Trying to use a Regex \/\*(.|\n)*?\*\/ to select every C multiline comment but its not working in SED

Question

I need to match the C styled multiline comment for eg.

#include <stdio.h>
int main()
{
    // this is a dummy function
    float sum = 0;
    // testing the sed commands
int x = 6; // single-line comment
x = x + 5;

char y = 'n'; /* end of c  *
file */

}

I need to delete all multi-line comments.

So I used sed s/\/\*\(.\|\n\)*\?\*\///, but it doesn't work. I tried replacing / with %: s%\/\*\(.\|\n\)*\?\*\/%% but it still isn't working.

Can anyone please help to put this regex \/\*(.|\n)*?\*\/ operation in sed command?

Answer 1

Sed works on "records" (lines) which are defined by the presence of a trailing newline (\n) character. This means you cannot match past a \n because as far as sed is concerned, the \n is the end of the record. You can get around this, in GNU sed, by using -z to slurp the file and treat the entire thing as a single record (unless your file has NULLs (\0) in it, in which case each \0 will define a record):

$ sed -zE 's|/\*.*\n.*\*/||' file.c 
#include <stdio.h>
int main()
{
    // this is a dummy function
    float sum = 0;
    // testing the sed commands
int x = 6; // single-line comment
x = x + 5;

char y = 'n'; 

}

However, this will fail if you have multiple multi-line comments in the same file because sed cannot do non-greedy matching, so it will always try and find the longest possible match which means it would match from the first /* to the last */. So use a tool that can do non-greedy matching, like perl:

$ perl -0777 -pe 's|/\*.*?\n.*?\*/||gs' file.c 
#include <stdio.h>
int main()
{
    // this is a dummy function
    float sum = 0;
    // testing the sed commands
int x = 6; // single-line comment
x = x + 5;

char y = 'n'; 

}

This, however, will fail if you have a single line /* */ comment. The safest way I can think of is to forget about trying to do this with regular expressions and instead write a little script that keeps count of opening and closing comment tags and deletes accordingly.

Another problem is that a string with /* or */ will also break it. For example, what if you have something like:

char foo [ ] = "A comment starts with /*";

At the end of the day, the only safe way of doing this will be something like this SO answer by Ed Morton which uses a C preprocessor:

If this is in a C file then you MUST use a C preprocessor for this in combination with other tools to temporarily disable specific preprocessor functionality like expanding #defines or #includes, all other approaches will fail in edge cases. This will work for all cases:

[ $# -eq 2 ] && arg="$1" || arg=""
eval file="\$$#"
sed 's/a/aA/g; s/__/aB/g; s/#/aC/g' "$file" |
          gcc -P -E $arg - |
          sed 's/aC/#/g; s/aB/__/g; s/aA/a/g'

Put it in a shell script and call it with the name of the file you want parsed, optionally prefixed by a flag like "-ansi" to specify the C standard to apply.

See https://stackoverflow.com/a/35708616/1745001 for details.