Search Results

Try using double square brackets..:)
for blank values of var you may need double square braces.
time=$1
var=$(echo "$time"|awk '{for(i=1;i<=NF;i++) if ( $i ~ ...

spacing around your tests (the square brackets need to be separate tokens

a=9
b=10
if [ a -gt b ]
then
echo what now
else
echo that makes sense
fi

Please use spaces between square brackets.

if [ $v_num -gt $comp_num ]

you should give the space in the condition..
if [ $v_num -gt $comp_num ]

Use sth like this..

line="Some text with ERR message"
index=$(echo $line | awk '{for(i=1;i<=NF;i++) if($i~/^ERR/){print i}}')
echo $index

awk -v x='11/20/12' '$0~x,EOF' datefiltertestfile.txt

print section of file from regular expression to end of file
awk '/2012\/20\/11/,EOF { print $0 }' filename

Hi

$ awk '{a[++i]=$0;}/ERR-/{for(j=NR-x;j<=NR;j++)print a[j];}' x=1 file
some other text
ERR-12345
some blah blah blah
ERR-56789

You can specify your value of N against the x.

Guru.

awk '/\[2012/{a=$0}/^ERR/{print a"-"$0}' input.txt
[2012-11-12 : 11:59-ERR-XXXXX
[2012-11-12 : 12:01-ERR-XXXXX


---------- Post updated at 10:01 AM ---------- Previous update was at 09:59 AM...

The -o option prints only the strings matching the pattern, each on a separate line. I think this is available only with GNU grep.
You may try this awk script:
awk...

grep -o "ERR-[0-9][0-9][0-9][0-9][0-9]" input_file| sort | uniq -c

sed -n '/10\/31\/2012/,/10\/31\/2012/p' file
sed -n '/11\/01\/2012/,/$/p' file