Improve non ascii checker

CONTRIBUTORS.txt

@@ -595,3 +595,6 @@ contributors:
 * Eero Vuojolahti: contributor
 
 * Kian-Meng, Ang: contributor
+
+* Carli* Freudenberg (CarliJoy): contributor
+  - Improve non-ascii-name checker

ChangeLog

@@ -14,6 +14,14 @@ Release date: TBA
 
   Closes #5588
 
+* Add ``non-ascii-identifier`` as replacement ``non-ascii-name``
+   to ensure really __all__ Python names are ASCII.
+   Checker now checks properly the names of imports (``non-ascii-module-import``) as
+   well for as of file names (``non-ascii-file-name``).
+   Non ASCII characters could be homoglyphs (look alike characters) and hard to
+   enter on a non specialized keyboard.
+   See `Confusable Characters in PEP 672 <https://www.python.org/dev/peps/pep-0672/#confusable-characters-in-identifiers>`_
+
 * When run in parallel mode ``pylint`` now pickles the data passed to subprocesses with
   the ``dill`` package. The ``dill`` package has therefore been added as a dependency.
 

doc/whatsnew/2.13.rst

@@ -14,8 +14,17 @@ New checkers
 
   Closes #5460
 
+* Add ``non-ascii-identifier`` as replacement ``non-ascii-name``
+   to ensure really __all__ Python names are ASCII.
+   Checker now checks properly the names of imports (``non-ascii-module-import``) as
+   well for as of file names (``non-ascii-file-name``).
+   Non ASCII characters could be homoglyphs (look alike characters) and hard to
+   enter on a non specialized keyboard.
+   See `Confusable Characters in PEP 672 <https://www.python.org/dev/peps/pep-0672/#confusable-characters-in-identifiers>`_
+
 Removed checkers
 ================
+* ``non-ascii-name`` has replaced by ``non-ascii-identifier``
 
 Extensions
 ==========

pylint/checkers/__init__.py

@@ -39,7 +39,14 @@
 15: stdlib
 16: python3
 17: refactoring
-18-50: not yet used: reserved for future internal checkers.
+.
+.
+.
+25: non-ascii-names
+25-50: not yet used: reserved for future internal checkers.
+This file is not updated. Use
+   script/get_unused_message_id_category.py
+to get the next free checker id.
 51-99: perhaps used: reserved for external checkers
 
 The raw_metrics checker has no number associated since it doesn't emit any

pylint/checkers/base.py

@@ -73,7 +73,7 @@
 import itertools
 import re
 import sys
-from typing import TYPE_CHECKING, Any, Dict, Iterator, Optional, Pattern, cast
+from typing import TYPE_CHECKING, Any, Dict, Iterable, Iterator, Optional, Pattern, cast
 
 import astroid
 from astroid import nodes
@@ -1719,7 +1719,31 @@ def _create_naming_options():
     return tuple(name_options)
 
 
-class NameChecker(_BasicChecker):
+class NameCheckerHelper:
+    """Class containing functions required by NameChecker and NonAsciiNamesChecker"""
+
+    def _check_name(
+        self, node_type: str, name: str, node: nodes.NodeNG, optional_kwarg: Any = None
+    ):
+        """Only Dummy function will be overwritten by implementing classes
+
+        Note: kwarg arguments will be different in implementing classes
+        """
+        raise NotImplementedError
+
+    def _recursive_check_names(self, args: Iterable[nodes.AssignName]):
+        """check names in a possibly recursive list <arg>"""
+        for arg in args:
+            if isinstance(arg, nodes.AssignName):
+                self._check_name("argument", arg.name, arg)
+            else:
+                # pylint: disable-next=fixme
+                # TODO: Check if we can remove this if branch because of
+                #       the up to date astroid version used
+                self._recursive_check_names(arg.elts)
+
+
+class NameChecker(_BasicChecker, NameCheckerHelper):
     msgs = {
         "C0103": (
             '%s name "%s" doesn\'t conform to %s',
@@ -1737,11 +1761,6 @@ class NameChecker(_BasicChecker):
                 ]
             },
         ),
-        "C0144": (
-            '%s name "%s" contains a non-ASCII unicode character',
-            "non-ascii-name",
-            "Used when the name contains at least one non-ASCII unicode character.",
-        ),
         "W0111": (
             "Name %s will become a keyword in Python %s",
             "assign-to-new-keyword",
@@ -1838,7 +1857,6 @@ def __init__(self, linter):
         self._name_hints = {}
         self._good_names_rgxs_compiled = []
         self._bad_names_rgxs_compiled = []
-        self._non_ascii_rgx_compiled = re.compile("[^\u0000-\u007F]")
 
