run_system.py

import csv
from subprocess import call
import os
import pandas as pd
import re

PATH = "./elf"
cur_path = os.getcwd()

try:
	call("cd ELF\ Miner && rm results.csv", shell=True)
except Exception as e:
	print e

flag = 0
for i in os.listdir(PATH):
	try:
		file = os.path.join(os.path.join(cur_path, PATH), i)
		call('objdump -M intel -D "' + file + '"' + ' > hello.txt', shell=True)
		if os.stat("hello.txt").st_size == 0:
			raise Exception("incorrect format")
		print file
		## ELF Miner
		call("cd ELF\ Miner && python ELFMiner.py " + "'" + file + "' && cp results.csv ../final.csv && cd " + cur_path, shell=True)
		
		## Syscalls
		headers = []

		with open('./Syscalls/features_less.txt', 'r') as f:
			for l in f:
				a = l.split(': ')[0]
				headers.append(a)

		df = pd.read_csv('final.csv')
		if flag == 0:
			for i in headers:
				df[i] = 0
			flag = 1

		s = re.compile('<.+>')
		symbols = re.compile('"\w+"')
		calls = dict()
		row = [0]*len(headers)
		with open('hello.txt', 'r') as f:
			for l in f:
				if 'call   ' in l.strip():
					ind = l.strip().index('call   ')
					new_str = l.strip()[ind:]
					m = re.findall(s, new_str)
					if m:
						new_str = m[0]
					if (('Base' in new_str and 'Base+' not in new_str) or ('plt' in new_str)) and new_str in headers:
						if new_str not in calls.keys():
							calls[new_str] = 1;
						else:
							calls[new_str] += 1

		for i in calls.keys():
			row[headers.index(i)] = calls[i]

		ind = df.shape[0]-1
		for i, h in enumerate(headers):
			df.loc[df.index[-1], h] = row[i]

		## Opcodes
		opcodes_list = ['mov', 'push', 'call', 'pop', 'cmp', 'jz', 'lea', 'test', 'jmp', 'add', 'jnz', 'retn', 'xor', 'and', 'bt', 'fdivp', 'fild', 'fstcw', 'imul', 'int', 'nop', 'pushf', 'rdtsc', 'sbb', 'setb', 'setle', 'shld', 'std', '(bad)']

		headers = []
		headers.extend(opcodes_list)
		headers.append('labels')

		if flag == 1:
			for i in headers:
				df[i] = 0
			flag = 2
		
		row = [0]*(len(headers)-1)
		# row[0] = file
		with open('hello.txt', 'r') as f:
			for l in f:
				if ':' in l:
					# print l
					for i, opcode in enumerate(opcodes_list):
						if opcode in l:
							row[i] += 1
		row.append('?')
		for i, h in enumerate(headers):
			df.loc[df.index[-1], h] = row[i]

		df.to_csv('final.csv', index=False)
		call('cp final.csv ./ELF\ Miner/results.csv', shell=True)

	except Exception as e:
		print e

call('python postprocessing.py', shell=True)