CODEWITHSUNDEEP
unixsed
Shashank Vivek
Shashank Vivek

Reputation: 17504

Searching a two digit number in a line of a file through Sed

I am trying to find out a line having a two digit number in it.Eg:

test.txt 

uuLinuxxx
Linux 2011
2011 Linux RedHat
Linux RedHat
2011
2013  2014
2010
/usr/bin
2
Ubuntu 20 world

Desired OP: Ubuntu 20 world

i am using 

sed -n '/[0-9]\{2\}/p' test.txt

but the problem is that it is printing all the lines having 2 or more digit number.

Upvotes: 2

Views: 10056

Answers (4)

potong
potong

Reputation: 58410

This might work for you (GNU sed):

sed -n '/\b[0-9]\{2\}\b/p' file

or:

sed -nr '/\b[0-9]{2}\b/p' file

or:

sed -r '\b[0-9]{2}\b/!d' file

or:

sed '/\<[0-9]\{2\}\>/!d' file

Upvotes: 5

jaypal singh
jaypal singh

Reputation: 77105

Use -w switch to match word (in your case digits).

grep -w '[0-9][0-9]' file

From man page:

-w, --word-regexp
     Select only those lines containing matches that form whole words.  
     The test is that the matching substring must either be at the beginning of the
     line,  or preceded by a non-word constituent character. Similarly, it must be either 
     at the end of the line or followed by a non-word constituent character. Word-
     constituent characters are letters, digits, and the underscore.

Upvotes: 2

NeronLeVelu
NeronLeVelu

Reputation: 10039

sed -n 's/.*/²&²/;/[^0-9][0-9]\{2\}[^0-9]/ s/.\(.*\)./\1/p' YourFile

Using a temporary border to allow only 1 check for extraction of the line with a 2 digit only number inside

Upvotes: 1

paxdiablo
paxdiablo

Reputation: 881423

The problem is that your regular expression is looking for two consecutive digits, which exist is the numbers 20 (good) and 99999999999999999999 (not so good).

What you need is a regular expression ensuring that there are no digits on either side of the two you find, such as:

[^0-9][0-9]{2}[^0-9]  # non-digit the two digits then non-digit

In addition, you need to trap those where the two digits are at the start or end of the line (or the only thing on the line). So you need multiple regular expressions separated by the or conjunction | or as separate -e arguments:

^[0-9]{2}[^0-9]       # at start of line
[^0-9][0-9]{2}[^0-9]  # in middle of line
[^0-9][0-9]{2}$       # at end of line
^[0-9]{2}[^0-9]$      # only thing on line

You may also want to choose a better tool for the job, such as grep. Using the slightly modified input file:

uuLinuxxx
Linux 2011
2011 Linux RedHat
Linux RedHat
2011
2013  2014
2010
/usr/bin
2
Ubuntu 20 world
99 at the start
at the end: 99
88

the following command (split for readability):

grep -E -e '[^0-9][0-9]{2}[^0-9]'
        -e '^[0-9]{2}[^0-9]'
        -e '[^0-9][0-9]{2}$'
        -e '^[0-9]{2}[^0-9]$' test.txt

gives you what you want:

Ubuntu 20 world
99 at the start
at the end: 99
88

Of course, if you have GNU grep with its Perl-based regular expressions, and you're after "words" which are two-digit numbers, this becomes much easier:

grep -P '\b\d{2}\b' test.txt

but, if you can guarantee that word restriction, the following will also work:

grep -Ew '[0-9]{2}' test.txt

Upvotes: 1

Related Questions