Maroon Lake's launcher watcher - a process that only spawns the game and waits - held 58 MB, 48 MB of
it MALLOC_SMALL. Measured with malloc stack logging it was not the state defaults but L_grow: every
program sized every property type's per-entity store to MAX_ENT (1024) slots at start, 1,583 stores,
entities or none. And every state record was made with its defaults in L_init_globals before Boot.
- emit_lazy.ludic: a program's (not the runtime's) state global is left out of L_init_globals, and
every read of it calls @S_<global>(), which makes it on first call from its own initializer - after
every registry and plain global, so a default may read them (the init-order crash cannot come back
through a state). What a getter makes is declared (@lp_fdecl): a state first touched in play is made
once and not judged as a frame's keep. A function value's trampoline calls the getter too.
- L_grow starts the stores at ECS_FIRST (8) and doubles as entities come, as it always did past MAX_ENT.
The watcher (with the game's watch step moved before the systems' defs): 57 MB -> 11 MB; the only
state it makes is UiState (14 KB). What is left: AppKit's window, opened by rt_init before main for any
windowed program (~4 MB), and the runtime's font, image and 2D inits (~1.2 MB).
Goldens: the arena, fence, value and json goldens; the 36 ui examples; 30 of the 32 ECS test cases
(sprite_render and sprite_atlas time out under a plain runner with the toolchain before this too).
Package tests: ludic.base, save, settings, i18n.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A state's field default reading a registry (jn_life_sp: []int = jn_life_species_new(), which reads
Species[sp].population) ran before the registry was filled, because globals were initialized in
declaration order: every gate scenario crashed in L_init_globals. Each global's initializer is now
followed - through the functions it calls and a record's field defaults - to the globals it reads,
and those are initialized first (a depth-first post-order; the source order kept between globals
that need nothing of each other, and in a cycle). Putting every state last is not enough: some
tables read a state's instance too. A test (a state whose default reads a registry declared after
it) crashes on d483c92 and prints '2 4' now; Maroon Lake's headless game loads and plays 180 frames.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Mem.scan() and R3D_ALLOC_SCAN=<frame> (at that frame's mark, when no function is running) walk
every heap block reachable from the program's globals - the states among them - conservatively: each
word that is a heap block's start (malloc_size says so) is followed, blocks made before tracking too.
A tracked block nothing reaches is a leak whatever a frame's totals say; they are summed by site and
printed ('alloc-scan: frame 39 - 32 bytes in 2 blocks ... reachable from no global or state', then
the sites). R3D_ALLOC_EXIT=1 runs the same scan as the program quits. A test: two records dropped in
frame 20 are the two found, the one pushed into a state is not.
The escape analysis now takes free(x) as giving x back (ES_FREED, flowing to what reached x) and a
print's argument as used up, so ludic deps --births lists only what is never given back.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every allocation the compiler emits goes through @lp_malloc/@lp_calloc/@lp_realloc/@lp_free, and a
Ludic-level one first stores its site (function, file, line, kind) in @lp_site. Off, that is one load
and a predictable branch (30 M allocations: 0.87-0.91 s against 0.87-0.90 s on leaks2).
On (the default in a headless build, and windowed under R3D_DEV), tracking starts at the first frame
on its own and judging once R3D_ALLOC_WARM frames in a row kept nothing (600) or R3D_ALLOC_WARM_MAX
after (re)start; Mem.play()/Mem.rewarm() sends a load back to its warm-up. A judged frame that ends
holding more than it began with is reported by site with its callers (the unwinder, taken only once
judging) and fails the run with exit 86 (R3D_ALLOC_FENCE=off|count|warn|fail). R3D_ALLOC_CENSUS
writes the totals and top sites at exit. The build's defaults are --fence=, --fence-warm=,
--fence-census= or a fence line in the program's package.ludic; the environment overrides them.
The runtime is IR (emit_fence_ir.ludic, generated from a template); tracking is a side table in one
calloc'd region, so no block carries a header and pointers crossing to natives stay safe. Examples
alloc_fence, alloc_fence_leak and alloc_fence_auto with cases in ludic-dev test; reseeded.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The left half of a + chain, a template's pieces and holes, a number's text and a side made only to be
compared are marked fresh and freed after the +, ==, != or print that reads them. lp_int_str and
lp_long_str move their digits to the start of the buffer, so the pointer they return is the one
malloc gave. Reseeded. examples/lang/string_temps.ludic: kept intermediates stay good, and 20,000
rounds grow the heap 0 bytes (2.9 MB before).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A template literal nested in a {...} hole crashed the parser: the outer
literal ended at the inner backtick. The lexer (and ludic-fmt's) now
reads a hole as code, taking strings, chars and templates in it whole.
