hacking: check for common double word typos

Adds a local hacking check for common double word typos
like "the the", "to to", etc.

Change-Id: I824a4a4b0f3bc18a16d5df50bd66cc626888ebaf

HACKING.rst

@@ -53,6 +53,7 @@ Nova Specific Commandments
 - [N340] Check nova.utils.spawn() is used instead of greenthread.spawn() and eventlet.spawn()
 - [N341] contextlib.nested is deprecated
 - [N342] Config options should be in the central location ``nova/conf/``
+- [N343] Check for common double word typos
 
 Creating Unit Tests
 -------------------

nova/cmd/manage.py

@@ -1277,7 +1277,7 @@ class CellV2Commands(object):
         if cell_uuid is None:
             raise Exception(_("cell_uuid must be set"))
         else:
-            # Validate the the cell exists
+            # Validate the cell exists
             cell_mapping = objects.CellMapping.get_by_uuid(ctxt, cell_uuid)
         filters = {}
         instances = objects.InstanceList.get_by_filters(

nova/db/sqlalchemy/api.py

@@ -1913,7 +1913,7 @@ def instance_get_all_by_filters(context, filters, sort_key, sort_dir,
 def instance_get_all_by_filters_sort(context, filters, limit=None, marker=None,
                                      columns_to_join=None, use_slave=False,
                                      sort_keys=None, sort_dirs=None):
-    """Return instances that match all filters sorted the the given keys.
+    """Return instances that match all filters sorted by the given keys.
     Deleted instances will be returned by default, unless there's a filter that
     says otherwise.
 

nova/hacking/checks.py

@@ -101,6 +101,8 @@ http_not_implemented_re = re.compile(r"raise .*HTTPNotImplemented\(")
 spawn_re = re.compile(
     r".*(eventlet|greenthread)\.(?P<spawn_part>spawn(_n)?)\(.*\)")
 contextlib_nested = re.compile(r"^with (contextlib\.)?nested\(")
+doubled_words_re = re.compile(
+    r"\b(then?|[iao]n|i[fst]|but|f?or|at|and|[dt]o)\s+\1\b")
 
 
 class BaseASTChecker(ast.NodeVisitor):
@@ -576,6 +578,19 @@ def check_config_option_in_central_place(logical_line, filename):
         yield(0, msg)
 
 
+def check_doubled_words(physical_line, filename):
+    """Check for the common doubled-word typos
+
+    N343
+    """
+    msg = ("N343: Doubled word '%(word)s' typo found")
+
+    match = re.search(doubled_words_re, physical_line)
+
+    if match:
+        return (0, msg % {'word': match.group(1)})
+
+
 def factory(register):
     register(import_no_db_in_virt)
     register(no_db_session_in_public_api)
@@ -605,3 +620,4 @@ def factory(register):
     register(check_no_contextlib_nested)
     register(check_greenthread_spawns)
     register(check_config_option_in_central_place)
+    register(check_doubled_words)

nova/tests/unit/compute/test_compute_mgr.py

@@ -2748,7 +2748,7 @@ class ComputeManagerUnitTestCase(test.NoDBTestCase):
                     uuids.instance, 'fake-mac',
                     start_period=0, use_slave=True)
             # NOTE(sdague): bw_usage_update happens at some time in
-            # the future, so what last_refreshed is is irrelevant.
+            # the future, so what last_refreshed is irrelevant.
             bw_usage_update.assert_called_once_with(self.context,
                     uuids.instance,
                     'fake-mac', 0, 4, 6, 1, 2,

nova/tests/unit/test_hacking.py

@@ -628,3 +628,14 @@ class HackingTestCase(test.NoDBTestCase):
                                 checks.check_config_option_in_central_place,
                                 filename="nova/cmd/serialproxy.py",
                                 expected_errors=errors)
+
+    def test_check_doubled_words(self):
+        errors = [(1, 0, "N343")]
+
+        # Artificial break to stop pep8 detecting the test !
+        code = "This is the" + " the best comment"
+        self._assert_has_errors(code, checks.check_doubled_words,
+                                expected_errors=errors)
+
+        code = "This is the then best comment"
+        self._assert_has_no_errors(code, checks.check_doubled_words)

nova/virt/driver.py

@@ -1351,7 +1351,7 @@ class ComputeDriver(object):
         :param nova.objects.aggregate.Aggregate aggregate:
             The aggregate which should add the given `host`
         :param str host:
-            The name of the host to add to the the given `aggregate`.
+            The name of the host to add to the given `aggregate`.
         :param dict kwargs:
             A free-form thingy...
 
@@ -1370,7 +1370,7 @@ class ComputeDriver(object):
         :param nova.objects.aggregate.Aggregate aggregate:
             The aggregate which should remove the given `host`
         :param str host:
-            The name of the host to remove from the the given `aggregate`.
+            The name of the host to remove from the given `aggregate`.
         :param dict kwargs:
             A free-form thingy...
 

nova/virt/hardware.py

@@ -1373,7 +1373,7 @@ def get_host_numa_usage_from_instance(host, instance, free=False,
 
     :param host: nova.objects.ComputeNode instance, or a db object or dict
     :param instance: nova.objects.Instance instance, or a db object or dict
-    :param free: if True the the returned topology will have it's usage
+    :param free: if True the returned topology will have it's usage
                  decreased instead.
     :param never_serialize_result: if True result will always be an instance of
                                    objects.NUMATopology class.

nova/virt/libvirt/utils.py

@@ -454,7 +454,7 @@ def get_instance_path(instance, forceold=False, relative=False):
 
 
 def get_instance_path_at_destination(instance, migrate_data=None):
-    """Get the the instance path on destination node while live migration.
+    """Get the instance path on destination node while live migration.
 
     This method determines the directory name for instance storage on
     destination node, while live migration.

nova/virt/vmwareapi/ds_util.py

@@ -40,7 +40,7 @@ DcInfo = collections.namedtuple('DcInfo',
                                 ['ref', 'name', 'vmFolder'])
 
 # A cache for datastore/datacenter mappings. The key will be
-# the datastore moref. The value will the the DcInfo object.
+# the datastore moref. The value will be the DcInfo object.
 _DS_DC_MAPPING = {}
 
 

nova/wsgi.py

@@ -330,7 +330,7 @@ class Application(object):
           res = exc.HTTPForbidden(explanation='Nice try')
 
           # Option 3: a webob Response object (in case you need to play with
-          # headers, or you want to be treated like an iterable, or or or)
+          # headers, or you want to be treated like an iterable, or ...)
           res = Response()
           res.app_iter = open('somefile')