Repository navigation
fix(gamepad-android): show real device name + type_hint on sokol-Android (#270) - #272
Conversation
The gamepad-android demo's `connected_pads.nameFor` still indexed the fixed `entries` array by the raw device id and returned a generic "Gamepad" for any `id >= MAX_GAMEPADS`. Android `InputDevice` ids are not dense 0..3 and routinely exceed 4 (e.g. 9), so the recorded real name was never surfaced — the HUD showed `Pad 9: Gamepad [...]`. `record`, `forget` and `typeHintFor` were already migrated to a by-id lookup (`findById`) for labelle-engine#261; `nameFor` was the one method left on the old slot-indexing. Switch `nameFor` to `findById` so the stored `InputDevice.getName()` shows for sparse ids, and add a host regression test covering a sparse id (9). Combined with labelle-core deriving the type_hint, the HUD now shows e.g. `Pad 9: Xbox Wireless Controller [xbox]`. Note: the `[xbox]` type_hint half of this fix depends on the matching labelle-core change (typeHintFromName on the Android path); see the linked core PR. On-device verification on the test tablet is pending (handled by the main session). Fixes #270
PR SummaryLow Risk Overview
Adds a host regression test that records id 9, checks Reviewed by Cursor Bugbot for commit 5c4adbe. Bugbot is set up for automated code reviews on this repo. Configure here. |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request fixes a bug in nameFor where sparse Android device IDs exceeding MAX_GAMEPADS caused out-of-bounds indexing or missed entries, by switching to a lookup via findById. It also adds a test case to verify this fix. The reviewer identified a potential out-of-bounds read/panic in nameFor when a gamepad name is exactly NAME_CAP characters long, and suggested declaring the name field in the Entry struct as a sentinel-terminated array to resolve it.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
| const e = findById(id) orelse return "Gamepad"; | ||
| if (e.name_len == 0) return "Gamepad"; |
There was a problem hiding this comment.
There is a potential out-of-bounds read/panic in nameFor if a gamepad name is exactly NAME_CAP (64) characters long.
When e.name_len == NAME_CAP, the expression e.name[0..e.name_len :0] (which is e.name[0..64 :0]) expects the sentinel 0 to be at e.name[64]. Since e.name is a standard array of size NAME_CAP ([64]u8), accessing index 64 is out of bounds and will trigger a runtime panic in safe build modes.
A clean and idiomatic way to fix this is to declare name in the Entry struct (line 27) as a sentinel-terminated array:
name: [NAME_CAP:0]u8 = [_:0]u8{0} ** NAME_CAP,This ensures that e.name always has a sentinel at the end, making e.name[0..e.name_len :0] safe even when e.name_len == NAME_CAP.
record() capped names at NAME_CAP, so name_len could equal NAME_CAP (64) and nameFor's sentinel slice e.name[0..name_len :0] read index 64 of a [64]u8 -> out-of-bounds panic. Cap record at NAME_CAP-1 (reserve the NUL) and clamp in nameFor; add a buffer-filling regression test. (Re-do of a #272-review fix that was lost to a detached-worktree push no-op.)
1 participant
Problem
The gamepad-android HUD showed
Pad 9: Gamepad [unknown]for an Xbox controller, where desktop raylib showedXbox Wireless Controller [xbox](#270). Two independent bugs, one per repo:examples/gamepad-android/scripts/connected_pads.nameForstill indexed the fixedentriesarray by the raw device id and returned a generic"Gamepad"for anyid >= MAX_GAMEPADS. AndroidInputDeviceids are not dense 0..3 and routinely exceed 4 (e.g. 9).record/forget/typeHintForwere already migrated to a by-id lookup (findById) for labelle-engine#261 —nameForwas the one method left on the old slot-indexing, so the recorded real name was never surfaced. This is why desktop (ids 0..3) worked and Android (id 9) didn't.type_hintfrom the name. Fixed in labelle-core#31.Fix (this PR)
Switch
nameFortofindById, matching the other accessors, so the storedInputDevice.getName()shows for sparse ids. Add a host regression test covering a sparse id (9).Tests
connected_pads.zig:record(9, "Xbox Wireless Controller", "xbox")→nameFor(9)/typeHintFor(9)resolve, unknown id falls back to"Gamepad"/"unknown",forgetclears. Passes.zig build-obj -target aarch64-linux-android examples/gamepad-android/scripts/connected_pads.zig: clean.sokol_clibAndroid C-compile step fails only because the local env lacks the Android NDK sysroot (pre-existing, unrelated to this change — no backend Zig/C touched here).Cross-repo dependency
The
[xbox]type_hint half requires labelle-core#31 (derivestype_hintfrom the name on the Android path). Merge core#31 first / repoint the core pin on its release; this PR's name fix is independently correct and testable. On-device verification on the test tablet is pending (main session).Fixes #270. Refs labelle-core#31, #261, #248, #268, #269.