workshopsoft/ludic
2
0
Fork 0
Code Issues Pull requests Projects Releases 26 Packages Wiki Activity Actions
ludic/changes
History
Orkuncakilkaya e4265f5fdb feat(lang): 27.1 - a state machine as data: @Machine(Record.field) on a registry of transitions 
A registry marked `@Machine(Deer.mood)` is the transitions of a machine over that enum field of the
records a state's Table<Deer> holds. Its record has from and to (the enum's variants), on: string (an
action's name, "" for a transition the tick asks), guard: fn(Row<Deer>, reads...) -> bool and
enter: fn(Row<Deer>, reads...) -> void; the states are the enum's variants and the start is the
field's default. The rows are data (an .lres or defs), the names the studio already edits.

Written by the compiler (machines.ludic, machines_write.ludic): for each action an `on` names, a row
reducer in the registry's file (named ..__machine__DeerSteps, so it sits beside the program's own
row reducer on the same action, after it): the row's state, the first transition from it on that
action whose guard passes, the field set, enter run - guards and enters called by name. When a row
leaves a state on a guard alone, `state DeerStepsMachine` (the kept row view) and
deer_steps_tick(m: mut DeerStepsMachine, s: mut Herd, reads...), one transition a row a tick.
Nothing allocates.

The table is the whole machine: the field written anywhere else - an assignment, or a `machine`
block's become over it - is a type error (check_stmt.ludic, ck_machine_write). Guards and enters take
the row first, are the record's module's, keep a row reducer's rules (and may be handed the row);
a guard writes nothing through it. The graph is checked, each error at its row (in the .lres when
the rows are there): a state never reached from the start, a state with no way out, an `on` naming
no action or an action with no @Target, a self-transition with no guard, two ways out of a state on
one trigger behind an unguarded first. Also refused: @Machine off a registry, a field that is not a
plain enum with a default, a @Column field, no table (or two) of the record, a transitions record of
another shape, a machine outside its table's state's module.

