🚀 go-pugleaf

Hi all,

I have a file with the following data in a file called blah.txt

Ams3vh45           1 0
Onm6io99           2 0
Nny5ev37           3 0
Yka5vj88           4 0
Ttr2yb07           5 0


Now, I wish to sort this file by the first field (I *_cannot_* use
memory - it *_has_* to be on disk), so my thinking is:


Read in first record of blah.txt - if first record write to index file
blah.inx.

Arrive at second line in blah.txt

Start reading blah.inx.
Determine the point in blah.inx I have to insert the second record.

Close blah.inx.

Start reading the records in blah.inx into a temporary file blah.tmp.

When I arrive at the point in blah.inx where the new record has to go,
I insert my new record into the file.

I add the rest of the records to blah.tmp.

I then overwrite blah.inx with blah.tmp, and move to the next record
in blah.txt.


I believe that this algorithm is correct - the only problem is that I
don't really know how to start translating it into java what with all
the FileBuffers, Buffered readers and all sorts of different I/O
classes that are out there.


Any help in this direction much appreciated.


Paul...