[gHashTag]

Closes #1787

master does not compile. cargo build --release in bootstrap/ fails with 27 errors. Every contributor is blocked. This PR restores compilation.

What happened

PR #1783 (batch-merge-wave-loops-w420-w459) resolved its conflicts by keeping the wave-loop call sites while dropping the declarations they depend on.

The failure was masked: build.rs panics on the stale FROZEN_HASH seal (also not updated by #1783) before rustc runs, so the build stops at a seal error and the real compile errors never surface. Resealing exposes them.

What this restores

Every item was recovered from its originating commit — none of it was reconstructed by hand.

Item	From	Was
6 VerilogCodegen fields + initializers in all 3 constructors	WL458	E0609 × 12
type_is_float, gen_verilog_let_destructuring, tuple_element_widths, is_simple_tuple_type	WL455/458	E0599
let type_array = Self::parse_array_type(..) at both consumer sites	WL455	E0425
base/idx bindings + W383 local-array flattening block	WL458	E0425
json_out plumbing (SuiteOptions field + --json CLI arg)	WL440	E0425
HashSet import, fast → opts.fast	—	E0433/E0425

One adaptation rather than a straight revert: WL455's parse_array_type returned Option<(usize, String)>, master's returns Option<(Vec<usize>, String)> (multi-dimensional — the newer, better API). The consumers are adapted to master's signature by folding the dims into a flat element count, rather than reverting the API.

Two silent behaviour losses also fixed

• parse_let_destructuring routed let (a, b) = expr to a StmtAssign + ExprArrayLiteral{extra_kind="tuple"} shape that only the Verilog backend reads. Rust/C/Zig read the older StmtLocal + extra_field shape, so Rust emitted vec![s, d] = dm(a, b); — uncompilable. Routed back through parse_local_decl, whose shape all four backends already lower.
• parse_type_annotation joined tuple elements with , instead of the canonical , , so -> (u32, u32) became -> (u32,u32).

Result

	before	after
cargo build --release	27 errors	0 errors
unit tests	could not run	1528 / 1535 pass

FROZEN_HASH resealed (M5), per FROZEN.md §5.

Still red — needs the wave-loop authors

Seven tests remain failing. They are feature-level losses from the same merge, all in Verilog hardware lowering:

tests_w457_ram_style::ram_style_block_pragma_emitted
tests_w457_ram_style::ram_style_distributed_pragma_emitted
tests_w458::array_param_read_emitted
tests_w458::no_translate_off_comments
tests_w458::tests_w459::array_param_bound_from_test_block
tests_w458::tests_w459::test_block_emits_real_function_call
tests_hir_pipeline_parity::test_verilog_keyword_local_and_module_escaped

I deliberately did not reconstruct these. Guessing at RTL lowering semantics produces output that looks plausible and synthesizes wrong — strictly worse than a visibly failing test. Leads are in #1787: for ram_style the pragma never reaches the module-level var declaration though every other piece survives; for array_param the binding is written and read but the parameter is never rewritten to the bound module array.

These seven fail on master today too — they simply could not be observed, because nothing compiled.

🤖 Generated with Claude Code

[github-actions]

📓 NotebookLM Notebook linked to this PR

• Notebook ID: b83263109fb055dc
• URL: https://notebooklm.google.com/notebook/b83263109fb055dc

This notebook contains session context, decisions, and artifacts for this work.

[github-actions]

PR Dashboard

Generated at: 2026-08-06 12:15:33 UTC

• [GOLD-RING] fix(compiler): restore declarations dropped by batch merge #1783 — master does not compile #1788 [GOLD-RING] fix(compiler): restore declarations dropped by batch merge #1783 — master does not compile | Created: 2026-08-06 | Updated: 2026-08-06 | 8 failing / 22 total
• feat(gen-verilog): spec-first ternary designs synthesize to real Artix-7 hardware (on_clock/on_comb data ports) (#1764) #1786 feat(gen-verilog): spec-first ternary designs synthesize to real Artix-7 hardware (on_clock/on_comb data ports) (#1764) | Created: 2026-08-06 | Updated: 2026-08-06 | 0 failing / 3 total
• fix(codegen): faithful Rust/C emit — don't run optimizer on source backends (#1455) #1456 fix(codegen): faithful Rust/C emit — don't run optimizer on source backends (#1455) | Created: 2026-07-10 | Updated: 2026-07-10 | 3 failing / 30 total

Summary

Status	Count
Total Open PRs	3
PRs with Failing Checks	1
PRs with All Checks Green	2
READY	1
FAILING	1
PENDING	0

Seal Status

• ⚠️ STALE -- sha256(compiler.rs)=66ce0d11508e != manifest seal=87e5cbd3ad94.
  The committed NMSE numbers were certified against an older compiler.rs.
  Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).