NetNews Usenet Archive 1992 #30/NN_1992_30.iso/spool/comp/lang/perl/7384

Text File | 1992-12-12 | 1.6 KB | 50 lines

Newsgroups: comp.lang.perl
Path: sparky!uunet!ftpbox!mothost!merlin.dev.cdx.mot.com!fendahl.dev.cdx.mot.com!mcook
From: [e-mail] (Michael Cook)
Subject: Re: C comment parser?
Message-ID: <[e-mail]>
Sender: [e-mail] (USENET News System)
Nntp-Posting-Host: fendahl.dev.cdx.mot.com
Organization: Motorola Codex, Canton, Massachusetts
References: <[e-mail]>
Date: Sun, 13 Dec 1992 00:54:54 GMT
Lines: 37

[e-mail] (Michael Bain) writes:

>I'm having difficulty developing the regular expression to remove C
>comments from a variable (say $_) that contains a C function.  (I'm
>hacking a McCabe metric and am reading in a function at a time)

>What I have so far seems to work *most* of the time, but oftentimes it
>misses an ending C comment delimiter (*/) and ends up removing a lot of
>code until it reaches the last delimiter.

>Here's the RE:
>    s?/\*(.*\n)*.*\*/??; # remove comments

>Does somebody have a better one that works?

Remember that RE's are maximally matched (in Perl, and most Unix tools).  If
$_ has two or more comments in it, that s/// is going to delete them all in
one swoop, and it'll also delete everything between the comments.

I can give you an RE that'll work.  Here:

 s%/\*([^*]|\*+[^/*])*\*+/%%g

But RE's are really the way to go.  The if comment is very big, Perl will dump
core.  (I tried it.)

You can do it with index():

  $pos = 0;
  while (($pos = index($_, "/*", $pos)) >= 0)
  {
    $end = index($_, "*/", $pos + 2);
    $end >= 0 || die "unterminated comment";
    substr($_, $pos, $end - $pos + 2) = '';
  }

Michael.