Nikaia
Menu
v0.0.1343pre-alpha GitHub

agent: builder model: opus max_turns: 110 budget_usd: 22 —

Brief: a view field of a lent struct is handed on as a view (#1063)

Score: blocks #928 step 4 (yaml’s events carries each node’s anchor and tag, ADR-yaml-1 D7; load copies the tree with its strings still views, D6). A compiler defect found by the architect probing step 4; fixed in this lane because only the yaml package needs it now. It changes crates/: name in the report what it decided (the supervisor raises the version), Changelog: one sentence.

Facts: Where things live, Traps already met

Problem

A struct whose String field is a view (built from a view of a parameter, ADR-353 D7, #1000) cannot hand that view on from a lent value: reading w.text of a ref Word (a ref parameter, a for element) into a new struct is refused with NK2106 (“w.text is stored in Word.text, but w is only borrowed here”). A view is not a part taken out: copying it takes nothing from w (in Rust &'a str is Copy; fn again<'a>(w: &Word<'a>) -> Word<'a>). Part I 6.5 refuses a part of a loan handed over; a view field is already a view, and “an argument that is already a view” is passed as it is.

In the yaml package the refusal also cascades: once a function builds a Node from a ref Node’s anchor/tag, every Node { .., anchor: p.anchor, tag: p.tag, .. } in the grammar is refused NK1106 (the inference decides the fields need text of their own) and parse’s limits: Limits = default_limits() gets NK1127.

Probes (checked by the architect on today’s target/fast/nikaia)

Probe A (nikaia run a.nika; today NK2106 on w.text; wanted: prints 1 hello):

struct Word { start: i64, text: String }

fn word_of(input: String) -> Word {
	return Word { start: 0, text: input.trim() }
}

fn again(w: ref Word) -> Word {
	return Word { start: w.start + 1, text: w.text }
}

fn main() {
	let s = " hello "
	let w = word_of(s)
	let v = again(w)
	print(f"{v.start} {v.text}")
}

Probe B (today NK2106 on w.text and on w.tag; wanted: prints 1 hello):

struct Word { start: i64, text: String, tag: String? }

fn word_of(input: String) -> Word {
	return Word { start: 0, text: input.trim(), tag: null }
}

fn again_each(ws: ref Vec[Word]) -> Vec[Word] {
	let mut out: Vec[Word] = []
	for w in ws {
		out.push(Word { start: w.start + 1, text: w.text, tag: w.tag })
	}
	return out
}

fn main() {
	let s = " hello "
	let ws = [word_of(s), word_of(s)]
	let vs = again_each(ws)
	print(f"{vs[1].start} {vs[1].text}")
}

Probe C, the package shape: copy packages/yaml (src/, nikaia.toml) to a scratch directory, append to src/main.nika

fn copy_node(n: ref Node, mut count: Count, limit: i64) -> Node sync throws {
	count.n += 1
	let v = match n.value {
		Value::Seq(items) => {
			let mut out: Vec[Node] = []
			for x in items {
				out.push(copy_node(x, count, limit))
			}
			Value::Seq(out)
		}
		other => other.clone()
	}
	return Node { start: n.start, end: n.end, anchor: n.anchor, tag: n.tag, value: v }
}

pub fn load(input: ref Bytes) -> Vec[Node] sync throws {
	let docs = parse(input)
	let mut out: Vec[Node] = []
	for d in docs {
		let mut count = Count { n: 0 }
		out.push(copy_node(d, count, 100))
	}
	return out
}

struct Count { n: i64 }

and a tests/load.nika that loads Bytes("- 1\n- 2\n") and prints {docs.len()} {docs[0].end} (.stdout: 1 7). Today: 14 NK1106 in the grammar, two NK2106, one NK1127; wanted: the test passes. Do not commit this probe into the package (load is a later brief).

Locations (suspected by the architect, medium confidence)

To build

  1. A view field read from a lent struct (a ref parameter, a for element) and stored in a struct, pushed or returned is a view of the same buffer, not a part taken out: Probes A and B print 1 hello, Probe C passes. Tests in tests/compiler/tests/: view_field_of_a_lent_struct_handed_on.nika (Probe A) and view_fields_of_for_elements_handed_on.nika (Probe B), each with .stdout; if Probe C needs more than A and B fixed, a test of its shape in a package read by another (use compiler_tests, builder-facts). Commit.
  2. An owned String field of a lent struct stays refused with NK2106: let_lent_field_kept and let_moved_ref_parameter keep their .stderr byte for byte; tests/compiler --both-settings green.

If the fix needs a language decision (the rule in Part I 6.5 does not settle it), stop: Result: blocked, Needs: architect decision. Stop about 15 turns before max_turns, commit what is finished, report partial.

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/y22-1063-view-field-copy.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.