Search Results

single quote the &&.

(GNU) sort on Linux has a -z option to sort NULL delimited, multi-line records:
sed 's/^2013/\x0&/' log-file |sort -z |tr -d '\0'

Nice solutions, although the perl solution can be shortened:

perl -le'$_=shift;s/\b(.)/\u$1/g;print' 'are utilities ready for hurricane sandy'
or
perl -ple's/\b./\u$&/g' <<<'are utilities ready...

nl does have the ability to only number lines matching (basic) regex. But it still leaves extra spaces which may not be acceptable to you. Notice the alignment of the second and last line.
nl -s' '...

bash$ set -- are utilities ready for hurricane sandy
bash$ echo "${@^}"
Are Utilities Ready For Hurricane Sandy

If you just want a quick and dirty estimation:
d=$(diff <(echo "$1" |sed 's/./&\n/g') <(echo "$2" |sed 's/./&\n/g') |grep -c '^[<>]')
echo $((100-100*d/(${#1}+${#2})))%

grep . *.phe.ps |sed 's/\.phe\.ps:/ /'

This is how you assign value to variable in bash and most other shells:
var=val
not
set var = val

I will assume you want to change the header but keep the rest of the 30 million lines. Since the new header has the same length, the following will open file for both reading and writing:
sed -Ee...

The offset in TZ for GNU date (or libc?) seemed can't be bigger than 24. You need to convert the day part into -d argument as you said. Either override your date command with a function:
function...

$value=(split(/:/,$req_user_info))[3];

Actually you CAN cat gz files and the result is a valid gzip file.

If you are loading into Oracle using sqlldr, you can specify what you want through a load control file without the need to change the input file. Other loading utilities may have similar capabilities.

A small nitpick, there is a small difference in performance between
for ...;do cmd >> file;done
and
for ...;do cmd; done > file
The latter only opens the output file once.
With GNU sed, you can...

gawk 'BEGIN {
for (i = 1; i < 13; ++i)
m[strftime("%b", mktime("2012 " i " 1 0 0 0"))] = i
FS = "[ :,]+"
}
/all:/ {
print mktime($8 " " m[$2] " " $3 " " $4 " " $5 " " $6)
}'

sftp -b - user@host <<< 'put file'
You can omit the "-b -" part, but then sftp will always exit with 0 even if it fails.

awk '
BEGIN {
n = split("1-9,11-36,38-49,51-56", a, "[^0-9]") - 1
for (i = 1; i <= n; i += 2)
l[i] = a[i+1] - (p[i] = a[i]) + 1
}
{
for (i = 1; i <= n; i += 2) {
x = substr($0,...

You can add:
grep -q 'U:\\FOLDER' "$f" || continue;before sed.

MY_STRING_VAR="QWERTU\
QWERTU\
QWERTU\
QWERTU"

If your file is of fixed length (meaning that each line has the same length), then you can seek to the position of the target line and read the desired number lines.

-v Cause hexdump to display all input data. Without the -v option,
any number of groups of output lines, which would be identical
to the immediately preceding group...

As pointed out from the beginning, you need a "/" before the "exit;". You claimed that it didn't make a difference. But it definitely has to be there.

date +%b%-d

So much interest in this topic, let's try another way for performance since his file is huge. This will edit the file in place and should be blazing fast.

linux$ cat t.c
#include <stdlib.h>...

Options to join used here are:
-t$'\t' use tab as input and output field separator
-a1 -a2 include unmatched lines from the first and second file
-e- use a "-" for missing fields
-j1 use the...