amlsf / hb-project Goto Github PK

SupplyCache is a web application that helps communities locate resources after a disaster occurs. It maps resources available by residents and the areas affected by a disaster. User input can be via text message or web form. An archived events page builds data visualization for post-disaster analysis. SupplyCache uses Python, Flask, SQLAlchemy, Jinja, JavaScript (Heatmap, Rickshaw), HTML/CSS, Leaflet API, Twilio API, CloudMade API.

Leaflet is an open source JavaScript library for maps: the map data is from OpenStreetMap, an open source map created by communities with an emphasis on local knowledge, and map tiles (style) are from CloudMade. The heatmap layer uses heatmap.js and is integrated into my project to display areas affected by a disaster. For example, the redder the area, the more severe the situation. Therefore, the user should avoid those areas when traveling to reach another person for a supply. The marker clustering function uses the Marker Clustering plugin. When zoomed out, the markers cluster together to better visualize the number of residents with supplies. User input is taken in via text message using Twilio or through the provided web form.

Archived events are accessed via a login due to the sensitivity of the information provided. Once logged in, users will see a list of past events. Each event has a "view summary" button that expands to show a graph that displays supply trends over time. In addition, the graph legend allows the user to toggle between the different supplies. The graph uses Rickshaw, which is a JavaScript graphing toolkit built on D3. In addition, below the graph is a table that contains a general summarization of the data. The "view map" button takes the user to a separate map that repopulates the archived database for further analysis. For example, a user can see the heaviest hit areas during a particular disaster and determine what can be done better next time.

User input is stored into a database using Python and SQLAlchemy. For demo purposes, the database has been seeded. The database is queried and used to populate the maps, graphs, and tables.