Demo Back-End Web Search Application
(coded in a few hours)
While this mini-project was done quickly, quality was not compromised: It was coded as a production-ready web application (save for making it pretty :). This means it is secure against hacking (eg, SQL-injection attacks, system/server attacks, XSS/Cross-Site Scripting attacks, buffer overflow attacks, etc), and is well-tested/debugged — even for corner cases. (I encourage you to try to break or hack it, and/or to send me a critique of my code if you write software. :).
"Production-ready" to me also means including error detection and reporting, an unbreakable UI, and being implemented with solid software engineering practices (well-structured, maintainable, modular and well-commented code that follows the Model-View-Controller design pattern). Of course, all the HTML and CSS validates and displays properly on all browsers. It represents the quality of database-driven, back-end functionality I could do for your project, even without using any web frameworks, in less than half a day.
How it works: Type in the URL of any website. When you click on "Parse Webpage," the textual content of that web page will be examined by removing all the code, graphics, etc. Then, a tally of all the words longer than 3 characters found on that page will be presented, along with a count of how many times each word appears. The results will also be stored in a MySQL database. When you click on the "Search" link above, you will be able to type in a word and see all the web pages in the database where that word appears, and how many times it appears on each of those pages. Since this is just a quick demo, only HTML-based pages are supported.