Issue #1081: non necessary raw strings

CHANGELOG.md

@@ -9,6 +9,14 @@ Semantic versioning in our case means:
 - Major releases inidicate significant milestones or serious breaking changes.
 
 
+## 0.15.0
+
+### Features
+
+- Forbids to use raw strings `r''` when not necessary
+
+### Bug fixes
+
 ## 0.14.0 aka The Walrus fighter WIP
 
 ### Features

tests/fixtures/noqa/noqa.py

@@ -347,7 +347,7 @@ def function(  # noqa: WPS320
                }  # noqa: WPS318
 
 
-string_modifier = R'(s)'  # noqa: WPS321
+string_modifier = R'(\n)'  # noqa: WPS321
 multiline_string = """abc"""  # noqa: WPS322
 
 
@@ -692,3 +692,5 @@ def consecutive_yields():
     'wrong',
 ]
 *numbers, = [4, 7]  # noqa: WPS356
+
+unnecessary_raw_string = r'this string does not contain any backslashes.' # noqa: WPS358

tests/test_checker/test_noqa.py

@@ -164,6 +164,7 @@
     'WPS355': 1,
     'WPS356': 1,
     'WPS357': 0,  # logically unacceptable.
+    'WPS358': 1,
 
     'WPS400': 0,  # defined in ignored violations.
     'WPS401': 0,  # logically unacceptable.

tests/test_visitors/test_tokenize/test_primitives/test_string_tokens/test_string_modifier.py

@@ -3,6 +3,7 @@
 import pytest
 
 from wemake_python_styleguide.violations.consistency import (
+    RawStringNotNeededViolation,
     UppercaseStringModifierViolation,
 )
 from wemake_python_styleguide.visitors.tokenize.primitives import (
@@ -12,10 +13,10 @@
 
 @pytest.mark.parametrize('modifier', [
     'r',
-    'b',
     'rb',
-    'f',
     'fr',
+    'b',
+    'f',
     '',  # special case, no modifier is used
 ])
 @pytest.mark.parametrize('primitive', [
@@ -40,7 +41,11 @@ def test_correct_prefix(
     visitor = WrongStringTokenVisitor(default_options, file_tokens=file_tokens)
     visitor.run()
 
-    assert_errors(visitor, [])
+    assert_errors(
+        visitor,
+        [],
+        ignored_types=(RawStringNotNeededViolation,),
+    )
 
 
 @pytest.mark.parametrize('modifier', [
@@ -71,8 +76,12 @@ def test_uppercase_prefix(
     visitor = WrongStringTokenVisitor(default_options, file_tokens=file_tokens)
     visitor.run()
 
-    assert_errors(visitor, [UppercaseStringModifierViolation])
-    assert_error_text(visitor, modifier)
+    assert_errors(
+        visitor,
+        [UppercaseStringModifierViolation],
+        ignored_types=(RawStringNotNeededViolation,),
+    )
+    assert_error_text(visitor, modifier, multiple=True)
 
 
 @pytest.mark.parametrize('modifier', [
@@ -106,4 +115,5 @@ def test_uppercase_prefix_multiple(
     assert_errors(visitor, [
         UppercaseStringModifierViolation,
         UppercaseStringModifierViolation,
+        RawStringNotNeededViolation,
     ])

tests/test_visitors/test_tokenize/test_primitives/test_string_tokens/test_string_multiline.py

@@ -3,6 +3,7 @@
 import pytest
 
 from wemake_python_styleguide.violations.consistency import (
+    RawStringNotNeededViolation,
     WrongMultilineStringViolation,
 )
 from wemake_python_styleguide.visitors.tokenize.primitives import (
@@ -76,7 +77,11 @@ def test_incorrect_multiline_strings(
     visitor = WrongStringTokenVisitor(default_options, file_tokens=file_tokens)
     visitor.run()
 
-    assert_errors(visitor, [WrongMultilineStringViolation])
+    assert_errors(
+        visitor,
+        [WrongMultilineStringViolation],
+        ignored_types=(RawStringNotNeededViolation,),
+    )
 
 
 @pytest.mark.parametrize('primitive', [

tests/test_visitors/test_tokenize/test_primitives/test_string_tokens/test_unicode_prefix.py

@@ -38,7 +38,7 @@ def test_unicode_prefix(
 
 @pytest.mark.parametrize('primitive', [
     '"name"',
-    'r"text"',
+    'r"text with escape carac \n"',
     "b'unicode'",
     '"u"',
     '"12"',

tests/test_visitors/test_tokenize/test_primitives/test_string_tokens/test_unnecessary_raw_strings.py

