Here's an example...
sysopen(DF,"test.txt", O_RDONLY | O_CREAT);
sysread(DF, $rec, -s DF);
# Split up records into array.
# Will lose \n on recs - add later.
# Do some work on @test
Work... Work... Work...
# Build up into a single record, putting \n back in
sysopen(DF,"test.txt", O_WRONLY | O_CREAT);
Does that help set the scene a little better? :)
So basically I want to make this as efficient as possible. I have to use SYSREAD and SYSWRITE to make the IO as optimal as possible. ie: Reading a 1 meg file would take 2000 IO processes if buffered!!!!! SYSREAD will in effect take 1!