fix: 修改正则匹配逻辑

detection/Regexdetection.py

@@ -34,6 +34,7 @@ def find_dangerous_functions(
             r"\bos\.kill\b": "high",
             r"\bos\.popen\b": "medium",
             r"\bos\.spawn\b": "medium",
+            r"\bsubprocess": "medium",
         },
     }
     risk_patterns = patterns.get(file_extension, {})
@@ -43,7 +44,9 @@ def find_dangerous_functions(
             clean_line = remove_comments(line, file_extension)
             if not clean_line:
                 continue
+            # 消除换行符，避免影响正则匹配
+            clean_line = clean_line.replace("\\n", "")
             for pattern, risk_level in risk_patterns.items():
-                if re.search(pattern, clean_line):
+                if re.search(pattern, clean_line, re.MULTILINE | re.DOTALL):
                     classified_results[risk_level].append((line_number, clean_line))
     return classified_results

detection/__main__.py

@@ -7,7 +7,8 @@ from reportlab.platypus import Paragraph, Spacer, SimpleDocTemplate
 
 from detection.pickle_detection import pickleDataDetection
 from .Regexdetection import find_dangerous_functions
-from .GPTdetection import detectGPT,GPTdetectFileList
+from .GPTdetection import detectGPT, GPTdetectFileList
+
 # from .cngptdetection import detectGPT,GPTdetectFileList
 from .pyc_detection import disassemble_pyc
 from .utils import *
@@ -30,6 +31,8 @@ ORDERS = [
     "__getattribute__",
     "getattr",
     "child_process",
+    "kill",
+    "fork",
 ]
 
 # Initialize colorama
@@ -146,8 +149,6 @@ def generate_text_content(results: Dict[str, List[Tuple[int, str]]]) -> str:
                 text_output += line_text
             text_output += "\n"
 
-
-
     return text_output
 
 
@@ -372,7 +373,7 @@ def checkModeAndDetect(mode: str, filePath: str, fileExtension: str, pycdc_addr:
 def process_path(
     path: str, output_format: str, mode: str, pycdc_addr: str, output_file=None
 ):
-    results = {"high": [], "medium": [], "low": [], "none": [],"pickles": []}
+    results = {"high": [], "medium": [], "low": [], "none": [], "pickles": []}
     if os.path.isdir(path):
         # 使用rglob获取所有文件
         all_files = [
@@ -383,15 +384,12 @@ def process_path(
         if mode == "llm":
             results = GPTdetectFileList(all_files)
         else:
-        # 扫描动画
+            # 扫描动画
             for file_path in tqdm(all_files, desc="Scanning files", unit="file"):
                 file_extension = file_path.suffix
-                if file_extension in [".pkl",".pickle"]:
+                if file_extension in [".pkl", ".pickle"]:
                     res = pickleDataDetection(str(file_path), output_file)
-                    results["pickles"].append({
-                        "file": str(file_path),
-                        "result": res
-                    })
+                    results["pickles"].append({"file": str(file_path), "result": res})
                     continue
                 file_results = checkModeAndDetect(
                     mode, str(file_path), file_extension, pycdc_addr
@@ -409,10 +407,7 @@ def process_path(
         file_extension = os.path.splitext(path)[1]
         if file_extension in [".pkl", ".pickle"]:
             res = pickleDataDetection(str(path), output_file)
-            results["pickles"].append({
-                "file": str(path),
-                "result": res
-            })
+            results["pickles"].append({"file": str(path), "result": res})
         elif file_extension in SUPPORTED_EXTENSIONS:
             file_results = checkModeAndDetect(mode, path, file_extension, pycdc_addr)
             if file_results is not None: