ludic/selfhost/backend
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
..
game a program's states made on first injection, and the ECS stores started at 8 slots 2026-09-29 14:14:05 +03:00
stdlib map packs: Fs.mount / Fs.unmount at run time, and a map's bakes under assets/baked/maps/<map>/ 2026-09-29 17:27:06 +03:00
emit_addr.ludic feat(lang): L11 views and templates - the UI is markup, not code 2026-09-24 15:58:47 +03:00
emit_call.ludic compiler: text keys (phase 26.2) - k"pause.resume" / kn"catch.count" of the builtin type Key (not a string, and no string one; == compares), its run-time value the key after a marker byte (1, or 2 for a plural); package.ludic's lang "assets/lang" en and a gettext .po reader; every key literal checked against en.po (a kn"" one wants a msgid_plural), trf / trn and a template's t('key', ...) against the English's holes, a component's template words and a @Text row's English as "English left" (warnings, english_left in ludic deps), a @Text Key field a compiled row leaves out filled with its derived key <registry>.<row>.<field> (or @TextKey's prefix; maps.<map>.<table>.<row>.<field> for a @PerMap row, checked), one English under several undescribed keys warned at en.po's line; the schema's "lang" (keys and their sites, unused, undescribed, split, and every other language's missing / fuzzy / extra); k"" in .lres data, @PerMap rows included (lres_key); ludic-fmt keeps a key literal whole; reseeded, bootstrap-cfree fixpoint holds 2026-09-29 23:33:52 +03:00
emit_core.ludic Os.heap_relief(): the allocator's freed-but-cached memory handed back, once after a load 2026-09-29 14:30:20 +03:00
emit_decl.ludic 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
emit_deps.ludic compiler: text keys (phase 26.2) - k"pause.resume" / kn"catch.count" of the builtin type Key (not a string, and no string one; == compares), its run-time value the key after a marker byte (1, or 2 for a plural); package.ludic's lang "assets/lang" en and a gettext .po reader; every key literal checked against en.po (a kn"" one wants a msgid_plural), trf / trn and a template's t('key', ...) against the English's holes, a component's template words and a @Text row's English as "English left" (warnings, english_left in ludic deps), a @Text Key field a compiled row leaves out filled with its derived key <registry>.<row>.<field> (or @TextKey's prefix; maps.<map>.<table>.<row>.<field> for a @PerMap row, checked), one English under several undescribed keys warned at en.po's line; the schema's "lang" (keys and their sites, unused, undescribed, split, and every other language's missing / fuzzy / extra); k"" in .lres data, @PerMap rows included (lres_key); ludic-fmt keeps a key literal whole; reseeded, bootstrap-cfree fixpoint holds 2026-09-29 23:33:52 +03:00
emit_escape.ludic 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
emit_expr.ludic compiler: text keys (phase 26.2) - k"pause.resume" / kn"catch.count" of the builtin type Key (not a string, and no string one; == compares), its run-time value the key after a marker byte (1, or 2 for a plural); package.ludic's lang "assets/lang" en and a gettext .po reader; every key literal checked against en.po (a kn"" one wants a msgid_plural), trf / trn and a template's t('key', ...) against the English's holes, a component's template words and a @Text row's English as "English left" (warnings, english_left in ludic deps), a @Text Key field a compiled row leaves out filled with its derived key <registry>.<row>.<field> (or @TextKey's prefix; maps.<map>.<table>.<row>.<field> for a @PerMap row, checked), one English under several undescribed keys warned at en.po's line; the schema's "lang" (keys and their sites, unused, undescribed, split, and every other language's missing / fuzzy / extra); k"" in .lres data, @PerMap rows included (lres_key); ludic-fmt keeps a key literal whole; reseeded, bootstrap-cfree fixpoint holds 2026-09-29 23:33:52 +03:00
emit_fence.ludic fence: every runtime call is its line's site, and site 0 comes back when it returns 2026-09-29 00:40:28 +03:00
emit_fence_ir.ludic intern: past its store or its table, the fence's overflow - said once 2026-09-29 00:37:56 +03:00
emit_float.ludic schema: ludicc --emit-schema / ludic schema, --check --diagnostics=json, and editor attributes 2026-09-29 18:06:54 +03:00
emit_fnval.ludic a program's states made on first injection, and the ECS stores started at 8 slots 2026-09-29 14:14:05 +03:00
emit_frame.ludic schema: ludicc --emit-schema / ludic schema, --check --diagnostics=json, and editor attributes 2026-09-29 18:06:54 +03:00
emit_head.ludic a program's states made on first injection, and the ECS stores started at 8 slots 2026-09-29 14:14:05 +03:00
emit_intrin.ludic fence: every runtime call is its line's site, and site 0 comes back when it returns 2026-09-29 00:40:28 +03:00
emit_intrin2.ludic a render3d program's window is opened by its renderer, not by the runtime before main 2026-09-29 14:19:10 +03:00
emit_lazy.ludic tests: the full suite's seven failures on lang/foundations - one compiler fault and six stale expectations 2026-09-29 21:19:00 +03:00
emit_mem.ludic fence (25.1): every allocation goes through the fence - sites, frame judging, census, callers 2026-09-28 15:35:29 +03:00
emit_new.ludic capacities (25.5a): @max(n) on a list field, and a full table is a failure 2026-09-28 16:43:02 +03:00
emit_resource.ludic owned fields (25.5e): @owns(Kind) on a record's field, and owned_leaks 2026-09-28 17:01:28 +03:00
emit_schema.ludic 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
emit_schema_code.ludic 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
emit_schema_code_game.ludic 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
emit_schema_hash.ludic Build.schema_hash(): FNV-1a 64 of the program's own schema (the bytes ludic schema prints), worked out only when a program names it, 0 under ludicc --release, which ludic bundle now passes (Mac and Windows); reseeded 2026-09-29 19:05:28 +03:00
emit_schema_lang.ludic compiler: text keys (phase 26.2) - k"pause.resume" / kn"catch.count" of the builtin type Key (not a string, and no string one; == compares), its run-time value the key after a marker byte (1, or 2 for a plural); package.ludic's lang "assets/lang" en and a gettext .po reader; every key literal checked against en.po (a kn"" one wants a msgid_plural), trf / trn and a template's t('key', ...) against the English's holes, a component's template words and a @Text row's English as "English left" (warnings, english_left in ludic deps), a @Text Key field a compiled row leaves out filled with its derived key <registry>.<row>.<field> (or @TextKey's prefix; maps.<map>.<table>.<row>.<field> for a @PerMap row, checked), one English under several undescribed keys warned at en.po's line; the schema's "lang" (keys and their sites, unused, undescribed, split, and every other language's missing / fuzzy / extra); k"" in .lres data, @PerMap rows included (lres_key); ludic-fmt keeps a key literal whole; reseeded, bootstrap-cfree fixpoint holds 2026-09-29 23:33:52 +03:00
emit_schema_ui.ludic ludic schema: a components section and a natives list (R4) 2026-09-29 19:36:04 +03:00
emit_stmt.ludic frame allocs (25.2): @frame on a step list's field, @alloc_ok on a statement, and on an exported function 2026-09-28 15:56:32 +03:00
emit_vis.ludic 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