Run mypy in Github continuous integration checks (#2324)

* run mypy in CI

* more type hint fixing

* Update requirements.txt

* google.cloud import workaround

python/mypy#10360

* Revert "google.cloud import workaround"

This reverts commit e03dbc7.

* mypy google.cloud workaround

python/mypy#12985

* fix regular tests

* remove --install-types flag

* last mypy fix

* type fixes for address_reasons

* move email lists creation

* update for newly merged changes

* Fix regular tests

* Changes suggested by @jrobbins

* Catch invalid requests

* small type hint

* Change methods to properly override

* revert to previous implementation

* revert test

* add back original validate request type method

* remove unused import and rearrange methods

* remove comment string casting; add test back

* add ndb ignore

* Update test_html_rendering.html

* mypy fix

* remove merge addition

.github/workflows/ci.yml

@@ -41,6 +41,9 @@ jobs:
       - name: lint
         run: npm run lint
 
+      - name: mypy
+        run: mypy --ignore-missing-imports --exclude cs-env/ --exclude appengine_config.py .
+
       - name: test
         run: npm test
 

api/accounts_api.py

@@ -15,7 +15,7 @@
 
 import logging
 
-from google.cloud import ndb
+from google.cloud import ndb  # type: ignore
 
 from framework import basehandlers
 from framework import permissions

api/comments_api.py

@@ -14,7 +14,7 @@
 # limitations under the License.
 
 import logging
-from typing import Any, Optional
+from typing import Any
 
 from framework import basehandlers
 from framework import permissions
@@ -24,7 +24,6 @@
 
 
 def comment_to_json_dict(comment: Activity) -> dict[str, Any]:
-
   return {
       'comment_id': comment.key.id(),
       'feature_id': comment.feature_id,
@@ -56,8 +55,8 @@ def do_get(self, **kwargs) -> dict[str, list[dict[str, Any]]]:
     is_admin = permissions.can_admin_site(user)
 
     # Filter deleted comments the user can't see.
-    comments = filter(
-      lambda c: self._should_show_comment(c, user.email(), is_admin), comments)
+    comments = list(filter(
+      lambda c: self._should_show_comment(c, user.email(), is_admin), comments))
 
     dicts = [comment_to_json_dict(c) for c in comments]
     return {'comments': dicts}
@@ -110,7 +109,7 @@ def do_post(self, **kwargs) -> dict[str, str]:
 
   def do_patch(self, **kwargs) -> dict[str, str]:
     comment_id = self.get_param('commentId', required=True)
-    comment: Optional[Activity] = Activity.get_by_id(comment_id)
+    comment: Activity = Activity.get_by_id(comment_id)
 
     user = self.get_current_user(required=True)
     if not permissions.can_admin_site(user) and (

api/features_api.py

@@ -27,7 +27,7 @@
 class FeaturesAPI(basehandlers.APIHandler):
   """Features are the the main records that we track."""
 
-  def get_one_feature(self, feature_id: int) -> dict:
+  def get_one_feature(self, feature_id: int) -> dict[str, Any]:
     features = core_models.Feature.get_by_ids([feature_id])
     if not features:
       self.abort(404, msg='Feature %r not found' % feature_id)
@@ -64,7 +64,7 @@ def do_search(self) -> dict[str, Any]:
         'features': features_on_page,
         }
 
-  def do_get(self, **kwargs) -> dict:
+  def do_get(self, **kwargs) -> dict[str, Any]:
     """Handle GET requests for a single feature or a search."""
     feature_id = kwargs.get('feature_id', None)
     if feature_id:

framework/basehandlers.py

@@ -18,13 +18,14 @@
 import logging
 import os
 import re
+from typing import Optional
 
 import flask
 import flask.views
 import werkzeug.exceptions
 
 import google.appengine.api
-from google.cloud import ndb
+from google.cloud import ndb  # type: ignore
 
 import settings
 from framework import csp
@@ -268,8 +269,8 @@ def require_signed_in_and_xsrf_token(self):
 
 class FlaskHandler(BaseHandler):
 
-  TEMPLATE_PATH = None  # Subclasses should define this.
-  HTTP_CACHE_TYPE = None  # Subclasses can use 'public' or 'private'
+  TEMPLATE_PATH: Optional[str] = None  # Subclasses should define this.
+  HTTP_CACHE_TYPE: Optional[str] = None  # Subclasses can use 'public' or 'private'
   JSONIFY = False  # Set to True for JSON feeds.
   IS_INTERNAL_HANDLER = False  # Subclasses can skip XSRF check.
 
