Shell Programming and Scripting

Can anyone help me in giving a script for the below scenario

I have file from the source increamenting in size...

I require to write a script witch will move the data to the new file once the file reaches 50MB of size .
This needs If the first file reaches 50MB then my script has to generate a new file which again store the data upto to 50MB size and so on..... untill the
end of the file.


1st file -- 50MB
2nd file -- 50MB
3nd file -- 50MB
------
------------
Thanks

Take a look at the 'split' command, such as,

Code:

split -b 50MB orig_file prefix

NOTE: not tested!

Man Page for split (OpenSolaris Section 1) - The UNIX and Linux Forums

Usually, the trick is disconnecting the file from the application that is writing it. Sometimes, make the file a named pipe and control copying from that pipe as you please. Named pipes are a little tricky, with different behaviors for open for read and write in different orders. Play with one in the shell to find which goes first -- I forget.

The way I understood is that you want to be in a loop where there is a main file that keeps having records inserted into it and
when it reaches 50Mb, you want to copy into another file, then empty the main file.

You decide:

Code:

#!/bin/ksh
typeset -i mLastLine mSize
mSize=$(wc -c < xxx)
if [[ ${mSize} -ge 50000000 ]]; then
  cp xxx xxx_50k
  mLastLine=$(wc -l < xxx_50k)+1
  tail +${mLastLine} xxx > xxx
fi