todays challenge: automate csplit of large file

[WiccaChic]

I am totally trying to leach of your brain matter. We run AIX and sometime we get huge text files that we need to split into say four equal (or near equal) parts. We use the csplit command to do this. For instance I just di the following for file BIGFILE:

$: wc BIGFILE
3322657 176809963 3623691354 BIGFILE

So I divided 3322657 by four, did some rounding and addition to get my csplit starting points and then did:
$: csplit pdc_txt_P0055.txt 830664 1661328 2491992

Which got me four fractions of the original data. Can someone show me how to script this out so my ops could just enter a file name and enter the number of parts to split it into?

 

[toolkit]

Instead of csplit, why not use split -b <bytes> ?

http://www.google.co.uk/search?num=20&hl=en&ie=UTF-8&q=unix+man+split&meta=

 

[PHV]

Something like this ?
read f?"File name "
read n?"Number of parts "
typeset -i l=$(wc -l < $f)
csplit -k $f $(((l+n-1)/n)) {$n}

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244

 

[WiccaChic]

PHV!!! U'da man! Or woman as far as I know. Thanks!!!