bears/general: Add RegexLintBear

Closes #1532

Author: bkhanale

bear-languages.yaml

@@ -329,6 +329,7 @@ RadonBear:
   - Python
   - Python 2
   - Python 3
+RegexLintBear:
 RuboCopBear:
   - Ruby
 RubyFastererBear:

bear-requirements.txt

@@ -31,6 +31,7 @@ pylint~=1.7.2
 pyroma~=2.2.0
 pyyaml~=3.12
 radon==1.4.0
+regexlint~=1.6
 restructuredtext-lint~=1.0
 rstcheck~=3.1
 safety~=1.8.2

bear-requirements.yaml

@@ -68,6 +68,8 @@ pip_requirements:
     version: ~=3.12
   radon:
     version: ==1.4.0
+  regexlint:
+    version: ~=1.6
   restructuredtext-lint:
     version: ~=1.0
   rstcheck:

bears/general/RegexLintBear.py

@@ -0,0 +1,55 @@
+import re
+
+from queue import Queue
+from sarge import run, Capture
+from contextlib import suppress
+
+from bears.general.AnnotationBear import AnnotationBear
+
+from coalib.bears.LocalBear import LocalBear
+from coalib.results.Result import Result
+from coalib.settings.Section import Section
+from coalib.settings.Setting import Setting
+from coalib.testing.LocalBearTestHelper import execute_bear
+
+from dependency_management.requirements.PipRequirement import PipRequirement
+
+
+class RegexLintBear(LocalBear):
+    LANGUAGES = {'All'}
+    REQUIREMENTS = {PipRequirement('regexlint', '1.6')}
+    AUTHORS = {'The coala developers'}
+    AUTHORS_EMAILS = {'[email protected]'}
+    LICENSE = 'AGPL-3.0'
+    CAN_DETECT = {'Formatting'}
+
+    def run(self, filename, file, language: str):
+        """
+        Bear for linting regex through regexlint.
+
+        :param language:
+            The programming language of the file(s).
+        """
+        section = Section('')
+        section.append(Setting('language', language))
+        bear = AnnotationBear(section, Queue())
+
+        with execute_bear(bear, filename, file) as result:
+            for src_range in result[0].contents['strings']:
+                src_line = src_range.affected_source({filename: file})[0]
+                regex = src_line[src_range.start.column:src_range.end.column-1]
+                with suppress(re.error):
+                    re.compile(regex)
+                    out = run('regexlint --regex "{}"'.format(regex),
+                              stdout=Capture()).stdout.text
+                    if out[-3:-1] == 'OK':
+                        continue
+                    yield Result.from_values(
+                        origin=self,
+                        message=out,
+                        file=filename,
+                        line=src_range.start.line,
+                        column=src_range.start.column,
+                        end_line=src_range.end.line,
+                        end_column=src_range.end.column,
+                    )

tests/general/RegexLintBearTest.py

@@ -0,0 +1,81 @@
+from os.path import abspath
+from queue import Queue
+
+from bears.general.RegexLintBear import RegexLintBear
+from coalib.results.Result import Result
+from coalib.settings.Section import Section
+from coalib.settings.Setting import Setting
+from coalib.testing.LocalBearTestHelper import (
+    LocalBearTestHelper, execute_bear)
+
+
+test_good_py_file = """
+some_regex = r'[a-zA-Z]]'
+"""
+
+test_bad_py_file = """
+some_regex = r'(else|elseif)'
+"""
+
+test_good_cpp_file = """
+char some_regex[] = "[a-zA-Z]]";
+"""
+
+test_bad_cpp_file = """
+char some_regex[13] = "(else|elseif)";
+"""
+
+test_re_error_file = """
+some_regex = r'*ab' # This should be skipped
+some_other_regex = r'[a-z]'
+"""
+
+BAD_MESSAGE = """
+E105:argv:root:0: Potential out of order alternation between 'else' and 'elseif'
+  '(else|elseif)'
+         ^ here
+""".lstrip()
+
+
+class RegexLintBearTest(LocalBearTestHelper):
+
+    def setUp(self):
+        self.section = Section('')
+        self.queue = Queue()
+        self.uut = RegexLintBear(self.section, self.queue)
+
+    def test_good_python_file(self):
+        self.section.append(Setting('language', 'python 3'))
+        self.check_validity(self.uut, test_good_py_file.splitlines())
+
+    def test_bad_python_file(self):
+        self.section.append(Setting('language', 'python 3'))
+        with execute_bear(self.uut, abspath('bad_python_file'),
+                          test_bad_py_file.splitlines()) as results:
+            self.assertEqual(len(results), 1)
+            self.assertEqual(results[0],
+                             Result.from_values(origin=self.uut,
+                                                message=BAD_MESSAGE,
+                                                file='bad_python_file',
+                                                line=2, column=15,
+                                                end_line=2, end_column=29))
+
+    def test_good_cpp_file(self):
+        self.section.append(Setting('language', 'cpp'))
+        self.check_validity(self.uut, test_good_cpp_file.splitlines())
+
+    def test_bad_cpp_file(self):
+        self.section.append(Setting('language', 'cpp'))
+        with execute_bear(self.uut, abspath('bad_cpp_file'),
+                          test_bad_cpp_file.splitlines()) as results:
+            self.assertEqual(len(results), 1)
+            self.assertEqual(results[0],
+                             Result.from_values(origin=self.uut,
+                                                message=BAD_MESSAGE,
+                                                file='bad_cpp_file',
+                                                line=2, column=23,
+                                                end_line=2, end_column=37))
+
+    def test_re_error_file(self):
+        self.section.append(Setting('language', 'cpp'))
+        self.check_validity(self.uut, test_re_error_file.splitlines())