Here is a rough idea for the website portion:
- Create a database of every URL on your net
- Your search app could then query the databse to get a url
- Then open a stream to the url and read in the html file
- Use a
string (or stream?) tokenizer to go through every
word - Keep track of each occurrence of the desired keyword
- Output the result to the user
You could implement a
servlet to do the actual work, this would make it easy to output results as an html page to the user
Anyone care to expand upon this?
------------------
- Ryan Burgdorfer
- Java Acolyte in
- Columbus, OH USA