Nikaia
Menu
v0.0.1343pre-alpha GitHub

agent: builder model: opus max_turns: 90 budget_usd: 22 —

Brief: #1037, a dependency’s struct with two view fields is named with one lifetime (ADR-353 D7)

Score: blocks #928 (yaml properties, y2c-properties.md run 2 is next in this batch); a defect of something specified (ADR-353 D7), so a compiler fix, not a package change.

Issue (#1037, copied)

A struct of another package with two or more fields that hold a view has one lifetime per field (ADR-353 D7: “the ledger’s tethered lists these fields in that order, and a package that reads the type takes its places from that list”). Named as a type in the reading package, e.g. a parameter n: ref two::Node, it is lowered with one lifetime argument, and rustc refuses it. The same program in one file runs. Not naming the type (a for n in two::make(..) with a match inline) runs.

Probe (package two, nikaia.toml [package] name = "two", src/main.nika):

pub struct Node {
	pub a: String?,
	pub b: String?,
	pub value: Value,
}

pub enum Value {
	Bool(bool),
	Str(String),
}

pub fn make(s: String) -> Vec[Node] {
	return [Node { a: s.strip_prefix("l"), b: null, value: Value::Bool(true) }, Node { a: null, b: s.trim(), value: Value::Str(s.trim()) }]
}

fn main() {
}

tests/named.nika, expected bool(true) then str(left):

use two

fn show(n: ref two::Node) -> String {
	return match n.value {
		two::Value::Bool(b) => f"bool({b})"
		two::Value::Str(s) => f"str({s})"
	}
}

fn main() {
	for n in two::make(" left ") {
		println(show(n))
	}
}

The architect ran it (scratch package, nikaia test --project ., today’s develop): rustc Struct takes 3 lifetime arguments but 1 lifetime argument was supplied (3:1), then Type bool cannot be dereferenced and twice The size for values of type str cannot be known (4:2). Traced: only the first. Whether the other three follow from it is unverified; done means the program prints both lines. The package ledger is right: [type."Node"] tethered = ["a", "b", "value"]. The yaml draft (origin/wip/y2-properties) fails 9 of 10 package tests with exactly this set of errors.

Locations

Read docs/builder-facts.md (its first trap: check the winnow-grammar rev of target/fast/nikaia before a probe).

To build (one commit)

  1. The reading package takes a dependency’s places from its ledger (ADR-353 D7) wherever the type is written: a parameter, ref, a result, a field of a local struct. Test: the compiler’s test project is itself a package (tests/compiler/nikaia.toml, name compiler_tests; src/main.nika holds an empty main today) and its tests reach its pub items with use compiler_tests, as packages/yaml/tests do with use yaml (the architect ran the probe in that layout, with fn main in the package: same failure). So put the probe’s Node, Value, make as pub struct Places, pub enum PlacesValue, pub fn places into tests/compiler/src/main.nika (keep its main), and tests/compiler/tests/a_dependencys_type_with_two_view_fields_keeps_its_places.nika + .stdout with the probe’s show/main (compiler_tests::Places); it fails on the unfixed binary (say so under Tests:).
  2. packages/yaml gate green (its tests name ref yaml::Node; Node has one place today).

Not here: #1038 (a dependency’s view field returned beside an owned text, NK1104): a separate defect in the checker or the tier pass; if your probe meets it, stop and name it, do not fix it in this commit.

Done when

The new test passes; NIKAIA_CACHE_LIMIT=2G target/fast/nikaia test --project tests/compiler > out.txt 2>&1 green (both settings as the gate asks); the package gate below green; committed and pushed with Fixes #1037. Files you may change: the emitter files above and their lowered twins, crates/nikaia/src/emit/mod.rs, crates/nikaia-std/src/tools/emit_tables.nika and twin, tests/compiler/src/main.nika, the new test. Stop and report if the fix needs a checker or ledger change (ledger writing, check_*.nika): name the fact and the field (docs/architect.md, An emitter fix that needs what the checker knows).

Rules (CLAUDE.md; this block is pasted into every brief, also a fix-forward)

Report (at most 15 lines, to exactly /home/user/Nikaia/docs/briefs/yaml/y13-1037-dependency-places.md.report.md)

Fields of docs/architect.md, Briefs and reports: Result: done|noop|partial|blocked (the bare word alone on that line: Result: done, never a summary after it, which cuts the batch, b869a 2026-10-27; Owner: is a question for the owner or none, never a role), Files, Commit, Body (optional, one paragraph for the commit message), Fixes, Pushed, Tests, Gate (green|red), Deviation, Found, Needs, Changelog, Owner, Disk freed, and for a move Rs2nika: (the kept N of M lines of rs2nika --score, then the three costliest hand edits; none for a task that moves no Rust, none (redesign) for a brief with Shape: redesign and no Convert: line). none where nothing, and then the bare word: Deviation: none (a note) cuts the batch (batch.nika compares exactly); a note goes into the commit message. Found: is only for a defect outside your change (it cuts the batch); a process event such as another session resetting or rewriting your files is no Found:, it goes under Needs: (m3a 2026-10-28); a converter draft that came out wrong is no Found:, it is one of the costliest edits under Rs2nika:, which the converter lane reads (e12-4b 2026-10-27); how you solved your own step goes into the commit message. Deviation: is only a step of the brief built otherwise than written or not built (a choice the brief left to you, such as keep what X needs and say so, is no deviation: it goes into the commit message; nor is building two steps in one commit when every step is built, e12-3a 2026-10-27, nor a step another commit on develop already built, shown by your test failing with that commit’s change reverted, b889 2026-10-27: name the commit in Body:, nor a test’s expected .stdout/.stderr taken from what the program really prints where the brief’s sketch of it was short or wrong, b934 2026-10-29: say so in Body:, nor a lowering that follows the ADR governing it where the brief’s expected lowering contradicts that ADR, b933 2026-10-29: name the ADR in Body:, nor a lowering other than the brief’s where rustc cannot take the brief’s, b1029 2026-10-31: say why in Body:, nor re-keying an allow-list entry (KNOWN) whose hash changed for the same program and the same reason, b-fixforward-sweep2 2026-10-28: say so in Body:, nor an existing test’s expected .stdout/.stderr updated because your change rightly changes the message it prints, b1012b 2026-10-31: name the test in Body:); how you verified (e.g. a test not run against the unfixed compiler) goes under Tests:. A noop says why under Tests: (what was probed), not under Deviation:. Final answer, one line: done <sha>, report <path> (or BLOCKED: ... / PUSH REJECTED ...).

Attribution: Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com Claude-Session: https://claude.ai/code/session_01UVxJDveRM5qKZKyCqq6Y24

This site is open source. Improve this page.