Hi.

I'm just going over a SED tutorial and specifically the '&' and '*' special characters.

I understand how it works:

The above is clear but let's say i run the following, i don't understand why there is a space before the output:

what is the '*' matching in this case and why is a space being added here ?

An asterisk means "zero or more occurrences of the previous character or group".

In other words, the expression 's/[0-9]*/& &/' reads "replace zero or more consecutive occurrences of the digits 0-9 with whatever was matched twice, separated by a space".

Now consider this:So sed looks at the string "foo bar", and what do you know, there are exactly zero occurrences of the digit group, right there at the beginning of the string! So it dutifully replaces nothing with nothing twice, separated by a space. The result is the string " foo bar".

1 members found this post helpful.

Hi.

Thanks for the reply.

So 'nothing' is 'something' ?

That's what confused me there.

Nothing certainly is the same as "zero or more". To be specific, it's zero.

And once sed has found what you've told it to look for, it won't look again without the greedy operator. That's why it turned "abc 123" into " abc 123", because the nothing between the start of the string and the first "a" matches '[0-9]*' perfectly.

1 members found this post helpful.

No - the fact that the "nothing" pattern matched is significant ("something" if you wish).
Unless you have a *VERY* good reason - and you have justification for it - don't use "*" in a pattern, but (almost) always prefer "+".
Make sure you understand the difference.

1 members found this post helpful.

Thanks again.

I understand that bit but i don't understand why duplicating 'nothing' using the '&&' actually creates a 'space'. Why is 'nothing' turned into a 'space'

Because you have a space between the two ampersands in the right hand side of the expression.

1 members found this post helpful.

Oh lol !

I didn't even think of that! Thank you.