Parse enterprise-numbers at run time

"enterprise-numbers" is converted to tab-separated values and renamed
"enterprises". Unused fields are stripped.

PENs are stored in a hash table loaded at run-time.

User "enterprises" file is loaded from the personal config dir.

Misc make-sminmpec.pl improvements and fixes.
Note: names of type "Entity (formerly ...)" have the formerly part commented out for a cleaner output.

Change-Id: I60c533afbe3e399077fbf432088064471ad3e1e2
Reviewed-on: https://code.wireshark.org/review/22246
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Reviewed-by: João Valverde <j@v6e.pt>

1 files changed, 27 insertions, 77 deletions

diff --git a/tools/make-sminmpec.pl b/tools/make-sminmpec.pl
index 339532f459..a30e717f71 100755
--- a/tools/make-sminmpec.pl
+++ b/tools/make-sminmpec.pl
@@ -1,5 +1,5 @@
 #!/usr/bin/perl -w
-# create the sminmpec.c file from
+# create the enterprises file from
 # http://www.iana.org/assignments/enterprise-numbers
 #
 # Wireshark - Network traffic analyzer
@@ -23,9 +23,9 @@
 use strict;
 use File::Spec;
 
-my ($vol, $script_dir, $file) = File::Spec->splitpath( __FILE__ );
-my $epan_dir = File::Spec->catpath($vol, $script_dir, '../epan');
-chdir($epan_dir) || die("Can't find $epan_dir");
+my ($vol, $script_dir) = File::Spec->splitpath( __FILE__ );
+my $root_dir = File::Spec->catpath($vol, $script_dir, "..");
+chdir($root_dir) || die("Can't find $root_dir");
 
 my $in = shift;
 
@@ -54,100 +54,50 @@ if($in =~ m/^http:/i) {
 	if ($result->code eq 200) {
 		warn "done fetching $in\n";
 		@in_lines = split /\n/, $result->content;
-		open ENFILE, "> enterprise-numbers";
-
-		for (@in_lines) {
-			chomp;
-			print ENFILE "$_\n";
-		}
-
-		close ENFILE;
-	} elsif ($result->code eq 304) {
-		warn "enterprise-numbers was up-to-date\n";
-		open IN, "< enterprise-numbers";
-		@in_lines = <IN>;
-		close IN;
 	} else {
 		die "request for $in failed with result code:" . $result->code;
 	}
-
 } else {
   open IN, "< $in";
   @in_lines = <IN>;
   close IN;
 }
 
-
 my $body = '';
 my $code;
-my $prev_code = -1;  ## Assumption: First code in enterprise file is 0;
-
-sub escape_non_ascii {
-    my $val = unpack 'C', $_[0];
-    return sprintf '\0%.3o',$val;
-}
-
+my $name;
+my $last_updated = "(last updated ???)";
+my $end_of_document = 0;
 
 for(@in_lines) {
-	s/[\000-\037]//g;
-	s/\\/\\\\/g;
-	s/"/\\"/g;
-	s/([\x80-\xFF])/escape_non_ascii($1)/ge;
+	chomp;
 
 	if (/^(\d+)/) {
-		$code = sprintf("%5d", $1);
-	} if (/^  (\S.*)/ ) {
-		my $name = $1;
-		if ($code < $prev_code) {
-			print STDERR ("Input 'Codes' not sorted in ascending order (or duplicate codes)): $prev_code $code\n");
-                        exit 1;
-		}
-		while ($code > ($prev_code+1)) {
-			$prev_code = sprintf("%5d", $prev_code+1);
-			$body .= "    { $prev_code, sminmpec_unknown },  /* (Added by Wireshark) */\n";
-		}
-		$prev_code = $code;
-		$body .= "    { $code, \"$name\" },\n";
-		$smi_total++;
+		$code = sprintf("%d", $1);
+	} elsif (/^   ?(\S.*)/ ) { # up to three spaces because of formatting errors in the source
+		$name = $1;
+		next if (/^\s*\(?\s*unassigned/i);
+		$name =~ s/\s+$//;
+		$name =~ s/ \((formerly .*)\)/\t# $1/;
+		$body .= "\n$code\t$name";
+	} elsif (/\(last updated/i) {
+		$last_updated = $_;
+	} elsif (/^ *End of Document/) {
+		$end_of_document = 1;
 	}
 }
 
-# If this happens check what IANA is serving.
-# XXX We already overwrote enterprise-numbers above.
-if ($smi_total < $smi_total) { die "Too few SMI entries ($smi_total)\n"; }
+die "\"End of Document\" not found. Truncated source file?" unless ($end_of_document);
 
-open OUT, "> sminmpec.c";
+open OUT, "> enterprises";
 
 print OUT <<"_SMINMPEC";
-/*
- * THIS FILE IS AUTOGENERATED, DO NOT EDIT
- * generated from http://www.iana.org/assignments/enterprise-numbers
- * run "tools/make-sminmpec.pl <infile> <outfile>" to regenerate
- *
- * Note 1: "Gaps" in the iana enterprise-numbers list have been "filled in"
- *       with "(Unknown)" as the name so that direct (indexed) access to
- *       the list is possible.
- *
- * Note 2: We should probably parse "enterprise-numbers" at program start
- *       instead of generating this file.
- */
-#include "config.h"
-
-#include <glib.h>
-
-#include <epan/value_string.h>
-#include <epan/sminmpec.h>
-
-static const gchar sminmpec_unknown[] = "(Unknown)";
-
-const value_string sminmpec_values[] = {
-$body    {    0, NULL}
-};
-
-#define array_length(x)	(sizeof x / sizeof x[0])
-
-value_string_ext sminmpec_values_ext = VALUE_STRING_EXT_INIT(sminmpec_values);
-
+#
+# generated from http://www.iana.org/assignments/enterprise-numbers
+# run "tools/make-sminmpec.pl [infile]" to regenerate
+#
+# $last_updated
+$body
 _SMINMPEC
 
 close OUT;