ludic schema's code section gains `machines` (registry, record, field, enum, table, start, states,
actions, tick, module, at); ludic deps names a machine's reducer `reducer Deer in Herd.deer on Spook
(machine DeerSteps)`. vocab @Machine; docs annot-machine, kw-machine; LANGUAGE.md "A machine as
data"; examples actions/machine (+ deer_steps.lres) and ten rejects; test.ludic feat, reject and
schema cases (not run); changes/machines.md. Reseeded; bootstrap-cfree fixpoint holds (317642
lines); Maroon Lake's `ludic build --check` is clean against this tree.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-30 16:37:37 +03:00
..
a-hidden-body-still-stands-in-the-sun.md fix(render3d): a body hidden from the camera still stands in the sun 2026-09-19 21:32:11 +03:00
a-leaf-is-not-matte.md fix(render3d): a leaf is not matte 2026-09-19 16:41:33 +03:00
a-lens-and-an-edge.md feat(render3d): anti-aliasing that exists, and a lens for the viewfinder 2026-09-19 17:38:25 +03:00
actions.md feat(lang): 0.R1 - actions and reducers 2026-09-25 19:13:46 +03:00
an-aspen-quakes.md feat(render3d): an aspen quakes, and a tree is no longer bent like a bow 2026-09-19 21:13:42 +03:00
anim-ozz.md feat(ludic.anim): ozz-animation 0.17.0 underneath - a skin's skeleton and each clip built at load, sampled to 1e-4 of anim_mix (phase 19.1) 2026-09-28 12:44:49 +03:00
asset-map.md @Asset(kind, map): a path under each map's directory - ludicc --check looks for it in every map (a @PerMap row's in its own, a game-wide row's in all), unless @Asset(kind, map, optional); the schema marks it scope map; reseeded 2026-09-29 22:13:14 +03:00
attrs-before-export.md parser: an attribute before export is kept - @ToClients export event E was a local event (the attributes read in front of export were dropped when the declaration was parsed afresh one call down), as were @Sync / @Owned / @Server / @On / @Public ... before export; and export @ToClients event was refused. Attributes and export now read in either order into one declaration, and a wrapper marks what it added exported (g_at_keep goes); examples/networking/net_export and three schema cases (added, not run); reseeded, bootstrap-cfree fixpoint holds 2026-09-30 01:03:44 +03:00
bake-expand.md ludic.base: bake_expand(inputs, map) and bake_maps(first_input) beside bake_inputs_hash - {map} put, globs expanded in whole-path byte order with dot-names out - so a bake's runner and the check expand a row one way; tests/baked_expand_test 2026-09-29 22:22:05 +03:00
bake-images.md render3d: an impostor's bake is BC7 with its mips - both atlases uploaded compressed, as baked 2026-09-29 19:18:48 +03:00
baked-textures.md render3d: baked textures before the PNG - a png_decode takes assets/baked/png, a cut-out load assets/baked/cutouts 2026-09-29 17:30:57 +03:00
bind-variable.md feat(lang): bind a port member that takes nothing to a variable 2026-09-25 13:02:01 +03:00
build-ir-in-temp.md fix(cli): build's IR goes to the run's temp directory, not the project 2026-09-23 17:29:55 +03:00
builtin-names.md checker: a function named like a built-in a call always takes is refused at its declaration - function words(st, k) compiled and every call became words(n) with a pointer for n (invalid IR, far from the cause); the table is selfhost/check/check_builtins.ludic, emit_call's built-ins no find_fn guard lets a declared function take, and ludic-dev syntax --check holds it to emit_call both ways; every other built-in (buffer, floats, double, ...) yields to a declared function in the checker as it already did in codegen. string_temps' keep and cross_heap_test's keep renamed (hand, keep_cell); rejected/builtin_call_name and functions/builtin_yield with their cases (added, not run); reseeded, bootstrap-cfree fixpoint holds 2026-09-30 01:10:24 +03:00
check-build.md feat(cli): ludic build --check / ludicc --check - check without building 2026-09-25 13:13:42 +03:00
check-stdin-file.md ludicc / ludic build: --stdin-file <path> checks an unsaved buffer in place of its file (R10) 2026-09-30 00:10:04 +03:00
chunked-keys.md chunked tables: a key is unique across its map and the slot's own, not interned - a slot keeps row i's key in its own buffer i, rewritten when the slot is refilled (interning ~92k map-unique keys as a player walks would fill the bounded intern table and keep them all); ludicc --check refuses a key written in two of a map's chunk files, naming both; reseeded 2026-09-29 21:43:20 +03:00
contact-darkening.md feat(render3d): the lake answers to a body standing in it 2026-09-19 20:23:27 +03:00
declared-namespaces.md feat(lang): L6 namespaces declared in Ludic - alias 2026-09-24 02:40:57 +03:00
def-from-resource.md feat(lang): def REGISTRY from "file.lres" - a game fills an open registry from its own file 2026-09-25 07:22:05 +03:00
defaults-views-templates.md feat(ui): ui_scale and ui_box, on-submit, zoom (and a length over a length in calc), on-hold 2026-09-25 03:31:38 +03:00
deps-alias-writes.md feat(cli): ludic deps --writes warns about writes through a local alias 2026-09-25 13:22:37 +03:00
