Search Results

Put the umask command to the user's login files
cd /home/shared
echo umask 002 >> .profile
echo umask 002 >> .login

Hello,

If I understand correctly, you can use umask option here, as for giving permissions like 755,
we can use command like umask 002 so permissions will be set by the inverse of the umask...

Perhaps a heavily commented and hard coded example?

#!/usr/bin/perl

use strict;
use warnings;

my $in="file";

open my $fh, '<', $in or die "Could not read $in: $!\n";

# start reading...

awk '{print} NR==9{print "newline"}' file

awk 'NR==10{print "line"}1' file

Yes and... by the way the sed statement proposed by RavinderSingh13 also removed the dot in the numbers :

The expected :
723.187 2662.000
became
723187 2662000

which - i guess - is not...

Your problem with this: awk '/unburnt:.*burnt:/{Tu=$6;Tb=$NF}END{print Tu, Tb}' star.log is that you do the print in the END section. This will just print one instance.
Change it to:
awk...

Did you try :
awk '/unburnt:.*burnt:/{print $6, $NF}' star.log

Supposing that you have 1-3 digit user numbers, you could also use:
awk '{print $2 }' ?.usr ??.usr ???.usr > data.txt
without needing to resort to sort and xargs. Obviously, you'll need more terms...

Don't blame awk. It's your shell doing the globbing in lexicographical order.

One way:
printf "%s\n" *.usr|sort -n|xargs awk '...'

spacebar is right, but to fulfill your requirement of many files, you need to supply a number of files, like awk '...' DI*.DAT. Depending on your directory structure, you may want to use e.g. find to...

You can 'print' your output found to a file, for example:
awk '/Sl.*thickness/ {Sl=$3;Tt=$NF}END{print FILENAME ":" Sl, Tt >> "data.out"}' DILAT.DAT

Assuming that you are interested in the last field of rows containing "Thermal thickness"
awk '/Thermal thickness/ { Tt = $NF }'

export the variables and use $ENV{'i'} and $ENV{'j'} in your perl script.

Use single quotes to get these variables to work:
perl -i -pe '/open/&&s/t\d+_p\d+/t$ENV{'j'}_p$ENV{'i'}/' interp.f

You can try:
awk 'BEGIN{ print "sL0", "dL0", "Tb","Tu"}/sL0.*dL0/{sL0=$3;dL0=$6}/Tb.*Tu/{ print sL0, dL0, $3,$6}'

The problem with decompilers is that things like variable names, function names, class members, structure members, symbolic constants, and the like aren't preserved when source code is compiled into...

Linux I assume. objdump can show what the program is doing. There are decompilers
which produce assembler code (google for disassembler). In general most of these kinds of tools are not very...

for file in t*.dat
do
plot "'$file'" ...
done

for i in {1..40};
do
str=$(printf %05d $i)
plot \'"t$str.dat"\' using 1:2
done

How about:-
printf "%s\n" t*.dat | sort | while read file
do
plot '"$file"' using 1:2
done

grep -v '^(\*064) 1' filename

sed '/^(\*064) 1/d' filename

You need to escape the *

Give a try..
find . -name '*.pdf' -type f | while read FILE
do
pdftotext $FILE | grep -q "search phrase"
[ $? -eq 0 ] && echo $FILE
done

If you create a file named grepPDF containing:
#!/bin/ksh
for i in "$@"
do
pdftotext "$i" - | grep 'search phrase' > grepPDF.$$ && \
printf "\nFollowing lines are in...