I believe what you are talking about is an inverted text index where the
word is used to lookup what document it occurs in. To create this index you
would have to make a program that takes a "document" (be it a row in the
database or a file or whatever), splits it up into words, inserts these
words into a table (assigning each unique word an index number).. and then
for each word index has a list of documents it occurs in. Getting the
results back would just be the inverse of this.
One webpage that I can refer you to is http://www.cs.mu.oz.au/mg/. Managing
gigabytes is a book that covers full text indexing and to aid the readers
they created source code that illustrates the points in the book. This is a
bit more complicated than what I described above.
For hit highlighting usually a straight search/replace is done.. replacing
the search words with <B>searchword</B>.
Hope this helps. I would be interested in your findings could you please
send me info when you finish gathering? Thanks.
ryan
> Could anyone suggest some (specific) resources (books, articles,
> urls, whatevers) about indexing a database? I think what I'm looking
> for is what do you store in the index; how does it get in there and
> then how do you get the information out again?
>
> For example:
> How does hit highlighting (your keywords in bold on the search result
> page) work? I understand that you look up a word in the index and the
> index tells you that the word is found somewhere in some document,
> but how does the results page know how to make your word bold?
>
> Again, my apologies if this is off-topic!! Thanks for all of your
> insight to date. I've learned a lot from reading other peoples'
> questions. :)
>
> emma
> PS All of my findings are going into a summary document that I'm
> going to be giving to my team at work. Let me know if you're
> interested in receiving a copy when I'm done. :)
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
