[GH-ISSUE #3202] NSEC3 validation cleanup #1153

New issue

Open

opened 2026-03-16 01:44:38 +03:00 by kerem · 1 comment

kerem commented

2026-03-16 01:44:38 +03:00

Owner

Originally created by @divergentdave on GitHub (Aug 13, 2025).
Original GitHub issue: https://github.com/hickory-dns/hickory-dns/issues/3202

Gracefully handle cases where a wildcard name would exceed 255 bytes. I think that for closest encloser names that are 254 or 255 bytes, we can probably assume that a wildcard does not exist. The alternate way to approach this would be to avoid the Name type when computing hashes of wildcard names to sidestep this length restriction.
We can eliminate the unreachable branch in closest_encloser_proof() by re-ordering the Some((0, _)) and Some((i, _)) if i > 0 branches.
We could remove uses of swap_remove() and make some variables immutable by hoisting the call to build_encloser_candidates_list() out of closest_encloser_proofs(), passing a reference to the list in, and only using references for these candidates.

Originally created by @divergentdave on GitHub (Aug 13, 2025). Original GitHub issue: https://github.com/hickory-dns/hickory-dns/issues/3202 - [ ] Gracefully handle cases where a wildcard name would exceed 255 bytes. I think that for closest encloser names that are 254 or 255 bytes, we can probably assume that a wildcard does not exist. The alternate way to approach this would be to avoid the `Name` type when computing hashes of wildcard names to sidestep this length restriction. - [x] We can eliminate the unreachable branch in `closest_encloser_proof()` by re-ordering the `Some((0, _))` and `Some((i, _)) if i > 0` branches. - [ ] We could remove uses of `swap_remove()` and make some variables immutable by hoisting the call to `build_encloser_candidates_list()` out of `closest_encloser_proofs()`, passing a reference to the list in, and only using references for these candidates.

kerem commented

2026-03-16 01:44:44 +03:00

Author

Owner

@djc commented on GitHub (Aug 16, 2025):

We can eliminate the unreachable branch in closest_encloser_proof() by re-ordering the Some((0, _)) and Some((i, _)) if i > 0 branches.

Took care of this in

#3208

@djc commented on GitHub (Aug 16, 2025): > * We can eliminate the unreachable branch in `closest_encloser_proof()` by re-ordering the `Some((0, _))` and `Some((i, _)) if i > 0` branches. Took care of this in - #3208