Fix line length (partial #5358)

.flake8/azul_flake8.py

@@ -253,13 +253,13 @@ def check_statement_order(self, node: EitherImport) -> None:
             if module_name not in self.expected_resolution_failures:
                 self._error(node, ImportErrors.unresolvable)
         else:
-            # The order in which NodeVisitor traverses the syntax tree is unspecified
-            # so we can't be sure which nodes have already been visited.
-            # To prevent a single out-of-order import from causing errors on every
-            # other line, we only compare each import statement with one other statement.
-            # To prevent the same line from being reported twice, the error is always
-            # reported for the node we're currently visiting, regardless of whether it
-            # comes first or second.
+            # The order in which NodeVisitor traverses the syntax tree is
+            # unspecified so we can't be sure which nodes have already been
+            # visited. To prevent a single out-of-order import from causing
+            # errors on every other line, we only compare each import statement
+            # with one other statement. To prevent the same line from being
+            # reported twice, the error is always reported for the node we're
+            # currently visiting, regardless of whether it comes first or second.
             pred = self._visited_predecessor(node)
             succ = self._visited_successor(node)
             if pred is not None and not self._is_correct_order(pred, ordered_import):
@@ -339,14 +339,19 @@ def _visited_predecessor(self, node: EitherImport) -> Optional[OrderedImport]:
         Scan the list of previously visited nodes for the node with the highest
         line number that is less than the provided node's line number.
         """
-        return max(filter(lambda t: t.node.lineno < node.lineno, self.visited_order_info),
-                   key=lambda t: t.node.lineno,
-                   default=None)
+        return max(
+            filter(lambda t: t.node.lineno < node.lineno, self.visited_order_info),
+            key=lambda t: t.node.lineno,
+            default=None
+        )
 
     def _filtered_tokens(self, linenno):
-        return [tokeninfo
-                for tokeninfo in self.line_tokens[linenno]  # 1-based indexing for source code lines
-                if tokeninfo.type not in (tokenize.COMMENT, tokenize.NEWLINE, tokenize.NL, tokenize.ENDMARKER)]
+        return [
+            tokeninfo
+            # 1-based indexing for source code lines
+            for tokeninfo in self.line_tokens[linenno]
+            if tokeninfo.type not in (tokenize.COMMENT, tokenize.NEWLINE, tokenize.NL, tokenize.ENDMARKER)
+        ]
 
 
 class AzulImports:

docs/hca_file_downloader.py

@@ -255,7 +255,8 @@ def download_file(self, url: str, output_path: str) -> None:
         """
         Download a file from the given URL.
         """
-        url = url.replace('/fetch', '')  # Work around https://github.com/DataBiosphere/azul/issues/2908
+        # Work around https://github.com/DataBiosphere/azul/issues/2908
+        url = url.replace('/fetch', '')
         response = requests.get(url, stream=True)
         response.raise_for_status()
         try:

scripts/cgm_adapter.py

@@ -440,7 +440,11 @@ def process_project(self, project: Mapping[str, Any]) -> bool:
         self.rows_completed.extend(rows_processed)
         return True
 
-    def get_blob(self, project_uuid: str, shortname: str, file_name: str) -> gcs.Blob:
+    def get_blob(self,
+                 project_uuid: str,
+                 shortname: str,
+                 file_name: str
+                 ) -> gcs.Blob:
         """
         Return the blob from the source bucket.
         """

scripts/envhook.py

@@ -219,7 +219,10 @@ def share_aws_cli_credential_cache(self):
 NV = TypeVar('NV')
 
 
-def zip_dict(old: Mapping[K, OV], new: Mapping[K, NV], missing=None) -> dict[K, tuple[OV, NV]]:
+def zip_dict(old: Mapping[K, OV],
+             new: Mapping[K, NV],
+             missing=None
+             ) -> dict[K, tuple[OV, NV]]:
     """
     Merge two dictionaries. The resulting dictionary contains an entry for every
     key in either `old` or `new`. Each entry in the result associates a key to

scripts/manage_requirements.py

@@ -72,7 +72,8 @@ def create(cls, req: Requirement) -> Optional['PinnedRequirement']:
             assert op == '=='
             return cls(name=req.name.lower(), versions=Versions(version))
         elif req.vcs:
-            reject(req.revision is None, 'VCS requirements must carry a specific revision', req)
+            reject(req.revision is None,
+                   'VCS requirements must carry a specific revision', req)
             return cls(name=req.name.lower())
         elif req.recursive:
             return None
@@ -191,7 +192,8 @@ def run(self):
         direct_runtime_reqs = self.get_direct_reqs(self.runtime)
         direct_build_reqs = self.get_direct_reqs(self.build)
         dupes = direct_build_reqs & direct_runtime_reqs
-        require(not dupes, 'Some requirements are declared as both run and build time', dupes)
+        require(not dupes,
+                'Some requirements are declared as both run and build time', dupes)
 
         all_reqs = self.get_reqs(self.build)
         build_reqs = all_reqs - pip_reqs
@@ -265,7 +267,10 @@ def get_direct_reqs(self, qualifier: Qualifier) -> PinnedRequirements:
         with open(path) as f:
             return self.parse_reqs(f)
 
