Generate consolidated Report

[Rahul_13]

Hi,

I have some reports like the one attached along with this thread. I need to write a script which will create a master report with the combined data from all the reports.

For instance in the attached report, I have

ID Attempts Success Failure

11 20 18 2

I need to add Attempts,Sucess, Failure of all the ID=11 from the different files and report to the master file. Similiarly, I need to report the other IDs also.

Also,I need to read the input file names from a configuration file.

Please help. Either PERL or UNIX script would do.

Thanks,
Rahul

[Jean-Luc Oudart]

Hi
this assumes all filenames *.txt and id 2 digits only.
Adapt when required.

###########################################
#!/bin/sh

grep ^[0-9][0-9]" " *.txt | cut -d":" -f2 | sort | awk '{
if($1==oldid) {
attempts+=$2;
success+=$3;
failure+=$4;
}
else {
print oldid,attempts,success,failure;
oldid=$1;
attempts=$2;
success=$3;
failure=$4;
}
}
END{print oldid,attempts,success,failure;}'

Regards
Jean-Luc

[Peter Nikitka]

Hi,
try the following:

awk 'NR==1 {nfield=NF;header=$0;next}
/^[0-9]/ && NF==nfield {ids[$1]=$1;for(i=2;i<=NF;i++) f[$1,i]+=$i }
END {print (header,"\n")
for (id in ids) {printf(id); for(i=2;i<=nfield;i++) printf("\t%s",f[id,i]); printf("\n")}}' file1 .... filen

This will ignore lines not having the identical number of fields of the first header.

mfG Peter

[Muthukumar_5]

You can try with PERL scripting as,

#!/usr/contrib/bin/perl
#master.pl
$i=0;
while (<>)
{
($id,$attempt,$success,$failure)=split;
$I_arr[$id]=$id;
$A_arr[$id]+=$attempt;
$S_arr[$id]+=$success;
$F_arr[$id]+=$failure;
}

foreach $var (@I_arr)
{
if ($var !~ /^$/)
{
printf "ID= $I_arr[$var] Attempt = $A_arr[$var] Success = $S_arr[$var] Failure = $F_arr[$var] \n";
}
}

# perl master.pl > master.pl

will make it.

hth.