git clone git://github.com/amesshaps/find-my-dress
cd find-my-dress
virtualenv env
source env/bin/activate
pip install -r requirements.txt
make serve
... is a giant pain in the ass, apparently. The hacks in this article
were helpful in getting set up, but they demonstrate a global installation. If you want to install into a local
virtualenv, just symlink the cv.py and cv2.so files into the virtualenv instead:
cd /path/to/virtualenv
ln -s /usr/local/Cellar/opencv/<version>/lib/python/site-packages/cv.py lib/python/site-packages/cv.py
ln -s /usr/local/Cellar/opencv/<version>/lib/python/site-packages/cv2.so lib/python/site-packages/cv2.so
(Honestly, the global install may be a better option if you can live with it, as numpy is also required and can be another pain to install.)
This process is similar to the quickstart above, but saves the frozen requirements and commits them at the end.
cd /path/to/find-my-dress
rm -rf env
virtualenv env
source env/bin/activate
pip install -r requirements.txt
# Symlink OpenCV as described above
# Test server, ensure that all functionality works
pip freeze > requirements.freeze.txt
git add requirements.freeze.txt
git commit -m "Updated frozen requirements"