-    def write_transitive_reqs(self, reqs: PinnedRequirements, qualifier: Qualifier) -> None:
+    def write_transitive_reqs(self,
+                              reqs: PinnedRequirements,
+                              qualifier: Qualifier
+                              ) -> None:
         self.write_reqs(reqs,
                         file_name=f'requirements{qualifier.extension}.trans.txt',
                         type='transitive')

scripts/manifest_404.py

@@ -74,7 +74,11 @@ def filter_projects():
     url = config.service_endpoint.set(path='/index/projects')
     resp = requests.get(url=str(url), params=dict(size=1000))
     projects = resp.json()['termFacets']['project']['terms']
-    return [one(project['projectId']) for project in projects if project['term'] != 'Tabula Muris']
+    return [
+        one(project['projectId'])
+        for project in projects
+        if project['term'] != 'Tabula Muris'
+    ]
 
 
 project_ids = filter_projects()

scripts/provision_credentials.py

@@ -71,7 +71,8 @@ def provision_google(self, build, email, secret_name):
 
     @classmethod
     def _random_hmac_key(cls):
-        # Even though an HMAC key can be any sequence of bytes, we restrict to base64 in order to encode as string
+        # Even though an HMAC key can be any sequence of bytes, we restrict to
+        # base64 in order to encode as string
         key = base64.encodebytes(os.urandom(48)).decode().replace('=', '').replace('\n', '')
         assert len(key) == 64
         return json.dumps({'key': key, 'key_id': str(uuid.uuid4())})
@@ -106,7 +107,8 @@ def _create_service_account_creds(self, service_account_email):
         key = iam.projects().serviceAccounts().keys().create(
             name='projects/-/serviceAccounts/' + service_account_email, body={}
         ).execute()
-        logger.info("Successfully created service account key for user '%s'", service_account_email)
+        logger.info("Successfully created service account key for user '%s'",
+                    service_account_email)
         return parse_google_key(key)
 
     def _destroy_aws_secrets_manager_secret(self, secret_name):
@@ -116,7 +118,8 @@ def _destroy_aws_secrets_manager_secret(self, secret_name):
                 ForceDeleteWithoutRecovery=True
             )
         except self.secrets_manager.exceptions.ResourceNotFoundException:
-            logger.info('AWS secret %s does not exist. No changes will be made.', secret_name)
+            logger.info('AWS secret %s does not exist. No changes will be made.',
+                        secret_name)
         else:
             assert response['Name'] == secret_name
             # AWS docs recommend waiting for ResourceNotFoundException:
@@ -144,7 +147,8 @@ def _destroy_service_account_creds(self, service_account_email, secret_name):
                 SecretId=config.secrets_manager_secret_name(secret_name)
             )
         except self.secrets_manager.exceptions.ResourceNotFoundException:
-            logger.info('Secret already deleted, cannot get key_id for %s', service_account_email)
+            logger.info('Secret already deleted, cannot get key_id for %s',
+                        service_account_email)
             return
         else:
             key_id = json.loads(creds['SecretString'])['private_key_id']

scripts/rename_resources.py

@@ -56,7 +56,8 @@ def main(argv: list[str]):
             else:
                 if new_name is None:
                     if args.dry_run:
-                        log.info('Found %r, would be removing it from the Terraform state', current_name)
+                        log.info('Found %r, would be removing it from the Terraform state',
+                                 current_name)
                     else:
                         log.info('Found %r, removing it from the Terraform state', current_name)
                         terraform.run('state', 'rm', current_name)

scripts/scrape_aws_service_model.py

@@ -23,13 +23,20 @@
 
 program_name, _ = os.path.splitext(os.path.basename(__file__))
 
-output_file_path = os.path.join(config.project_root, 'terraform', 'gitlab', 'aws_service_model.json.gz')
+output_file_path = os.path.join(config.project_root,
+                                'terraform',
+                                'gitlab',
+                                'aws_service_model.json.gz')
 
