empty_comment errors reduces reported score by too much

[orSolocate]

Steps to reproduce

1. open empty_comment,py unit-test data file path in terminal at: tests/extensions/data/empty_comment.py

2. run Pylint with the empty-comment extension Add checker for empty comments #3870:
   pylint --load-plugins=pylint.extensions.empty_comment empty_comment.py

3. Watch Pylint output in the console.

Current behavior

************* Module empty_comment
empty_comment.py:2:0: R2044: Line with empty comment (empty-comment)
empty_comment.py:3:0: R2044: Line with empty comment (empty-comment)
empty_comment.py:5:0: R2044: Line with empty comment (empty-comment)
empty_comment.py:7:0: R2044: Line with empty comment (empty-comment)

---

Your code has been rated at 3.33/10

--> Score is very low!

Expected behavior

Since this error is of type Refactor, you would expect it would not reduce the score by that much.

pylint --version output

pylint 2.8.0-dev1
astroid 2.5
Python 3.7.5 (tags/v3.7.5:5c02a39a0b, Oct 15 2019, 00:11:34) [MSC v.1916 64 bit (AMD64)]

[orSolocate]

I wrote the checker so I could fix it. Just let me know what priority value to put. right now the priority level is -1 which is very high. should -5 be good enough?

[orSolocate]

@Pierre-Sassoulas you agree with the change?

[Pierre-Sassoulas]

Yes, but I think this is something that should be fixed with #2399. I don't think there is a specific checker to do ? Also what does the priority have to do with the score ? Isn't it the message type that has an influence ? (Error being *5 right now and all the other * 1)

[orSolocate]

@Pierre-Sassoulas Yeah you are right the priority is for the linter to figure out how to order the checkers in a 'run queue'. doesn't affect the score. The message type R seems to be right for me for this checker. Might be related to the score calculation bug you mentioned.
You could close this issue.

[Pierre-Sassoulas]

Thanks for your input, I'm closing in favor of #2399