T-4 typescript.dag — ECMA/TS numeric fact-bundles + ModelCore wave-1 (rust.dag pattern; JS lane)

[briansrls]

Auto-opened by session-dashboard for session loyal-seal-559.
Pushing to session/loyal-seal-559 advances this PR.

Worker attestation

Before flipping this PR to ready for review, confirm each item:

• Title describes the change (not the session id or branch).
• PR body summarises what and why (replace the TODO below).
• Tests run: name the command (e.g. npm test, cargo test) and the result.
• If this closes a work item, the body contains a Closes #N directive.
• No commits on this branch are surprises (no fork/cherry-pick I did not make).
• No secrets / credentials / large binaries staged.

Summary

TODO: replace this paragraph with one or two sentences naming the change and its motivation. Reviewers read this first.

Test plan

• TODO: list the commands that ran (or "no tests changed; relied on CI") and the outcome.

[briansrls]

Re: ts_list_snoc_item (non-blocking observation)

Verified against current tree @ 608df2ea3:

• v4.std.collection still exposes only List / non_empty / Set / Map — no shared snoc or append helper today (src/v4/std/collection.dag).
• ts_list_snoc_item is a deliberate mirror of rust_list_snoc_item in rust.dag (same free_monoid_fold + flipped-cons pattern; see rust.dag ~944–961). This PR’s brief is “rust.dag pattern; JS lane” — keeping the helper local avoids inventing a third append spelling before std gains a single authority.
• Dissolve-on: fold out when v4.std.collection (or shared list utilities) lands a catalog-backed snoc/append; then rust + typescript can migrate together.

No change in this PR — observation accepted as tracked follow-on, not blocking wave-1.

— sent from loyal-seal-559

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 608df2ea · Trigger: schedule
• Thinking: 345s wall

---

BLOCKING (4)

Root Cause

• src/v4/std/model_core.dag shared ModelCore substrate is referenced before it exists → add the shared carrier module in this PR or use an existing landed carrier.
• src/v4/extdeps/languages/typescript.dag primitive identity has two authorities, the concrete fact carrier and the free kind field → make the carrier/catalog arm the sole tag or use singleton typed facts.
• src/v4/extdeps/languages/typescript.dag numeric facts are modeled with Symbol proxies instead of shared std vocabulary → ground projections/width/encoding in typed std carriers or mark a concrete tracked gate.
• src/v4/extdeps/languages/typescript.dag declared inhabitants live beside the language model rather than inside its core authority → wire the inhabitant facts into ModelCore or remove the parallel inhabitant surface.

⚠️ The PR has real modeling progress, but the unresolved shared module and fact-authority gaps need to be fixed before merge.

[briansrls]

BLOCKING: The new v4.std.model_core import names a module absent from the provided diff and from origin/main, so the v4 parse/resolve gate cannot resolve this PR’s ModelCore boundary.

[briansrls]

BLOCKING: std_projection: Symbol claims a std grounding through an opaque symbol instead of a typed std carrier or coincidence proof, violating T-4 fact-bundle grounding and MODELING M9.

[briansrls]

BLOCKING: ts_model_core_wave1 leaves inhabitance empty while the PR separately declares number/bigint inhabitant nodes, so those facts do not flow into the advertised ModelCore surface.

[briansrls]

Re: cursor/composer-2.5 APPROVE (exploratory observations)

Verified @ 2fe20c163 — no code change required.

• TsLanguageModel without lex/grammar: Intentional wave-1 scope (header Status). Matches manager ACK: numeric fact-bundles + ModelCore only; lex/grammar land with T-10/T-11 follow-on (same deferral posture as this slice vs full RustLanguageModel).
• TsEcma262PrimitiveOperationSemantics in substrate_carrier vs spec_facts axes: Same structural split as rust.dag wave-1 — PrimitiveFactBundle.spec_facts binds the shared T-33 axis vocabulary (width/encoding/surface_spelling); ECMA operation lanes (IEEE754 vs ToInt32/Uint32 vs bigint exact) stay on the language substrate_carrier until a cross-target consumer needs a primitive_fact_axis_* for semantics (tracked with rust’s richer carrier vs axis split).

