Search Results

With sed (with some assumptions):
echo "$str"|sed -n '/[^{]*{aaID=\([^;]*\);/{;s//\1\
/;P;D;}'
producing
z_701
S_300
x_500

Not a necessity if you have <<-EOF instead of <<EOF. In the former case, EOF may be preceded by tabs.

It starts off very much like your original attempt.

\|~DRG\|(\d+): Begins by looking for |~DRG|<some number>. The number is captured.

.*?: ... followed by a non-greedy wildcard, so it doesn't...

This works with several variations I created from your sample data:
perl -lpe 's/\|~DRG\|(\d+).*?(?=\|~(?!DCT))/$1>15?"":$&/ge' file

Regards,
Alister

Note: with read you can use IFS local to the read command, so that you do not need to globally set/reset IFS :
IFS=: read field1 field2 field3 field4 < x.dat

Word boundaries are not specified in POSIX and are not universally available in grep. Most greps do have a -w switch:
$ echo bla bla connected | grep '\<connected\>'
$ echo bla bla connected |...

In the main script can you try this :

instead

ssh faunus:/home/dslmain/scripts/senddcmcadefttest.sh
this :

ssh user@remote_machine "/home/dslmain/scripts/senddcmcadefttest.sh ; exit \$?"
...

$ cat floattest2
#!/bin/sh
a=-0.5
b=1.3

if [ $a -le 0 ] || [ $b -ge 1 ]; then
echo Hello
else
echo Goodbye
fi


a=-0.5
b=1.3

if [ $a -lt 0 ] || [ $b -gt 1 ]; then
echo Hello 2...

It would be safer to assign the values to an associative array making eval unnecessary and allowing for fairly simple verification of a valid name entered.


#!/usr/bin/env ksh

function prompt...

Search 'RFC2822 regex' - the regular expression for the official standard for addresses is, um, long :).
...

I would be inclined to prefer [^ \t@] to \w for the bit before the @, since a surprising range of character are allowed in the local part of email addresses:
Email address - Wikipedia, the free...

I used the following command and it works fine for me

awk '{print >("input_" int((NR+4999)/5000))}' input.txt

[m]intty is both a valid regular expression for grep and a valid sh pattern for pathname expansion. If you don't quote that argument, and if the current working directory has a matching item, the...

What about your solution?
We appreciate people saying their issue is solved, we like even more when someone saying found a solution, show what they found...

You redirect stdin to read the file. 'select' reads an option from stdin. So 'select' ends up reading from the file.. Open it into another fd, and read from that.

I'm not sure if it's -u5 or -u...

@gary_w if you strip out the path part of $0 you could report wrong file eg: ./ls would report /usr/bin/ls instead of /home/garyw/ls

On ksh it's probably best to use one of : realpath $(whence...

One way of achieving the effect of "-maxdepth 1" is to work from the directory above the one you are actually interrogating. Then apply the "-prune" to stop find searching deeper.

#!/bin/ksh...

Using "-prune" is essentialy the same as using "-maxdepth 1"

I'm assuming "full_path_to_dir_to_search" already contains a splat in order to create the list?


find...

Hi gary_w,

As I understand, the option you are looking for is:

-maxdepth 1
which doesn't descend into directories.

Substitute '-prune' with '-maxdepth 1', and an absolute path doesn't mind....

exactly!

:%s/\(.*\)/'\1'/ =>
% = for all lines in the current vim window
s = substitute
/ = start with the search expression hereafter (s/// syntax)
\( = escape character + begin...

Take a close look at your quoting. You cannot include a single-quote in a single-quoted string. Usually, that's done by closing the quoted string, escaping a single quote, and re-opening a quoted...

@pludi: controlling SQL*Plus commands like spool with PL/SQL is hard, but SQL*Plus can be tricked to act conditional.

column spool_destination noprint new_value spooldestination
SELECT...

From crontab you can do :
minute hour day_of_month month day_of_week your_script.sh%1%yes%n%64

Jean-Pierre.

awk ' /^header/ && FNR > 1 {next} {print $0} ' infile > newfile