This repository contains the following GSLab Python libraries:
- gslab_make
- gslab_fill
- gslab_scons
- gencat
Information about each of these packages is available in its internal documentation.
- Python 2.7
setuptools
(installation instructions)mock
(> 2.0.0) andcoverage
in order to run the unit tests
The preferred installation method is to use pip:
pip install git+ssh://git@github.com/gslab-econ/gslab_python.git@master
or
pip install git+https://git@github.com/gslab-econ/gslab_python.git@master
which are the SSH and HTTPS protocol versions.
The package at any tagged release, branch, or commit can be installed with the same commands, just changing master
to the desirved target e.g.,
pip install git+ssh://git@github.com/gslab-econ/gslab_python.git@<tagg, branch name, or commit hash>
Note that this installation procedure may require obtaining machine privileges through,
say, a sudo
command.
Alternatively, one may install the local version of gslab_python by running (from the root of the repository)
pip install .
We do not reccommend that these packages be installed by executing
python setup.py install
This method of installation uses egg files rather than Wheels, which can cause conflicts with previous versions of gslab_tools
. If this method of installation is executed, some files need to be removed from the directory with a clean
argument. clean
removes /build
,/dist
, and GSLab_Tools.egg-info
, which are built upon installation. This argument can be called by executing
python setup.py clean
We recommend that users use coverage
to run this repository's unit tests. Upon installing coverage (this can be done with
pip using the command pip install coverage
), one may test gslab_python
's contents
and then produce a code coverage report the commands:
python setup.py test [--include=<paths>]
Here, the optional --include=
argument specifies the files whose test results
should be included in the coverage report produced by the command.
It works as coverage
's argument of the same name does. The command should be
run without this option before committing to gslab_python
.
See here.
Q: What if I want to install a different branch called dev
of gslab_python
rather than master
?
A: Either git checkout dev
that branch of the repo before installing, or change @master
to @dev
in the pip install
instruction.