Commit Graph
7 Commits
Author SHA1 Message Date
adminandClaude Opus 5.5 b0d0868f55 Give a list one struct type so every item keeps its shape (#254)
game-maker / gate (push) Successful in 3m28s
game-maker / suite (push) Successful in 10m38s
A list of structs used to let every item carry its own fields. Adding
an item meant typing the same fields again, and nothing kept the items
in step with the struct asset from #253. A list can now name a struct
type for its items, and the asset keeps them in shape.

- A list variable and a list field get the row Items with Mixed, every
  struct asset and Make struct type. Make struct type collects every
  field any item has, so no item loses a field.
- A typed item carries its type under the key `$struct`
  (ITEM_TYPE_KEY). listItems does not count it as a field.
- A list that holds anything but structs refuses a type, also when
  the offending item sits in an override. The console names the list
  and the count, and the list stays as it was.
- In a typed list `+ Item` adds an item with the defaults, without the
  kind menu. Field names are locked, values stay editable.
- The rows of a typed item take their kind from the struct asset
  (declaredKind in InlineList.ts). An asset field shows the asset
  picker instead of a text field, as reported in the issue.
- A struct item in a mixed list gets its own Type row. Switching a
  list back to Mixed keeps the type on each item.
- syncStructTypes brings marked items into shape and carries renames
  into them, also inside overrides and nested lists. One undo step.
- The delete dialog of a struct asset also counts typed lists.
- The game never sees the key. World and sharedState strip it with
  withoutItemTypes and bagWithoutItemTypes, so a struct stays a
  plain object at runtime.
- Add item and Set item get the setting Type. With a type they only
  accept a struct with exactly its field names (fitsType in
  SetItem.ts). Anything else is left out and logged.
- The item indent from #252 is back to its old width, the narrow one
  squeezed the value fields.

Not done: writing a variable does not check the type, and the List
node offers no type.

Found on the way: a test helper in list-of-structs.test.ts called
nodes without params. It now passes the same arguments as the
interpreter.

Tests:
- ./check.sh: 183 files, 2679 tests, CHECK EXIT 0
- e2e GM_STEPS=1,21,22,32,54,55: ALL PASSED, 150 checks
- e2e GM_STEPS=1-6,21,22,32,54,55 before the follow-up: 404 checks
- New: tests/core/typed-lists.test.ts, e2e module 55 with 13 checks
- Counter-checks bite in unit and browser, originals restored by cmp

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-28 07:23:00 +02:00
adminandClaude Opus 5.5 b90b62886b Put one item into a list, a struct staying whole, with Add item (#250)
A list could only grow through Join lists, and that takes its b apart:
a struct becomes one item per field and a list is joined. A list of
structs such as loot could not grow at all.

- Add item returns a new list with value as exactly one more item. A
  struct or a list goes in whole. The list that came in is left alone,
  like with Set item.
- index -1 or unset appends. Otherwise the item goes in before that
  index. An index outside the list appends, so the value is not lost.
- The rule that an item takes the kind of its slot moved out of Set item
  into fitToSlot. Both nodes use it, Set item behaves as before.
- No e2e check and no browser look. The node is covered by unit tests.

Tests:
- ./check.sh: 180 files, 2634 tests, CHECK EXIT 0
- 7 catalog cases, 4 cases in list-of-structs.test.ts, and index in the
  whole-number inputs of texture-kind.test.ts
- Counter-check: taking the value apart like Join lists fails 3 tests
- This commit alone: tsc and the three touched unit files pass
  (337 tests)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-27 21:07:04 +02:00
adminandClaude Opus 5.5 0c5ac32b59 Let structs and lists nest to any depth (#239)
game-maker / gate (push) Successful in 3m52s
game-maker / suite (push) Successful in 10m24s
Structs and lists were capped at two fixed levels (list of structs,
struct with a list field). Anything deeper was dropped silently on
load, in node settings and when written to a variable. The cap came
from an early design choice, not from a technical need, so it is
lifted completely instead of adding a third fixed level.

- One kind list NESTED_KINDS replaces ListItemKind and
  StructFieldKind. The flat special cases in listItems.ts are gone,
  every value goes through one recursive path.
- A struct inside a struct is declared, like a top-level struct.
  Values coming in over an edge are brought into that shape. Struct
  items in lists still take their fields from the value.
- New valuePath.ts addresses any nested spot by keys and indices.
  ListSlot.field is now ListSlot.path.
- Field commands (add, rename, remove) take the path to the struct.
  They change declaration, own value and all overrides in one undo
  step, as before on the top level.
- Inspector: "+ Field" offers Struct and List on every level. A
  nested struct is a header row with its own "+ Field", its fields
  are indented one step further.
- List tab: a list item shows "n items" and opens one level deeper.
  The title shows the path, e.g. "Listy · nums · [3]".
- Pack in Pack now writes a struct into a struct field instead of
  text. Objects are kept at any depth and stripped at every exit.
- A list edit in a struct field without an existing object override
  now starts from the shown value instead of an empty struct.
- Old projects load unchanged. Save, load and save again gives the
  same result.

