awk in UNIX script producing wrong results

[pluscarden]

Hi...

Can anyone help me with the following please...

I have a UNIX script that does something along the following lines...

#!/usr/bin/ksh
...
...
...
...
for i in client_file
do
awk -f myfile.awk -v client=$i report
done
...
...
...

-----------------------------------------------

My problem is that the 'for...done' command works fine when I run it on the command line (I get the expected results from the awk script) but it doesn't work properly when I run the UNIX script itself.

Any ideas?

Thanks in advance

 

[pluscarden]

P.S.

A bit more info....

The awk script itself is just doing something along the following lines...

BEGIN {
client_found=0
category=0
}

/Category 1/ {category=1}

/Category 2/ {category=2}

/Category 3/ {category=3}

$0 ~ client {
if ( category == 1 ) {
bad[client]=""
client_found=1
}
else if ( category == 2 ) {
bad[client]=""
client_found=1
}
else if ( category == 3 ) {
if ( client_found == 0 ) {
good[client]=""
client_found=1
}
}

END {
if ( client_found == 0 ) {
bad[client]=""
}
for ( b in bad ) {
print b >> "badfile"
}

for (g in good ) {
print g >> "goodfile"
}
}


----------------------------

The report file looks something like this...

Category 1
client1
client2
client3

Category 2
client2
client4

Category 3
client1
client5
client6

------------------------------------------------

and clientfile contains the following...

client1
client2
client3
client4
client5
client6
client7
client8

-------------------------------------------------

Expected result is that "badfile" should end up containing the following :

client1
client2
client3
client4
client7
client8

and that "goodfile" should end up contaning the following...

client5
client6

------------------------------------------------------

as I said, this works fine when I run it on the ommand line. But when I run it via the UNIX script, it just puts every single client into the "badfile" and oesn't even bother creating a "goodfile"

 

[PHV]

> for i in client_file
Have you tried this:

for i in $(<client_file)

Hope This Help
PH.

 

[aigles]

You read the report file for each client.
The following awk program, proceed all clients in a single awk program execution.

awk -f my_file.awk report client_file

[tt]
#
# GoodBad - Memorize clients from report file
# Args: report_file
# Result: Good[] = Clients from category 3
# Bad[] = Other clients
#

function GoodBad (report_file, good_category) {
good_category = 0;
while (getline < report_file) {
if ($0 ~ /^Category /) {
good_category = ($2 == 3);
} else {
if (good_category && $1 != &quot;&quot
Good[$0] = &quot;&quot;;
else
Bad[$0] = &quot;&quot;;
}
}
}

#
# Initialisations - Read report_file
#

BEGIN {
GoodBad(ARGV[1]);
ARGV[1] = &quot;&quot;;
if (ARGC == 2) ARGV[ARGC++] = &quot;-&quot;;
}

#
# Read client file -
# Separate the good clients from bad
#

{
if ($0 in Bad || ! ($0 in Good))
print $0 >> &quot;badfile&quot;;
else
print $0 >> &quot;goodfile&quot;;
}
[/tt]

Jean Pierre.