diff options
Diffstat (limited to 'main.py')
| -rw-r--r-- | main.py | 64 |
1 files changed, 64 insertions, 0 deletions
@@ -0,0 +1,64 @@ +# 输入一系列pcap和pcap对应的标签，输出一个dataframe格式的csv
+from time import time
+from configparser import ConfigParser
+from multiprocessing import cpu_count
+import pandas as pd
+import sys
+import os
+from concurrent.futures import ThreadPoolExecutor
+from threading import Lock
+import re
+
+lock = Lock()
+pcap_name_complier = re.compile(".*/(?P<domain>.*)_(?P<doh_uri>.*)_(?P<browser>.*)_(?P<time>.*)\.pcap")
+
+
+def multi_write(input_filename, id):
+ match_result = pcap_name_complier.match(input_filename)
+ if match_result:
+ domain = match_result["domain"]
+ doh_uri = match_result["doh_uri"]
+ browser = match_result["browser"]
+ else:
+ domain = None
+ doh_uri = None
+ browser = None
+ lengths, time_lags, directions = trans_pcap_to_row(input_filename)
+ if len(lengths) == 0 or sum(time_lags) < 1e-7:
+ return
+ flow_serial = [lengths[i] * directions[i] for i in range(len(lengths))]
+ time_serial = [time_lags[i] * directions[i] for i in range(len(time_lags))]
+ features = extract_statistic_feature(lengths, time_lags, directions)
+
+ lock.acquire()
+ df.loc[id] = [domain, doh_uri, browser, flow_serial, time_serial, features]
+ lock.release()
+
+
+def run(input_filepath):
+ files = os.listdir(input_filepath)
+ with ThreadPoolExecutor(max_workers=30) as thread_pool:
+ id = 0
+ for i in range(len(files)):
+ file = files[i]
+ full_file = os.path.join(input_filepath, file)
+ if not full_file.endswith(".pcap"):
+ continue
+ # multi_write(full_file, id)
+ thread_pool.submit(multi_write, full_file, id)
+ id += 1
+ print(id)
+
+
+if __name__ == "__main__":
+ # input_filepath = sys.argv[1]
+ # print(input_filepath)
+ # output_filename = sys.argv[2]
+ input_filepath = "E:\\doh&web\\201102\\win10\\web_traffic_after_predeal"
+ output_filename = "./result/web_feature_2.csv"
+ start_time = time()
+ df = pd.DataFrame(columns=["domain", "doh_uri", "browser", "flow_serial", "time_serial", "features"])
+ run(input_filepath)
+ df.to_csv(output_filename)
+ end_time = time()
+ print("总耗时：", end_time - start_time)
|