Search Results

Give this perl script a try...

#!/usr/bin/perl

use Time::Local;

while (<>) {
@f = split /\|/, $_;

($y1, $o1, $d1) = (substr($f[3], 0, 4), substr($f[3], 5, 2), substr($f[3], 8,...

With recent shells with brace expansion:var=7-15
eval echo {${var/-/..}}
7 8 9 10 11 12 13 14 15
The usual caveats for using eval apply.

Good to hear :)

1.
newvar=$(printf "%e\n" "$(bc -l <<< "${EV/[eE]/*10^}*${DV/[eE]/*10^}" )")

2.
$ printf "%e\n" "$(bc -l <<< "${EV/[eE]/*10^}*${DV/[eE]/*10^}" )"
5.000000e-04
$ printf...

How about this:

IPS="10.66.1.133 10.66.6.133 10.66.7.133 10.66.0.133
10.66.1.5 10.66.2.133 10.66.4.133 10.66.3.133
10.66.5.133 10.66.2.5 10.66.3.5 10.66.7.5
...

Write that as multi-liner, and it is visibly easy to extend
sed '
s/WEND/W_END/
s/CLOSE/C_LOSE/
' f.txt

Yes definitely it reads the file only once.

Run that line with the -x option set to see what causes the trouble. It may be sufficient to escape the $ signs.

Embedding the remote code in local code denotes a second evaluation.
Instead have the script code in a separate file script.sh and run it remotely with
for server in server1 server2
do
ssh -qx -l...

Restricting RSS feeds has shown to be counter productive to our search engine rankings and directly (negatively) impacted traffic.

Therefore, effective immediately, I have removed the RSS...

Rather than matching whole line why no just match on the "snmpd stop" string:

sed -i '/snmpd stop/d' /var/spool/cron/root

tr and sed have no concept of 'column', and trying to brute-force a regex that does can be confusing.

awk supports columns directly.

awk -F"|" -v OFS="|" '{ for(N=1; N<=NF; N++) if($N == "@")...

Never overwrite your originals; that's a recipe for disaster. One mistake and all your code is blown away.

I'd suggest renaming your '.m' files into '.pm' files instead. Create '.m' files out...

If you don't get any responses, you might be better off posting this at python-forum (http://www.python-forum.org/pythonforum/index.php).

Try:
sed '/^command="/r file1' file2

cat file | tr "\n" " "

Hello Jaysun,

I posted a sample script to show how to increment the variable and use it in find. You might need to adopt the same technique and do the changes accordingly.

Regards
Ravi

See if this works for you:
#!/usr/bin/ksh
typeset -i mNbr=1
while [[ ${mNbr} -le 24 ]]
do
mDirA="/mnt/raw_vendor_data/raw_${mNbr}_hist/primary/"
...

$ for i in {2..24}; do mkdir ${i}_MAY_2011 ; done

I think you want "Source3" in all the "tags" in the last record. Try this script: #!/usr/bin/perl
open I,"$ARGV[0]";
local $/;
$_=<I>;
$/="\n";
chomp $_;
@x=split /\n\n/,$_;
$x[$#x+1]=$x[$#x];...

What kind of data is this? Many small files, many large ones, or mixed? If it's many small files, it will take longer as rsync has to build an index of all files in both directories.

Generally,...

OK, fired up my old RHEL5.3 harddisk and had a look for hard links. The following hard links exists in /usr/bin

-rwxr-xr-x 2 root root 20044 Jan 5 2009 iconvconfig
-rwxr-xr-x 2 root...

Hard links are harder to spot since they don't tell you where they all are.

One use I've abused hard links for is renaming a file that's being written to. Since the file still exists through all...

Interesting. Obviously a bad example for that strain of Linux, but still a valid example for mainstream unix.
In unix when the links count for a file is greater than value 1 this signifies a hard...

Common example is the "ls" command.

First find the inode <number> of the ls command.

ls -lisad /usr/bin/ls

The inode number is in the first column.


Then use "find" to find every file...

Basically, a soft link is a special kind of file that points to a different file somewhere on the available (mounted) filesystems. A hard link is a regular file entry in a directory, only that it...