This source is licensed under GPLv3, see LICENSE.txt (note: metacademy's content is CC BY SA 3.0).
OCTAL is based upon Metacademy: https://github.com/metacademy/metacademy-application With a live demo at http://www.metacademy.org
The following instructions bring up a development version of OCTAL. Look at the production directions for creating a production-level OCTAL server.
-
python 2.7X
: NB: installpython-devel
if you're using a package manager such as apt-get. The following command should print a path (if this command raises an exception, make sure that you have thepython-devel
version installed):python -c 'from distutils.sysconfig import get_makefile_filename as m; print m()'
-
gcc
: Install gcc (OSX users: installing the OSX developer tools is probably the easiest way to do this). The following command should not throw an error:gcc -v
-
pip
Install pip; here are pip's installation instructions. The following command should not throw an error:pip -V
-
node.js
(needed to run tests): Install node The following command should not throw an error:node -v
-
Make sure you have installed the following packages:
freetype-devel gcc-gfortran (Linux) freetype gfortran (Mac OS X, via homebrew) easy_install virtualenv (both)
-
On Mac OS X, setup freetype:
ln -s /usr/local/opt/freetype/include/freetype2 /usr/local/include/freetype
-
create a top-level octal directory in a desired location, e.g.
mkdir -p ~/MyProjects/octal
-
go to the top-level octal directory
cd ~/MyProjects/octal
-
from your top-level octal directory, clone the application repo:
git clone https://github.com/danallan/octal-application.git
-
go to the octal-application directory
cd octal-application
-
install the octal application (note: this project uses a virtual environment for development):
make
If you are on OS X and have a version of Xcode >= 5.1, you might need to prefix the above command with this:
ARCHFLAGS=-Wno-error=unused-command-line-argument-hard-error-in-future make
python app_server/manage.py createsuperuser
OCTAL currently depends scipy
. Scipy can be tricky to install (pip install scipy
won't work unless you have all of the systems-level dependencies), but Google might help here. Once that's done, link its site-packages folder to to your virtual environment's lib/python2.7/site-packages/
folder. One example (you might have to adapt to your install):
ln -s /usr/local/lib/python2.7/site-packages/scipy lib/python2.7/site-packages/
-
start the virtual environment (you must do this for each new session)
source ../meta_venv/bin/activate
-
start the test servers
./run_test_servers.sh
-
Open localhost:8080 with a modern javascript-enabled browser
Note: the initial search may take some time to load (the backend must load the graph into memory), but all subsequent pages should load much quicker.
Since this installation is based upon metacademy, if you have problems with the installation please be sure to look at the issues at https://github.com/metacademy/metacademy-application/issues?state=open