A decimal literal past 2147483647, or a hex one of more than eight
digits, was wrapped into a negative int; it is a long with its value
now, and giving one to an int is refused. Reseed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A port is a record of function values a module calls through
(Clock.now()) without naming the module that answers. A port used and
never bound, a bind missing a required member or naming one the port
lacks, and a second bind are refused; the binder must see the port, and
what it binds is checked from its own file. Reseed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
registry NAME of RECORD [as PREFIX] is a global table; def NAME key { ... } in
any file is one entry, collected in source order and filled before any code
runs. Each entry gets an index constant (PREFIX_KEY), the table PREFIX_COUNT,
and a record with a key field gets it filled and a NAME_find(key). A record
literal naming a field its record lacks is now an error everywhere; a global's
initializer is lowered as its own file's code (its errors, and what its module
may see, were whichever statement came last).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The typed buffers are slices: words/floats/fixeds/doubles/pointers(n) make
zeroed, bounds-checked []int/[]float/... and the type names mean them. buffer(n)
is a []byte, with text_of, Fs.read_bytes/write_bytes and view(xs, start, n).
bytes(), indexing a raw pointer or bytes, free, resize, Memory.*, raw file calls,
data_of and C externs are refused outside unsafe { } / unsafe function, and a
project's own files may write unsafe only with --unsafe; the runtime and packages
are the platform. A slice passed to an extern goes as its data.
What the change found: Sync's atomics on a slice header, words(n) uninitialised,
input's fixed axes in ints, truetype's fixed outlines as ints, skin matrices
typed int, gl_shader's source table made from raw bytes. render3d gets safe
entry points (safe_api.ludic). Rendering is byte-identical; a frame costs the same.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Input.text, App.monitor_count / window_to_monitor / window_fixed,
gl_sleep_us, and the grass and collide changes, uncommitted on main and
depended on by the game; carried here so the language work starts from
what the game actually uses. main's working tree is untouched.
Crypto.random_* and Uuid.* read /dev/urandom, which Windows lacks, so every
byte was zero; the Windows target now calls RtlGenRandom (advapi32).
ludicc takes --title, which ludic build/run/bundle pass from package.ludic's
app name, so rt_init opens the window under that name instead of the
program name.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
`float` (32-bit) and `double` (64-bit) with + - * / %, comparisons and unary minus.
Decimal literals take their type from context and stay `fixed` elsewhere; int and long
promote implicitly (LUDIC_WARN_FLOAT_PROMOTE=1 lists every promotion). float(), double(),
int(), long() and fixed() convert; floats(n)/doubles(n) buffers; float fields, globals,
constants and parameters; Math.* computes in float for float arguments; string/print
write the shortest round-tripping decimal; float_bits/float_from_bits expose the bits.
@deterministic code may not use floats. The f_* runtime helpers stay as they are.
Editors know the new type words; the JetBrains plugin is 1.5.0.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The Windows runtime keyed the held set by the character MapVirtualKeyA gave a
virtual key, so a game's WASD belonged to whatever the active layout put there,
and with an input method on every letter arrived as VK_PROCESSKEY. The typing
block is now read from the scancode (the arrows, numpad and F-keys still by
virtual key); macOS reads keyCode the same way. Input.key_label(key) names a key
in the player's own layout (Windows) or as its US character elsewhere.
Verified on Windows with SendInput into a live window: VK_Z carrying W's scancode
holds 'w', VK_PROCESSKEY carrying A's holds 'a', Caps Lock changes nothing.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
--target <triple> (default: the host, from OS=Windows_NT) selects the Windows
runtime. emit_win.ludic defines the POSIX names the backend already calls over
the UCRT and Win32 in IR, so rename replaces an existing file, ftell is 64-bit,
and fopen is binary. Known folders follow %APPDATA% / %LOCALAPPDATA% / %TEMP%,
and the driver speaks cmd.exe, writes .exe outputs and finds LLVM's clang.
Verified: macOS three-stage fixpoint, ludic-dev test 135/135, and on Windows
the Mac-emitted Windows IR and the Windows-built compiler's IR are identical
through two generations.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Three things a game could not say, each of which had been worked around.
