fix: address PR review feedback (regex optimization, typos, duplicate sigs)

Author: EdoardoAllegrini

capa/features/extractors/ts/engine.py

@@ -29,6 +29,11 @@
 )
 from capa.features.extractors.ts.tools import LANGUAGE_TOOLKITS, BaseNamespace, CSharpNamespace, LanguageToolkit
 
+_RE_CSHARP_PAGE = re.compile(r'@ .*Page Language\s*=\s*"C#".*'.encode(), re.IGNORECASE)
+_RE_ASPX_IMPORT_DIRECTIVE = re.compile(r"@\s*Import Namespace=".encode(), re.IGNORECASE)
+_RE_ASPX_NAMESPACE = re.compile(r'@\s*Import namespace="(.*?)"'.encode(), re.IGNORECASE)
+_RE_RUNAT_SERVER = re.compile(r'runat\s*=\s*"server"'.encode())
+
 
 class TreeSitterBaseEngine:
     buf: bytes
@@ -227,29 +232,13 @@ def get_namespaces(self) -> Iterator[BaseNamespace]:
         yield from self.get_imported_namespaces()
 
     def is_c_sharp(self, node: Node) -> bool:
-        return bool(
-            re.match(
-                r'@ .*Page Language\s*=\s*"C#".*'.encode(),
-                self.get_byte_range(node),
-                re.IGNORECASE,
-            )
-        )
+        return bool(_RE_CSHARP_PAGE.match(self.get_byte_range(node)))
 
     def is_aspx_import_directive(self, node: Node) -> bool:
-        return bool(
-            re.match(
-                r"@\s*Import Namespace=".encode(),
-                self.get_byte_range(node),
-                re.IGNORECASE,
-            )
-        )
+        return bool(_RE_ASPX_IMPORT_DIRECTIVE.match(self.get_byte_range(node)))
 
     def get_aspx_namespace(self, node: Node) -> Optional[BaseNamespace]:
-        match = re.search(
-            r'@\s*Import namespace="(.*?)"'.encode(),
-            self.get_byte_range(node),
-            re.IGNORECASE,
-        )
+        match = _RE_ASPX_NAMESPACE.search(self.get_byte_range(node))
         return CSharpNamespace(match.group(1).decode("utf-8"), node) if match is not None else None
 
 
@@ -295,4 +284,4 @@ def identify_language(self, node: Node) -> str:
         return LANG_JS
 
     def is_server_side_c_sharp(self, node: Node) -> bool:
-        return bool(re.findall(r'runat\s*=\s*"server"'.encode(), self.get_byte_range(node)))
+        return bool(_RE_RUNAT_SERVER.search(self.get_byte_range(node)))

capa/features/extractors/ts/signatures/cs.json

@@ -33,7 +33,6 @@
         "System.IO.File.GetLastAccessTime",
         "System.IO.File.GetLastWriteTime",
         "System.IO.File.ReadAllBytes",
-        "System.IO.File.ReadAllBytes",
         "System.IO.File.ReadAllBytesAsync",
         "System.IO.File.ReadAllLines",
         "System.IO.File.ReadAllLinesAsync",
@@ -45,7 +44,6 @@
         "System.IO.File.SetLastAccessTime",
         "System.IO.File.SetLastWriteTime",
         "System.IO.File.WriteAllBytes",
-        "System.IO.File.WriteAllBytes",
         "System.IO.File.WriteAllBytesAsync",
         "System.IO.File.WriteAllLines",
         "System.IO.File.WriteAllLinesAsync",

capa/features/extractors/ts/tools.py

@@ -50,7 +50,7 @@ def join(self, name: str) -> str:
         """import subprocess ; subprocess.Popen => subprocess.Popen
         from threading import Timer (threading.Timer) => Timer
         """
-        toolkit = LANGUAGE_TOOLKITS[LANG_CS]
+        toolkit = LANGUAGE_TOOLKITS[LANG_PY]
         qualified_names = toolkit.split_name(self.name)
         if len(qualified_names) < 2:
             return name
@@ -224,8 +224,8 @@ class PythonToolkit(LanguageToolkit):
     property_query_type: str = "attribute"
     string_delimiters: str = "\"'"
     integer_prefixes: List[Tuple[Union[str, Tuple[str, ...]], int]] = [
-        (("0b, 0B"), 2),
-        (("0o, 0O"), 8),
+        (("0b", "0B"), 2),
+        (("0o", "0O"), 8),
         (("0x", "0X"), 16),
     ]
     integer_suffixes: Tuple[str, ...] = ()