Web scraping of json feeds to build an average value and insert into xml file

I need a web script to scrape data of 500 - 600 json feeds similar to this [login to view URL] It needs to collect the item ID e.g "item":10938,", the buyout value e.g "buyout":300000 and the quantity e.g "quantity":10. The buyout value needs to be divided by the quantity to give the price of a single unit of the item ID. All the different prices need to be scraped throughout the json to build an average price for each item ID and inserted into an xml file. The web scraper needs to scan hourly and hold the data for up to a week. Experience using amazon web service to store the data would be a bonus. An example of the average price for the item ID 10938 {"auc":1966885197,"item":10938,"owner":"Dimetri","bid":200100,"buyout":300000,"quantity":10,"timeLeft":"VERY_LONG","rand":0,"seed":1439290368}, {"auc":1967673596,"item":10938,"owner":"Renyaa","bid":33750,"buyout":400000,"quantity":15,"timeLeft":"VERY_LONG","rand":0,"seed":1036371072}, {"auc":1966194443,"item":10938,"owner":"Morpanda","bid":37307,"buyout":45000,"quantity":1,"timeLeft":"MEDIUM","rand":-27,"seed":1066795030}, Average price would be 101667 This would then be inserted into lines similar to these (MaxBuyout being the average price);