 
 @lru_cache(maxsize=1000)
 def get(url_path):
     assert url_path.startswith('/')
-    cache_file_path = os.path.join(config.project_root, '.cache', program_name, url_path[1:], 'cache.json')
+    cache_file_path = os.path.join(config.project_root,
+                                   '.cache',
+                                   program_name,
+                                   url_path[1:],
+                                   'cache.json')
     if os.path.exists(cache_file_path):
         with open(cache_file_path, 'r', encoding='utf-8') as f:
             return json.load(f)
@@ -41,7 +48,10 @@ def get(url_path):
         response_json = response.json()
         cache_dir_path = os.path.dirname(cache_file_path)
         os.makedirs(cache_dir_path, exist_ok=True)
-        f = tempfile.NamedTemporaryFile(mode='w+', dir=cache_dir_path, encoding='utf-8', delete=False)
+        f = tempfile.NamedTemporaryFile(mode='w+',
+                                        dir=cache_dir_path,
+                                        encoding='utf-8',
+                                        delete=False)
         try:
             json.dump(response_json, f, indent=4)
         except BaseException:

src/azul/__init__.py

@@ -342,7 +342,10 @@ def dss_deployment_stage(self, dss_endpoint: str) -> str:
     def dss_direct_access(self) -> bool:
         return self._boolean(self.environ['AZUL_DSS_DIRECT_ACCESS'])
 
-    def dss_direct_access_role(self, lambda_name: str, stage: Optional[str] = None) -> Optional[str]:
+    def dss_direct_access_role(self,
+                               lambda_name: str,
+                               stage: Optional[str] = None
+                               ) -> Optional[str]:
         key = 'AZUL_DSS_DIRECT_ACCESS_ROLE'
         try:
             role_arn = self.environ[key]
@@ -909,7 +912,10 @@ def is_hca_enabled(self, catalog: Optional[str] = None) -> bool:
     def is_anvil_enabled(self, catalog: Optional[str] = None) -> bool:
         return self._is_plugin_enabled('anvil', catalog)
 
-    def _is_plugin_enabled(self, plugin_prefix: str, catalog: Optional[str]) -> bool:
+    def _is_plugin_enabled(self,
+                           plugin_prefix: str,
+                           catalog: Optional[str]
+                           ) -> bool:
         def predicate(catalog):
             return any(
                 plugin.name.split('_')[0] == plugin_prefix
@@ -929,7 +935,11 @@ def integration_test_catalogs(self) -> Mapping[CatalogName, Catalog]:
             if catalog.is_integration_test_catalog
         }
 
-    def es_index_name(self, catalog: CatalogName, entity_type: str, aggregate: bool) -> str:
+    def es_index_name(self,
+                      catalog: CatalogName,
+                      entity_type: str,
+                      aggregate: bool
+                      ) -> str:
         return str(IndexName(prefix=self._index_prefix,
                              version=2,
                              deployment=self.deployment_stage,
@@ -1032,7 +1042,8 @@ class BrowserSite(TypedDict):
         real_path: str
 
     @property
-    def browser_sites(self) -> Mapping[str, Mapping[str, Mapping[str, BrowserSite]]]:
+    def browser_sites(self
+                      ) -> Mapping[str, Mapping[str, Mapping[str, BrowserSite]]]:
         import json
         return json.loads(self.environ['azul_browser_sites'])
 
@@ -1148,9 +1159,10 @@ def aggregation_lambda_timeout(self, *, retry: bool) -> int:
 
     api_gateway_timeout = 29
 
-    # The number of seconds to extend the timeout of a Lambda fronted by API Gateway so that API Gateway times out
-    # before the Lambda. We pad the Lambda timeout so we get consistent behaviour. Without this padding we'd have a
-    # race between the Lambda being killed and API Gateway timing out.
+    # The number of seconds to extend the timeout of a Lambda fronted by
+    # API Gateway so that API Gateway times out before the Lambda. We pad the
+    # Lambda timeout so we get consistent behaviour. Without this padding we'd
+    # have a race between the Lambda being killed and API Gateway timing out.
     #
     api_gateway_timeout_padding = 2
 

src/azul/azulclient.py

@@ -131,15 +131,20 @@ def reindex(self, catalog: CatalogName, prefix: str) -> int:
         self.index(catalog, notifications)
         return len(notifications)
 
-    def index(self, catalog: CatalogName, notifications: Iterable[JSON], delete: bool = False):
+    def index(self,
+              catalog: CatalogName,
+              notifications: Iterable[JSON],
+              delete: bool = False
+              ):
         errors = defaultdict(int)
         missing = []
         indexed = 0
         total = 0
         path = (catalog, 'delete' if delete else 'add')
         indexer_url = config.indexer_endpoint.set(path=path)
 
