Code

packet creation and dumping hacks used by check_ica* and check_lotus
authorStanley Hopcroft <stanleyhopcroft@users.sourceforge.net>
Tue, 25 Jan 2005 09:12:47 +0000 (09:12 +0000)
committerStanley Hopcroft <stanleyhopcroft@users.sourceforge.net>
Tue, 25 Jan 2005 09:12:47 +0000 (09:12 +0000)
git-svn-id: https://nagiosplug.svn.sourceforge.net/svnroot/nagiosplug/nagiosplug/trunk@1100 f882894a-f735-0410-b71e-b25c423dba1c

contrib/packet_utils.pm [new file with mode: 0755]

diff --git a/contrib/packet_utils.pm b/contrib/packet_utils.pm
new file mode 100755 (executable)
index 0000000..cd1e830
--- /dev/null
@@ -0,0 +1,120 @@
+package packet_utils;
+
+# $Id$
+
+# $Log$
+# Revision 1.1  2005/01/25 09:12:47  stanleyhopcroft
+# packet creation and dumping hacks used by check_ica* and check_lotus
+#
+# Revision 1.1  2005-01-25 15:28:58+11  anwsmh
+# Initial revision
+#
+       
+require Exporter;
+@ISA = qw(Exporter);
+@EXPORT_OK = qw(tethereal pdump);
+       
+sub tethereal {
+       my ($tethereal_dump, $start_byte) = @_ ;
+       
+       # Return a string or array (depending on context) containing the characters from a tethereal trace.
+       # Skip all stuff until the first byte given by $start_byte in the first row ..
+       
+       &outahere("Invalid tethereal dump:", substr($tethereal_dump, 0, 71), 'fails to match m#\d\d\d\d  \S\S(?: \S\S){1,15}#')
+               unless $tethereal_dump =~ m#\d\d\d\d  \S\S(?: \S\S){1,15}# ;
+       
+       my @tethereal_dump      = split(/\n/, $tethereal_dump) ;
+       my $first_line          = shift @tethereal_dump ;
+       $first_line                     = unpack('x6 a48', $first_line) ;
+                                                                                               # Take one extra space (after hex bytes) to use a template of 'a2 x' x 16 
+                                                                                               # instead of 'a2 x' x 15 . 'a2'
+       my $last_line           = pop @tethereal_dump ;
+       $last_line                      = unpack('x6 a48', $last_line) ;
+       
+       my $idx = index($first_line, $start_byte) ;
+       
+       &outahere(qq(Invalid tethereal dump: "$start_byte" not found in first line - "$first_line")) 
+               if $idx == -1 ;
+       
+       $first_line = substr($first_line, $idx) ;
+       
+       my ($dump, @dump) = ('', ()) ;
+       
+       my $bytes = 0 ;
+       $bytes++
+               while $first_line =~ m#\b\S\S#g ;
+       push @dump, unpack('a2x' x $bytes, $first_line) ;
+       
+       push @dump, unpack('x6 ' . 'a2x' x 16, $_) 
+               foreach @tethereal_dump ;
+       
+       $bytes = 0 ;
+       $bytes++
+               while $last_line =~ m#\b\S\S#g ;
+
+                                                                                               # Be more cautious with the last line; the ASCII decode may
+                                                                                               # have been omitted.
+
+       push @dump, unpack(('a2x' x ($bytes - 1)) . 'a2', $last_line) ;
+       
+       return wantarray ? map pack('H2', $_), @dump : pack('H2' x scalar @dump, @dump) ;
+       # return wantarray ? map hex($_), @dump : pack('H2' x scalar @dump, @dump) ;
+       
+}
+
+sub pdump {
+       my ($x) = shift @_ ;
+       my (@bytes_in_row, $row, $dump) ;
+
+       my $number_in_row = 16 ;
+       my $number_of_bytes = length $x ;
+       my $full_rows = int( $number_of_bytes / $number_in_row ) ;
+       my $bytes_in_last_row = $number_of_bytes % $number_in_row ;
+       my $template = "a$number_in_row " x $full_rows ;
+       my $nr = 0 ;
+                                                                                               # Output format styled on tethereal.
+       foreach $row ( unpack($template, $x) ) {
+               @bytes_in_row = unpack('C*', $row) ;
+               $row =~ tr /\x00-\x1f\x80-\xff/./ ;
+               $dump .= join('  ', sprintf('%4.4x', $nr * 0x10), join(' ', map { sprintf "%2.2x", $_} @bytes_in_row), $row) ;
+               $dump .= "\n" ;
+               $nr++ ;
+       }
+
+       if ( $bytes_in_last_row ) {
+               my $number_of_spaces = ($number_in_row - $bytes_in_last_row)*3  - 2 ;
+
+                                                                                               # 3 spaces (2 digts + 1 space) for each digit printed
+                                                                                               # minus two spaces for those added by the join('  ',) below.
+
+               $row = substr($x, -$bytes_in_last_row) ;
+               @bytes_in_row = unpack('C*', $row) ;
+               $row =~ tr /\x00-\x1f\x80-\xff/./ ;
+                                                                                               # my $bytes = join(' ', map { sprintf "%2.2x", $_} @bytes_in_row) ;
+                                                                                               # See comment below.
+               my $spaces = ' ' x $number_of_spaces ;
+               $dump .= join('  ', sprintf("%4.4x", $nr * 0x10 ), join(' ', map { sprintf "%2.2x", $_} @bytes_in_row), $spaces, $row) ;
+               $dump .= "\n" ;
+       }
+
+       print STDERR $dump, "\n" ;
+
+=begin comment
+
+tsitc> perl -MBenchmark -e 'timethese(1_00_000, { printf => q<printf "%2.2x %2.2x %2.2x\n", 61, 62, 63>, sprintf => q<$x = sprintf "%2
+.2x %2.2x %2.2x\n", 61, 62, 63; print $x>} )' | perl -ne 'print if /intf/'
+
+Benchmark: timing 100000 iterations of printf, sprintf...
+    printf:  1 wallclock secs ( 0.55 usr +  0.00 sys =  0.55 CPU)
+   sprintf:  0 wallclock secs ( 0.11 usr +  0.00 sys =  0.11 CPU)
+
+so having sprintf in a loop seems more rational than a printf for the line ...
+
+=comment
+
+=cut
+  
+}
+
+       
+