Add unspecified-encoding checker pylint-dev#3826

This adds an unspecified-encoding checker that adds a warning
whenever open() is called without an explicit encoding argument.
This closes pylint-dev#3826

CONTRIBUTORS.txt

@@ -521,3 +521,5 @@ contributors:
 * Yilei Yang: contributor
 
 * Marcin Kurczewski (rr-): contributor
+
+* Daniel van Noord (DanielNoord): contributor

ChangeLog

@@ -6,8 +6,9 @@ What's New in Pylint 2.10.0?
 ============================
 Release date: TBA
 
-..
-  Put new features here and also in 'doc/whatsnew/2.10.rst'
+* Added ``unspecified-encoding``: Emitted when open() is called without specifying an encoding
+
+  Closes #3826
 
 
 What's New in Pylint 2.9.6?

doc/whatsnew/2.10.rst

@@ -12,7 +12,9 @@ Summary -- Release highlights
 New checkers
 ============
 
+* Added ``unspecified-encoding``: Emitted when open() is called without specifying an encoding
 
+  Closes #3826
 
 Other Changes
 =============

pylint/checkers/stdlib.py

@@ -30,6 +30,7 @@
 # Copyright (c) 2021 Marc Mueller <30130371+cdce8p@users.noreply.github.com>
 # Copyright (c) 2021 Matus Valo <matusvalo@users.noreply.github.com>
 # Copyright (c) 2021 victor <16359131+jiajunsu@users.noreply.github.com>
+# Copyright (c) 2021 Daniel van Noord <13665637+DanielNoord@users.noreply.github.com>
 
 # Licensed under the GPL: https://www.gnu.org/licenses/old-licenses/gpl-2.0.html
 # For details: https://github.com/PyCQA/pylint/blob/main/LICENSE
@@ -425,6 +426,13 @@ class StdlibChecker(DeprecatedMixin, BaseChecker):
             "deprecated-decorator",
             "The decorator is marked as deprecated and will be removed in the future.",
         ),
+        "W1514": (
+            "Using open without explicitly specifying an encoding",
+            "unspecified-encoding",
+            "It is better to specify an encoding when opening documents. "
+            "Failing to do so can create problems when opening documents on other systems. "
+            "See https://www.python.org/dev/peps/pep-0597/",
+        ),
     }
 
     def __init__(self, linter=None):
@@ -485,6 +493,7 @@ def _check_shallow_copy_environ(self, node):
         "subprocess-popen-preexec-fn",
         "subprocess-run-check",
         "deprecated-class",
+        "unspecified-encoding",
     )
     def visit_call(self, node):
         """Visit a Call node."""
@@ -496,6 +505,8 @@ def visit_call(self, node):
                 if inferred.root().name == OPEN_MODULE:
                     if getattr(node.func, "name", None) in OPEN_FILES:
                         self._check_open_mode(node)
+                    if getattr(node.func, "name", None) == "open":
+                        self._check_open_encoded(node)
                 elif inferred.root().name == UNITTEST_CASE:
                     self._check_redundant_assert(node, inferred)
                 elif isinstance(inferred, astroid.ClassDef):
@@ -573,6 +584,17 @@ def _check_open_mode(self, node):
             ):
                 self.add_message("bad-open-mode", node=node, args=mode_arg.value)
 
+    def _check_open_encoded(self, node):
+        """Check that the encoded argument of an open call is valid."""
+        try:
+            open_arg = utils.get_argument_from_call(node, keyword="encoding")
+        except utils.NoSuchArgumentError:
+            self.add_message("unspecified-encoding", node=node)
+        if open_arg:
+            open_arg = utils.safe_infer(open_arg)
+            if isinstance(open_arg, astroid.Const) and open_arg.value is None:
+                self.add_message("unspecified-encoding", node=node)
+
     def _check_env_function(self, node, infer):
         env_name_kwarg = "key"
         env_value_kwarg = "default"

tests/functional/u/unspecified-encoding_py3.py

@@ -0,0 +1,7 @@
+"""Warnings for using open() without specifying an encoding"""
+# pylint: disable=consider-using-with
+
+NAME = "foo.bar"
+open(NAME, "wb", encoding="utf-8")
+open(NAME, "w", encoding=None)  # [unspecified-encoding]
+open(NAME, "rb")  # [unspecified-encoding]

tests/functional/u/unspecified-encoding_py3.txt

@@ -0,0 +1,2 @@
+bad-open-mode:6:0::"Using open without explicitly specifying an encoding"
+bad-open-mode:7:0::"Using open without explicitly specifying an encoding"