Developers’ Guide

See also the following pages from Developers’ resources:

Guidelines for contributing

When preparing contributions, please follow the guidelines in contribution. Also:

  • If the planned changes are substantial or will be backward-incompatible, it’s best to discuss them on the claw-dev Google group before starting.

  • Make sure all tests pass and all the built-in examples run correctly.

  • Be verbose and detailed in your commit messages and your pull request.

  • It may be wise to have one of the maintainers look at your changes before they are complete (especially if the changes will necessitate modifications of tests and/or examples).

  • If your changes are not backward-compatible, your pull request should include instructions for users to update their own application codes.

Reporting and fixing bugs

If you find a bug, post an issue with as much explanation as possible on the appropriate issue tracker (for instance, the PyClaw issue tracker is at https://github.com/clawpack/pyclaw/issues. If you’re looking for something useful to do, try tackling one of the issues listed there.

Developer communication

Developer communication takes place on the google group at http://groups.google.com/group/claw-dev/, and (increasingly) within the issue trackers on Github.

Installation instructions for developers

Cloning the most recent code from Github

You can create a read-only development version of Clawpack via:

git clone https://github.com/clawpack/clawpack.git
cd clawpack
git submodule init
git submodule update

Note: The https://github.com… form of specifying a remote clones the repository in a form that does not allow pushing to it (unlike the git@github.com:… form). This is good practice, so you do not accidently try to push to the main clawpack repository rather than to your own fork (see Adding your fork as a remote below if you need your own fork, e.g. for issuing pull requests).

The commands above download the following clawpack modules as subrepositories checked out at specific commits (as opposed to the tip of a branch).

This should give a snapshot of the repositories that work well together. (Note that there are many inter-dependencies between code in the repositories and checking out a different commit in one repository may break things in a different repository.)

Now install this version of Clawpack using:

pip install -e .

The -e flag means that this is an editabl version of the Clawpack code (rather than installing the original version in the site-packages directory).

If you want to use the Fortran versions in classic, amrclaw, geoclaw, etc., you need to set environment variables and proceed as described at Set environment variables.

Checking out the master branch on each repository

Following the instructions above gives you a top level $CLAW directory that is checked out to the tip of the master branch, and each subrepository will be checked out to a particular commit as specified by this master branch. For development work, You probably want to check out each subrepository to the master branch as well. The shell script $CLAW/pull_all.sh can be used to do this for all subrepositories (or look at this file to see how to do it more selectively). At a shell prompt, type:

source pull_all.sh

which will check out master on each repository and then do a git pull to make sure it is up to date. If you do this shortly after cloning all the repositories, they should all have been up to date already.

Updating to the latest master branch

The script pull_all.sh can be used at any time to check out all subrepositories to master and do a git pull. This is handy if you want to make sure your version of master is up to date in every repository.

You should first make sure that you do not have uncommitted changes in any repository that might conflict with the git checkout master or git pull commands. You can do this easily with the command:

python $CLAW/clawutil/src/python/clawutil/claw_git_status.py

and then check the files claw_git_status.txt and claw_git_diffs.txt, which summarize the status of each subrepository.

Never commit to master

You should never commit to master, only to a feature branch, so the master branch should always reflect what’s in the master branch on the primary Github repositories.

You can update master to reflect any changes via the above approach (for all subrepositories at once), or do git checkout master and then git pull within any of the subrepositories separately.

Adding your fork as a remote

If you plan to make changes and issue pull requests to one or more repositories, you will need to do the following steps for each such repository:

  1. Go to http://github.com/clawpack and fork the repository to your own Github account. (Click on the repository name and then the Fork button at the top of the screen.)

  2. Add a remote pointing to your repository. For example, if you have forked the amrclaw repository to account username, you would do:

    cd amrclaw
git remote add username git@github.com:username/amrclaw.git

    provided you have ssh keys set up, or else:

    if you don’t mind having to type your password whenever you push or pull.

    You should push only to this remote, not to origin, e.g.:

    git push username

You might also want to clone some or all of the following repositories:

These are not brought over by cloning the top clawpack super-repository. You can get one of these in read-only mode by doing, e.g.:

git clone https://github.com/clawpack/doc.git

Then go through the above steps to add your own fork as a remote if you plan to modify code and issue pull requests.

Note: The https://github.com… form of specifying a remote clones the repository in a form that does not allow pushing to it (unlike the git@github.com:… form). This is good practice, so you do not accidently try to push to the main clawpack repository rather than to your own fork.

