Cfrtjryk

I have a list of values like:

1

2

3

4

Ak123

Ak23

Ak147

1Apple

2Apricot

3Mango

4Orange

I just want to execute a simple grep command to list me only the numbers. i.e. 1, 2 , 3 and 4.
I tried this command -

grep -Ein --color '^s*[0-9]' test.txt

but it returns the alphanumeric also. Instead I tried omitting the character by this command:

grep -Ein --color '^s*[0-9][^A-Z]' test.txt

but it gives 0 results.

Assuming each line only contains a single word with no flanking whitespace:

grep -x -E '[0-9]+'

or

grep -x -E '[[:digit:]]+'

This would extract any line that contained only digits. The -x option to grep forces the pattern to match across a complete line. I'm using -E to enable extended regular expression to be able to use +.

The pattern [[:digit:]]+ would match at least one digit. The type of digit that it matches may depend on your locale.

To allow for whitespace before and after:

grep -x -E '[[:blank:]]*[0-9]+[[:blank:]]*'

or

grep -x -E '[[:blank:]]*[[:digit:]]+[[:blank:]]*'

The pattern [[:blank:]]* matches zero or more tabs or spaces.

Your expression ^s*[0-9] matches lines that may start with a space character (assuming s matches a space character, it's really a PCRE) and then has a digit. After the digit, any character may occur, as in the string 2Apricot.

Your expression ^s*[0-9][^A-Z] is similar, but [^A-Z] forces the matching of a second non-alphabetic non-uppercase character. This matches things like 33, 1a, and 5- and longer strings, but would not match the single character string 3.

Try also

$ grep -v '[^ 0-9]' file

1

2

3

4