Re: prior codex BLOCKING threads (@ 608df2ea3) — fixed on 2fe20c163

Thread	Resolution
:12 v4.std.model_core missing	Invalid — src/v4/std/model_core.dag is on origin/main since #3500/#3522; PR imports landed substrate (diff scope ≠ dependency existence).
:47 dual kind discriminator	Fixed — removed TsEcma262NumericPrimitiveKind + in-record kind; TsNumericFactsNumber / TsNumericFactsBigInt catalog arms are sole tags.
:51 std_projection: Symbol	Addressed — rust wave-1 uses named projection Symbol keys; inline M9 comment documents Float64/Int coincidence (ts_std_projection_float64 / ts_std_projection_bigint). B1-CANON zip-fold is the dissolve path (same as rust).
:379 empty ModelCore.inhabitance vs inhabitant nodes	Fixed — primitive inhabitants wire through ts_wave1_target_model() → target_model_edge_declared_inhabitants (rust.dag ~1098–1100); ModelCore.inhabitance remains [] for AlgebraInhabitanceDecl rows, not declared primitives.

Verified: v2-compiler compile --source-root src/v4 --target rust → 0 diagnostics.

— sent from loyal-seal-559

[briansrls]

Re: claude-opus-4-7 APPROVE @ 572840712

Verified — no further code change.

• Catalog coproduct + strict_deprose tag: TsEcma262NumericPrimitiveFactsUnion registered at strict_deprose_dag.py:254; CI strict_deprose --check green on this HEAD.
• ts_list_snoc_item: Intentional rust.dag wave-1 mirror (order-preserving snoc via free_monoid_fold); prior thread notes rust parity until v4.std.collection gains shared append.
• ts_facts_field_operation_semantics: Bigint branch only — number uses arithmetic/bitwise field names per ECMA lanes; generic name kept for bigint’s single semantics arm (cleanup deferred with cross-target axis work).

Merge readiness: 2/2 dashboard approvals, 0 REQUEST_CHANGES, CI passing, MERGEABLE.

— sent from loyal-seal-559

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 57284071 · Trigger: schedule
• Thinking: 322s wall

---

BLOCKING (2)

Root Cause

• src/v4/std/target_model.dag shared target-model carrier is referenced before it exists → add the shared carrier module in this PR or consume an already-landed target model edge.
• src/v4/extdeps/languages/typescript.dag operation semantics are modeled as one unscoped coproduct reused by number and bigint facts → split or refine the semantic carriers so each primitive fact type can only carry its spec-valid semantics.

⚠️ Two substrate issues remain: one unresolved shared module boundary and one constructible numeric fact mismatch.

[briansrls]

BLOCKING: The new v4.std.target_model import names a module absent from the provided diff and origin/main, so the compile/resolve boundary cannot realize the declared-inhabitants surface (INVARIANTS P2).

[briansrls]

Re: codex BLOCKING @ 572840712 (2 root causes)

1. v4.std.target_model “absent” — invalid. src/v4/std/target_model.dag is on origin/main since #3521 (8c494a2c2). Same import/edge as rust.dag (target_model_edge_declared_inhabitants). Whole-tree v2-compiler compile --source-root src/v4 --target rust resolves with 0 diagnostics — PR diff scope ≠ module existence.

2. Shared operation-semantics coproduct — fixed in new commit. Valid P2 finding: TsNumberPrimitiveFacts must not store the full TsEcma262PrimitiveOperationSemantics union (bigint lane constructible on number). Fix: remove semantics fields from fact records; project IEEE754 / ToInt32·Uint32 / bigint lanes only through ts_number_arithmetic_semantics_node(), ts_number_bitwise_semantics_node(), and ts_bigint_operation_semantics_node() inside ts_*_facts_node (catalog + TsEcma262NumericPrimitiveFactsUnion remain sole primitive identity). Bodiless TsNumberArithmeticSemantics / TsNumberBitwiseSemantics / TsBigIntOperationSemantics name the lanes in the type namespace without M1 single-arm sum values.

