The architect approved the four ADR-fix tickets. All four change what the architecture documents claim. No code changes. ADR-0001, ADR-C-001: the worked example claimed the POC has no real backend. It rewrites against `backend/src/BigRegister.Api`. Every path it named is repointed. The out-of-scope list drops two discharged bullets: 33 `parse*` boundaries exist, and `npm run gen:api` is real. ADR-0001, ADR-C-003: a new section states that the generated client is the wire contract. A hand-written `contracts/*.dto.ts` is the exception for two cases only. The four survivors stay, because NSwag emits every property as optional and flattens `RegistrationStatusDto` into five optional strings. The `parse*` trust boundary stays mandatory, because a generated type is a compile-time claim about the wire and not a runtime guarantee. ADR-0003, ADR-C-007: four paths moved in WP-67 and are repointed. Point 4 kept the principle and changed its example to `skeleton` and `spinner`. Two of its claims were false and the amendment says so: `app-alert` wraps the vendored `.feedback` classes, and `site-header` composes the vendored `.titlebar`. ADR-0004, ADR-C-009: the exception section states a four-part test instead of one named exception. `OrgTemplateStore` and `FeatureFlagStore` both pass it. RB-07 gated this ticket, because clause 4 needs an audited allow path. RB-07 landed that, so the ADR does not ratify a control that the code lacks. Three tickets need a matching CLAUDE.md correction in the same diff. CLAUDE.md section 2 loses the false `alert` example. Section 4 gets the generated-client rule and the four-part test. Two findings were wrong. ADR-C-001 asked to keep an out-of-scope bullet that reads "SessionStore is in-memory". The session persists to `localStorage` now, so the bullet covers multi-tab sync only. ADR-C-007 flagged one half of point 4 and missed that the other half is equally false. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
4.4 KiB
ADR-C-003 — state that the generated client is the wire contract
Status: implemented · 2026-08-27 · Source finding: 06-adr-conformance.md §ADR-C-003
What was wrong
ADR-0001 set "one source of truth that generates types for both sides" as the target state.
The code reached it. CLAUDE.md §4 still stated the pre-codegen rule — "DTO lives in
contracts/" — as standing law, so §4 could be cited to justify both deleting the four
survivors and adding new hand-written DTOs for already-generated endpoints.
What changed
| File | Change |
|---|---|
docs/reference/architecture/0001-...md |
new §"Where the contract lives, after codegen" |
CLAUDE.md §4 |
the flat "DTO lives in contracts/" rule replaced with the generated-client rule + the two exceptions |
No code changed. Per CLAUDE.md's own precedence rule, the ADR was amended first and CLAUDE.md corrected to match, in one diff.
The decision the finding asked for: the four survivors stay
ADR-C-003 required an explicit, recorded decision on the four remaining hand-written DTOs. They stay, all four under exception case 2 ("the generator types the shape too loosely"). This is not a preference — adopting the generated shapes would violate CLAUDE.md §3.
Evidence. NSwag emits every property as optional, and flattens a discriminated union into a bag of optional fields:
generated (api-client.ts) |
hand-written (dashboard-view.dto.ts) |
|
|---|---|---|
DashboardViewDto |
registration?, person?, decisions? — all optional (:2017) |
all three required |
RegistrationDto |
six optional fields (:2202) |
six required fields |
RegistrationStatusDto |
one flat record of five optional strings, tag?: string (:2211) |
a real union of three variants, tag: 'Geregistreerd' | 'Geschorst' | 'Doorgehaald', per-variant fields required |
The generated RegistrationStatusDto makes { tag: 'Geregistreerd', doorgehaaldOp: '…' }
representable. That is precisely the illegal state CLAUDE.md §3 exists to forbid, and the
parse* boundary would have to reconstruct the union by hand anyway.
The ADR therefore records that retiring these four is not a cleanup to schedule. It becomes correct only if the backend annotates its DTOs so the generator emits required properties and real unions — which names the actual prerequisite instead of leaving the question open.
Verified counts, not carried over from the finding
- Hand-written
contracts/*.dto.ts: 4 —apps/ssp/src/app/registratie/contracts/{brp-address,dashboard-view,duo-diplomas}.dto.tsandlibs/beheer/src/contracts/stamdata.dto.ts. - All four duplicate generated types by the same names:
BrpAddressDto(:1997),DashboardViewDto(:2017),DuoLookupDto(:2056),DuoDiplomaDto(:2047),PolicyQuestionDto(:2168),ManualDiplomaPolicyDto(:2106),StamdataColumnDto(:2246),StamdataTableDto(:2253),StamdataTableSummaryDto(:2261). None is a codegen gap — the finding's "case 1" has no occupant today, which is worth knowing. - The
parse*boundary is restated as mandatory regardless of type provenance. The amendment says why in one line: a generated type is a compile-time claim about the wire, not a runtime guarantee.
Gate released
ADR-C-003 blocked any ticket that would delete the four contracts/*.dto.ts files or add a
hand-written DTO for a generated endpoint. No open ticket needed it. The rule is now written
down, so a future one can be judged against it rather than against a stale §4.