Is there a way to substitute a text that has regex formatted number ranges and replace with all combinations?

Question

Let's say I have a file that looks like:

host[1-2]lx
host[1-4]win

what I want is an easy way to get

host1lx
host2lx
host1win
host2win
host3win
host4win

Is there a neat way to automate that process?

Edit: What I was hoping for is something like using a substitution on host[1-4]win using the line number, where I can repeat the replacement using the [1-4] part. Imagine something like

:110s_host\[1-4\]win_host1win\rhost2win\rhost3win\rhost4win\r_

But is there a way to not spell out the four hosts, instead have sed repeat it based on what was found in the [...]?

Answer 1

I doubt a simple solution exist for any regular expression but for you case you could:

Define the following function:

function! GenerateLineRange(start, end, i1, i2)
    let ret = ''
    for i in range(a:i1, a:i2)
        let ret = ret . a:start . i . a:end
        if i != a:i2
          let ret = ret . "\n"
        endif
    endfor
    return ret
endfunction

Use it in a substitution:

:%s/\v(.*)(\[(\d+)-(\d+)\])(.*)/\=GenerateLineRange(submatch(1), submatch(5), submatch(3), submatch(4))