     def open(self):
         self.linter.stats.reset_bad_names()
@@ -1878,7 +1896,7 @@ def _create_naming_rules(self):
 
         return regexps, hints
 
-    @utils.check_messages("disallowed-name", "invalid-name", "non-ascii-name")
+    @utils.check_messages("disallowed-name", "invalid-name")
     def visit_module(self, node: nodes.Module) -> None:
         self._check_name("module", node.name.split(".")[-1], node)
         self._bad_names = {}
@@ -1904,19 +1922,15 @@ def leave_module(self, _: nodes.Module) -> None:
             for args in warnings:
                 self._raise_name_warning(prevalent_group, *args)
 
-    @utils.check_messages(
-        "disallowed-name", "invalid-name", "assign-to-new-keyword", "non-ascii-name"
-    )
+    @utils.check_messages("disallowed-name", "invalid-name", "assign-to-new-keyword")
     def visit_classdef(self, node: nodes.ClassDef) -> None:
         self._check_assign_to_new_keyword_violation(node.name, node)
         self._check_name("class", node.name, node)
         for attr, anodes in node.instance_attrs.items():
             if not any(node.instance_attr_ancestors(attr)):
                 self._check_name("attr", attr, anodes[0])
 
-    @utils.check_messages(
-        "disallowed-name", "invalid-name", "assign-to-new-keyword", "non-ascii-name"
-    )
+    @utils.check_messages("disallowed-name", "invalid-name", "assign-to-new-keyword")
     def visit_functiondef(self, node: nodes.FunctionDef) -> None:
         # Do not emit any warnings if the method is just an implementation
         # of a base class method.
@@ -1944,14 +1958,12 @@ def visit_functiondef(self, node: nodes.FunctionDef) -> None:
 
     visit_asyncfunctiondef = visit_functiondef
 
-    @utils.check_messages("disallowed-name", "invalid-name", "non-ascii-name")
+    @utils.check_messages("disallowed-name", "invalid-name")
     def visit_global(self, node: nodes.Global) -> None:
         for name in node.names:
             self._check_name("const", name, node)
 
-    @utils.check_messages(
-        "disallowed-name", "invalid-name", "assign-to-new-keyword", "non-ascii-name"
-    )
+    @utils.check_messages("disallowed-name", "invalid-name", "assign-to-new-keyword")
     def visit_assignname(self, node: nodes.AssignName) -> None:
         """check module level assigned names"""
         self._check_assign_to_new_keyword_violation(node.name, node)
@@ -1991,14 +2003,6 @@ def visit_assignname(self, node: nodes.AssignName) -> None:
                 else:
                     self._check_name("class_attribute", node.name, node)
 
-    def _recursive_check_names(self, args):
-        """check names in a possibly recursive list <arg>"""
-        for arg in args:
-            if isinstance(arg, nodes.AssignName):
-                self._check_name("argument", arg.name, arg)
-            else:
-                self._recursive_check_names(arg.elts)
-
     def _find_name_group(self, node_type):
         return self._name_group.get(node_type, node_type)
 
@@ -2039,13 +2043,9 @@ def _name_disallowed_by_regex(self, name: str) -> bool:
             pattern.match(name) for pattern in self._bad_names_rgxs_compiled
         )
 
+    # pylint: disable-next=arguments-renamed
     def _check_name(self, node_type, name, node, confidence=interfaces.HIGH):
         """check for a name using the type's regexp"""
-        non_ascii_match = self._non_ascii_rgx_compiled.match(name)
-        if non_ascii_match is not None:
-            self._raise_name_warning(
-                None, node, node_type, name, confidence, warning="non-ascii-name"
-            )
 
         def _should_exempt_from_invalid_name(node):
             if node_type == "variable":