@@ -0,0 +1,57 @@
+# -*- coding: utf-8 -*-
+
+import pytest
+
+from wemake_python_styleguide.violations.consistency import (
+    RawStringNotNeededViolation,
+)
+from wemake_python_styleguide.visitors.tokenize.primitives import (
+    WrongStringTokenVisitor,
+)
+
+
+@pytest.mark.parametrize('raw_strings', [
+    r"r'some text\\'",
+    r"r'some text\''",
+    r"r'some text\"'",
+    r'r"some text\'"',
+    r"r'some text\t'",
+    r"r'some text\a'",
+    r"r'some text\n'",
+    r"r'some text\u041b'",
+    r"r'some text\043'",
+    r"r'some text\x23'",
+])
+def test_necessary_raw_string(
+    parse_tokens,
+    assert_errors,
+    default_options,
+    raw_strings,
+):
+    """Ensures that correct usage of raw string works."""
+    file_tokens = parse_tokens(raw_strings)
+
+    visitor = WrongStringTokenVisitor(default_options, file_tokens=file_tokens)
+    visitor.run()
+
+    assert_errors(visitor, [])
+
+
+@pytest.mark.parametrize('raw_strings', [
+    "r'No escaped caracter'",
+    "r'Here neither'",
+    "r'''Not here as well'''",
+])
+def test_unnecessary_raw_string(
+    parse_tokens,
+    assert_errors,
+    default_options,
+    raw_strings,
+):
+    """Ensures that usage of raw string is forbidden if no backslash."""
+    file_tokens = parse_tokens(raw_strings)
+
+    visitor = WrongStringTokenVisitor(default_options, file_tokens=file_tokens)
+    visitor.run()
+
+    assert_errors(visitor, [RawStringNotNeededViolation])

wemake_python_styleguide/constants.py

@@ -357,7 +357,7 @@
 STDIN: Final = 'stdin'
 
 # Used as a special name patterns for unused variables, like _, __:
-UNUSED_VARIABLE_REGEX: Final = re.compile(r'^_+$')
+UNUSED_VARIABLE_REGEX: Final = re.compile('^_+$')
 
 # Used to specify as a placeholder for `__init__`:
 INIT: Final = '__init__'

wemake_python_styleguide/violations/consistency.py

@@ -83,6 +83,7 @@
    BracketBlankLineViolation
    IterableUnpackingViolation
    LineCompriseCarriageReturnViolation
+   RawStringNotNeededViolation
 
 Consistency checks
 ------------------
@@ -145,6 +146,7 @@
 .. autoclass:: BracketBlankLineViolation
 .. autoclass:: IterableUnpackingViolation
 .. autoclass:: LineCompriseCarriageReturnViolation
+.. autoclass:: RawStringNotNeededViolation
 
 """
 
@@ -2143,3 +2145,29 @@ class LineCompriseCarriageReturnViolation(TokenizeViolation):
 
     error_template = r'Found a ``\r`` (carriage return) line break'
     code = 357
+
+
+@final
+class RawStringNotNeededViolation(TokenizeViolation):
+    r"""
+    Forbid the use of raw strings when there is no backslash in the string.
+
+    Reasoning:
+        Raw string are only needed when dealing with ``\`` in the string.
+
+    Solution:
+        Do not prefix the string with ``r``. Use a normal string instead.
+
+    Example::
+        # Correct:
+        r'This is a correct use \n'
+
+        # Wrong:
+        r'This string should not be prefixed with r.'
+
+    .. versionadded:: 0.15.0
+
+    """
+
+    error_template = 'Error template here'
+    code = 358

wemake_python_styleguide/visitors/tokenize/primitives.py

@@ -165,12 +165,14 @@ def visit_string(self, token: tokenize.TokenInfo) -> None:
             WrongMultilineStringViolation
             ImplicitRawStringViolation
             WrongUnicodeEscapeViolation
+            RawStringNotNeededViolation
 
         """
         self._check_correct_multiline(token)
         self._check_string_modifiers(token)
         self._check_implicit_raw_string(token)
         self._check_wrong_unicode_escape(token)
+        self._check_unnecessary_raw_string(token)
 
     def _check_correct_multiline(self, token: tokenize.TokenInfo) -> None:
         _, string_def = split_prefixes(token)
@@ -231,6 +233,14 @@ def _check_wrong_unicode_escape(self, token: tokenize.TokenInfo) -> None:
             # character can never be the start of a new backslash escape.
             index += 2
 
+    def _check_unnecessary_raw_string(self, token: tokenize.TokenInfo) -> None:
+        modifiers, string_def = split_prefixes(token)
+
+        if 'r' in modifiers.lower() and '\\' not in string_def:
+            self.add_violation(
+                consistency.RawStringNotNeededViolation(token),
+            )
+
 
 @final
 class WrongStringConcatenationVisitor(BaseTokenVisitor):