deps-reach.md feat(cli): 0.R4 - ludic deps sees through fn values (widest_reach) and lists the widest functions (--widest N, --reach N) 2026-09-26 06:24:46 +03:00
deps-write-reach.md feat(cli): 0.R5 - ludic deps says what a function can come to change (widest_write_reach, --wreach N); a port member and a registry field reach only themselves 2026-09-26 23:19:54 +03:00
devlink-ui.md ludic.devlink: the interface's verbs (ui_screen, ui_model, ui_tree, ui_override) through a DevlinkUi port 2026-09-29 22:48:34 +03:00
devlink.md ludic.devlink: an editor's live link into a running dev build (protocol v1, stage 1) 2026-09-29 19:09:59 +03:00
dilate-parallel.md render3d: cut-out padding on every core - tex_dilate's rows per pass through Job.parallel_for, bytes unchanged 2026-09-29 17:11:21 +03:00
dispatch-kept-records.md changes: a dispatched action's record kept by the queue 2026-09-28 02:09:35 +03:00
duplicate-definitions.md feat(lang): a name is defined once, a local once per block, and a result is always returned 2026-09-23 23:55:20 +03:00
ecs-growable-stores.md feat(compiler): the built-in ECS's stores grow - 100 000 entities, where 1024 was the wall 2026-09-27 22:53:24 +03:00
editor-schema.md ludic schema: a components section and a natives list (R4) 2026-09-29 19:36:04 +03:00
engine-alias-shadow.md fix(lang): a function named like an engine namespace method's target is refused 2026-09-25 05:14:45 +03:00
expect-eq-strings.md fix(test): expect_eq on strings compares their text and prints both on a failure 2026-09-25 19:23:38 +03:00
expect-floats.md fix(test): expect_eq and expect_near compare floats as floats 2026-09-25 05:06:15 +03:00
fmt-editor-output.md ludic fmt for editors: --lint --json, and a buffer on stdin (R9) 2026-09-30 00:24:43 +03:00
friend-of.md feat(lang): friend module lab of fishing, data - a friend of the modules it names 2026-09-25 04:12:47 +03:00
function-values.md feat(lang): functions are values (L2), and render3d takes its scene as callbacks 2026-09-24 00:27:26 +03:00
gb-window-fill.md render3d: the blades' density window filled apart from its upload - gb_window_fill (the tiles round the camera's, zeros off the map, the corner) and gb_frame sends it; the same bytes, now readable by a test without a GPU 2026-09-30 02:03:43 +03:00
generics.md feat(lang): L5 generic records and functions 2026-09-24 02:21:15 +03:00
ground-painted-zones.md render3d: a README note on painted ground layers, and the changeset 2026-09-30 15:24:23 +03:00
http-text-copy.md Http.text and Http.header return copies - they handed back the handle's own buffer (and on macOS the response's string), which Http.free released: a text read before the free and used after it was garbage or empty (maroon-lake's maps.json was written with 0 bytes) 2026-09-29 20:57:44 +03:00
i18n-error-mode.md i18n 26.9, error mode: English left is an error under a lang line (a template's words, a text attribute's, a choice that reads as words, a @Text row's English), ludic deps still counting english_left; hole counts and undescribed splits stay warnings. ludic.ui's own words are keys - ui_tk(ui_st, k"ui.right_click", plain) for the key field's Right / Middle / Left click and press a key..., its plain text for a program with no translator; the examples' en.po / tr.po carry ui.*, i18n_ui checks clean, rejected/i18n_keys/english_left holds the error, data_missing counts 2; LANGUAGE.md and the changeset; on lang/foundations 2288feeb; reseeded, bootstrap-cfree fixpoint holds 2026-09-30 02:45:15 +03:00
i18n-fill-nested.md compiler: text keys below a row, trf padding and tr's cast (ECS's 26.4 gaps) - a @Text Key in a record nested in a row, or in each item of a list of them, is filled with its derived key <registry>.<row>.<field>.<i>.<field> as a top-level one is, and a @Text []Key a row leaves out takes <...>.0, .1, ... for as many as en.po has (so no .lres spells a key; the manifest and en.po are now read before the parse that fills the rows); field: null is no text; trf / trn's trailing "" literals are padding, not counted against the English's holes; and string(x) of a string or a Key is x itself to the escape analysis, no allocation site, so tr(key) passes arena strict (a template's lone hole still copies); examples/lang/i18n_nested and its feat_case (added, not run); LANGUAGE.md and the changeset; reseeded, bootstrap-cfree fixpoint holds 2026-09-30 00:41:05 +03:00
i18n-keys-only.md ludic.i18n: the English path is out of L (26.9) - a plain string is drawn as it is in every language (a player named Settings stays Settings), a key, a key glued into text and a bracketed line are made as before. The pattern tables, the sentence and padding lookups, Ln, i18n_pattern_count and kr_words' English-argument lookup are gone; a language's .po is read for keys and plurals. Tests move to keys; README and a changeset say so. 2026-09-30 02:58:13 +03:00
