C source emitter Stage 0: author minimal c_target_model + run_c_smoke harness — emit 'int add(int,int){return x+y;}' + main, cc-compile, run, assert, --inject-fault RED

[briansrls]

Auto-opened by session-dashboard for session merry-heron-533.
Pushing to session/merry-heron-533 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.

[cursor]

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

[gunbai-bot]

Addressed review 39074 (coverage-completeness counter) in 10bd135 / 00126feb:

Finding verified — valid. emit_host_gate_declared_smoke_count had been bumped to 6, but emit_host_present_target_count() still counts only TargetModels with a configured runtime_row (rust, python, go, typescript, c → 5). cpp was demoted to target_emit_host_runtime_row_unconfigured; c replaces it rather than adding a sixth present target. The gate runs six smokes (including node_http, which is not a TargetModel runtime row), but the completeness invariant is present-target cardinality, not literal gate smoke invocations — same as on main where gate-only node_http and gate-excluded cpp cancelled to 5 == 5.

Fix: reverted emit_host_gate_declared_smoke_count to 5 in target_coverage_completeness_test.dag:87.

Also landed in the same commits (parent review follow-up, not blocking 39074):

• c_add_emit_translate_test.dag — emit/serialize translate witness + RED controls (missing translation rules, wrong node)
• c_cpp_realization_relationship_note on c.dag carrier

[gunbai-bot]

review 39075 — findings verified on current branch:

1. §3 nicknaming (c_source_text ≡ cpp_source_text) — accurate; already tracked on carrier via c_cpp_realization_relationship_note (c.dag:288) with dissolution trigger naming cpp re-home onto c / shared c_subset. No additional change in this PR — intentional 🟡 frontier, not a silent fork.

2. Byte-width RED coverage regression — valid flag. The deleted cpp_add_emit_host_run_test carried nonzero_exit + byte_width_mismatch wire-discipline REDs; initial c test only had green + inject-fault. Fixed in latest commit: ported both REDs to c_add_emit_host_run_test.dag via c_fixed_source_row (same cc/FiveWidthBytes pattern as the former cpp test). cpp remains honestly host_smoked: false with target_emit_host_runtime_row_unconfigured.

3. Transport vs execution test duplication — confirmed; matches established emit-host pattern (gate transport in dag/tools/emit_host_transport.dag, execution witness in src/v2/test/claim/execution/*_emit_host_run_test.dag). Not introduced here.

Thanks for the APPROVE.

[gunbai-bot]

review 39081 — findings verified:

1. §2 fixture duplication (emit_host_transport.dag vs c_add_emit_host_run_test.dag) — accurate; intentional gate-side Bool wrapper vs executor-side test fn split, same shape as other emit-host languages. Agreed this is Stage-1 consolidation when a second cc-runtime language reuses the fixture surface — not blocking here.

2. Weak inject-fault discrimination — valid. The value != 5 guard made add(2,2) exit before writing stdout, so RED passed via nonzero-exit rather than octet mismatch (despite c_emit_host_smoke_red_doc claiming the latter). Fixed in latest commit: removed the guard from c_emit_host_inject_fault_main_suffix so fault path emits [0,4,0,0,0] vs green oracle [0,5,0,0,0]; inject-fault witnesses now refuse on nonzero-exit alone.

Thanks for the APPROVE.

[gunbai-bot]

review 39087 — verified on current branch (0d8b232):

All affirmations hold: C Stage-0 mirrors the established emit-host pattern; inject-fault RED discriminates via stdout octet mismatch (add(2,2) → [0,4,0,0,0] vs green oracle [0,5,0,0,0]); main suffix carries c_emit_host_main_suffix_dissolution_trigger; cpp demoted to target_emit_host_runtime_row_unconfigured with host_smoked: false; §3 nicknaming tracked in c_cpp_realization_relationship_note. No code changes requested — acknowledged.

— sent from merry-heron-533

[gunbai-bot]

review 39088 — verified on current branch (0d8b232):

All affirmations hold: c_target_model owns grammar rows, lex, binding_spellings, and cc runtime_row; cpp correctly uses target_emit_host_runtime_row_unconfigured; scaffolds carry named dissolution triggers; smoke/execution witnesses include inject-fault octet mismatch, nonzero-exit, and byte-width RED controls. No merge-blocking issues identified — acknowledged.

— sent from merry-heron-533

[gunbai-bot]

review 39092 — verified on current branch (7292c5f):

No findings to address. All affirmations hold post-CI fix: grammar/lex/binding on c_target_model, cpp unconfigured, wet C smoke via emit_host_gate_passes/run_c_smoke() (green + inject-fault octet RED), hermetic translate tests in manual/, execution witnesses in offline execution/ lane, dissolution triggers + c_cpp_realization_relationship_note in place. The hermetic c_emit_host_smoke_test.dag witness was removed in 7292c5f (gate is the wet consumer).

— sent from merry-heron-533

[gunbai-bot]

review 39093 — verified on current branch (7292c5f):

No findings to address. Affirmations hold: c_cpp_realization_relationship_note (c.dag:620) names the §3 nicknaming risk + dissolution condition; cpp uses target_emit_host_runtime_row_unconfigured; gate signature/coverage/inject-fault RED consistent; main-suffix scaffolds carry c_emit_host_main_suffix_dissolution_trigger (c.dag:628).

— sent from merry-heron-533