more information about the project on the make.opendata wiki
(execute all following commands in the git repository)
install flask and PIL:
virtualenv env source env/bin/activate pip install Flask pil
test your installation:
python app/test.py
run the application:
python app/main.py
- install Python 2.7 from http://www.python.org/download/releases/2.7.3/
- install easy_install as described here: http://flask.pocoo.org/docs/installation/#windows-easy-install
- use easy_install to install PIP: to do this, open a Windows command prompt and execute 'easy_install pip'
- use PIP to install FLASK and PIL: to do this, open a Windows command prompt and execute: pip install flask pip install pil
- installing PIL may terminate on error message "error: Unable to find vcvarsall.bat" if this occurs, download a precompiled version PIL-fork-1.1.7.win-amd64-py2.7.exe from http://www.lfd.uci.edu/~gohlke/pythonlibs/ and run this executable (installer file)
- to test whether both have been successfully installed, start the Python GUI/Shell and enter import flask import Image (Note: if 'import Image' fails, try 'from PIL import Image') If Python accepts these commands without error message, flask and Image were successfully installed.
- run the application: double-click on appmain.py
To calculate the pixels from the swiss coordinates, we use the following approximation formula (this conversion is only valid for the provided images!):
x_i = 1.4 * x_c - 658.2 y_i = -1.42 * y_c + 423.32
Where x_i,y_i
are the pixel coordinates and x_c/y_c
the swiss coordinates
divided by 1000 (i.e Bern is at 600/200).