regex - Linux 上的复杂文本解析

Question

我有一些具有以下结构的文件：

main test1
slave something1 and something2
  something1
  something2

slave something3 and something4
  something3
  something4

main test2
slave something5 and something6
  something5
  something6

如何将此文件修改为：

  test1something1 and test1something2
  test1something1
  test1something2

  test1something3 and test1something4
  test1something3
  test1something4

  test2something5 and test2something6
  test2something5
  test2something6

somethingM而somethingN不是qeual，它们可以完全不同，例如
something1可以apple
something2可以banana

与sed// awl_perl

Answer 1

awk '$1=="main" {prefix=$2; next} {gsub(/something/, prefix "&"); print}' <<END
main test1
  something1 and something2
  something1
  something2

main test2
  something3 and something4
  something3
  something4
END

  test1something1 and test1something2
  test1something1
  test1something2

  test2something3 and test2something4
  test2something3
  test2something4