Audio.play_at(id, gain:, pitch:, pan:) fires a one-shot with its own gain,
pitch and stereo position. Audio.volume and Audio.pitch are global - they are
the options screen - so a game placing a sound in the world was fighting them,
and distance attenuation was simply not expressible. The backend already took
volume and rate per call; this adds setPan: alongside them and stops routing
through the master state.
ludic.render3d gains outline_model(model, mat, width, r, g, b): a rim around
something that is not an Actor. The outline pass walked the actor list and
stopped, so instanced scatter - a forest - could not be highlighted at all. It
is a queue flushed by the same pass, which is what gets the depth test right
when the caller does not control pass order.
And terrain_coast(cx, cz, margin, fall), the other way to make an island:
the sea around the survey's own edge rather than cut out of the middle of it.
terrain_island measures a radius from a centre, which drowns two thirds of a
real survey to make an island of the rest; this measures inward from the
boundary, so everything the data covers stays land and the coast is where the
data runs out. Both modes gained a strand - the last few metres of height
either side of the water line compressed, which stretches a cliff plunge out
into beach and shallows.
132 regression tests and the self-host fixpoints pass.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
- `[a, b, c]` list literals (E_LIST → emit_list); static_type learns
slice-element, `new T`, list, string and literal kinds
- `x op= y` lowers through the same path as `x = x op y` (emit_bin_vals):
fixed `*=`/`/=` use the Q16.16 64-bit paths, string `+=` concatenates,
int→long widens; unary `-` keeps a fixed operand's type (arith_ty)
- one `unescape()` table for "strings", 'chars' and `interpolation`;
`'\''`, `'\\'`, `'\"'` no longer read as 0; unterminated char literals
and unexpected characters are errors instead of silently skipped
- every diagnostic is `file:line: error: msg` (g_parse_file / g_err_file,
Node.file + Node.line set by node()); tok_desc() in expectation errors;
duplicate `function` names and unknown `phase` names are reported in
source terms (phase_id used to default unknown phases to Overlay)
- interpolation holes skip braces inside string literals
- hand-IR preludes move from the user `@fn_` prefix to `@lp_` so a user
`is_ws` / `str_eq` / `path_join` no longer collides at link time
- `@ClearColor(expr)` accepts any constant expression; `Os.pid()` added
(docs page + inventory); `str_starts()` in support/str
- main.ludic: `else if` flag ladder, char literals, stale script comments
- examples/lang/operators.ludic covers all of the above; os.ludic covers
Os.pid; docs pages for Os.pid and the Overlay phase; ten changesets
- reseeded: selfhost/ludicc.seed.ll is the new compiler's own fixpoint
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Closes the two open issues and lands the pending unreleased batch:
- #90: `Sprite { atlas: 1 }` routes esys_sprite through atlas_draw_ex
(scale/flip/tint), so cell / cell_span / strip ids of any size draw
through the engine sprite-render system. examples/library/sprite_atlas
is the pixel-readback regression.
- #91: `become` from an @On(Event) listener / global handler / plain
function no longer segfaults the compiler; it emits @L_scene_leave()
(a dispatch on the live scene id) so the leaving scene's on-exit runs.
UI_* handles are readable from any code (widget table built on first
use). examples/library/scene_menus covers it.
- fix: a windowed `ludicc -o` build that reaches the audio runtime only
through the atlas/Assets preload import now links audio.ll +
AVFoundation (the audio backend link was gated on a game-level
Audio.* call, so any windowed game declaring Sprite failed to link).
- the hand-written "Unreleased" CHANGELOG section is converted to
changesets under changes/ so `x release` generates it.
- plus the batch: engine-driven retained UI + UiClicked event, Overlay
phase, TileSkin tilemap-render system, Key.* constants, Font/Ui/File
namespaces, Sprite.strip, prefabs, managers, countdown fields,
enum-typed machines, layer @Queries, ludic.prefs / ludic.dungeon
packages, Ai.seek pathing, Solids.solid2, cursor confine (mode 3)
fix, shooter centre-aim fix, reserved-word function diagnostic.
Verified: x test (124/124), x test-tools, check-impl, check-vocabulary,
check-docs, docs-gen + docs-check, bootstrap-cfree (seed is a fixpoint).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
A windowed action game can hide the OS cursor and lock/confine the mouse to the
window. Input.cursor_mode(mode): 0 normal, 1 hidden (draw your own reticle),
2 locked (hidden + dissociated — the mouse feeds relative motion via
Input.mouse_dx/dy and Input.mouse_x/y is a clamped virtual cursor, FPS/twin-stick
aim), 3 confined (dissociated but visible; the mouse can't leave the window).
