MetOffice#20 LabelledExit unit test and remove plural

source/stylist/fortran.py

@@ -203,7 +203,7 @@ def examine_fortran(self, subject: FortranSource) -> List[Issue]:
         return issues
 
 
-class LabelledExits(FortranRule):
+class LabelledExit(FortranRule):
     """
     Catches cases where a construct is exited but not explicitly named.
     """

system-tests/fortran/configuration.ini

@@ -16,8 +16,8 @@ rules = MissingPointerInit
 [style.wrong_kind]
 rules = KindPattern(integer=r'.*_jazz', real=r'.*_metal')
 
-[style.labelled_exits]
-rules = LabelledExits
+[style.labelled_exit]
+rules = LabelledExit
 
 [style.multiple]
 rules = FortranCharacterset, MissingImplicit(require_everywhere=True), MissingPointerInit

system-tests/fortran/expected.labelled_exit.txt

@@ -0,0 +1,5 @@
+stdout
+Found 2 issues
+stderr
+$$/labelled_exit.f90: 4: Usage of "exit" without label indicating which do construct is being exited from.
+$$/labelled_exit.f90: 10: Usage of "exit" without label indicating which do construct is being exited from.

unit-tests/fortran_labelled_exit_test.py

@@ -0,0 +1,78 @@
+#!/usr/bin/env python
+##############################################################################
+# (c) Crown copyright 2018 Met Office. All rights reserved.
+# The file LICENCE, distributed with this code, contains details of the terms
+# under which the code may be used.
+##############################################################################
+"""
+Test of the rule for missing exit labels.
+"""
+from typing import List
+
+import pytest  # type: ignore
+# ToDo: Obviously we shouldn't be importing "private" modules but until pytest
+#       sorts out its type hinting we are stuck with it.
+#
+from _pytest.fixtures import FixtureRequest  # type: ignore
+
+import stylist.fortran
+from stylist.source import FortranSource, SourceStringReader
+
+
+class TestLabelledExit(object):
+    """
+    Tests the checker of missing exit labels.
+    """
+    @pytest.fixture(scope='class',
+                    params=['', 'foo'])
+    def do_construct_name(self, request: FixtureRequest) -> str:
+        """
+        Parameter fixture giving do construct names to exit statement
+        """
+        return request.param
+
+    def test_exit_labels(self,
+                          do_construct_name: str) -> None:
+        """
+        Checks that the rule reports missing exit labels correctly.
+        """
+        template = '''
+program test
+contains
+    function function1()
+        foo : do
+            exit {do_construct_name}
+        end do foo
+    end function function1
+    
+    function function2()
+        foo : do
+            if (.true.) exit {do_construct_name}
+        end do foo
+    end function function2
+end program test
+'''
+
+        expectation: List[str] = []
+        message = '{line}: Usage of "exit" without label indicating which do construct is being exited from.'
+        if do_construct_name == '':
+            expectation.extend([
+                message.format(line=6),
+                message.format(line=12)
+            ])
+        expectation.sort(key=lambda x: (int(x.split(':', 1)[0]),
+                                        x.split(':', 1)))
+
+        text = template.format(
+            do_construct_name=do_construct_name)
+        print(text)  # Shows up in failure reports, for debugging
+        reader = SourceStringReader(text)
+        source = FortranSource(reader)
+        unit_under_test = stylist.fortran.LabelledExit()
+        issues = unit_under_test.examine(source)
+        issue_descriptions = [str(issue) for issue in issues]
+        issue_descriptions.sort(key=lambda x: (int(x.split(':', 1)[0]),
+                                               x.split(':', 1)))
+        print(issue_descriptions)
+        print(expectation)
+        assert issue_descriptions == expectation