i18n-keys.md ludic.i18n: keys (phase 26.1) - en.po as the base, tr/trf/trn marked strings made into text by L, keyed plurals, the fallback to English then the key ([[key]] loud in a dev build), the English form beside it 2026-09-29 23:16:12 +03:00
i18n-listener-keys.md i18n: a key used only in a listener counts as used - the key check walked prog but not the bodies kept beside it (@On listeners, the lifecycle hooks, tests, computed fields, scenes), so their keys were never checked against en.po and ludic schema listed them unused (the game's 15 itemuse.* keys, emitted through ItemPlace); they are walked now as privates.ludic walks them. examples/lang/i18n_listen and its schema case (added, not run); reseeded, bootstrap-cfree fixpoint holds 2026-09-30 03:12:25 +03:00
inflate-context.md inflate: its context in a caller-owned record, allocation-free per block; the state-backed calls kept 2026-09-29 17:25:16 +03:00
int-min-literal.md compiler: the most negative int is emitted as itself - itoa took its digits off -v, which overflows back to itself, so 0x80000000 in an int was written as a bare "-" (invalid IR); digits now come off v as it is, and the buffer holds a long's 20 2026-09-29 22:00:49 +03:00
json-escape-linear.md Json.parse decodes an escaped string in one pass into one buffer - joined a character at a time, every shorter copy was kept, and a long escaped text took gigabytes (ui-preview: 6.5 GB in 3 s); protocol-v1.md states the @import path rule and the key a file override answers to 2026-09-29 21:53:57 +03:00
json-unicode.md Json.parse decodes \uXXXX to UTF-8 - a surrogate pair joined, a lone surrogate or bad hex as U+FFFD - and \t \r \b \f, where it dropped the backslash and kept the hex as text; examples/lang/json_unicode checks it byte by byte 2026-09-29 21:07:10 +03:00
json-write-file.md runtime: Json.write_file, and Json.encode through a kept buffer 2026-09-28 15:32:36 +03:00
lab-png-16-bit.md ludic.lab: 16-bit PNGs (lab_png_write16_from), so a test can write a height map render3d reads back as R16 2026-09-29 22:08:22 +03:00
leaves-light-and-the-bole.md fix(runtime): the wheel on macOS did nothing, or everything 2026-09-20 01:24:37 +03:00
lexer-literals.md fix(lang): a template inside another's hole, and integer literals past 2^31 - 1 2026-09-25 08:10:46 +03:00
light-air-and-the-hour.md feat(render3d): the air and the light are the hour's, not one constant apiece 2026-09-19 14:42:30 +03:00
light-you-can-see.md feat(render3d): light you can see - sun shafts and valley mist 2026-09-19 16:06:16 +03:00
lint.md feat(tools): L10 ludic-fmt enforces a project's style 2026-09-24 14:38:35 +03:00
listener-return.md fix(events): a listener's return ends that listener, not the dispatch 2026-09-23 17:42:18 +03:00
ludic-deps.md feat(cli): ludic deps - the module graph as the compiler resolved it 2026-09-25 11:16:34 +03:00
ludic-lab.md feat(ludic.lab): the visual lab as a package - a scene on a plate, never a world 2026-09-25 07:48:43 +03:00
machines.md feat(lang): 27.1 - a state machine as data: @Machine(Record.field) on a registry of transitions 2026-09-30 16:37:37 +03:00
map-scoped-tables.md docs: map-scoped tables in LANGUAGE.md, the maps manifest key, a changeset 2026-09-29 21:03:56 +03:00
maps-entry-only.md ludic build --check reads the maps only for the package's entry - a partial program (a unit test, a molecule, a bake's runner) lacks the game's constants and cannot judge them; --maps reads them anyway, --no-maps never; ludicc unchanged 2026-09-29 22:14:53 +03:00
math-result-types.md fix(check): a Math.* call has the type the emitter gives it 2026-09-28 14:12:28 +03:00
model-release.md feat(render3d): 23.4 - model_release, shared textures counted, and scatter layers sized to what they hold 2026-09-27 23:34:14 +03:00
module-layers.md feat(lang): layers - module flow in layer app uses base, items 2026-09-25 12:54:15 +03:00
module-scope.md feat(lang): L3 module scope - module, export, friend module 2026-09-24 00:52:12 +03:00
module-uses.md feat(lang): module NAME uses A, B - a module reaches only the modules it names 2026-09-25 04:06:45 +03:00
moltenvk-no-command-pooling.md Os.platform/arch uname once; render3d primes a new buffer's Metal buffer 2026-09-28 13:34:45 +03:00
native-libraries.md feat(pkg): phase 15 - a package can carry a native library 2026-09-26 23:35:45 +03:00
native-notices.md feat(bundle): a native library's licence ships with it - each linked package's native/LICENSE* beside the .exe, in Contents/Resources on macOS 2026-09-27 14:43:16 +03:00
near-fade-spirv.md feat(render3d): Vulkan in a macOS window through MoltenVK; window built-ins take a slice's elements; the NEAR_FADE SPIR-V 2026-09-27 00:52:45 +03:00
numbers-float-strict.md feat(lang): strict numbers in float files; render3d on float 2026-09-23 17:13:25 +03:00
numbers-float.md feat(lang): numbers float - a module whose decimal literals are float 2026-09-23 16:02:11 +03:00
