print first few lines, then apply regex on a specific column to print results.

Shell Programming and Scripting
1

abc.dat

tty cpu
tin tout us sy wt id
0 0 7 3 19 71
extended device statistics 
r/s w/s   kr/s kw/s  wait actv wsvc_t asvc_t %w %b device
0.0 133.2 0.0 682.9  0.0  1.0  0.0    7.2    0 79 c1t0d0
0.2 180.4 0.1 5471.2 3.0  2.8  16.4  15.6   15 52 aaaaaa1-xx

I want to skip first 5 line headers. Then select rows with column 11 starts with 'aaaaaa1'.

 
# Print first 5 rows + all rows from 6th row whose column 11 starts with
#  'aaaaaa1' string.
cat abc.dat | awk '(NR<6) {print}; (NR>5), $11 ~ /^aaaaaa1/'

Apparently last part of awk is not filtering 6th row even though its 11 th column does not start with 'aaaaaa1' string.

Please correct it..

2 
$
$
$ cat abc.dat
tty cpu
tin tout us sy wt id
0 0 7 3 19 71
extended device statistics
r/s w/s   kr/s kw/s  wait actv wsvc_t asvc_t %w %b device
0.0 133.2 0.0 682.9  0.0  1.0  0.0    7.2    0 79 c1t0d0
0.2 180.4 0.1 5471.2 3.0  2.8  16.4  15.6   15 52 aaaaaa1-xx
$
$
$ perl -lane '$.>5 && print if $F[10] =~ /^aaaaaa1/' abc.dat
0.2 180.4 0.1 5471.2 3.0  2.8  16.4  15.6   15 52 aaaaaa1-xx
$
$
$ awk 'NR>5 && $11 ~ /^aaaaaa1/' abc.dat
0.2 180.4 0.1 5471.2 3.0  2.8  16.4  15.6   15 52 aaaaaa1-xx
$
$

Ok, seems like I misread it.
If you want to print the first 5 lines and apply the condition from line 6 onwards then -

$
$
$ perl -lane 'print if $.>5 && $F[10] =~ /^aaaaaa1/ || $.<=5' abc.dat
tty cpu
tin tout us sy wt id
0 0 7 3 19 71
extended device statistics
r/s w/s   kr/s kw/s  wait actv wsvc_t asvc_t %w %b device
0.2 180.4 0.1 5471.2 3.0  2.8  16.4  15.6   15 52 aaaaaa1-xx
$
$
$
$ awk 'NR<=5 || NR>5 && $11 ~ /^aaaaaa1/' abc.dat
tty cpu
tin tout us sy wt id
0 0 7 3 19 71
extended device statistics
r/s w/s   kr/s kw/s  wait actv wsvc_t asvc_t %w %b device
0.2 180.4 0.1 5471.2 3.0  2.8  16.4  15.6   15 52 aaaaaa1-xx
$
$

tyler_durden

3

Thanks. That works.. I will go with last command..
I also want to use a variable in place of search string 'aaaaaa1'. For some reason awk's variable substitution does not seem to working,,

# since /bin/awk is older version,, it does not recongnize -v flag, so I am using awk from /usr/xpg4/bin/ library.

var1='aaaaaa1'; /usr/xpg4/bin/awk -v var2=$var1 'NR<=5 || NR>5 && $11 ~ /^var2/' abc.dat

tty cpu
tin tout us sy wt id
0 0 7 3 19 71
extended device statistics 
r/s w/s   kr/s kw/s  wait actv wsvc_t asvc_t %w %b device
<nothing>

Please let me know how to correct the above awk to recognize string variable in place of regex expression..

4

Try:

var1='aaaaaa1'; /usr/xpg4/bin/awk -v var2=$var1 'NR<=5 || NR>5 && $11 ~ "^"var2' abc.dat
5

Perfect.. Thanks a lot.