Nikaia
Menu
v0.0.1343pre-alpha GitHub

agent: builder model: opus max_turns: 80 budget_usd: 20 —

Brief: #1026, a view field read from a struct a grammar label binds stays a view

Score: a compiler defect that blocks #928 step 2b (yaml properties and aliases, y2c-properties.md, next in this batch); bugs before moving.

Issue (#1026, copied)

A rule that returns a struct of views, bound by a label in another rule (p:P), hands its fields on as text of its own: p.a going into the struct the outer rule returns is refused with NK1106, and with it every other alternative that puts a view into the same field. The same read through a function of the program works (#803: let p = props() then Pair { a: p.a, .. } runs; the architect ran that too). ADR-282 D13: “a function of the program answers with its declared or decided result; a read of an element hands on what the position is”, and a grammar’s rule is a function of its grammar for this walk (declare_rules).

Probe (the architect ran it on 9d92e474):

struct Props { a: String }
pub struct Pair { pub a: String, pub b: String }

grammar G {
	rule WS = "" -> { }
	rule Word = digit+ -> { }
	entry rule top -> Pair = x:text(Word) " " y:text(Word) eof -> { Pair { a: x, b: y } } | p:P " " y:text(Word) eof -> { Pair { a: p.a, b: y } }
	rule P -> Props = "&" x:text(Word) -> { Props { a: x } }
}

fn main() {
	for t in ["12 34", "&12 34"] {
		let b = Bytes(t.bytes().collect())
		let r = G::top(b) catch {
			println("error")
			continue
		}
		println(r.a)
		println(r.b)
	}
}

Today: two NK1106 (Pair.a holds String, but you're giving it ref String) at x (7:76) and at p.a (7:130). Without the second alternative it runs. The y2 builder saw the same with a String? field, with let an = p.a, and with match p { Props { a: x } => .. }. Expected output: 12 34 12 34, one per line.

Cause (architect, read on 9d92e474)

bind_grammar_pattern (crates/nikaia-std/src/tools/text_tiers.nika, Pattern::Bind arm, about line 5231) binds a label with plain_local(kinds), so the local’s of (the struct it holds) stays null. A field read finds its position through tier_struct_of -> struct_of_local -> local.of (about lines 2962, 3199); with none, field_place is null and field_kinds answers owned_kinds(), so Pair.a sees a view and own text. For a call called_struct (about 3022, #803) names the struct; for a loop’s binding walked_struct (local.of = walked, about 2315).

Locations

To build (one commit)

  1. A label bound to a rule of the grammar (directly, or through a group) whose declared result is a plain struct of the program binds a local whose of is that struct, so its fields read as their positions. Test tests/compiler/tests/grammar_a_view_field_read_from_a_bound_struct_stays_a_view.nika with its .stdout: the probe above, plus a String? field, a let an = p.a read, a match p { Props { a: x } => .. }, and a rule named as the struct it returns (rule Props -> Props, the shape yaml uses), each printing its view. Check the test fails on the unfixed compiler (say so under Tests:).

Done when

The new test passes (target/fast/nikaia test on it), the tests whose names start with grammar_ and the text-tier tests you touched pass, the gate of .claude/agents/builder.md (#852) is green, the commit says Fixes #1026. Changelog: one sentence (this changes crates/).

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/y8-1026-grammar-label-struct.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 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:); 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.