one-moltenvk.md runtime (macOS): MoltenVK opened first, so a process holds one - the SDK's loader in /usr/local/lib loaded its own MoltenVK beside the linked one and the program drew through it; the loader only when a layer is asked for, pinned to ours (VK_DRIVER_FILES, lib/macos-arm64/MoltenVK_icd.json) unless a driver is named. steady's stream round: a 300-cell warm-up, then the least of three 150-cell windows 2026-09-30 00:41:27 +03:00
one-wind.md feat(render3d): one wind, and everything in the valley is moved by it 2026-09-19 18:24:43 +03:00
open-registry.md feat(lang): open registry - other modules' defs, in an order imports cannot change 2026-09-25 04:28:34 +03:00
os-platform-once.md Os.platform/arch uname once; render3d primes a new buffer's Metal buffer 2026-09-28 13:34:45 +03:00
package-export-clash.md fix(lang): 0.R4 - ludic.ui's UiAct is its own, and a name that meets a package's export says whose it is 2026-09-26 06:57:35 +03:00
physics-jolt.md feat(physics): phase 16 - ludic.physics over Jolt Physics v5.6.0 2026-09-26 23:41:37 +03:00
pin-icd-always.md runtime (macOS): with a layer asked for, the loader is pinned to our MoltenVK whatever the shell set - the SDK's setup-env.sh exports VK_DRIVER_FILES and VK_ICD_FILENAMES at its own, and the one left set loaded a second MoltenVK; VK_DRIVER_FILES overwritten, VK_ICD_FILENAMES cleared, R3D_VK_ANY_DRIVER=1 to keep the shell's 2026-09-30 00:47:07 +03:00
pkg-remove-get-json.md ludic remove <module> and ludic get --json (R9) 2026-09-30 00:24:49 +03:00
ports.md feat(lang): ports - port Clock { ... } in a module, bind Clock { ... } in the app 2026-09-25 04:20:12 +03:00
private-kinds.md feat(lang): a module's private records and events are its own 2026-09-25 11:02:48 +03:00
private-per-module.md feat(lang): a module's private names are its own 2026-09-25 07:08:05 +03:00
prune-late-states.md fix(migrate): 0.R4 - --prune keeps a state declared after a plain parameter; a parameter named twice is refused 2026-09-26 05:32:08 +03:00
queue-owns-count.md feat(base): 0.R4 - a queue keeps its own count, so every package verb takes only its own state; ludic migrate state --prune 2026-09-26 04:19:56 +03:00
r3d-fault-loud.md render3d: a renderer that cannot draw says so and stops - LUDIC_HOME's leading ~ expanded (an unexpanded one left a run without shaders, logging 27 missing variants and drawing on), the SPIR-V manifest ends with E <count> and a cut or miscounted one is refused, and every such failure is a fault: r3d_open / r3d_load_step fail on it, r3d_fault() / r3d_fault_exit(code) for the game; tests/manifest_test 2026-09-29 22:12:39 +03:00
r3d-readback-flush.md render3d: gpu_tex_read_all and gpu_tex_write_all flush the frame first - the read-back is a submit of its own and took the image before the draws that fill it (the sky bake's BRDF table was all zeros at one width in three) 2026-09-29 23:18:05 +03:00
r3d-variants-all.md render3d: heightgen's plain and SMOOTH variants compiled (61 programs) - a smoothed or noise-only terrain had no height field on Vulkan; run-time defines as pure functions (program_defs.ludic) and manifest_test holds every program the source can ask for, literal or computed, to the manifest 2026-09-29 22:30:35 +03:00
rain-that-leaves-something.md feat(render3d): rain that leaves something behind 2026-09-19 19:43:29 +03:00
README.md refactor(cli)!: split the contributor tool out of the ludic CLI 2026-09-05 23:15:12 +03:00
redeclared-local-missing-return.md feat(lang): a name is defined once, a local once per block, and a result is always returned 2026-09-23 23:55:20 +03:00
reducer-reads.md feat(lang): 0.R5 - a reducer writes one state and may read others, declared between its state and the action 2026-09-26 18:23:48 +03:00
registries.md feat(lang): L8 registries by declaration - registry and def 2026-09-24 13:34:51 +03:00
render3d-bc-not-a-target.md fix(render3d): nothing draws into a BC7 texture - sampler binding binds, and compressed images get no mips 2026-09-28 10:34:23 +03:00
render3d-bc-textures.md feat(render3d): 23.3 - a .dds beside a .png is uploaded BC-compressed with its whole mip chain 2026-09-28 01:23:45 +03:00
render3d-blade-density.md render3d: the GPU blades grow by a painted density (a layer of the ground densities, grass_density_layer) where a map has one - a 5x5-tile window round the camera in a storage buffer the cull samples - and by today's rules where not; grass_rule_at is the rules on the CPU for the migration 2026-09-29 21:14:22 +03:00
render3d-chunk-heights.md feat(render3d): terrain_chunk_heights - the ground's B-spline heights a chunk at a time (23.5) 2026-09-28 11:25:21 +03:00
render3d-gltf-cached.md changes: gltf_cached 2026-09-28 11:39:49 +03:00
render3d-grass-kind.md render3d: a grass kind - the GPU blades' per-kind numbers are a GrassKind record the game hands over (grass_kind_set), not constants 2026-09-29 19:15:39 +03:00
render3d-height-format.md perf(render3d): the terrain's height and normal in R32F + RG16F, 128 MB where one RGBA32F was 256 2026-09-28 02:31:18 +03:00
render3d-hooks.md feat(lang): functions are values (L2), and render3d takes its scene as callbacks 2026-09-24 00:27:26 +03:00
