Skip to content

Preserve surrogate code points in registry values #9597

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
youknowone merged 2 commits into from
Nov 4, 2026
Merged

Preserve surrogate code points in registry values #9597

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b206783
Preserve surrogate code points in registry values
youknowdot Nov 3, 2026
1e75d00
Make registry value conversion infallible
youknowdot Nov 3, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 13 additions & 16 deletions crates/vm/src/stdlib/winreg.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -519,7 +519,7 @@ mod winreg {

// Slice the data buffer to the actual size returned.
let data_slice = &ret_data_buf[..current_data_size as usize];
let py_data = reg_to_py(vm, data_slice, reg_type)?;
let py_data = reg_to_py(vm, data_slice, reg_type);

// Return tuple (value_name, data, type)
return Ok(vm
Expand Down Expand Up @@ -646,7 +646,7 @@ mod winreg {
let wide_name = WideCString::from_str(name).map_err(|e| e.to_pyexception(vm))?;
let (ret_buf, typ) = host_winreg::query_value_bytes(hkey, &wide_name)
.map_err(|err| os_error_from_windows_code(vm, err as i32))?;
let obj = reg_to_py(vm, &ret_buf, typ)?;
let obj = reg_to_py(vm, &ret_buf, typ);
// Return tuple (value, type)
Ok(vm.ctx.new_tuple(vec![obj, vm.ctx.new_int(typ).into()]))
}
Expand Down Expand Up @@ -696,22 +696,22 @@ mod winreg {
}
}

fn reg_to_py(vm: &VirtualMachine, ret_data: &[u8], typ: u32) -> PyResult {
fn reg_to_py(vm: &VirtualMachine, ret_data: &[u8], typ: u32) -> PyObjectRef {
match typ {
REG_DWORD => {
// If there isn’t enough data, return 0.
let val = ret_data
.first_chunk::<4>()
.copied()
.map_or(0, u32::from_ne_bytes);
Ok(vm.ctx.new_int(val).into())
vm.ctx.new_int(val).into()
}
REG_QWORD => {
let val = ret_data
.first_chunk::<8>()
.copied()
.map_or(0, u64::from_ne_bytes);
Ok(vm.ctx.new_int(val).into())
vm.ctx.new_int(val).into()
}
REG_SZ | REG_EXPAND_SZ => {
let u16_slice = host_winreg::bytes_as_wide_slice(ret_data);
Expand All @@ -720,13 +720,12 @@ mod winreg {
.iter()
.position(|&c| c == 0)
.unwrap_or(u16_slice.len());
let s = String::from_utf16(&u16_slice[..len])
.map_err(|e| vm.new_value_error(format!("UTF16 error: {e}")))?;
Ok(vm.ctx.new_str(s).into())
let s = Wtf8Buf::from_wide(&u16_slice[..len]);
vm.ctx.new_str(s).into()
}
REG_MULTI_SZ => {
if ret_data.is_empty() {
Ok(vm.ctx.new_list(vec![]).into())
vm.ctx.new_list(vec![]).into()
} else {
let u16_slice = host_winreg::bytes_as_wide_slice(ret_data);
let u16_count = u16_slice.len();
Expand All @@ -742,27 +741,25 @@ mod winreg {
let mut start = 0;
for i in 0..len {
if u16_slice[i] == 0 {
let s = String::from_utf16(&u16_slice[start..i])
.map_err(|e| vm.new_value_error(format!("UTF16 error: {e}")))?;
let s = Wtf8Buf::from_wide(&u16_slice[start..i]);
strings.push(vm.ctx.new_str(s).into());
start = i + 1;
}
}
// Handle last string if not null-terminated
if start < len {
let s = String::from_utf16(&u16_slice[start..len])
.map_err(|e| vm.new_value_error(format!("UTF16 error: {e}")))?;
let s = Wtf8Buf::from_wide(&u16_slice[start..len]);
strings.push(vm.ctx.new_str(s).into());
}
Ok(vm.ctx.new_list(strings).into())
vm.ctx.new_list(strings).into()
}
}
// For REG_BINARY and any other unknown types, return a bytes object if data exists.
_ => {
if ret_data.is_empty() {
Ok(vm.ctx.none())
vm.ctx.none()
} else {
Ok(vm.ctx.new_bytes(ret_data.to_vec()).into())
vm.ctx.new_bytes(ret_data.to_vec()).into()
}
}
}
Expand Down
Loading