python: Convert tcpdump into text2pcap readable format

Question

Recently there was a requirement for me to convert the textual output of "tcpdump -i eth0 -neXXs0" into a pcap file. So I wrote a python script which converts the information into an intermediate format understandable by text2pcap. Since this is my first program in python there would obviously be scope for improvement. I want knowledgeable folks to weed out any descrepancy and/or enhance it.

Input

The tcpdump output is in the following format:

20:11:32.001190 00:16:76:7f:2b:b1 > 00:11:5c:78:ca:c0, ethertype IPv4 (0x0800), length 72: 123.236.188.140.41756 > 94.59.34.210.45931: UDP, length 30

0x0000:  0011 5c78 cac0 0016 767f 2bb1 0800 4500  ..\x....v.+...E.
0x0010:  003a 0000 4000 4011 812d 7bec bc8c 5e3b  .:..@[email protected]{...^;
0x0020:  22d2 a31c b36b 0026 b9bd 2033 6890 ad33  "....k.&...3h..3
0x0030:  e845 4b8d 2ba1 0685 0cb3 70dd 9b98 76d8  .EK.+.....p...v.
0x0040:  8fc6 8293 bf33 325a                      .....32Z

Output

enter code here

Format understandable by text2pcap:

20:11:32.001190

0000: 00 11 5c 78 ca c0 00 16 76 7f 2b b1 08 00 45 00   ..\x....v.+...E. 
0010: 00 3a 00 00 40 00 40 11 81 2d 7b ec bc 8c 5e 3b   .:..@[email protected]{...^; 
0020: 22 d2 a3 1c b3 6b 00 26 b9 bd 20 33 68 90 ad 33   "....k.&...3h..3
0030: e8 45 4b 8d 2b a1 06 85 0c b3 70 dd 9b 98 76 d8   .EK.+.....p...v. 
0040: 8f c6 82 93 bf 33 32 5a   .....32Z 

Following is my Code.

import re
# Identify time of the current packet.
time = re.compile ('(..:..:..\.[\w]*) ')
# Get individual elements from the packet. ie. offset, hexdump, chars
all = re.compile('[ |\t]+0x([\w]+:) +(.+)  +(.*)')
# Regex for two spaces
twoSpaces = re.compile('  +')
# Regex for single space
singleSpace = re.compile(' ')
# Single byte pattern.
singleBytePattern = re.compile(r'([\w][\w])')

# Open files.
f = open ('pcap.txt', 'r')
outfile = open ('ashu.txt', 'w')

for line in f:
  result = time.match (line)
  if result:
    # If current line contains time format dump only time
    print result.group()
    outfile.write (result.group() + '\n')
  else:
    print line,
    # Split line containing hex dump and tokenize into list elements.
    result = all.split (line)
    if result:
      i = 0
      for values in result:
        if (i == 2):
          # Strip off additional spaces in hex dump
          # Useful when hex dump does not end in 16 bytes boundary.
          val = twoSpaces.sub ('', values)

          # Tokenize individual elements seperated by single space.
          byteResult = singleSpace.split (val)
          for twoByte in byteResult:
            # Identify individual byte
            singleByte = singleBytePattern.split(twoByte)
            byteOffset = 0
            for oneByte in singleByte:
              if ((byteOffset == 1) or (byteOffset == 3)):
               # Write out individual byte with a space char appended
               print oneByte, 
               outfile.write (oneByte+ ' ')
              byteOffset = byteOffset + 1
        elif (i == 3):
          # Write of char format of hex dump
          print "  "+values,
          outfile.write ('  ' + values+ ' ')
        elif (i == 4):
          outfile.write (values) 
        else:
          print values,
          outfile.write (values + ' ')
        i=i+1
    else:
      print "could not split"
f.close ()
outfile.close ()

Answer 1

Use the -w option of tcpdump to write to a pcap format file

tcpdump -w filename.pcap

Wireshark should be able to read it.