4 Script that takes a file (output by wireshark/tshark, in JSON format) and analyze
5 the variety of packet sizes of a certain device at a certain time.
11 from collections import defaultdict
12 from dateutil import parser
14 JSON_KEY_SOURCE = "_source"
15 JSON_KEY_LAYERS = "layers"
18 JSON_KEY_ETH_DST = "eth.dst"
19 JSON_KEY_ETH_SRC = "eth.src"
20 JSON_KEY_FRAME = "frame"
21 JSON_KEY_FRAME_TIME = "frame.time"
22 JSON_KEY_FRAME_LENGTH = "frame.len"
23 TABLE_HEADER_X = "Timestamp (hh:mm:ss)"
24 TABLE_HEADER_Y = "Packet sizes (bytes)"
25 INCOMING_APPENDIX = "_incoming"
26 OUTGOING_APPENDIX = "_outgoing"
27 FILE_APPENDIX = ".dat"
30 def save_to_file(tblheader, dictionary, filenameout):
31 """ Show summary of statistics of PCAP file
33 tblheader: header for the saved table
34 dictionary: dictionary to be saved
35 filename_out: file name to save
37 # Appending, not overwriting!
38 f = open(filenameout, 'a')
39 # Write the table header
40 f.write("# " + tblheader + "\n")
41 f.write("# " + TABLE_HEADER_X + " " + TABLE_HEADER_Y + "\n")
42 # Write "0 0" if dictionary is empty
46 print "Writing zeroes to file: ", filenameout
49 # Iterate over dictionary and write (key, value) pairs
50 for key in sorted(dictionary):
52 f.write(str(key) + " " + str(dictionary[key]) + "\n")
54 print "Writing output to file: ", filenameout
61 print "Usage: python", sys.argv[0], "<input_file> <output_file> <device_name> <mac_address>"
63 # Parse the file for the specified MAC address
64 timefreq_incoming = parse_json(sys.argv[1], sys.argv[4], True)
65 timefreq_outgoing = parse_json(sys.argv[1], sys.argv[4], False)
66 # Write statistics into file
67 print "====================================================================="
68 print "==> Analyzing incoming traffic ..."
69 save_to_file(sys.argv[3] + INCOMING_APPENDIX, timefreq_incoming, sys.argv[2] + INCOMING_APPENDIX + FILE_APPENDIX)
70 print "====================================================================="
71 print "==> Analyzing outgoing traffic ..."
72 save_to_file(sys.argv[3] + OUTGOING_APPENDIX, timefreq_outgoing, sys.argv[2] + OUTGOING_APPENDIX + FILE_APPENDIX)
73 print "====================================================================="
76 # Convert JSON file containing DNS traffic to a map in which a hostname points to its set of associated IPs.
77 def parse_json(filepath, macaddress, incomingoutgoing):
78 """ Show summary of statistics of PCAP file
80 filepath: path of the read file
81 macaddress: MAC address of a device to analyze
82 incomingoutgoing: boolean to define whether we collect incoming or outgoing traffic
83 True = incoming, False = outgoing
85 # Maps timestamps to frequencies of packets
87 with open(filepath) as jf:
89 # data becomes reference to root JSON object (or in our case json array)
91 # Loop through json objects in data
92 # Each entry is a pcap entry (request/response (packet) and associated metadata)
94 # p is a JSON object, not an index
95 layers = p[JSON_KEY_SOURCE][JSON_KEY_LAYERS]
97 frame = layers.get(JSON_KEY_FRAME, None)
98 datetime = frame.get(JSON_KEY_FRAME_TIME, None)
99 length = frame.get(JSON_KEY_FRAME_LENGTH, None)
100 # Get into the Ethernet address part
101 eth = layers.get(JSON_KEY_ETH, None)
102 # Skip any non DNS traffic
104 print "[ WARNING: Packet has no ethernet address! ]"
106 # Get source and destination MAC addresses
107 src = eth.get(JSON_KEY_ETH_SRC, None)
108 dst = eth.get(JSON_KEY_ETH_DST, None)
109 # Get just the time part
110 datetimeobj = parser.parse(datetime)
111 timestr = str(datetimeobj.time())
112 print str(timestr) + " - src:" + str(src) + " - dest:" + str(dst)
113 # Get and count the traffic for the specified MAC address
115 if dst == macaddress:
116 # Put the time frequency in the dictionary
117 packetsize[timestr] = length
119 if src == macaddress:
120 # Put the time frequency in the dictionary
121 packetsize[timestr] = length
126 if __name__ == '__main__':