自动生成jsonl文件

ida_print.py

@@ -1,19 +0,0 @@
-import os
-import subprocess
-
-
-directory = './'
-
-
-if __name__ == '__main__':
-    cmd = 'D:\IDA_Pro_v6.8\idaq.exe -c -S"raw-feature-extractor/preprocessing_ida.py --path ./store/" '
-
-    for filename in os.listdir(directory):
-        if filename[-3:] == 'exe':
-            process = subprocess.Popen(["powershell", cmd+filename], stdout=subprocess.PIPE)
-
-            output = process.communicate()[0]
-
-
-
-

raw-feature-extractor/print_test.py

@@ -1,4 +1,5 @@
 # -*- coding: UTF-8 -*-
+import os
 import sys
 from matplotlib import pyplot as plt
 import networkx as nx
@@ -7,11 +8,6 @@ import hashlib
 import json
 
 
-def print_obj(obj):
-    "打印对象的所有属性"
-    print(obj.__dict__)
-
-
 def calc_sha256(file_path):
     with open(file_path, 'rb') as f:
         bytes = f.read()
@@ -25,10 +21,20 @@ import pickle
 # sub_10F20 308  反编译代码有字符串，但是这个特征提取里没有字符串 constant，可能是间接引用的，不识别。看了下所有函数的特征，几乎都没有字符串常量，可能都是写在别的地方然后引用的。
 # sub_166C4 393
 if __name__ == '__main__':
-    file_path = '../3c580f5beca53b6599e5f04d3aa68a34bd50521d7ec5d7163849eb69f53a4150.exe'
+    file_name_list = os.listdir('../A2C/')
-    testpath = '../store/3c580f5beca53b6599e5f04d3aa68a34bd50521d7ec5d7163849eb69f53a4150.exe.ida'
+    res_file = "../sample.jsonl"
+    sample_file = open(res_file, mode='a')
+    for file_name in file_name_list:
+        print file_name
+        file_path = '../A2C/' + file_name
+        testpath = '../store/' + file_name + '.ida'
+        if os.path.exists(testpath) and os.path.splitext(file_path)[-1].lower() == '.exe':
             fr = open(testpath, 'r')
             data1 = pickle.load(fr)  # 一个二进制文件的acfgs
+            # funtion num
+            function_number = len(data1.raw_graph_list)
+            if function_number == 0:
+                continue
             # function_edges
             function_edge_start = []
             function_edge_end = []
@@ -39,8 +45,7 @@ if __name__ == '__main__':
             fun_name_temp = []
             # function hsah
             file_hash = calc_sha256(file_path)
-    # funtion num
+
-    function_number = len(data1.raw_graph_list)
             acfg_list = []
             # 函数级特征
             for i in range(len(data1.raw_graph_list)):
@@ -74,7 +79,8 @@ if __name__ == '__main__':
                     # total instructions
                     block_features.append(temp_G.node[temp]['numIns'])
                     # string or integer constants
-            block_features.append(len(temp_G.node[temp]['strings']) if len(temp_G.node[temp]['strings']) != 0 else len(
+                    block_features.append(
+                        len(temp_G.node[temp]['strings']) if len(temp_G.node[temp]['strings']) != 0 else len(
                             temp_G.node[temp]['consts']))
                     # offspring
                     block_features.append(temp_G.node[temp]['offs'])
@@ -85,15 +91,15 @@ if __name__ == '__main__':
                     edge_list_start.append(item[0])
                     edge_list_end.append(item[1])
                 block_edges = [edge_list_start, edge_list_end]
-        acfg_list_item = {"block_number": block_number, "block_edges": block_edges, "block_features": acfg_list_item_feature}
+                acfg_list_item = {"block_number": block_number, "block_edges": block_edges,
+                                  "block_features": acfg_list_item_feature}
                 acfg_list.append(acfg_list_item)
 
-    json_temp = {"function_edges": function_edges, "acfg_list": acfg_list, "function_names": fun_name_temp, "hash": file_hash, "function_number": function_number}
+            json_temp = {"function_edges": function_edges, "acfg_list": acfg_list, "function_names": fun_name_temp,
+                         "hash": file_hash, "function_number": function_number}
             json_str = json.dumps(json_temp)
-    print json_str
+            sample_file.write(json_str)
-
+        else:
-
+            print "删除文件" + file_path
-
+            os.remove(file_path)
-
+    sample_file.close()
-
-