f-string position is miscomputed in multiple cases

[addisoncrump]

The position of errors is miscomputed in f-strings:

$ ruff --no-cache test2.py --show-source
error: Failed to parse test2.py:1:5: f-string: single '}' is not allowed
test2.py:1:5: E999 SyntaxError: f-string: single '}' is not allowed
  |
1 | f' }'
  |     ^ E999
  |

Found 1 error.

Moreover, it is not sound in the presence of carriage return + newline:

$ $ cat -e test3.py 
f'''^M$
^M$
^M$
^M$
^M$
 }'''^M$
$ ruff test3.py --fix --show-source --no-cache
error: Failed to parse test3.py:4:2: f-string: single '}' is not allowed
test3.py:4:2: E999 SyntaxError: f-string: single '}' is not allowed
  |
4 |   
5 | | 
  | |_^ E999
6 |    }'''
  |

Found 1 error.

In rare cases, this may lead to a panic if multi-byte UTF-8 characters are present (carriage return + newline followed by UTF-8 char before mismatched }):

$ ruff --no-cache minimized-from-8e22bc8c0f8f652f62a9a858e232b0bbf18e702b
warning: Linting panicked minimized-from-8e22bc8c0f8f652f62a9a858e232b0bbf18e702b: This indicates a bug in `ruff`. If you could open an issue at:

https://github.com/astral-sh/ruff/issues/new?title=%5BLinter%20panic%5D

with the relevant file contents, the `pyproject.toml` settings, and the following stack trace, we'd be very appreciative!

panicked at 'byte index 10 is not a char boundary; it is inside '〈' (bytes 8..11) of `f'''

〈}'''
`'

Locator::after slices a source code snippet at the byte level. This may lead to an invalid UTF-8 slice if the index is in the middle of a code point, which occurs during f-string computation. This occurs in multiple f-string error types.

I am still searching for a root cause.

[charliermarsh]

Any idea if this is still true?

[MichaReiser]

I believe this is fixed. At least the offsets in the playground look correct https://play.ruff.rs/957af831-fd05-416e-8cff-e23cab5cca2a

[addisoncrump]

I don't have an example testcase presently, but if I come across one I'll reopen this.