@@ -362,7 +363,6 @@ def get(self, *args, **kwargs):
       location = self.request.url.replace('www.', '', 1)
       logging.info('Striping www and redirecting to %r', location)
       return self.redirect(location)
-
     handler_data = self.get_template_data(*args, **kwargs)
     users.refresh_user_session()
 

framework/basehandlers_test.py

@@ -528,19 +528,19 @@ def test_get_headers(self):
         actual)
 
   def test_get_template_data__missing(self):
-    """Every subsclass should overide get_template_data()."""
+    """Every subclass should overide get_template_data()."""
     self.handler = basehandlers.FlaskHandler()
     with self.assertRaises(NotImplementedError):
       self.handler.get_template_data()
 
   def test_get_template_path__missing(self):
-    """Subsclasses that don't define TEMPLATE_PATH trigger error."""
+    """Subclasses that don't define TEMPLATE_PATH trigger error."""
     self.handler = basehandlers.FlaskHandler()
     with self.assertRaises(ValueError):
       self.handler.get_template_path({})
 
   def test_get_template_path__specified_in_class(self):
-    """Subsclasses can define TEMPLATE_PATH."""
+    """Subclasses can define TEMPLATE_PATH."""
     actual = self.handler.get_template_path({})
     self.assertEqual('test_template.html', actual)
 
@@ -551,7 +551,7 @@ def test_get_template_path__specalized_by_template_data(self):
     self.assertEqual('special.html', actual)
 
   def test_process_post_data__missing(self):
-    """Subsclasses that don't override process_post_data() give a 405."""
+    """Subclasses that don't override process_post_data() give a 405."""
     self.handler = basehandlers.FlaskHandler()
     with self.assertRaises(werkzeug.exceptions.MethodNotAllowed):
       self.handler.process_post_data()

framework/cloud_tasks_helpers.py

@@ -26,7 +26,7 @@
 if not settings.UNIT_TEST_MODE:
   import grpc  # See requirements.dev.txt.
   from google.api_core import retry
-  from google.cloud import tasks
+  from google.cloud import tasks  # type: ignore
 
 
 

framework/csp.py

@@ -17,7 +17,6 @@
 import copy
 import logging
 import os
-import six
 
 import flask
 
@@ -101,7 +100,7 @@ def get_csp_header_key():
 def build_policy(policy):
   """Builds the CSP policy string from the internal representation."""
   csp_directives = [
-      k + ' ' + ' '.join(v) for k, v in six.iteritems(policy) if v is not None
+      k + ' ' + ' '.join(v) for k, v in policy.items() if v is not None
   ]
   if REPORT_URI:
     csp_directives.append('report-uri %s' % REPORT_URI)

framework/csp_test.py

@@ -52,11 +52,11 @@ def test_get_default_policy__strict(self):
     """We can get the regular strict policy."""
     policy = csp.get_default_policy(nonce='12345')
     self.assertCountEqual(list(csp.DEFAULT_POLICY.keys()), list(policy.keys()))
-    self.assertIn('strict-dynamic', policy['script-src'])
+    self.assertIn('\'strict-dynamic\'', policy['script-src'])
     self.assertIn("'nonce-12345'", policy['script-src'])
 
   @mock.patch('framework.csp.USE_NONCE_ONLY_POLICY', True)
-  def test_get_default_policy__strict(self):
+  def test_get_default_policy__strict_two(self):
     """We can get the even stricter nonce-only policy."""
     policy = csp.get_default_policy(nonce='12345')
     self.assertCountEqual(list(csp.NONCE_ONLY_POLICY.keys()), list(policy.keys()))
@@ -71,7 +71,7 @@ def test_get_csp_header_key__enforced(self):
         csp.get_csp_header_key())
 
   @mock.patch('framework.csp.REPORT_ONLY', True)
-  def test_get_csp_header_key__enforced(self):
+  def test_get_csp_header_key__enforced_two(self):
     """We can get the header used when only reporting violations."""
     self.assertEqual(
         csp.HEADER_KEY_REPORT_ONLY,

framework/secrets.py

@@ -20,7 +20,7 @@
 import string
 import time
 
-from google.cloud import ndb
+from google.cloud import ndb  # type: ignore
 
 
 # For random key generation

internals/approval_defs.py

@@ -35,7 +35,7 @@
     'main/third_party/blink/API_OWNERS?format=TEXT')
 
 ApprovalFieldDef = collections.namedtuple(
-    'ApprovalField',
+    'ApprovalFieldDef',
     'name, description, field_id, rule, approvers')
 
