I need to make finding precise words or phrases in Word documents (I put this document as plain text to a field text type of a table in my database). I'm working with PostgreSQL 8.3.1 on Linux; before, working with tsearch2-in version 8.2.x-could find presence or absence of a word in the text field, in reality what I need is to count the number of occurrences in this field a word or phrase perfect (two or more words that appear in the same order). I repeat: what we need is "counting" how many times you see a pattern in that field type text.
Be one of you who has done something similar to me of some kind to improve what I am doing?.