+   - `git checkout -b issue42_function-template-documentation master` to
+     create a local feature branch off of `master`.
+   - Write your edits, updates, or feature code.
+   - `git status`, `git add -u`, `git status`, and `git commit`,
+     in that order, to add your changes. If something is amiss, follow
+     the terminal guidance to fix it.
+   - Write a concise commit message (first line), then in-depth commentary below
+     using the keywords ["Addresses" or "Closes"][_ghkey] where appropriate.
+   - `git push origin issue42_summarize-usage-in-pseudocode` to push the
+     *working branch* to your fork on GitHub.
+5. Visit GitHub and make the pull request official. Please fill out the pull
+   request template and assign a reviewer &emdash; `@tkphd` if noone else suits.
+
+Obvious fixes will be merged quickly. Enhancements may undergo a code review,
+which we typically conduct using [reviewable][_review].
+
+## Happy Coding!
+
+[_mmsp]:    https://github.com/mesoscale/mmsp
+[_branch]:  http://nvie.com/posts/a-successful-git-branching-model/
+[_ci]:      https://docs.travis-ci.com/
+[_conduct]: https://github.com/mesoscale/mmsp/blob/master/CODE_OF_CONDUCT.md
+[_doxy]:    https://www.stack.nl/~dimitri/doxygen/manual/docblocks.html
+[_gcc]:     https://gcc.gnu.org/
+[_ghhelp]:  https://help.github.com/
+[_ghkey]:   https://help.github.com/articles/closing-issues-using-keywords/
+[_git]:     https://git-scm.com/
+[_gitter]:  https://gitter.im/mesoscale/mmsp
+[_goog]:    https://google.github.io/styleguide/cppguide.html
+[_issue]:   https://github.com/mesoscale/mmsp/issues
+[_pr]:      https://help.github.com/articles/about-pull-requests/
+[_review]:  https://reviewable.io/reviews/mesoscale/mmsp
+[_tut]:     https://egghead.io/courses/how-to-contribute-to-an-open-source-project-on-github

README.md

@@ -1,6 +1,7 @@
 Mesoscale Microstructure Simulation Project
 ====
 [![Build Status](https://travis-ci.org/mesoscale/mmsp.svg?branch=develop)](https://travis-ci.org/mesoscale/mmsp)
+[![DOI](https://zenodo.org/badge/19985417.svg)](https://zenodo.org/badge/latestdoi/19985417)
 
 The goal of the Mesoscale Microstructure Simulation Project (MMSP) is to provide a simple,
 consistent, and extensible programming interface for all grid and mesh based microstructure
@@ -42,18 +43,21 @@ MMSP is nothing more than a collection of C++ header files that declare a number
 *MMSP requires:*
 
  * Minimal programming experience
- * An ISO compliant C++ compiler (e.g. GCC 2.95 or later)
- * zlib libraries for data compression (e.g. zlib 1.2.7)
- * libpng headers for mmsp2png image generation utility (e.g. libpng12-dev)
- * MPI libraries if compiling parallel programs (e.g. Open MPI)
+ * An ISO compliant C++ compiler (e.g. `gcc` 2.95 or later)
+ * zlib libraries for data compression (e.g. `zlib` 1.2.7)
+ * libpng headers for mmsp2png image generation utility (e.g. `libpng12-dev`)
+ * ParaView VTK headers for VTI and PVD visualization utilities (e.g. `paraview` and `paraview-dev`)
+ * MPI libraries if compiling parallel programs (e.g. `openmpi`)
 
 *Documentation*
 
 The MMSP manual is a work in progress. It is currently the only source for detailed documentation about MMSP.
 
 *Contact us*
 
-The administrators for the MMSP source code are Jason Gruber (gruberja@gmail.com) and Trevor Keller (trevor.keller@gmail.com). Please do not hesitate to send questions or comments.
+The administrators for the MMSP source code are Jason Gruber (gruberja@gmail.com), Trevor Keller (trevor.keller@gmail.com) and Dan Lewis (lucentdan@gmail.com). Please do not hesitate to send questions or comments. Please cite using the following DOI:
+
+[![DOI](https://zenodo.org/badge/19985417.svg)](https://zenodo.org/badge/latestdoi/19985417)
 
