Build OCR Web Scavenging Program

Build a small program using either javascript or python that scavenges price data presented as images on the web and converts the images into text using the OCR tessaract engine(opensource). The program will then analyze and prioritize the data using predetermined criteria. The program will then manipulate the important data slightly and then output it as a text file. The program will upload the data to a website. This program should be automated to run about twice daily(this may change depending on how long it takes to process the entire price lists). Candidates should have a strong background in: Html and scavenging html pages Databases(this project will likely involve 20k+ data points) Best practices for programming