[SOLVED] grep to select lines with M in last word

David the H. · 03-07-2011, 04:47 PM

It's generally good to give a variety of solutions, even if only for educational purposes, which is the main reason I posted what I did. I agree that in this particular case awk may indeed be the more efficient solution, but I was more concerned with demonstrating how regular expressions work than in what works "best" here. And it may help some reader down the line working on something similar; someone who doesn't have access to or understand how to use awk, for example, or who needs a regex solution specifically.

colucix · 03-07-2011, 05:01 PM

Dark Helmet, your arguments are reasonable and well phrased. Evidently, the slightly sarcastic tone you used in your previous post has not been well accepted by Telengard and he overreacted to criticism. Now that your reasons are more clear, we hope that the does not lead to anything discussion will terminate and that you will shake your hands.

colucix · 03-07-2011, 05:06 PM

David the H., thank you for clarifying.

fancylad · 03-07-2011, 07:26 PM

what about using perl?

Quote:

matt@amd:~$echo -e "Fable Mabel\nHairy Mary \nMary Martian" | perl -lane 'print if $F[1] =~/^M/'
Fable Mabel
Hairy Mary
Mary Martian

kurumi · 03-07-2011, 08:08 PM

Quote:

Originally Posted by Telengard

awk already considers tabs and normal space characters to be whitespace. My awk command is only 13 characters, while your egrep weighs in at a whopping 35 characters. How is your egrep different or better than this?

Code:

awk '$NF~"M"'

use // instead of ". and your regex is wrong.

Code:

awk '$NF~/^M/'

colucix · 03-08-2011, 02:57 AM

Quote:

Originally Posted by kurumi

use // instead of ". and your regex is wrong.

Code:

awk '$NF~/^M/'

Please let's stop to address other members' solutions as wrong or worse. The difference between

Code:

'$NF ~ "M"'

and

Code:

'$NF ~ /^M/'

is that the former uses a so-called dynamic regexp (even if it's actually a string constant), the latter uses a regexp constant. In both cases the expression is very simple and shouldn't bring to any problem, but there are some situations where the usage of one is preferred to the other. See http://www.gnu.org/software/gawk/man...mputed-Regexps for an insight. Furthermore the additional anchor ^ is not really needed as per OP's requirements, since the M (if present) will be the first character in the last word. Checking if it is present should be enough.

Please note the thread has been marked as solved by the OP, therefore let's add further information only if they are really valuable and give an actual improvement to the technical discussion. Thanks.