Tests:
- ./check.sh: 172 files, 2498 tests, CHECK EXIT 0
- e2e 21 + 22: ALL PASSED, 90 checks (new 21d-iii, 22f-ii,
  21d and 22f updated)
- 20 new unit tests in nested-values.test.ts and
  nested-list-slot.test.ts, three tests that pinned the old cap
  turned around
- 6 counter-probes, each fails 1 to 8 tests, originals restored
  and checked with cmp
- Not covered: a nested struct inside a struct item in the list
  tab has no e2e click

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-26 22:33:20 +02:00
adminandClaude Opus 5 ffdf0a8ddb The new node "Raycast all" answers with every collider along the ray instead of only the nearest one: it carries the same inputs and settings as Raycast (Ignore owner, Hit triggers, Layers from #158) plus "max hits" where 0 means all, and it hands back parallel lists sorted by distance - objects, colliders, points x/y, normals x/y, distances - next to count, so index i of every list speaks about the same hit; a collider counts once no matter how many convex parts its shape is built from, while a tilemap counts per cell it is hit in, and the filter sits in Solids.raycastAll like the one of #158, the one path the editor run and Play both take, so all 12 new unit tests run through describe.each against both systems; the node needed the value system to learn something it could not do - a graph list carried numbers, text, booleans and bundles but no object handles, asList turned an object into its name, so objects/colliders would have been lists of names, which loses the object when two carry the same name, and collider children usually do; Value now knows a travelling list that may hold handles and "Item at" gives such an item out as an object, ready for Destroy, Apply force or Name of, while the single door into a stored file - writeTyped/asTypeOf - runs the list through the new storedList() so every object becomes its name again and nothing living reaches a project file, which is its own unit test over JSON.stringify; three list nodes follow the handles: "Index of" compares objects by identity because a name is not the object of that name, "Set item" and "Fill" write an object as an object, and "Join text" prints the names instead of the innards; this decision was mine to make because the question ran into the MCP timeout - the alternative was a list of names plus "Object by name", which is the smaller change but lossy; 4 new e2e checks in the new block 43e (shield and floor in one go reading 2 and "145, 280", the second object answering to its name, which only a real object in a list can do, "max hits" 1 stopping at the nearest, and the node showing the row) and 18 new unit tests, 12 in raycast.test.ts, 4 on the value system and 2 in the node catalog; two counter-checks hold: not counting tilemap cells one by one drops 2 unit tests, and answering with the nearest hit only drops module 43 with 2 FAILs; one finding on the way - the node read "maxHits" while every other input carries a spoken label, because int() took the key as its label, so it has a third parameter now, and the e2e check caught that, no unit test did (#157)
game-maker / gate (push) Successful in 2m4s
game-maker / suite (push) Successful in 7m26s
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-20 18:54:12 +02:00
rootandClaude Fable 5.1 8a7185948d Make texture and colour full kinds and drop the load-time migrations
- `Texture` and `Color` can be picked as a variable kind, a struct field
  kind and a declared port kind; a new `Texture` node hands a picture id
  into the graph. Declared, a texture is its bare id – the `$texture` tag
  stays a list-only thing.
- The `String` node is labelled "Text" like its kind; the type id stays.
- Whole-number inputs (`Item at`, `Set item`, `Sub list`, `Fill`,
  `Repeat`, `Substring`, `To text`) show no decimals (`int()` in shared.ts).
- All server projects are reseeded, so `legacyNodes.ts`,
  `migrateDeclaredLists` and the `ownerId` fallback go, with their tests.
- Note `reseed-because-of-list` moved to done/; new note `refactor-types`.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-12 14:06:14 +02:00
rootandClaude Fable 5.1 11c900c998 Drop the per-list item declaration; every item carries its own kind
A list no longer says what its items are. ListDef.items, ListDef.fields,
ListFieldDef.items and SetListItemsCommand are gone; the single seam is
src/core/properties/listItems.ts. JSON already tells number, text, boolean
and struct apart, so only two kinds need a tag: {"$color": n} and the new
{"$texture": id}. Field names cannot start with "$", so a tag never
collides with a struct field. On a data link a texture item arrives as
its bare id; the graph has no images.

List tab: the mark in front of each row is the kind picker, "+ Item"
asks for the kind, both are ordinary value edits (one undo step). A list
inside a struct field only offers the flat kinds because the depth rule
forbids a struct item there (ListAccess.kinds); the menu used to offer
it and silently do nothing.

Legacy lists that declared items: 'color' are tagged on load.

Actor demo: the four walk animations and four standing frames live in
two list variables of texture items indexed by facing (Item at), so the
graph can be rebuilt in the editor. The stand frames get their grid ids
back; 44 nodes instead of 46.

Also: the user's edits to on-start-during-play.md and the new note
reseed-because-of-list.md.

Verified: check.sh 794 tests, full e2e 631 checks, smokes 24/19/20.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-12 13:35:02 +02:00
rootandClaude Opus 5 a9c3111377 Let a variable hold a list
A variable could only be a number, a text or a yes/no, so every set of values had
to be smuggled through a string - which is where Tetris got its unreadable
"03066326048763714" colours from. A list is now a value of its own, with an item
kind (number, text, yes/no, colour) that decides how the inspector reads the one
comma-separated line it is typed as, and eight nodes to work with: List, Fill,
Count, Item at, Set item, Sub list, Join lists, Index of.

Two traps came with it. An array is typeof 'object', so every place that told a
value from an entity had to learn the difference, or a list would have been
passed around as a game object. And a value was compared by identity in three
places - with lists that marks everything as an overridden property after each
reload, so those comparisons now compare contents.

There is no list of objects: a list is written to the project file and an object
is a live handle that no reload survives. A set of objects stays a list of names.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-09 21:26:31 +02:00