 # Note: This can be requested manually through the UI, but it is not
@@ -193,7 +193,8 @@ def set_vote(
         set_on=now, set_by=set_by_email)
     new_vote.put()
 
-  update_gate_approval_state(gate)
+  if gate:
+    update_gate_approval_state(gate)
 
 def get_gate_by_type(feature_id: int, gate_type: int):
   """Return a single gate based on the feature and gate type."""

internals/core_enums.py

@@ -14,6 +14,7 @@
 
 
 import collections
+from typing import Optional
 
 
 WEBCOMPONENTS = 1
@@ -149,43 +150,43 @@
 GATE_SHIP = 4
 
 # Prototype stage types for every feature type.
-STAGE_TYPES_PROTOTYPE = {
+STAGE_TYPES_PROTOTYPE: dict[int, Optional[int]] = {
     FEATURE_TYPE_INCUBATE_ID: STAGE_BLINK_PROTOTYPE,
     FEATURE_TYPE_EXISTING_ID: STAGE_FAST_PROTOTYPE,
     FEATURE_TYPE_CODE_CHANGE_ID: None,
     FEATURE_TYPE_DEPRECATION_ID: None
   }
 # Dev trial stage types for every feature type.
-STAGE_TYPES_DEV_TRIAL = {
+STAGE_TYPES_DEV_TRIAL: dict[int, Optional[int]] = {
     FEATURE_TYPE_INCUBATE_ID: STAGE_BLINK_DEV_TRIAL,
     FEATURE_TYPE_EXISTING_ID: STAGE_FAST_DEV_TRIAL,
     FEATURE_TYPE_CODE_CHANGE_ID: STAGE_PSA_DEV_TRIAL,
     FEATURE_TYPE_DEPRECATION_ID: STAGE_DEP_DEV_TRIAL
   }
 # Origin trial stage types for every feature type.
-STAGE_TYPES_ORIGIN_TRIAL = {
+STAGE_TYPES_ORIGIN_TRIAL: dict[int, Optional[int]] = {
     FEATURE_TYPE_INCUBATE_ID: STAGE_BLINK_ORIGIN_TRIAL,
     FEATURE_TYPE_EXISTING_ID: STAGE_FAST_ORIGIN_TRIAL,
     FEATURE_TYPE_CODE_CHANGE_ID: None,
     FEATURE_TYPE_DEPRECATION_ID: STAGE_DEP_DEPRECATION_TRIAL
   }
 # Origin trial extension stage types for every feature type.
-STAGE_TYPES_EXTEND_ORIGIN_TRIAL = {
+STAGE_TYPES_EXTEND_ORIGIN_TRIAL: dict[int, Optional[int]] = {
     FEATURE_TYPE_INCUBATE_ID: STAGE_BLINK_EXTEND_ORIGIN_TRIAL,
     FEATURE_TYPE_EXISTING_ID: STAGE_FAST_EXTEND_ORIGIN_TRIAL,
     FEATURE_TYPE_CODE_CHANGE_ID: None,
     FEATURE_TYPE_DEPRECATION_ID: STAGE_DEP_EXTEND_DEPRECATION_TRIAL
   }
 # Shipping stage types for every feature type.
-STAGE_TYPES_SHIPPING = {
+STAGE_TYPES_SHIPPING: dict[int, Optional[int]] = {
     FEATURE_TYPE_INCUBATE_ID: STAGE_BLINK_SHIPPING,
     FEATURE_TYPE_EXISTING_ID: STAGE_FAST_SHIPPING,
     FEATURE_TYPE_CODE_CHANGE_ID: STAGE_PSA_SHIPPING,
     FEATURE_TYPE_DEPRECATION_ID: STAGE_DEP_SHIPPING
   }
 
 # Mapping of original field names to the new stage types the fields live on.
-STAGE_TYPES_BY_FIELD_MAPPING = {
+STAGE_TYPES_BY_FIELD_MAPPING: dict[str, dict[int, Optional[int]]] = {
     'finch_url': STAGE_TYPES_SHIPPING,
     'experiment_goals': STAGE_TYPES_ORIGIN_TRIAL,
     'experiment_risks': STAGE_TYPES_ORIGIN_TRIAL,

internals/core_models.py

@@ -22,7 +22,7 @@
 import re
 from typing import Any, Optional
 
-from google.cloud import ndb
+from google.cloud import ndb  # type: ignore
 
 from framework import rediscache
 from framework import users
@@ -205,7 +205,7 @@ def format_for_template(self, version=2) -> dict[str, Any]:
     self.migrate_views()
     logging.info('In format_for_template for %s',
                  repr(self)[:settings.MAX_LOG_LINE])
-    d = self.to_dict()
+    d: dict[str, Any] = self.to_dict()
     is_released = self.impl_status_chrome in RELEASE_IMPL_STATES
     d['is_released'] = is_released
 
@@ -494,7 +494,8 @@ def filter_unlisted(self, feature_list: list[dict]) -> list[dict]:
 
   @classmethod
   def get_by_ids(
-      self, feature_ids: list[int], update_cache: bool=False) -> list[dict]:
+      self, feature_ids: list[int],
+      update_cache: bool=False) -> list[dict[str, Any]]:
     """Return a list of JSON dicts for the specified features.
 
