Shell Programming and Scripting

hi I would like to print those lines and their next lines which begin with ( in second column

example

input

Code:

panda       (-,xxx,xxx.co.uk)\
12
patrols         patrols1\
2
patrols1         (xxx,-,xxx.co.uk)\
4
patrols2         (-,xxx,xxx.co.uk)\
14
canada_patrol       (-,xxx,xxx.co.uk)\
12
simpsons      (-,xxx,xxx.co.uk)\
2
cool         cool1\
2
cool1        (-,xxx,xxx.co.uk)\
18

desired output:

Code:

panda       (-,xxx,xxx.co.uk)\
12
patrols1         (xxx,-,xxx.co.uk)\
4
patrols2         (-,xxx,xxx.co.uk)\
14
canada_patrol       (-,xxx,xxx.co.uk)\
12
simpsons      (-,xxx,xxx.co.uk)\
2
cool1        (-,xxx,xxx.co.uk)\
18

I tried so far:

Code:

awk  '{ if ( $2~"^\\(.*" ) printf $0; getline; print $0 }' file

but did not work.

Thanks a lot

Code:

sed -n '/.[^(]*[ \t]*(/{p;n;p}' file

Hi, wakatana:

Your attempt was close. Changing the printf to a print and enclosing the statements within curly braces to make them all part of the if statement's block fixes it:

Code:

awk  '{ if ( $2~"^\\(.*" ) {print $0; getline; print $0} }'

However, you could do away with the if statement and just use AWK's native pattern-action syntax:

Code:

awk  '$2~"^\\(.*" {print $0; getline; print $0 }'

If you like, you can simplify it a bit by using a regular expression literal instead of a string (so as to avoid having to escape through the string parsing layer before arriving at the regular expression parser:

Code:

awk  '$2~/^\(/ {print; getline; print}'

Regards,
Alister

---------- Post updated at 03:45 AM ---------- Previous update was at 03:38 AM ----------

This can fail if there is an open parenthesis in the first field. The negated open parenthesis class need not occur and the same goes for the blank class with the space and tab, allowing the literal "(" in the pattern to fall within the first field.

A much better approach would be:

Code:

sed -n '/^[^ ]*  *(/{p;n;p;}' file

Regards,
Alister

Thanks a lot guys, you helped