Pressing Import ROM on iOS takes the whole app down:
src/import/RomImporter.lua: attempt to call field 'pickFile' (a nil value)
love.system.pickFile is a NATIVE BRIDGE, not part of LOVE. It exists only on
builds that compiled one -- Android, and iOS builds patched by
mobile/ios/patch_love_src.py -- so on a build without it the field is simply
nil. RomImporter:546 routes iOS down the same path as Android
(`mobileOS == "Android" or mobileOS == "iOS"`), and all three mobile pick
sites called the field unguarded.
That is why the reports say "any version": nothing about it is version
specific. Red, Blue and Yellow all reach the same call.
Every one of those call sites already handles a device with no document
picker -- Choose falls back to "No picker available, copy your ROM into:"
plus the save directory, and the mod / save rows have their own notices --
and love.system.createFile at its single call site was already guarded this
way. These three were not, so the fallback that was written for exactly this
case could never be reached.
Route them through one small helper that answers false when the bridge is
absent. A build without a picker now degrades to the copy-into-the-save-folder
flow, which on iOS is a working path: the Files app exposes the app's
Documents folder and GRBootstrap sweeps what lands there into the save dir.
tests/rom_importer_no_picker_test.lua covers Import ROM, Import mod and
Import save with the bridge missing, and asserts the picker is still used
when it is present. Reverting the fix reproduces the reported error exactly.
Reported in #482 (confirmed by three people) and #512.
Introduce a tuned .luacheckrc and scripts/lint.sh so the engine has a
standing static-analysis baseline -- the tool that would have caught both
bugs in the previous commit before they shipped.
The config is high-signal by design: it keeps the categories that catch
real defects (undefined globals/locals, unused values, unreachable code)
and mutes the cosmetic ones the codebase deliberately lives with (a self/dt
an interface requires but a method ignores, documented empty fall-through
branches, long lines). It marks `love` mutable (games assign callbacks onto
it) and teaches it LuaJIT's table.unpack.
.luacheckrc is tracked via a .gitignore exception, matching how .github and
.gitignore opt out of the blanket dotfile ignore.
`luacheck src` now reports 7 benign warnings and 0 errors, down from 185.
Also drop one dead `require` (ItemEffects loaded src.pokemon.Pokemon and
never used it).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q6bFAiQyZ5jDmewsbB4LG9
Both are code paths that never run in a green test today but crash or
misbehave the moment a mod or a link failure exercises them.
1. Music.lua: applyVolume built its `music.volume` hook context from the
private `state` table, but was defined *above* `local state = {...}`, so
those reads bound to the nil global `state`. Any mod registering the
music.volume hook crashed with "attempt to index a nil value (global
'state')" the first time a volume was applied. Forward-declare `state`
above applyVolume. Regression test drives a file-backed song through the
hook and asserts the context resolves.
2. Tournament.lua: `local battle, why = isHost and newHost() or newGuest()`
had two defects. The and/or idiom truncates a call to its first result,
so `why` (the specific failure reason) was always dropped and every link
failure showed the generic "Link battle can't start" instead of e.g.
"same mods on both games". Worse, when a host's newHost() returned nil,
the `or` fell through and wrongly called newGuest() as the host. Split
into an explicit if/else so the reason is preserved and each role calls
its own constructor.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q6bFAiQyZ5jDmewsbB4LG9
Introduce an OPTIONS -> PERFORMANCE setting that scales the port's
optional presentation extras down for weaker hardware, so older/lower-end
devices can run the game smoothly.
The tier governs the three heaviest non-faithful extras -- the 3D TILT,
the GBC FX post-process shader, and survey ZOOM (which renders connected
neighbor maps) -- plus a hard FPS ceiling. It never touches game logic,
which is fixed-step off dt, so every tier plays identically.
- src/core/Performance.lua: tiers (auto/high/balanced/low), a conservative
device auto-detect (ARM handhelds -> low, phones -> balanced, normal
desktops -> high), per-tier caps, and the option-row cycle. Zero
requires, like GameVersion.
- Game:applyOptions clamps the *live* presentation state against the tier
without rewriting stored options, so a lower tier hides the player's
TILT/GBC FX/ZOOM/FPS choices and a higher tier restores them exactly.
- Zoom.offsetRange floors the range at FIT when survey is disallowed, so
the option row, hotkey, and mouse wheel all stop at close-up on LOW.
- New save.options.performance default "auto"; OPTIONS row heads the
display group and re-applies live.
- Tests: tests/engine/performance_tiers.lua (ROM-free); mod_ui_tests row
golden updated for the spliced row.
AUTO resolves to HIGH on a normal desktop and on every options.lua that
predates the option, so the common case is unchanged.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q6bFAiQyZ5jDmewsbB4LG9
Battle commands toggled directly between rows and columns. An outward press therefore moved the cursor to the opposite command instead of leaving it at the edge.
The selected badge list stayed on the stack while its description opened. Cancelling the next list only closed that newer menu, leaving the NPC conversation open.
On compact Android viewports the item picker can end up with no list
space. Passing its negative height to setScissor crashes LÖVE.
Use an empty clip region instead.
Follow-up to #452, from dogfooding the merged build on a real iPhone.
A mod fused into game.love sits in the read-only app bundle: the mod
manager's Delete can't remove it and it reappears every launch, and an
imported newer version can never fully replace it. iOS now packs no
mods, matching every other platform - mods install as .zips at runtime
with a real install/upgrade/delete lifecycle.
With nothing fusing it, mods/pokewalker moves out of this tree to its
canonical home, https://github.com/mresnick67/Gen1ReComp-Pokewalker
(already at 0.2.0 there with move learning and a paged credit dialog;
in-tree 0.1.0 was stale). docs/ios-install.md points at it. The
event:save.loaded coverage-debt entry returns to gate_meta_coverage:
its coverage came from the mod's test suite and leaves with it.
Local reproduction of the ci workflow on this branch: all tiers, mod
lint, fingerprint + meta-coverage gates green.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
bryanthaboi
hernan0078
Jake Eaker
Claude
johnjohto
Myles Resnick