render3d-moltenvk.md feat(render3d): 22.10 - ludic.render3d carries MoltenVK, so a Mac program has its Vulkan driver 2026-09-28 01:16:26 +03:00
render3d-plan25-vulkan.md render3d: VkAllocationCallbacks counted (R3D_ALLOC_VK), pipeline create infos freed, actor pool at init 2026-09-28 15:46:19 +03:00
render3d-recycle.md render3d: recycle texture and buffer ids, free a released model whole, actor_release 2026-09-28 12:00:53 +03:00
render3d-small-window.md render3d: r3d_small, a renderer sized for a small UI window 2026-09-29 14:23:00 +03:00
render3d-texture-scratch.md render3d: free a texture's create infos and the .dds path it looked for; steady's model at 0 2026-09-28 12:25:02 +03:00
render3d-vulkan-only.md wip(render3d): plan 22.14 - the OpenGL backend removed (suite 306/306, not yet handed over) 2026-09-28 00:30:11 +03:00
render3d-zero-per-frame.md fix(render3d): nothing allocated in the steady state - the Vulkan allocator reuses its records 2026-09-28 11:14:56 +03:00
resources.md feat(lang): L9 resource files - registry ... from 2026-09-24 14:00:20 +03:00
row-reducers.md feat(lang): 27.3 - a reducer on a table's row: reducer T in S.table on A, @Target, Row<T>, @RowVerb, @Column 2026-09-30 16:14:59 +03:00
runtime-on-call.md fix(runtime): Random.* in any program; value_* called directly brings the value tree 2026-09-25 04:59:48 +03:00
runtime-temporaries.md runtime: Log, DateTime.format, Input.text, Path, Mime, Fs, Os and Text keep nothing per call 2026-09-28 13:55:49 +03:00
schema-code.md compiler: ludic schema gains a "code" section (R7) - the code map from the compiler, so the studio's scan is only a fallback: modules (package, layer, uses and where, friend, files), states, actions, reducers, every dispatch, events (cancellable, net) with every emit and their @On listeners, ports (members with fn types, defaults, required) and binds (port, member, fn, value, at), handlers (phase, hook and target, @Public, @Server / @Predicted, @Queries with filters as written, scene and layer), models (@Owned, @Sync), prefabs, scenes (start, public, shows, lasts / then, loads, enter / exit, layers) and fn_refs - every fn name in code or a resource file with the slot it fills (registry field and row, port member or default, record field, emit field, call argument, assignment, let). Places are "file:line:col"; sorted by name, sites by place; schema_version stays 1 (additive). The parser records emits, fn refs, handlers and scenes' layers / lasts / loads for it (emit_schema_code.ludic, emit_schema_code_game.ludic); LANGUAGE.md and the changeset say so; schema_case lines added (not run); reseeded, bootstrap-cfree fixpoint holds 2026-09-30 00:19:24 +03:00
session-rest-least.md ludic.session: ply_rest_least, the party's rest for the clock (the valley's selftest57 down into the packages): the roster's freed slot reused and ply_leave_all, the nearest player alone, and ludic.shop's week of demand leaving the shared Random stream untouched 2026-09-29 22:51:01 +03:00
state.md feat(lang): 0.S - components take states in their header; migrate writes them, never inside a name, never outside the programs given, never into a package's state 2026-09-25 17:11:15 +03:00
still-water.md fix(render3d): the meadow is vegetation, and still water reads as water 2026-09-19 18:07:01 +03:00
stream-arena.md render3d: a stream's cache is made with the stream - records and an arena - and evicts in place 2026-09-28 13:26:59 +03:00
string-temporaries.md Os.heap_bytes: the heap without the renderer; string_temps reads it 2026-09-28 12:31:42 +03:00
syntax-vocabulary.md changes: syntax-vocabulary (ludic syntax, ludic-dev syntax, the language server's outline and attribute arguments) 2026-09-29 23:59:47 +03:00
template-copy.md templates: {s} with one string hole and nothing else is a new string, not s itself - it was the one template that passed its string through (string() does), and code that wrote it as a copy kept what it then freed; the mark rides the callee, since escape analysis writes the call's uns; reseeded 2026-09-29 20:14:58 +03:00
test-generics.md fix(test): a generic function called from a test block works 2026-09-25 04:46:08 +03:00
test-handler-rt-init.md compiler: a test program may hold a phase handler - the test runner called @rt_init whenever the program had systems, and a handler alone makes none (LLVM: use of undefined value '@rt_init'); it calls it only when it exists, as an entry program's main does; a ludic test case with a handler (added, not run); reseeded, bootstrap-cfree fixpoint holds 2026-09-30 00:34:34 +03:00
test-parallel.md feat(test): ludic test -j N runs tests side by side 2026-09-25 11:38:09 +03:00
test-per-process.md feat(test): ludic test <dir>, each test block in a process of its own 2026-09-25 04:39:55 +03:00
