james-willis commented on code in PR #1127:
URL: https://github.com/apache/sedona-db/pull/1127#discussion_r3726319276
##########
c/sedona-proj/src/proj_dyn_bindgen.rs:
##########
@@ -81,8 +81,21 @@ pub struct PJ_INFO {
pub release: *const c_char,
pub version: *const c_char,
pub searchpath: *const c_char,
+ pub paths: *const *const c_char,
+ pub path_count: usize,
}
+// `proj_info` returns `PJ_INFO` by value, so this hand-written layout must
match
+// PROJ's ABI exactly: a struct that is too small makes PROJ write past the
+// caller's return buffer, corrupting adjacent stack memory. When the
`proj-sys`
+// feature is available its generated bindings are the source of truth, so pin
+// our size and alignment to them to catch any future drift at compile time.
+#[cfg(feature = "proj-sys")]
+const _: () = {
+ assert!(std::mem::size_of::<PJ_INFO>() ==
std::mem::size_of::<proj_sys::PJ_INFO>());
+ assert!(std::mem::align_of::<PJ_INFO>() ==
std::mem::align_of::<proj_sys::PJ_INFO>());
+};
Review Comment:
not sure if we really need this. LMK if we should dispatch of this.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
