This project writes a simple web page crawler, computes a simple version of Google's Page Rank algorithm and visulizes the resulting network. spider.py: read pages from the web, get the link followed and about to be followed spider.sqlite: database that stores the link information sprank.py: compute the page rank of the data inside spider.sqlite spreset.py: reset the data spjson.py: dump data from spider.sqlite into force.js force.html: visualize page rank data spdump.py: dump data into text format Run: follow the running process shown in the schematic