 
 This work was supported in part by the US NSF under award #1056704 through the Metals and Metallic Nanostructures Program, Division of Materials Research. 

algorithms/generators.cpp

@@ -1,4 +1,3 @@
-// generators.cpp
 // Implementations of methods to initialize MMSP grids
 #ifndef _GENERATORS_CPP_
 #define _GENERATORS_CPP_
@@ -81,9 +80,9 @@ MMSP::vector<int> getPosition(const DistanceVoxel& dv) {
 
 template<int dim, typename T>
 void exact_voronoi(MMSP::grid<dim, MMSP::sparse<T> >& grid, const std::vector<std::vector<MMSP::vector<int> > >& seeds) {
-	int id=0;
+	int id = 0;
 	#ifdef MPI_VERSION
-	id=MPI::COMM_WORLD.Get_rank();
+	MPI_Comm_rank(MPI_COMM_WORLD, &id);
 	#endif
   // Exact Voronoi tessellation from seeds, based on Euclidean distance function. Runtime is O(Nseeds*L*W*H).
   // seeds must contain every seed from every rank.
@@ -193,7 +192,7 @@ void approximate_voronoi(MMSP::grid<dim, MMSP::sparse<T> >& grid, const std::vec
   #endif
   int id = 0;
 	#ifdef MPI_VERSION
-  id = MPI::COMM_WORLD.Get_rank();
+	MPI_Comm_rank(MPI_COMM_WORLD, &id);
 	#endif
   // Perform the tessellation, using fast-marching fanciness
   if (dim == 2) {
@@ -400,7 +399,9 @@ void approximate_voronoi(MMSP::grid<dim, MMSP::sparse<T> >& grid, const std::vec
   }
 	#ifdef DEBUG
 	#ifdef MPI_VERSION
-	if (MPI::COMM_WORLD.Get_rank()==0)
+	int id = 0;
+	MPI_Comm_rank(MPI_COMM_WORLD, &id);
+	if (id==0)
 	#endif
 	std::cout<<"Completed approximate tessellation ("<<time(NULL)-tstart<<" sec)."<<std::endl;
 	#endif
@@ -409,23 +410,24 @@ void approximate_voronoi(MMSP::grid<dim, MMSP::sparse<T> >& grid, const std::vec
 void seedswap(std::vector<std::vector<MMSP::vector<int> > >& seeds)
 {
 	#ifdef MPI_VERSION
-  int id=MPI::COMM_WORLD.Get_rank();
-  int np=MPI::COMM_WORLD.Get_size();
+    int id=0, np=0;
+    MPI_Comm_rank(MPI_COMM_WORLD, &id);
+    MPI_Comm_size(MPI_COMM_WORLD, &np);
   // Exchange seeds between all processors
   int send_size=3*seeds[id].size(); // number of integers
   int* send_buffer = new int[send_size]; // number of integers
   send_size = seeds_to_buffer(seeds[id], send_buffer);
 	int* seed_sizes = new int[np];
-  MPI::COMM_WORLD.Barrier();
-	MPI::COMM_WORLD.Allgather(&send_size, 1, MPI_INT, seed_sizes, 1, MPI_INT);
+    MPI_Barrier(MPI_COMM_WORLD);
+	MPI_Allgather(&send_size, 1, MPI_INT, seed_sizes, 1, MPI_INT, MPI_COMM_WORLD);
 	int total_size=0;
 	for (int i=0; i<np; ++i) total_size+=seed_sizes[i];
 	int* offsets = new int[np];
 	offsets[0]=0;
 	for (int i=1; i<np; ++i) offsets[i]=seed_sizes[i-1]+offsets[i-1];
 	int* seed_block = new int[total_size];
-  MPI::COMM_WORLD.Barrier();
-	MPI::COMM_WORLD.Allgatherv(send_buffer, send_size, MPI_INT, seed_block, seed_sizes, offsets, MPI_INT);
+    MPI_Barrier(MPI_COMM_WORLD);
+	MPI_Allgatherv(send_buffer, send_size, MPI_INT, seed_block, seed_sizes, offsets, MPI_INT, MPI_COMM_WORLD);
 	delete [] send_buffer; send_buffer=NULL;
 