     Because the cache may rarely have stale data, this should only be
@@ -514,7 +515,7 @@ def get_by_ids(
         result_dict[feature_id] = feature
 
     for future in futures:
-      unformatted_feature = future.get_result()
+      unformatted_feature: Optional[Feature] = future.get_result()
       if unformatted_feature and not unformatted_feature.deleted:
         feature = unformatted_feature.format_for_template()
         feature['updated_display'] = (
@@ -525,7 +526,7 @@ def get_by_ids(
         result_dict[unformatted_feature.key.integer_id()] = feature
 
     result_list = [
-        result_dict.get(feature_id) for feature_id in feature_ids
+        result_dict[feature_id] for feature_id in feature_ids
         if feature_id in result_dict]
     return result_list
 
@@ -632,7 +633,7 @@ def getSortingMilestone(feature):
 
   @classmethod
   def get_in_milestone(self, milestone: int,
-      show_unlisted: bool=False) -> dict[int, list[dict[str, Any]]]:
+      show_unlisted: bool=False) -> dict[str, list[dict[str, Any]]]:
     """Return {reason: [feaure_dict]} with all the reasons a feature can
     be part of a milestone.
 
@@ -648,7 +649,7 @@ def get_in_milestone(self, milestone: int,
     if cached_features_by_type:
       features_by_type = cached_features_by_type
     else:
-      all_features: dict[int, list[Feature]] = {}
+      all_features: dict[str, list[Feature]] = {}
       all_features[IMPLEMENTATION_STATUS[ENABLED_BY_DEFAULT]] = []
       all_features[IMPLEMENTATION_STATUS[DEPRECATED]] = []
       all_features[IMPLEMENTATION_STATUS[REMOVED]] = []
@@ -782,10 +783,11 @@ def get_in_milestone(self, milestone: int,
 
   def crbug_number(self) -> Optional[str]:
     if not self.bug_url:
-      return
+      return None
     m = re.search(r'[\/|?id=]([0-9]+)$', self.bug_url)
     if m:
       return m.group(1)
+    return None
 
   def new_crbug_url(self) -> str:
     url = 'https://bugs.chromium.org/p/chromium/issues/entry'
@@ -1132,7 +1134,7 @@ def __init__(self, *args, **kwargs):
 
   @classmethod
   def get_feature_entry(self, feature_id: int, update_cache: bool=False
-      ) -> FeatureEntry:
+      ) -> Optional[FeatureEntry]:
     KEY = feature_cache_key(FeatureEntry.DEFAULT_CACHE_KEY, feature_id)
     feature = rediscache.get(KEY)
 
@@ -1174,7 +1176,7 @@ def get_by_ids(self, entry_ids: list[int], update_cache: bool=False
     procesing a POST to edit data.  For editing use case, load the
     data from NDB directly.
     """
-    result_dict = {}
+    result_dict: dict[int, int] = {}
     futures = []
 
     for fe_id in entry_ids:
@@ -1192,9 +1194,8 @@ def get_by_ids(self, entry_ids: list[int], update_cache: bool=False
         rediscache.set(store_key, entry)
         result_dict[entry.key.integer_id()] = entry
 
-    result_list = [
-        result_dict.get(fe_id) for fe_id in entry_ids
-        if fe_id in result_dict]
+    result_list = [result_dict[fe_id] for fe_id in entry_ids
+                   if fe_id in result_dict]
     return result_list
 
   # Note: get_in_milestone will be in a new file legacy_queries.py.

internals/data_backup.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import logging
+from typing import Any
 
 from googleapiclient.discovery import build
 from googleapiclient.discovery_cache.base import Cache
@@ -20,7 +21,7 @@
 from framework import basehandlers
 
 class MemoryCache(Cache):
-    _CACHE = {}
+    _CACHE: dict[Any, Any] = {}
 
     def get(self, url):
         return MemoryCache._CACHE.get(url)