How to exclude parent Unix processes from grepped output from ps - unix

Question

I have got a file of pids and am using ps -f to get information about the pids.
Here is an example..
ps -eaf | grep -f myfilename
myuser 14216 14215 0 10:00 ? 00:00:00 /usr/bin/ksh /home/myScript.ksh
myuser 14286 14216 0 10:00 ? 00:00:00 /usr/bin/ksh /home/myScript.ksh
where myfilename contains only 14216.
I've got a tiny problem where the output is giving me parent process id's as well as the child. I want to exclude the line for the parent process id.
Does anyone know how I could modify my command to exclude parent process keeping in mind that I could have many process id's in my input file?

Answer 1

Hard to do with just grep but easy to do with awk.
Invoke the awk script below from the following command:
ps -eaf | awk -f script.awk myfilename -
Here's the script:
# process the first file on the command line (aka myfilename)
# this is the list of pids
ARGIND == 1 {
pids[$0] = 1
}
# second and subsequent files ("-"/stdin in the example)
ARGIND > 1 {
# is column 2 of the ps -eaf output [i.e.] the pid in the list of desired
# pids? -- if so, print the entire line
if ($2 in pids)
printf("%s\n",$0)
}
UPDATE:
When using GNU awk (gawk), the following may be ignored. For other [obsolete] versions, insert the following code at the top:
# work around old, obsolete versions
ARGIND == 0 {
defective_awk_flag = 1
}
defective_awk_flag != 0 {
if (FILENAME != defective_awk_file) {
defective_awk_file = FILENAME
ARGIND += 1
}
}
UPDATE #2:
The above is all fine. Just for fun, here's an alternate way to do the same thing with perl. One of the advantages is that everything can be contained in the script and no pipeline is necessary.
Invoke the script via:
./script.pl myfilename
And, here's script.pl. Note: I don't write idiomatic perl. My style is more akin to what one would expect to see in other languages like C, javascript, etc.:
#!/usr/bin/perl
master(#ARGV);
exit(0);
# master -- master control
sub master
{
my(#argv) = #_;
my($xfsrc);
my($pidfile);
my($buf);
# NOTE: "chomp" is a perl function that strips newlines
# get filename with list of pids (e.g. myfilename)
$pidfile = shift(#argv);
open($xfsrc,"<$pidfile") ||
die("master: unable to open '$pidfile' -- $!\n");
# create an associative array (a 'hash" in perl parlance) of the desired
# pid numbers
while ($pid = <$xfsrc>) {
chomp($pid);
$pid_desired{$pid} = 1;
}
close($xfsrc);
# run the 'ps' command and capture its output into an array
#pslist = (`ps -eaf`);
# process the command output, line-by-line
foreach $buf (#pslist) {
chomp($buf);
# the pid number we want is in the second column
(undef,$pid) = split(" ",$buf);
# print the line if the pid is one of the ones we want
print($buf,"\n")
if ($pid_desired{$pid});
}
}

Answer 2

Use this command:
ps -eaf | grep -f myfilename | grep -v grep | grep -f myfilename