 	for (int i=0; i<np; ++i) {
@@ -437,8 +439,8 @@ void seedswap(std::vector<std::vector<MMSP::vector<int> > >& seeds)
 	delete [] seed_block; seed_block=NULL;
   int vote=1;
   int total_procs=0;
-  MPI::COMM_WORLD.Barrier();
-  MPI::COMM_WORLD.Allreduce(&vote, &total_procs, 1, MPI_INT, MPI_SUM);
+  MPI_Barrier(MPI_COMM_WORLD);
+  MPI_Allreduce(&vote, &total_procs, 1, MPI_INT, MPI_SUM, MPI_COMM_WORLD);
   #ifdef DEBUG
   if (id==0) std::cout<<"Synchronized "<<np*seeds[id].size()<<" seeds on "<<total_procs<<" ranks."<<std::endl;
   #endif
@@ -448,11 +450,10 @@ void seedswap(std::vector<std::vector<MMSP::vector<int> > >& seeds)
 template<int dim>
 void seeds_from_poisson_process(const int x0[dim], const int x1[dim], const int g0[dim], const int g1[dim], const int& nseeds, std::vector<std::vector<MMSP::vector<int> > >& seeds)
 {
-  int id=0;
-  int np=1;
+    int id=0, np=1;
 	#ifdef MPI_VERSION
-  id=MPI::COMM_WORLD.Get_rank();
-  np=MPI::COMM_WORLD.Get_size();
+  MPI_Comm_rank(MPI_COMM_WORLD, &id);
+  MPI_Comm_size(MPI_COMM_WORLD, &np);
 	#endif
   std::vector<MMSP::vector<int> > local_seeds; // blank for now
   seeds.clear();
@@ -482,7 +483,7 @@ void seeds_to_file(const int g0[dim], const int g1[dim], const std::vector<std::
 {
   int id=0;
 	#ifdef MPI_VERSION
-  id=MPI::COMM_WORLD.Get_rank();
+  MPI_Comm_rank(MPI_COMM_WORLD, &id);
 	#endif
 	if (id==0) {
 		std::ofstream output(filename);
@@ -496,18 +497,17 @@ void seeds_to_file(const int g0[dim], const int g1[dim], const std::vector<std::
 		output.close();
 	}
 	#ifdef MPI_VERSION
-	MPI::COMM_WORLD.Barrier();
+	MPI_Barrier(MPI_COMM_WORLD);
 	#endif
 }
 
 template<int dim>
 void seeds_from_file(const int x0[dim], const int x1[dim], const int g0[dim], const int g1[dim], const char* seedfilename, std::vector<std::vector<MMSP::vector<int> > >& seeds)
 {
-  int id=0;
-  int np=1;
+    int id=0, np=1;
 	#ifdef MPI_VERSION
-  id=MPI::COMM_WORLD.Get_rank();
-  np=MPI::COMM_WORLD.Get_size();
+  MPI_Comm_rank(MPI_COMM_WORLD, &id);
+  MPI_Comm_size(MPI_COMM_WORLD, &np);
 	#endif
 	std::ifstream input(seedfilename);
 	if (!input) {
@@ -547,11 +547,10 @@ void seeds_from_file(const int x0[dim], const int x1[dim], const int g0[dim], co
 template<int dim>
 void honeycomb_seeds(const int x0[dim], const int x1[dim], const int g0[dim], const int g1[dim], const int a, std::vector<std::vector<MMSP::vector<int> > >& seeds)
 {
-  int id=0;
-  int np=1;
+    int id=0, np=1;
 	#ifdef MPI_VERSION
-  id=MPI::COMM_WORLD.Get_rank();
-  np=MPI::COMM_WORLD.Get_size();
+  MPI_Comm_rank(MPI_COMM_WORLD, &id);
+  MPI_Comm_size(MPI_COMM_WORLD, &np);
 	#endif
   std::vector<MMSP::vector<int> > local_seeds; // blank for now
   seeds.clear();
@@ -597,7 +596,7 @@ void tessellate(MMSP::grid<dim, MMSP::sparse<T> >& grid, const int& nseeds) {
 	approximate_voronoi<dim,T>(grid, seeds);
 	#else
   exact_voronoi<dim,T>(grid, seeds);
-	MPI::COMM_WORLD.Barrier();
+  MPI_Barrier(MPI_COMM_WORLD);
 	#endif
 } // tessellate