William Brogden wrote:Of course there is a way - you end up reinventing the wheel but you will learn a lot.
You will need to invent:
1. method for identifying files and relating the identity to:
2. parsing words or phrases or whatever you want to index on and:
3. creating a dictionary of words with links to the files they were found in.
(its called an inverted index)
4. creating a lookup mechanism to get from user input to the words in the index and eventually to the files.
Since my first try at this in 1979 (ah CPM, the good old days), I have found multiple solutions to the above - it is very educational.
Bill
Ulf Dittmer wrote:
I dont want to use lucene.
Why not?