使用sed将文件替换为字符串，而不使用两个sed命令？

Question

In the process of automating a license header, I have a place holder: /* @@LICENSE_HEADER@@ */ , which I can replace with a file named license with the following:

sh-3.2$ sed -e $'/\/\* @@LICENSE_HEADER@@ \*\//{r license\n}' \
-e $'/\/\* @@LICENSE_HEADER@@ \*\//d' file.c

I've cobbled this together from other sources. Is it possible to replace the string, rather than issuing a second command to delete the line the string appears in?

Answer 1

尝试sed -e $'/\\/\\* @@LICENSE_HEADER@@ \\*\\//{r license\\n;d;}' file.c

Answer 2

Sed is sort of a programming language for editing, so you can use more than one command per line.

By the way, I suggest you change the license string to something that doesn't contain any special regular expression characters. Here, I just made it @@@LICENSE NUMBER@@@ , so I don't have to keep escaping the regular expression characters. After all, it's your string, so you can make it anything you want:

sed  '/@@@LICENSE NUMBER@@@/ {
r license.txt
d
}' file.c