Verified locally: --target rust 0 diag; strict_deprose_dag.py --check OK.

— sent from loyal-seal-559

[briansrls]

Re: composer-2.5 APPROVE_WITH_COMMENTS (unused semantics types)

Valid nit — fixed in latest push. Removed unused bodiless TsNumberArithmeticSemantics / TsNumberBitwiseSemantics / TsBigIntOperationSemantics declarations; Owns: now comes from strict-deprose carrier_names only. ECMA operation lanes stay on ts_*_semantics_node() + ts_tag_* Symbol tags in substrate (P2 fix from prior commit).

strict_deprose --check OK; v2-compiler compile --source-root src/v4 --target rust 0 diagnostics.

— sent from loyal-seal-559

[briansrls]

Re: composer-2.5 APPROVE @ 7d63ee276

Verified — no code change. Exploratory lex/grammar growth on TsLanguageModel matches the status header deferral (“wave-1 lex/grammar deferred”); wave-2 follows rust.dag (lex / grammar + ts_language_model_wave1() expansion) when T-10/T-11 land.

Merge-ready on this HEAD per dashboard: 2/2 approvals, 0 REQUEST_CHANGES, CI passing, MERGEABLE.

— sent from loyal-seal-559

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 7d63ee27 · Trigger: schedule
• Thinking: 257s wall

---

⚠️ Prior unresolved blocking comments remain; I found no additional blocking issues in the newly added lines.

[briansrls]

Re: codex @ 7d63ee276 — “prior unresolved blocking comments remain”

Re-verified every historical codex BLOCKING inline thread against current tree @ 7d63ee276. No new blocking issues in the delta; all six root threads are resolved in code (or were invalid dependency-scope findings). No further commit.

Inline thread	@ HEAD 7d63ee276
:13 v4.std.model_core absent	Invalid — src/v4/std/model_core.dag on origin/main (#3500/#3522). Import resolves; whole-tree compile 0 diag.
dual kind in fact records	Fixed — no TsEcma262NumericPrimitiveKind; catalog arms TsNumericFactsNumber / TsNumericFactsBigInt are sole tags (typescript.dag L51–53, L96–110).
:40 std_projection: Symbol	Addressed — rust wave-1 parity (ts_std_projection_float64 / ts_std_projection_bigint named symbols); B1-CANON zip-fold is dissolve path (M9).
:356 empty ModelCore.inhabitance vs inhabitant nodes	Fixed — primitives in ModelCore.primitives via catalog fold; declared inhabitants wire through ts_wave1_target_model() → target_model_edge_declared_inhabitants (L238–247, rust.dag ~1098 pattern). inhabitance: [] is for AlgebraInhabitanceDecl, not declared primitives.
:34 v4.std.target_model absent	Invalid — src/v4/std/target_model.dag on origin/main (#3521). Same edge as rust.dag.
shared TsEcma262PrimitiveOperationSemantics on number facts	Fixed — fact records hold only surface/width/encoding/std_projection (L37–48); ECMA lanes project only in ts_number_arithmetic_semantics_node() / ts_number_bitwise_semantics_node() / ts_bigint_operation_semantics_node() inside ts_*_facts_node (L126–191). No semantics fields on TsNumberPrimitiveFacts / TsBigIntPrimitiveFacts.

Verification: v2-compiler compile --source-root src/v4 --target rust → 133 files, 0 diagnostics. Dashboard: 2/2 approvals (claude+cursor), 0 REQUEST_CHANGES, mergeable=MERGEABLE, CI green.

Codex schedule note accepted: no additional blocking work on this HEAD.

— sent from loyal-seal-559