Hi All,
I have come across a command
echo "123 abc" | sed 's/[0-9]*/& &/'
output is
123 123 abc
then i tried in different ways to get 123 abc abc as output.
Can u please explain me the importance of & in the above command.
Thank you
- Chanakya
This is from the sed man page:
An ampersand (&) appearing in the replacement
will be replaced by the string matching the
RE.
This means that the sed will replace & with 123 (the string that matches the regular expression in this example). Since the & occurs twice, it is placed there twice.
Thank you blowtorch ..
Expecting the same i tried
echo 123 abc | sed 's/*[a-z]/& &/' to get 123 abc abc as output but i could not ..can u guide me how to get the desired output
try this:
echo 123 abc | sed 's/\([a-z].*\)/& &/'
Try this :
echo "123 abc" | sed 's/[a-z]\{1,\}/& &/'
Another example of sed :
echo "++++ 123 abc ----" | sed 's/\([0-9]\{1,\}\)[[:space:]]*\([a-z]\{1,\}\)/Number=\1 Text=\2/'
Output:
++++ Number=123 Text=abc ----
Jean-Pierre.
hi aigles,
thank you.. I got the required output. Can u explain me what is wrong in
echo 123 abc | sed 's/\([a-z].*\)/& &/' and if possible explain ur command also echo "123 abc" | sed 's/[a-z]\{1,\}/& &/'
-Chanakya
The regex [a-z].* matches one alphabetic character and all following characters.
For example, the command :
echo 123 abc 456 | sed 's/\([a-z].*\)/& &/'
gives the ouput:
123 abc 456 abc 456
echo "123 abc" | sed 's/[a-z]\{1,\}/& &/'
The regex [a-z]\{1,\} matches one or more alphabtic character.
Some commands accept the following syntax [a-z]+
Jean-Pierre.
thank you aigles .. i got it