The platform auto-releases (shows + reconnects) while the window is not key
(Cmd-Tab) and on close, so the cursor is never left captured. Headless it is a
no-op (DCE'd).
Native macOS impl in cocoa.ll: [NSCursor hide]/[unhide] (ref-counted, toggled
only on change so the count stays balanced across focus changes),
CGAssociateMouseAndMouseCursorPosition, and CGGetLastMouseDelta for the relative
virtual cursor, behind a new win_cursor_mode intrinsic. Windowed-only behaviour
(not in the headless golden suite); example compiles headless and links
windowed. Full suite 115/0, fixpoint holds.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The heart of Tiled support: load a map and draw it.
- rt_tmap model (Tmap/TmLayer/TmTileset in tiled.ludic): map header + ordered
layers (dense int32 GID arrays, heap-allocated to w*h, lifting the 96x64 cap)
+ tilesets. Built from the intermediate Value tree, so the TMX and TMJ paths
both feed it.
- GID resolver (Tiled.resolve): gid -> (tileset, localId, flipH/V/D); the three
flip flags masked off before the local-id lookup, returned alongside. gid==0
is empty.
- Image-backed render (Tiled.draw): every visible tile layer in file order,
blitting each tile from its tileset image with flips applied at draw.
- Compatibility projection (Tiled.project / auto on load): a designated
collision layer projects to the legacy byte tilemap ('#' solid, '=' one-way
via the oneway property, ' ' empty) so Grid.*/Path.*/esys_move are unchanged.
- Tiled.load resolves external tilesets + images relative to the map file and
auto-projects a collision/solids/walls layer.
- The grid and physics_tiles demos now run off a loaded map (grid_maze.tmx /
physics_map.tmx) instead of hand-authored Map.row strings, byte-identically.
Two compiler fixes fell out of this (see the changeset):
- emit_index_addr set g_addr_ty before evaluating the index, so slice[obj.field]
came back mis-typed; set it last, like the raw-pointer branches.
- @strcmp was declared by both the world table and the fs prelude; centralise
it in the head prelude so a game that uses Fs/Path links.
Proven by library/tiled_p1.ludic (14 assertions: loads+renders Kenney map
identically from .tmx and .tmj, flip mirroring) + the converted grid/
physics_tiles demos. x test: 92 passed; self-host bootstrap fixpoint intact.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds the Http.* namespace and its transport, the HTTP client from #6. The JSON
companion the proposal called for already shipped as Json.* (#44).
- runtime/native/http.ll: the macOS transport — NSURLConnection driven through
the objc runtime C ABI (no ObjC/C source), run on a detached pthread so the
frame never blocks; TLS is the system's, on by default. A fixed slot pool holds
each in-flight request; the worker publishes status/body/response behind an
atomic done flag (release/acquire). Spliced + Foundation linked only when a
program uses Http.*.
- runtime/native/http.ludic: the Http.* runtime — get/post/request, the
open/set/body/send builder, poll/status/ok/text/body_len/header/free, plus a
pure-Ludic response parser (Http.parse + case-insensitive header lookup) that
is transport-independent and portable.
- compiler: Http.* dispatch, g_uses_http splice, hs_* transport intrinsics +
declarations, the conditional Foundation link, and a new \r string/char escape
the protocol needs.
- docs: a full docs/language/http section (16 pages); check-impl/check-docs green.
- test: examples/library/http.ludic self-asserts the parser offline (Darwin-gated
build via the canonical path, since it links Foundation).
Verified end to end against real endpoints: HTTPS GET (200 + headers + body) and
POST (body + custom header). HTTP is out-of-band and never feeds the
deterministic sim. Reseeded; suites green (81 + 29).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds the Audio.* namespace and its platform backend, the audio subsystem #22 was
blocked on.
- runtime/native/audio.ll: the macOS backend, AVAudioPlayer driven through the
objc runtime C ABI (no ObjC/C source), same style as cocoa.ll — snd_load /
play / stop / playing / set_volume / set_rate. Spliced and linked with
AVFoundation only when a windowed build actually uses Audio.* (needed_framework,
since AVAudioPlayer is reached by name).
- runtime/native/audio.ludic: the Audio.* runtime — a handle table, master
volume/pitch, a single music channel. load/play/play_sound/play_music/stop/
stop_music/stop_all/volume/pitch/is_playing. Every native call is
is_windowed()-guarded, so a headless build carries the API as no-ops (load
returns 0, is_playing false) and needs no audio device.
