One Liner, print multiple regex matches

cipher has asked for the wisdom of the Perl Monks concerning the following question:

I am trying to print multiple regex matches using Perl one liner but it does not work. Here is the data from where I am trying to extract info. name=abc dep=HR type=permanent age=35 name=xyz dep=Sales type=contract age=31 If I try to only print name it works fine.

perl -nle ' print $1 if /name=(\w+)/' filename
[download]

But I need information from other fields too. I tried the following command but there is no output

perl -nle ' print $1 $2 if /name=(\w+).*age=(\d+)/' filename
[download]

Also tried.

perl -nle ' print $1 if /name=(\w+)/ && print $2 if /age=(\d+)/' filen
+ame
[download]

Help is appreciated.

Comment on One Liner, print multiple regex matches Select or Download Code

Replies are listed 'Best First'.
Re: One Liner, print multiple regex matches by toolic (Bishop) on Nov 16, 2012 at 21:45 UTC
Add quotes around $1 and $2: `perl -nle ' print "$1 $2" if /name=(\w+).age=(\d+)/' filename` [download] warnings can also be used with one-liners to help you debug: `perl -nlew ' print $1 $2 if /name=(\w+).age=(\d+)/' filename Can't open print $1 $2 if /name=(\w+).*age=(\d+)/: No such file or di +rectory.` [download] Please update your post to add "code" tags around your input file data.	[reply] [d/l] [select]
Re^2: One Liner, print multiple regex matches by cipher (Acolyte) on Nov 16, 2012 at 22:24 UTC
Thank's It worked! I simply added the quotes and its showing data for both regex matches. Working code: `perl -nle ' print "$1$2" if /name=(\w+).*age=(\d+)/' filename` [download]	[reply] [d/l]
Re: One Liner, print multiple regex matches by kennethk (Abbot) on Nov 16, 2012 at 21:43 UTC
In both those cases, you've got syntax errors. You could modify the second one to use two look-aheads (see Looking ahead and looking behind in perlretut) to grab the fields: `perl -nle ' print "$1$2" if /(?=name=(\w+))(?=age=(\d+))/' filename` Note how your omission of quotes or a comma from your second case means that $1 is being treated at a filehandle. Your third version fails because you are only allowed one if per statement (see Compound Statements). You could make this work using `&&` and `\|\|` like: `perl -nle ' /name=(\w+)/ && print($1) \|\| /age=(\d+)/ && print($1) ' filename` or, use two different statements to make it even easier: `perl -nle 'print $1 if /name=(\w+)/; print $1 if /age=(\d+)/' filename` Note that you are using two different regular expressions, so each one would set the `$1` buffer if it matches. As a last comment, assuming you are running a reasonably recent version of perl, your output might be easier to deal with if you use `-E` in place of `-e` and swap your prints to says (see perlrun). Oh, and please wrap input data in `<code>` tags, as white space is mangled on display if you don't. #11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.	[reply] [d/l] [select]
Re: One Liner, print multiple regex matches by Kenosis (Priest) on Nov 16, 2012 at 22:51 UTC
Another option with your data is to capture the text on either side of the equals sign: `perl -nle "print qq{$1 => $2} while /(\S+)=(\S+)/g" data.txt` [download] Output: `name => abc dep => HR type => permanent age => 35 name => xyz dep => Sales type => contract age => 31` [download]	[reply] [d/l] [select]
Re^2: One Liner, print multiple regex matches by cipher (Acolyte) on Dec 18, 2012 at 22:45 UTC
Thanks Kenosis. :)	[reply]
Re^3: One Liner, print multiple regex matches by Kenosis (Priest) on Dec 18, 2012 at 23:13 UTC
You're most welcome, cipher!	[reply]