feat: compile regexes

Author: azmeuk

doc/changelog.rst

@@ -7,6 +7,7 @@ Changelog
 Added
 ^^^^^
 - Support for Python 3.14.
+- Compile regexes.
 
 Removed
 ^^^^^^^

scim2_models/resources/schema.py

@@ -35,10 +35,12 @@
 
 T = TypeVar("T", bound=BaseModel)
 
+_NON_WORD_OR_LEADING_DIGIT = re.compile(r"\W|^(?=\d)")
+
 
 def _make_python_identifier(identifier: str) -> str:
     """Sanitize string to be a suitable Python/Pydantic class attribute name."""
-    sanitized = re.sub(r"\W|^(?=\d)", "", identifier)
+    sanitized = _NON_WORD_OR_LEADING_DIGIT.sub("", identifier)
     if sanitized in RESERVED_WORDS:
         sanitized = f"{sanitized}_"
 

scim2_models/utils.py

@@ -21,6 +21,10 @@
     # Python 3.9 has no UnionType
     UNION_TYPES = [Union]
 
+_UNDERSCORE_ALPHANUMERIC = re.compile(r"_+([0-9A-Za-z]+)")
+_NON_WORD_UNDERSCORE = re.compile(r"[\W_]+")
+_VALID_PATH_PATTERN = re.compile(r'^[a-zA-Z][a-zA-Z0-9._:\-\[\]"=\s]*$')
+
 
 def _int_to_str(status: int | None) -> str | None:
     return None if status is None else str(status)
@@ -86,7 +90,7 @@ def _to_camel(string: str) -> str:
     '$ref' stays '$ref'.
     """
     snake = to_snake(string)
-    camel = re.sub(r"_+([0-9A-Za-z]+)", lambda m: m.group(1).title(), snake)
+    camel = _UNDERSCORE_ALPHANUMERIC.sub(lambda m: m.group(1).title(), snake)
     return camel
 
 
@@ -97,7 +101,7 @@ def _normalize_attribute_name(attribute_name: str) -> str:
     """
     is_extension_attribute = ":" in attribute_name
     if not is_extension_attribute:
-        attribute_name = re.sub(r"[\W_]+", "", attribute_name)
+        attribute_name = _NON_WORD_UNDERSCORE.sub("", attribute_name)
 
     return attribute_name.lower()
 
@@ -121,7 +125,7 @@ def _validate_scim_path_syntax(path: str) -> bool:
 
     # Cannot contain invalid characters (basic check)
     # Allow alphanumeric, dots, underscores, hyphens, colons (for URNs), brackets
-    if not re.match(r'^[a-zA-Z][a-zA-Z0-9._:\-\[\]"=\s]*$', path):
+    if not _VALID_PATH_PATTERN.match(path):
         return False
 
     # If it contains a colon, validate it's a proper URN format