- compiler: Audio.* namespace dispatch, g_uses_audio splice, snd_* intrinsics +
declarations, and the conditional AVFoundation link in both the canonical
(main.ludic) and dev-runner (x app) paths.
- docs: a full docs/language/audio section (10 method pages); check-impl green.
- test: examples/library/audio.ludic self-asserts the headless no-op path.
Playback is out-of-band and never feeds the deterministic sim, but triggers are
frame-driven so replays fire the same sounds. Reseeded; suites green (80 + 29).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Wire the macOS platform side of the #50 device layer, feeding the same state
buffers the read APIs consume — no API changes, purely OS glue.
- mouse: cocoa.ll reads the live cursor via mouseLocationOutsideOfEventStream,
converted to framebuffer pixels and y-flipped, so windowed games get
Input.mouse_x/y without injection (W_mx/W_my were never written before).
- gamepad: win_pad polls GCController.controllers each frame, packing extended-
gamepad buttons (SDL_GameControllerButton order) and thumbsticks (16.16 fixed,
Y negated for SDL convention) into in_pad_*. Windowed builds now load
GameController via -needed_framework (its classes are reached by name, so a
plain -framework link dead-strips it); DCE'd in headless builds.
- touch: the view's NSTouch phase handlers snapshot the touching set into
in_touch_* (normalizedPosition -> framebuffer pixels).
Web platform.js gains zero-fill stubs for win_held/mouse/pad/touch so a windowed
wasm build resolves the device-layer imports. New test asserts the windowed link
loads GameController. Reseeded; full + selfhost suites green (79 + 29).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The raw device layer the input proposal sketched, over the action maps +
record/replay of #7. Beyond one key per frame, gameplay can read:
- Multiple simultaneous held keys: Input.key_down / key_pressed / key_released,
with clean rising/falling edges (hold left AND jump).
- Analog from keys: Input.axis(neg, pos) and a normalized Input.vector(l,r,u,d)
(diagonals scaled by 1/sqrt(2)), plus Input.strength(action).
- Mouse: Input.mouse_x/y, mouse_dx/dy (per-frame delta), mouse_down(btn), wheel.
- Gamepads: Input.pad_connected/pad_button/pad_axis (SDL-order buttons, -1..1
sticks); touch: Input.touch_count/touch_x/touch_y.
The held set is fed by the platform when windowed — cocoa.ll now tracks
keyDown/keyUp into a 256-bit held-key bitset (win_held) and the mouse
buttons/wheel (win_mouse), gated so headless builds DCE the native calls — and
by the Input.press / Input.set_mouse / Input.set_pad / Input.set_touch injection
on every target (Godot-style action injection: replays, AI, network-fed input).
Input.record / replay now snapshot the full per-frame device state (held set +
mouse), extending #7's single-key tape.
Everything is integer and deterministic, so the same inputs reproduce the same
frame on every run and headless. The gamepad/touch native hardware bindings
(GameController.framework / NSTouch) feed the same injected state and are the one
remaining platform-glue follow-up; the software layer, semantics and replay are
complete and driven deterministically today.
Worked example + regression: examples/library/input_device.ludic
(1 1 0 1 0 1 71 -71 5 1 3 1 2 1 0 0 1, injection-driven headless). 23 new
docs/language/input pages. Full suite 78 passed, self-host C-free fixpoint
intact, no golden drift; cocoa.ll assembles and a windowed build links.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Split the flat 38-file selfhost/ into concern-based subdirectories:
frontend/ lex, parse, parse_game, ast
support/ str, buf, io
backend/ core IR + expression/statement lowering
backend/game/ ECS/scene/event/world lowering
backend/stdlib/ the namespaced Math.*/Text.*/Crypto.*/… intrinsics
and split the three oversized emitters at responsibility boundaries so
no file mixes concerns:
emit_game.ludic -> + emit_world.ludic (reflection world table,
tick helpers, @main synthesis)
emit_expr.ludic -> + emit_call.ludic (namespaced builtins, call
lowering, expr dispatch)
emit_text.ludic -> + emit_text_prelude.ludic (emitted string-builder runtime)
FRAGS in tools/x/selfhost.ludic is updated to the new paths with the link
order preserved, and the Python doc/vocabulary tooling is updated to walk
the new layout. Because the build is a plain in-order concatenation and
every split lands on a blank-line boundary, the regenerated seed is
byte-identical: `x reseed` leaves selfhost/ludicc.seed.ll unchanged,
`x bootstrap-cfree` still reaches its fixed point, and both `x test` (56)
and `x selfhost-test` (29, incl. golden renders) stay green.
Closes#29
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
