To run the TimeGrouper backend localy in a python virtual environment, follow these steps:

1. Run ./setup.sh on Linux/Mac and setup.bat on Windows. This will install all python packages needed in a virtual environment in flask/

2. Activate the virtual environment source flask/bin/activate

3. Run python app.py

The RESTful service should now be running at: http://localhost:5000/

Service Endpoints:

• GET: http://localhost:5000/getSimMatrix/{_appname_}

• POST: http://localhost:5000/getSimMatrix -- (Note: If #patches > 100 the summary of the similarity matrix is included in the response as the first json array.)

1. with {simMetric : 'some-similarity-Metric', cAlgorithm : 'clustering-algorithm'} these are required
2. with {filter : 'chrome' , filter : 'firefox' ...etc} in the request to filter by app name
3. with {updateMech : 'SU', updateMech : 'PD' ...} to filter by update mechanism
4. {exploitable='true'} to filter by verified exploitability (true if patch has been exploited in the past)

• GET: http://localhost:5000/getPatches/{_patch_id_}

• POST: http://localhost:5000/getPatches

1. with {patchId : 'chrome_001' , patchId : 'firefox_003' ...etc} in the request
2. with {highlight : 'chrome', updateMech : 'firefox' ...} to highlight by app name (highlights the similarity matrix positions that correspond to a similarity for two patches that both fit the highlighting condition). Returns [i,j] pairs for the boxes that should be highlighted