Skip RUF039 when \n, \t, ... are involved

[alexprengere]

The latest Ruff 039 rule is a bit noisy (see #14527 too), and I think there are false positives: running it on a private codebase, I had a lot of:

_RE_STRIP = re.compile("[ \t\r\n]+")
                        ^^^^^^^^^^ RUF039
= help: Replace with raw string

Unless I missed something, there is no easy way to convert that one to a raw string, as r"\n" is semantically different to "\n" (but I agree that is a good general rule, when possible).

[MichaReiser]

Converting to a raw string should be possible for all escape sequence except \b

source

You can also verify this using the repl

>>> re.match("\n", "\n")
<re.Match object; span=(0, 1), match='\n'>
>>> re.match(r"\n", "\n")
<re.Match object; span=(0, 1), match='\n'>
>>> 

[alexprengere]

Thanks for the reply! I guess CPython implemented this to make people's life easier with the raw strings.
It feels a bit weird as it blurs some lines between what is escaped by re and what is escaped when building a string, but I guess as this works, I will close the issue.