-        with ThreadPoolExecutor(max_workers=self.num_workers, thread_name_prefix='pool') as tpe:
+        with ThreadPoolExecutor(max_workers=self.num_workers,
+                                thread_name_prefix='pool') as tpe:
 
             def attempt(notification, i):
                 log_args = (indexer_url, notification, i)
@@ -181,7 +186,11 @@ def handle_future(future):
             for future in futures:
                 handle_future(future)
 
-        printer = PrettyPrinter(stream=None, indent=1, width=80, depth=None, compact=False)
+        printer = PrettyPrinter(stream=None,
+                                indent=1,
+                                width=80,
+                                depth=None,
+                                compact=False)
         logger.info("Sent notifications for %i of %i bundles for catalog %r.",
                     indexed, total, catalog)
         if errors:

src/azul/chalice.py

@@ -463,7 +463,8 @@ def _controller(self, controller_cls: Type[C], **kwargs) -> C:
         return controller_cls(app=self, **kwargs)
 
     def swagger_ui(self) -> Response:
-        swagger_ui_template = self.load_static_resource('swagger', 'swagger-ui.html.template.mustache')
+        swagger_ui_template = self.load_static_resource('swagger',
+                                                        'swagger-ui.html.template.mustache')
         base_url = self.base_url
         redirect_url = furl(base_url).add(path='oauth2_redirect')
         deployment_url = furl(base_url).add(path='openapi')

src/azul/changelog.py

@@ -40,12 +40,14 @@ def title_first(item):
 
 def write_changes(output_dir_path):
     """
-    Write the change log as a Python literal to a module in the given directory. We're using Python syntax because it
-    can be looked up and loaded very easily. See changes().
+    Write the change log as a Python literal to a module in the given directory.
+    We're using Python syntax because it can be looked up and loaded very
+    easily. See changes().
     """
     with write_file_atomically(os.path.join(output_dir_path, module_name + '.py')) as f:
-        # Write each change as a single line. I tried pprint() but it reorders the keys in dictionaries and its line
-        # wrapping algorithm is creating a non-uniform output.
+        # Write each change as a single line. I tried pprint() but it reorders
+        # the keys in dictionaries and its line wrapping algorithm is creating a
+        # non-uniform output.
         f.write(variable_name + ' = [\n')
         for change in changelog()[variable_name]:
             f.write('    ' + repr(change) + ',\n')

src/azul/collections.py

@@ -49,7 +49,8 @@ def dict_merge(dicts: Iterable[Mapping]) -> Mapping:
 V = TypeVar('V')
 
 
-def explode_dict(d: Mapping[K, Union[V, list[V], set[V], tuple[V]]]) -> Iterable[dict[K, V]]:
+def explode_dict(d: Mapping[K, Union[V, list[V], set[V], tuple[V]]]
+                 ) -> Iterable[dict[K, V]]:
     """
     An iterable of dictionaries, one dictionary for every possible combination
     of items from iterable values in the argument dictionary. Only instances of

src/azul/compliance/fedramp_inventory_service.py

@@ -123,7 +123,11 @@ class Mapper(metaclass=ABCMeta):
     def map(self, resource: ResourceConfig) -> Iterable[InventoryRow]:
         raise NotImplementedError
 
-    def _common_fields(self, resource: ResourceConfig, *, id_suffix: Optional[str] = None) -> dict:
+    def _common_fields(self,
+                       resource: ResourceConfig,
+                       *,
+                       id_suffix: Optional[str] = None
+                       ) -> dict:
         return dict(
             asset_tag=resource.name,
             location=resource.region,
@@ -496,7 +500,9 @@ def get_resources(self) -> Iterator[ResourceConfig]:
                     yield from map(ResourceConfig.from_response, items)
                     resource_keys = response['unprocessedResourceKeys']
 
-    def get_inventory(self, resources: Iterable[ResourceConfig]) -> Iterable[InventoryRow]:
+    def get_inventory(self,
+                      resources: Iterable[ResourceConfig]
+                      ) -> Iterable[InventoryRow]:
         rows_by_mapper: defaultdict[Mapper, list[InventoryRow]] = defaultdict(list)
         resource_counts = Counter()
         row_counts = Counter()

src/azul/deployment.py

@@ -261,7 +261,11 @@ def dss_checkout_bucket(self, dss_endpoint: str) -> str:
         return self._dss_bucket(dss_endpoint, 'checkout', lambda_name='service')
 
     @_cache
-    def _dss_bucket(self, dss_endpoint: str, *qualifiers: str, lambda_name: str) -> str:
+    def _dss_bucket(self,
+                    dss_endpoint: str,
+                    *qualifiers: str,
+                    lambda_name: str
+                    ) -> str:
         with self.direct_access_credentials(dss_endpoint, lambda_name):
             stage = config.dss_deployment_stage(dss_endpoint)
             name = f'/dcp/dss/{stage}/environment'