text-keys-data.md packages: the registry text a game fills is keys (Maroon Lake's phase 26.4) - ThingKind.name, GearKind blurb/howto/names, HintCard's words, JobDef title/text/how, NpcLine and NpcChoice text, WeatherKind.name are @Text Key fields (a compiled row that leaves one out takes its derived key, the English is the game's en.po); what a package hands back as text is the key's marked text, "" for none; NpcName.name, NpcChoice.line and SettingDef.text are not text and lose @Text; ludic.i18n's tr/trf/trn/key_text say @alloc_ok (tr is a cast); the packages' tests with key literals (built, not run); changes/text-keys-data.md 2026-09-30 00:29:03 +03:00
text-keys.md compiler: text keys' follow-up (Master's rulings) - a Key in a template literal's hole is an error (the parser marks each hole's string(), TPL_HOLE; write trf(k"...", ...)), trn takes only a plural key kn"..." and tr / trf refuse one (with or without en.po), and a program with no lang line (a package test) uses key literals unchecked and silently - the warning stays for a lang line whose en.po is missing; rejected/i18n_keys gains key_hole (2) and key_rules (3), key_missing's plural moves to trn; LANGUAGE.md and the changeset say so; reseeded, bootstrap-cfree fixpoint holds 2026-09-29 23:46:48 +03:00
text-to-float.md feat(text): Text.to_float; a void function's return <value> is an error 2026-09-23 19:19:28 +03:00
the-ground-up-close.md fix(render3d): the ground keeps more of its own colour, and the rock keeps all of it 2026-09-19 16:29:17 +03:00
the-world-knows-you-are-there.md feat(render3d): the lake answers to a body standing in it 2026-09-19 20:23:27 +03:00
toolchain-r3-fixes.md fix(lang): the action drain is the program's; --check runs every check a build makes; a registry key named count is refused; name lookups are tables 2026-09-25 23:34:43 +03:00
type-checker.md feat(lang): L4 type checker between parse and emit 2026-09-24 01:34:49 +03:00
udp-open-local.md Udp.open_local(port): a socket bound to 127.0.0.1 alone (udp.ll, udp_win.ll), for a port only this machine's tools reach - a dev link - where Udp.open binds every interface 2026-09-29 19:15:30 +03:00
ui-mounted.md ludic.ui: ui_mounted(out) - the classes of the components the last frame showed, in tree order and each once, into the caller's list - and ui_model_of(cls), the first mounted instance's model read in place; examples/library/ui_mounted 2026-09-29 22:51:33 +03:00
ui-override.md ludic.ui: an override's path is compared normalised, and an absolute path matches the relative one it ends with - a lab build registers its components as lab/../src/ui/..., so the studio's src/ui/... (or absolute) path matched nothing and the override changed nothing; nothing is normalised while no override stands; examples/library/ui_override_up registers under a .. path as the lab does 2026-09-29 23:00:12 +03:00
ui-press-drain.md fix(ui): 0.R4 - an action a UI button dispatches is reduced before the frame is presented 2026-09-26 07:08:26 +03:00
ui-preview-hold.md ui-preview: hold <id or key> <pseudo> - hover, active, focus, focus-visible, checked or disabled held on from the next frame on top of the real input, by the element's key (so through model and load), let go with "" or reset; ludic.ui's held.ludic keeps them and the matcher asks it first; protocol-v1.md and smoke.txt 2026-09-29 22:04:20 +03:00
ui-preview.md ludic ui-preview: ludic.ui components previewed for a studio over stdio (R5) 2026-09-29 19:53:26 +03:00
ui-t-key.md ludic.ui: t(key, holes...) in a template - the key marked (byte 1, a marked key kept as it is) and each hole's value after a byte 31, the text the translator decodes at draw time; joined through the text memo, so nothing is made while it reads the same (phase 26) 2026-09-29 23:16:05 +03:00
unknown-type-names.md fix(check): 0.R4 - a misspelled type in a parameter, a result or a field is refused where it is written 2026-09-26 05:58:35 +03:00
unsafe.md feat(lang): L7 memory is safe unless it says unsafe 2026-09-24 12:53:27 +03:00
uses-packages.md feat(lang): packages count under uses; a port of defaults may go unbound 2026-09-25 05:36:11 +03:00
vehicles-spec.md vehicles: a kind's numbers are a VehicleSpec the game hands over (vehicles_spec), not constants 2026-09-29 20:22:02 +03:00
void-return-error.md feat(text): Text.to_float; a void function's return <value> is an error 2026-09-23 19:19:28 +03:00
vulkan-leaks.md fix(render3d): the Vulkan renderer stops leaking as it loads and draws 2026-09-27 23:03:38 +03:00
vulkan-on-macos.md feat(render3d): Vulkan in a macOS window through MoltenVK; window built-ins take a slice's elements; the NEAR_FADE SPIR-V 2026-09-27 00:52:45 +03:00
wildlife-painted.md ludic.wildlife: painted habitats and ranges through the port (habitat_in, range_at), the rule where none is painted 2026-09-29 20:35:13 +03:00
window-buffers.md feat(render3d): Vulkan in a macOS window through MoltenVK; window built-ins take a slice's elements; the NEAR_FADE SPIR-V 2026-09-27 00:52:45 +03:00
worker-reads-states.md feat(compiler): a Job.parallel_for worker may read a state but not change one 2026-09-27 22:04:59 +03:00
zones.md ludic.zones: a disc per player from their fog, merged into clusters; the simulating packages ask whether a place is simulated 2026-09-29 15:53:23 +03:00

