added a script that helps find and fix duplicate objects that happen when user copied standard objects into their data file by hand

2025-11-06 18:52:58 +01:00 · 2010-04-29 20:32:13 +00:00 · 2010-04-29 20:32:13 +00:00 · 8c80ef00b0
commit 8c80ef00b0
parent 8ba49b53cb
1 changed files with 120 additions and 0 deletions
--- a/src/tools/resolve_duplicate_std_objects.py
+++ b/src/tools/resolve_duplicate_std_objects.py
@ -0,0 +1,120 @@
+#!/usr/bin/python
+#
+# The purpose of this script is to find obejcts that normally belong
+# to the Standard Objects library but were copied to the user-defined
+# library thus creating duplicate IDs. Script finds such objects in
+# the specified data file and prints sed commands that change the IDs
+# to make them unique. The script does not modify the data file in any
+# way.
+#
+# If a copy of a duplicate object in the user-defined library is
+# identical to the object in the Standard library, it is removed. If
+# the copy has been modified, the id is changed to be unique and all
+# references to it are corrected. Obejcts are compared as text lines
+# regardless of the contents.
+#
+# This script does not use fwbuilder DTD and operates on xml files as
+# plain text, it can be used with files of old or different versions
+# regardless of fwbuilder DTD.
+
+import getopt
+import time
+import re
+import sys
+
+NEW_ID_PREFIX = '%s' % int(time.time())
+generated_commands = []
+
+def usage():
+    print '%s data_file.fwb objects_init.xml' % sys.argv[0]
+
+def command_to_delete(id):
+    generated_commands.append("""sed '/id="%s"/d'""" % id)
+
+def command_to_change(id):
+    new_id = '%s_%s' % (id, NEW_ID_PREFIX)
+    generated_commands.append("""sed 's/id="%s"/id="%s"/'""" % (id, new_id))
+    generated_commands.append("""sed 's/ref="%s"/ref="%s"/g'""" % (id, new_id))
+
+def main():
+    try:
+        data_file = sys.argv[1]
+        std_file = sys.argv[2]
+    except IndexError,e:
+        usage()
+        sys.exit(1)
+
+    std_objects = {}
+
+    objects_of_interest = ['TCPService', 'UDPService', 'ICMPService', 'IPService',
+                           'Network', 'Host', 'IPv4', 'IPv6',
+                           'ObjectGroup', 'ServiceGroup', 'IntervalGroup',
+                           'Library']
+    objects_we_never_delete = ['ObjectGroup', 'ServiceGroup', 'IntervalGroup', 'Library']
+
+    object_type_re = re.compile('<([^ ]+) ')
+    object_id_re = re.compile('id="([^\"]+)"')
+    object_ref_re = re.compile('ref="([^\"]+)"')
+    object_comment_re = re.compile('comment="([^\"]*)"')
+
+    # first, read standard objects and catalog all lines that have "id=" in them
+    for line in open(std_file):
+        line = line.strip()
+        m = object_id_re.search(line)
+        if m:
+            id = m.group(1)
+            std_objects[id] = line
+
+    # now scan the data file, find all object definitions and compare
+    for line in open(data_file):
+        line = line.strip()
+        m = object_type_re.match(line)
+        if m:
+            obj_type = m.group(1)
+            if obj_type in objects_of_interest:
+                m = object_id_re.search(line)
+                if m:
+                    id = m.group(1)
+                    if id in std_objects:
+                        print 32 * '-'
+                        print 'Find duplicate object:'
+                        print 'Object in the data file:'
+                        print line
+                        print 'Object in the standard objects file:'
+                        print std_objects[id]
+                        # One of the typical cases is when only comment has changed
+                        file_line_no_comment = None
+                        std_line_no_comment = None
+                        m1 = object_comment_re.match(line)
+                        m2 = object_comment_re.match(std_objects[id])
+                        if m1:
+                            file_comment = m1.group(1)
+                            file_line_no_comment = line.replace('comment="%s"' % file_comment, '')
+                        if m2:
+                            std_comment = m2.group(1)
+                            std_line_no_comment = line.replace('comment="%s"' % std_comment, '')
+
+                        if line == std_objects[id]:
+                            if obj_type not in objects_we_never_delete:
+                                print 'Objects are IDENTICAL'
+                                command_to_delete(id)
+                            else:
+                                print 'Objects are IDENTICAL but we cant delete'
+                                command_to_change(id)
+
+                        elif (file_line_no_comment and
+                              std_line_no_comment and
+                              file_line_no_comment == std_line_no_comment):
+                            print 'Only comments are DIFFERENT'
+                            print file_line_no_comment
+                            print std_line_no_comment
+                            command_to_change(id)
+
+                        else:
+                            print 'Objects are DIFFERENT'
+                            command_to_change(id)
+
+    print 'cat %s | %s' % (data_file, ' | '.join(generated_commands))
+
+if __name__ == '__main__':
+    main()