Trying to figure out a efficent way to do this...

Question

Ok, I'm given a dictionary.txt file which holds 234,941 entries, seperated line by line. I need to pick X random entries from the file and stick them in an array (or possibly another type of data).
currently, I'm using a BufferedReader to read each line of the file, then sticking each word in to a temporary array of size 234,941
Then I pick X random #'sfrom 0 to 234,941 and stick the elements in to my new array (what I'm returning)
I'm thinking that there must be a more efficient way to do this. Plus I want to add flexibility if the dictionary file has more or less than the expected 234,941 entries. I know I could add an enlarge method to enlarge the array by making a new one (of larger size) and copying the contents of the old one into the new one - but that would be grossly inefficient I'd think...
any ideas? :)

Answer 1

* If all words are the same length (or are padded with spaces to make the start of each word at a predictable byte number) then you can just divide the file size by the known fixed length to get number of words, pick some random numbers in that range, and then read those words from the file.
* If you can know the number of words ahead of time (either because you build the file in the first place, or because you reuse the same file multiple times and you've already been through it once) then again, you can pick the random numbers first, then read the file until you've gotten those words.
* You could just get the length of the file in bytes, and then pick a random number between 0 and that number. Read to that byte (either line by line, checking each time if you've passed it, or directly with RandomAccessFile) and then move forward or back to the previous or next beginning of a word. If you have short or long words clustered at one end of the file or the other, this could skew the randomness some, but depending on what you're doing, it may be okay.