regex101: PCRE (Legacy) Match Workspace

\(+the literal ( (40₁₀ / 50₈ / 28₁₆) , repeated at least once

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

Non-Capturing Group(?:(?:\(+\*+(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*

*repeats the group contents below any number of times:

1st Alternative(?:\(+\*+(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)

Non-Capturing Group(?:\(+\*+(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)

\(+the literal ( (40₁₀ / 50₈ / 28₁₆) , repeated at least once

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

Non-Capturing Group(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*

*repeats the group contents below any number of times:

1st Alternative(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)

Non-Capturing Group(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)

\(+the literal ( (40₁₀ / 50₈ / 28₁₆) , repeated at least once

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

Non-Capturing Group(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*

*repeats the group contents below any number of times:

1st Alternative[^*(]

Negated Character Class[^*(]

*the literal * (42₁₀ / 52₈ / 2A₁₆)

(the literal ( (40₁₀ / 50₈ / 28₁₆)

2nd Alternative(?:\*+[^)*])

Non-Capturing Group(?:\*+[^)*])

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

Negated Character Class[^)*]

)the literal ) (41₁₀ / 51₈ / 29₁₆)

*the literal * (42₁₀ / 52₈ / 2A₁₆)

3rd Alternative(?:\(+[^*(])

Non-Capturing Group(?:\(+[^*(])

\(+the literal ( (40₁₀ / 50₈ / 28₁₆) , repeated at least once

Negated Character Class[^*(]

*the literal * (42₁₀ / 52₈ / 2A₁₆)

(the literal ( (40₁₀ / 50₈ / 28₁₆)

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

\)+the literal ) (41₁₀ / 51₈ / 29₁₆) , repeated at least once

2nd Alternative[^*(]

Negated Character Class[^*(]

*the literal * (42₁₀ / 52₈ / 2A₁₆)

(the literal ( (40₁₀ / 50₈ / 28₁₆)

3rd Alternative(?:\*+[^)*])

Non-Capturing Group(?:\*+[^)*])

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

Negated Character Class[^)*]

)the literal ) (41₁₀ / 51₈ / 29₁₆)

*the literal * (42₁₀ / 52₈ / 2A₁₆)

4th Alternative(?:\(+[^*(])

Non-Capturing Group(?:\(+[^*(])

\(+the literal ( (40₁₀ / 50₈ / 28₁₆) , repeated at least once

Negated Character Class[^*(]

*the literal * (42₁₀ / 52₈ / 2A₁₆)

(the literal ( (40₁₀ / 50₈ / 28₁₆)

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

\)+the literal ) (41₁₀ / 51₈ / 29₁₆) , repeated at least once

2nd Alternative[^*(]

Negated Character Class[^*(]

*the literal * (42₁₀ / 52₈ / 2A₁₆)

(the literal ( (40₁₀ / 50₈ / 28₁₆)

3rd Alternative(?:\*+[^)*])

4th Alternative(?:\(+[^*(])

\*+the literal * (42₁₀ / 52₈ / 2A₁₆) , repeated at least once

\)+the literal ) (41₁₀ / 51₈ / 29₁₆) , repeated at least once

Global pattern flags

g modifier: global. Finds all matches instead of stopping after the first

Search reference

Regular Expression

\(+\*+(?:(?:\(+\*+(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

New

The following regex handles any (* *) style string. Let's call it p{1}

    p{1} = \(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

The targeted string can include the characters {*, (, )}, but not the character sequences "(*" or "*)" (on which the regex terminates).

Examples:

  (* simple one *)
  (* more * ) ( * () **( difficult * ))()*(( one *)
  (* r *( e*a) () * *  l ()*  * ( ) )   l   * ( ) ) * **( y bad *)

Explanation:

    \(+\*+           # begin with a (* sequence (or some variation like ((* or ((**, etc.)
      (?:            # begin comment content
        [^*(]        # allow any non-* non-( characters (which begin open/close brackets)
        |            #   also
        (?:\*+[^)*]) # allow * 1+ times in a row ONLY if it's not immediately followed by ) or *
        |            #  also
        (?:\(+[^*(]) # allow ( 1+ times in a row ONLY if it's not immediately followed by * or (
      )*             # allow any number of these characters / character sequences
    \*+\)+           # then close the comment with a *) (or some variation like **) or *))), etc.)

To capture un-nested and nested comments, simply allow comments inside of comments. ie:

    p{2} = \(+\*+(?:(?:p{1})|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

in other words

    p{2} = \(+\*+(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

You'll see that this works on the following examples:
  (* test (* one *) *)
  (* a bit * ( ) *  harder (*  ) * () (( *) *)
  (* r *( (( e **( a ) ((* l *() l *) y * bad * ( *)

To capture up to triply-nested comments, follow the pattern set by p{2}:

    p{3} = \(+\*+(?:(?:p{2})|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

    p{3} = \(+\*+(?:(?:\(+\*+(?:(?:\(+\*+(?:[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+)|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

Examples:
  (* an (* easy (* one *) *) *)
  (* only (* some (* levels (* captured *) because *) 4x *) nested *)

The pattern can be followed to allow any depth of nested comments to be captured, by defining

    p{N} = \(+\*+(?:(?:p{N-1})|[^*(]|(?:\*+[^)*])|(?:\(+[^*(]))*\*+\)+

for N > 1

Regular Expressions 101

Regular Expression

Substitution