`acknowledge-existing-violations` omits colon after `noqa`, which ignores all errors, not specific errors

[bkeryan]

ni-python-styleguide acknowledge-existing-violations adds noqa comments without a colon, like this one:

    def GetMetadata(  # noqa N802: function name 'GetMetadata' should be lowercase (auto-generated noqa)
        self, request: v1_measurement_service_pb2.GetMetadataRequest, context: grpc.ServicerContext
    ) -> v1_measurement_service_pb2.GetMetadataResponse:

https://www.alpharithms.com/noqa-python-501210/ says

Multiple codes can be given, separated by a comma. The noqa token is case insensitive, the colon before the list of codes is required otherwise the part after noqa is ignored

I noticed this because I added type annotations and ni-python-styleguide fix did not wrap the extremely long line.

Related: ni/measurement-plugin-python#250

[mshafer-NI]

Oh, this is going to make acknowledging multiple on a single line a lot more complicated...

-class Cheese_Shop:  # noqa D101: Missing docstring in public class (auto-generated noqa)  # noqa N801: class name 'Cheese_Shop' should use CapWords convention (auto-generated noqa)
+class Cheese_Shop:  # noqa: D101, N801 Missing docstring in public class (auto-generated noqa)

[mshafer-NI]

Fix released in 0.4.1

[bkeryan]

Unfortunately, acknowledge-existing-violations doesn't leave the code in a state where lint passes.

For example, acknowledge-existing-violations adds this noqa:

    def RegisterService(self, request):  # noqa: N802 - function name 'RegisterService' should be lowercase (auto-generated noqa)

lint says the noqa makes black want to reformat the line:

.\tests\utilities\fake_discovery_service.py:14:25: BLK100 Black would make changes.

Previously, the blanket noqa prevented any reformatting.

fix runs black, rewriting it as:

    def RegisterService(
        self, request
    ):  # noqa: N802 - function name 'RegisterService' should be lowercase (auto-generated noqa)

lint says the noqa is on the wrong line:

.\tests\utilities\fake_discovery_service.py:14:10: N802 function name 'RegisterService' should be lowercase

acknowledge-existing-violations rewrites it as:

    def RegisterService(  # noqa: N802 - function name 'RegisterService' should be lowercase (auto-generated noqa)
        self, request
    ):  # noqa: N802 - function name 'RegisterService' should be lowercase (auto-generated noqa)

Now you have two noqas.

[mshafer-NI]

Unfortunately, acknowledge-existing-violations doesn't leave the code in a state where lint passes.

...

    def RegisterService(  # noqa: N802 - function name 'RegisterService' should be lowercase (auto-generated noqa)
        self, request
    ):  # noqa: N802 - function name 'RegisterService' should be lowercase (auto-generated noqa)

Now you have two noqas.

This is what the fix --aggresssive option addresses. It will iteratively acknowledge issues, run black (if changes were made) remove existing statements and go again.

0.4.1 was released on Friday with a patch on how the comments are generated. Please try fix --aggressive with that version.