I've found this following routine useful. Basically it allows me to just get nonblocking full lines of input from a filehandle (using the select/vec solution - and heavily based on tadman's code above).

# An non-blocking filehandle read that returns an array of lines read
# Returns:  ($eof,@lines)
my %nonblockGetLines_last;
sub nonblockGetLines {
  my ($fh,$timeout) = @_;

  $timeout = 0 unless defined $timeout;
  my $rfd = '';
  $nonblockGetLines_last{$fh} = ''
        unless defined $nonblockGetLines_last{$fh};

  vec($rfd,fileno($fh),1) = 1;
  return unless select($rfd, undef, undef, $timeout)>=0;
    # I'm not sure the following is necessary?
  return unless vec($rfd,fileno($fh),1);
  my $buf = '';
  my $n = sysread($fh,$buf,1024*1024);
  # If we're done, make sure to send the last unfinished line
  return (1,$nonblockGetLines_last{$fh}) unless $n;
    # Prepend the last unfinished line
  $buf = $nonblockGetLines_last{$fh}.$buf;
    # And save any newly unfinished lines
  $nonblockGetLines_last{$fh} =
        (substr($buf,-1) !~ /[\r\n]/ && $buf =~ s/([^\r\n]*)$//)
            ? $1 : '';
  $buf ? (0,split(/\n/,$buf)) : (0);
}
[download]

I've tested it with a pipe I opened using an IO::File, YMMV. Example usage:

$fh = new IO::File;
open($fh,"$cmd 2>&1 |");

do {
    ($eof,@lines) = nonblockGetLines($fh);
    foreach my $line ( @lines ) {
        print "Pipe saw:  [$line]\n";
    }
} until $eof;

close $fh;
[download]

I don't know enough about vec() to know why tadman called it twice, don't know if this is a race condition or if it's unneeded, and at this point don't care enough to figure it out - I've got other code to write! :) Enjoy!

---

---

• Are you posting in the right place? Check out Where do I post X? to know for sure.
• Posts may use any of the Perl Monks Approved HTML tags. Currently these include the following:

  <code> <a> <b> <big> <blockquote> <br /> <dd> <dl> <dt> <em> <font> <h1> <h2> <h3> <h4> <h5> <h6> <hr /> <i> <li> <nbsp> <ol> <p> <small> <strike> <strong> <sub> <sup> <table> <td> <th> <tr> <tt> <u> <ul>

• Snippets of code should be wrapped in <code> tags not <pre> tags. In fact, <pre> tags should generally be avoided. If they must be used, extreme care should be taken to ensure that their contents do not have long lines (<70 chars), in order to prevent horizontal scrolling (and possible janitor intervention).
• Want more info? How to link or How to display code and escape characters are good places to start.