Search Results

This string: ". /tmp/bin/myenv.sh; echo ${LOG_DIR}" substitutes ${LOG_DIR} before su is run. Use single-quotes ' ' instead to prevent it from substituting.

That code is vulnerable to pattern matching metacharacters. For this approach to work with arbitrary text, it is necessary to double-quote the nested parameter expansion.

$ s=*****a
$ echo "$s"...

You need to understand how "variable expansion" works: lets assume we have a variable assigned (i suggest you try the examples yourself at the shell prompt and play around a bit with them to get...

paste -s -d, f1.txt

Guru.

I'll bite:
find . -type f ! -perm 554 -exec chmod 554 {} \;

If you want pure POSIX compliance, use pure POSIX, not KSH. Pure POSIX shell code will work in SH and KSH, while things using KSH features(arrays, etc) might only work in KSH...

Kornshell 88 is not 100% POSIX compliant

Try to run it as sh:
#!/usr/bin/sh
# (I think it is in /usr/bin or was it #!/bin/posix/sh ??)
It will still use ksh, but it should emulate a POSIX...

I think either of these are acceptable in Ksh or bash depending on your preference/need for single or double bracketed expressions:


if ! [ $1 == "foo" ] || [ $1 == "bar" ]
then
echo true...

sort < infile | uniq -c

Associative arrays are just live savers in these needs

awk '{arr[$1]++}END{for (i in arr) {print i "-" arr[i]}}' uniq.dat

You can try this:


awk '{c[$1]++;} END { s=""; for( x in c ){ printf( "%s%s=%d", s, x, c[x]); s=","; } printf( "\n" ); }' input-file

A translation into sed:
sed -n 'H; /^CHECK/h; ${g;p;}' f1.txt

Regards,
Alister

If you turn off 'printing' by default (-n) and then put a print command after your address pair, it will work as you intended.


sed -n '/START2/,/STOP/p' input-file

user~$ PARM='V01'
user~$ grep -A1 '^\$'$PARM'\$' f1.txt
$V01$S
this is 1

If you still want a sed, here's one:
sed -n '/\$'$PARM'\$/{N;p;}' f1.txt

Try in this way ..

$ cat infile
1|2|3|4|5|6|7
$
$ a='$2 $3 $4'
$
$ echo $a
$2 $3 $4
$
$ echo "awk -F\| '{print $a}' infile" | sh
234
$

You'd really need to base it upon some evaluation that you can control. You'd test the inputs, and based on behavior, you'd fork off in whatever direction you need. The following function provides...

Not all of your records/lines have the same number of columns/fields as stated in the header.
Try this:

BEGIN {
n=split(fld, t, " ")
for(i=1;i<=n;i++)
fldN[t[i]]
}
FNR==1{
...