正则表达式替换（sed或perl）变量捕获

Question

I need to do a regex capture/repalce on a file with content like the following. While I'm okay at matching things, I'm having trouble doing a capture replace.

I'd like to do the following:

1. Capture which will be different in every instance.
2. Add a new line before the copy line that references the table name.

So, for example, copy mytable( would become hello mytable\\ncopy mytable( .

---

Sample Input

copy tablename(
        preferredid= 'c0»',
        qid= 'c0»' with null(''),
        qpi= 'c0»',
        ptid= 'c0»' with null(''))...

into '/idata2/backup/core/eq.ingres'

Answer 1

尝试这个 ：

sed 's/^copy \([^(]*\)($/hello \1\ncopy \1(/' 

Answer 2

使用perl单线程

perl -i -pe 's/^(?=copy (\w+)\()/hello $1\n/' file