README.md

Changesets

A changeset is one small Markdown file describing a single user-facing change, dropped in this directory. ludic-dev release consumes every changeset here into a new CHANGELOG.md section, bumps VERSION, and deletes the consumed files.

Format

bump: minor
type: feat
One or more lines describing the change, in the past-agnostic imperative used in
the changelog. Markdown is fine.
  • bump: — major, minor, or patch (SemVer). The release version is bumped by the highest level among the pending changesets (unless ludic-dev release <level> overrides it).
  • type: — the Conventional Commit type (feat, fix, perf, docs, …). It decides which group the change lands in: feat → Features, fix → Fixes, perf → Performance, and so on, in that order. A type with no known heading gets one named after itself.

Writing the body

The body is markdown and reaches the changelog as markdown: it becomes one list item, with continuation lines indented to stay inside it. Nested bullets, blank lines between paragraphs and inline code all survive.

bump: minor
type: feat
**Tiled map support** — load and draw Tiled maps.

- **TMX/TSX** — the XML formats, decoded to the same intermediate as JSON.
- **Collision** — the `collision` layer projects onto the engine tilemap.

Lead with the thing that changed, not with the mechanism. A reader scanning the release should be able to stop after your first clause.

Adding one

Create a file with a short, unique name, e.g. changes/regex-namespace.md. Any filename works except this README.md, which the release step always skips.

Preview how the next release will read before cutting it — this writes nothing:

ludic-dev release --dry-run