Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@crates/capi/src/object.rs`:
- Around line 87-89: The code currently uses
CStr::from_ptr(attr_name).to_str().expect(...) which will panic on invalid
UTF-8; instead decode the C string and handle errors by converting the UTF-8
error into a Python exception and returning the appropriate error indicator
(e.g., null pointer or error code) from the surrounding FFI function. Replace
the .expect() on to_str() with a match or map_err that calls the Python C-API to
set an exception (e.g., PyErr_SetString(PyExc_UnicodeDecodeError, ... ) or the
project's helper for raising Python errors), include the invalid data/error
message, and then return early from the function using the function's error
return convention; update uses of the local variable name accordingly after
successful decoding.
- Around line 70-74: The code currently calls
CStr::from_ptr(attr_name).to_str().expect(...), which will panic on invalid
UTF-8; replace the expect with explicit error handling by matching the Result
from to_str(): on Ok(s) assign s to name, on Err(e) set a Python exception (e.g.
PyErr_SetString(PyExc_ValueError, format!("invalid UTF-8 in attribute name: {}",
e).as_ptr()) or using pyo3::exceptions::PyValueError::new_err(...)) and return
the function's error sentinel (NULL for pointer returns or -1 for int returns).
Update the block around CStr::from_ptr(attr_name) / to_str() to use match or
map_err so attr_name decoding never panics and the function returns a
Python-level error instead; reference the variables/functions attr_name, name,
CStr::from_ptr and to_str when making the change.