Shell Programming and Scripting

Hello,

I am new here, so first of all I want say hello to everyone.

I am newbie on script but you may be able to help me on this :

I am on solaris (ksh)
I need to compare 2 files (one_trash.log / two_arch.log)
On the first file I've got like 1000 entries files name
Ex of one_trash.log :
one.tar
two.tar ..etc

On the second file I've got the repertories and the a lots of files name ~ almost 5000 entries
Ex of two_arch.log :
Rep1/rep2/rep3/one.tar
Rep10/rep21/rep32/two.tar ..etc

Here is my problem :
I need to create a file (ex match.log) that will contain all entries found on the one_trash.log that match with the two_arch.log. This file must contain the repertories and the file name

ex : Rep1/rep2/rep3/one.tar

I need this to see if the file name given on the first file one_trash.log exist on the two_arch.log and where this file is exactly stored.

here is an example of the two files :

one_trash.log

NCBO1T1_RTSTAT_20001227_20010103_S20010117
UK_IPSWITCHPRE_20031217_1452_1.TAR
UK_LONDONPSSPRE_20020327_0731_1.TAR


two_arch.log

OPE/NCB1/RTSTAT/NCBO1T1_RTSTAT_20010307_20010314_S20010328
OPE/NCB1/RTSTAT/NCBO1T1_RTSTAT_20010620_20010627_S20010704


I am sorry about my english.

Thanks a lot for your help. I lost on this. PLEASE

Try:

Code:

fgrep -f one_trash.log two_trash.log

The suggested fgrep -f could incorrectly match a filename from one_trash.log to a directory name in two_arch.log.

Try this:

Code:

awk -f aswex.awk one_trash.log two_arch.log

where aswex.awk contains:

Code:

BEGIN { FS = "/"}
FNR == 1 { fn++ } # track the file number
fn == 1 {
  a[$0] = 1
  }
fn == 2 && a[$NF]

That's a good point.

I'll just note that the awk code could be a bit simpler and easily on the command line:

Code:

awk -F/ 'FNR == NR { a[$0]++; next } a[$NF]'  \
one_trash.log two_arch.log

Dear Friends,

I am so sorry to be so late to reply to your posts. Thanks a lot for your help.

I am coming back to you because none of your script or command line seems to work for me.

For sure I am not good enough to make them work and I am unable to find a way to compare this 2 log files.

The error code is :

awk: syntax error near line 1
awk: bailing out near line 1

here is what I've tried :
from the shell "ksh" (I am on solaris 9)

awk -f/ 'FNR == NR { a[$0]++; next } a[$NF]' \
what else ? > one_trash.log two_arch.log

And i've got the errors.

If you need more info, just let me know.

Thanks again for all your help

use 'nawk' (or /usr/xpg4/bin/awk) instead of 'awk' on Solaris.

Hi,

do you mind tell me what i need to do ? Here is the error code I've got

nawk: can't open file FNR == NR { a[$0]++; next } a[$NF]
source line number 1
context is
>>> <<<


I've done this :

nawk -f/ 'FNR == NR { a[$0]++; next } a[$NF]' \
what else ? > one_trash.log two_arch.log