Modifying code

Before making changes, make sure master is up to date:

git checkout master
git pull

Then create a new branch based on master for any new commits:

git checkout -b new_feature master

Now make changes, add and commit them, and then push to your own fork:

# make some changes
# git add the modified files
git commit -m "describe the changes"

git push username new_feature

If you want these changes pulled into master, you can issue a pull request from the github page for your fork of this repository (make sure to select the correct branch of your repository).

Note: If you accidentally commit to master rather than creating a feature branch first, you can easily recover:

git checkout -b new_feature

will create a new branch based on the current state and history (including your commits to master) and you can just continue adding additional commits.

The only problem is your master branch no longer agrees with the history on Github and you want to throw away the commits you made to master. The easiest way to do this is just to make sure you’re on a different branch, e.g.,

git checkout new_feature

and then:

git branch -D master
git checkout -b master origin/master

This deletes your local branch named master and recreates a branch with the same name based on origin/master, which is what you want.

Issuing a pull request

Before issuing a pull request, you should make sure you have not broken anything:

  1. Make sure you are up to date with master:

    git checkout master
git pull

    If this does not say “Already up-to-date” then you might want to rebase your modified code onto the updated master. With your feature branch checked out, you can see what newer commits have been added to master via:

    git checkout new_feature
git log HEAD..master

    If your new feature can be added on to the updated master, you can rebase:

    git rebase master

    which gives a cleaner history than merging the branches.

  2. Run the appropriate regression tests. If you have modified code in pyclaw or riemann, then you should run the pyclaw tests. First, if you have modified any Fortran code, you need to recompile:

    cd clawpack/
pip install --user -e .

    Then run the tests:

    cd pyclaw
nosetests

    If any tests fail, you should fix them before issuing a pull request.

To issue a pull request (PR), go to the Github page for your fork of the repository in question, select the branch from which you want the pull request to originate, and then click the Pull Request button.

Testing a pull request

To test out someone else’s pull request, follow these instructions: For example, if you want to try out a pull request coming from a branch named bug-fix from user rjleveque to the master branch of the amrclaw repository, you would do:

cd $CLAW/amrclaw   # (and make sure you don't have uncommitted changes)
git checkout master
git pull  # to make sure you are up to date

git checkout -b rjleveque-bug-fix master
git pull https://github.com/rjleveque/amrclaw.git bug-fix

This puts you on a new branch of your own repository named rjleveque-bug-fix that has the proposed changes pulled into it.

Once you are done testing, you can get rid of this branch via:

git checkout master
git branch -D rjleveque-bug-fix

Top-level pull requests

The top level clawpack repository keeps track of what versions of the subrepositories work well together.

If you make pull requests in two different repositories that are linked, say to both pyclaw and riemann, then you should also push these changes to the top-level clawpack repository and issue a PR for this change:

cd $CLAW   # top-level clawpack repository
git checkout master
git pull
git checkout -b pyclaw-riemann-changes
git add pyclaw riemann
git commit -m "Cross-update pyclaw and riemann."
git push username pyclaw-riemann-changes

Git workflow

See git-resources for useful links.

Catching errors with Pyflakes and Pylint

Pyflakes and Pylint are Python packages designed to help you catch errors or poor coding practices. To run pylint on the whole PyClaw package, do:

cd $PYCLAW
pylint -d C pyclaw

The -d option suppresses a lot of style warnings, since PyClaw doesn’t generally conform to PEP8. To run pylint on just one module, use something like:

pylint -d C pyclaw.state

Since pylint output can be long, it’s helpful to write it to an html file and open that in a web browser:

pylint -d C pyclaw.state -f html > pylint.html

Pyflakes is similar to pylint but aims only to catch errors. If you use Vim, there is a nice extension package pyflakes.vim that will catch errors as you code and underline them in red.

Checking test coverage

You can use nose to see how much of the code is covered by the current suite of tests and track progress if you add more tests

nosetests --with-coverage --cover-package=pyclaw --cover-html

This creates a set of html files in ./cover, showing exactly which lines of code have been tested.

Trouble-Shooting Tips

If you are having trouble installing or building Clawpack try out some of the following tips:

  • Check to see if you have set the environment variable FFLAGS which may be overriding flags that need to be set. This is especially important to check when building the PyClaw fortran libraries as a number of flags must be set for the Python bindings and will override the defaults.