Problem with reading a huge text file efficiently

[buntha Choudhary]

I have a huge text file and my job is to read the content of the file and add those records in a JTable.
Simple reading of such a huge file is very inefficient.
What is the most efficient possible way to solve this issue and please provide an example.

[Christian Dillinger]

What is "huge"? How many lines? How many columns? Do you have to show everything in your table or just some fields from each line?

[buntha Choudhary]

consider it is about 50000 lines with having 3 columns.
I need to show all these records but can't use any general reading technique.

[Stephan van Hulst]

We're talking what 200 - 500 kb here?

Why can't you use a simple BufferedReader? Have you tried it? Doesn't it perform well enough?

[Christian Dillinger]

If it's 100 byte per line we are talking about 5 MB! But that's by far not to much hold in memory... I'd read to a List<String[]> and put that list into a simple TableModel.

[Steve Luke]

You might find using the NIO mapped files a means of speeding up the I/O functions.

    FileInputStream fis = ...;
    FileChannel channel = fis.getChannel();
    MappedByteBuffer buffer = channel.map(FileChannel.MapMode.READ_ONLY, 0, channel.size());

[William Brogden]

Wait a second

You are trying to present a swing JTable with 50,000 lines??

How in the name of sanity do you expect somebody to navigate in that?

Isn't there some kind of overall structure you can use to present smaller JTables? Maybe some sort of partial match or alphabetic selection?

Bill

[J. Insi]

i thought that what he did was,
50,000 lines as a total.

But in the application runs... it was divided into several pieces... 100 lines per page perhaps?
well, not sure.

But yes, using the buffer method works fine for reading at the time IO works...