找到匹配项时使用 sed 替换行首 [英] Using sed to replace beginning of line when match found

问题描述

我有一个 Java 文件.我想注释包含匹配项的任何代码行:

I have a Java file. I want to comment any line of code that contains the match:

 myvar

我认为 sed 应该可以帮助我

I think sed should help me out here

 sed 's/myVar/not_sure_what_to_put_here/g' MyFile.java

我不知道该放什么:

not_sure_what_to_put_here

在这种情况下，我不想替换 myVar 但我想插入

as in this case, I don't want to replace myVar but the I want to insert

//

到 myVar 出现的任何行的开头.

to the beginning of any line myVar occurs on.

任何提示

推荐答案

捕获包含 myvar 的整行:

$ sed 's/\(^.*myvar.*$\)/\/\/\1/' file

$ cat hw.java
class hw {
    public static void main(String[] args) {
        System.out.println("Hello World!"); 
        myvar=1
    }
}

$ sed 's/\(^.*myvar.*$\)/\/\/\1/' hw.java
class hw {
    public static void main(String[] args) {
        System.out.println("Hello World!"); 
//        myvar=1
    }
}

使用 -i 选项将更改保存在文件 sed -i 's/\(^.*myvar.*$\)/\/\/\1/'文件.

Use the -i option to save the changes in the file sed -i 's/\(^.*myvar.*$\)/\/\/\1/' file.

说明:

(      # Start a capture group
^      # Matches the start of the line 
.*     # Matches anything 
myvar  # Matches the literal word 
.*     # Matches anything
$      # Matches the end of the line
)      # End capture group 

所以这会查看整行，如果找到了 myvar，结果存储在第一个捕获组中，称为 \1.因此，我们将整行 \1 替换为前面有 2 个正斜杠 //\1 的整行，当然正斜杠需要转义以免混淆 sed 所以 \/\/\1 还要注意括号需要转义，除非你使用 sed 的扩展正则表达式选项.

So this looks at the whole line and if myvar is found the results in stored in the first capture group, referred to a \1. So we replace the whole line \1 with the whole line preceded by 2 forward slashes //\1 of course the forwardslashes need escaping as not to confused sed so \/\/\1 also note that brackets need escaping unless you use the extended regex option of sed.

这篇关于找到匹配项时使用 sed 替换行首的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！