Use sed to uncomments line in c++ file

Question

I have the following lines in a file on WSL

// title
// (
//     "line"
// )

// SomeOtherTitle
// (
//     "otherLine"
// )

I would like to use the sed command to uncomment the lines.

// title
// (
//     "line"
// )

And leave the rest commented.

I am going to use this in a bash script to uncomment the information several files. How can this be achieved?

I have tried:

sed 's#^// [ \t]$title#title#g' file
sed 's#^// [ \t]$(#(#g' file
sed 's#^// [ \t]$"line"#"line"#g' file # this one did not work for me
sed 's#^// [ \t]$)#)#g' file

Kind regards

Answer 1

First a minor point: using g is useless if you are anchoring the regex with ^ since there can only ever be one match. Now, for the main issue, none of the commands you show would ever match anything since they all have a $ in the left hand side. The $ means "end of the line" so nothing will be matched after that. In addition, you're using [ \t] which means one of space or tab. What you want is "0 or more" (specifically, you want one or more but 0 or more should work here), so you wanted [ \t]*. These should do what you want:

sed 's#^//[ \t]title#title#' file.c 
sed 's#^//[ \t](#(#' file.c 
sed 's#^//[ \t]*"line"#"line"#' file.c
sed 's#^//[ \t])#)#' file.c 

But this isn't a good idea since that would replace all cases of commented ( or ) if they are the first non-space or tab character on the line.

However, if you can be sure that you always want to uncomment the line // title and the three lines after that, you could do:

perl -pe 'if(m#^\s*//\s*title#){ $k=1 } 
          $k=0 if $k==3;
          s#^\s*//## if $k' file.c

On your input file, that produces:

$ perl -pe 'if(m#^\s*//\s*title#){ $k=1 } 
          $k=0 if $k==3;
          s#^\s*//## if $k' file.c
 title
 (
     "line"
 )

 SomeOtherTitle
 (
     "otherLine"
 )