Commit Graph

8 Commits

Author SHA1 Message Date
bb9242fd6e refactor(form-field): resources + snapshot + validator read form_field_options
Atomic reader switch. All call paths that previously read
form_fields.options / form_field_library.options from the JSON column
now read through FormFieldOptionService::toJsonShape() via the
morphMany relation:

  - FormFieldResource + FormFieldLibraryResource +
    PublicFormSchemaResource emit the rich-shape array
  - FilterRegistryController emits rich shape uniformly (no flat-array
    carve-out for filter-UI compatibility — preflight scan confirmed
    zero portal/app consumers, S5 territory)
  - FormFieldRuleBuilder plucks values from the relation for in:options
    rule construction
  - FormSubmissionService::buildSnapshot writes rich-shape options into
    snapshots and strips translations.{locale}.options from each field's
    translations bag (defensive — commit 2 backfill already did the
    bulk strip)
  - Four FormFieldRequest variants accept array-of-spec-objects,
    validate shape in after() via FormFieldOptionService::assertSpecsValid,
    and hand off to FormFieldOptionService::replaceOptions for writes
  - FormFieldService::create + update extract option specs from the
    request data and route through the service after the FormField row
    is persisted

FormField and FormFieldLibrary $casts no longer include 'options'; the
JSON column is no longer cast. Options removed from $fillable on both
models so ::create() / ::fill() / mass assignment can no longer touch
the legacy column. Both models gain a getOptionsAttribute() accessor
that resolves $model->options to the eager-loaded morphMany collection
— required because Eloquent's getAttribute() prefers a real DB column
over a relation method, and the JSON column lives on the table until
WS-5d commit 5 drops it.

Activity log — dual emit per §6.7 / §17.4.2 / §17.6.3:
  - field.updated carries old.options / new.options diff via
    toJsonShape() reconstruction, byte-equal JSON compare to avoid
    cosmetic false positives. Field updates that don't touch options
    omit the key entirely
  - field.options_replaced emits inside replaceOptions() on FormField
    subject only; library subject writes silent (mirrors the WS-5b /
    WS-5c convention)

JSON columns (form_fields.options, form_field_library.options) remain
present but unread — column drops land atomically in commit 5.

Two pre-existing test fixtures that seeded options via the JSON column
(FormFieldApiTest + PublicFormValidationTest) migrated to the
spec-array path: FormField::factory()->withOptions([...]) where the
options live on the field, or explicit spec-array request bodies for
HTTP tests.

Tests: 1193 → 1206 green (+13 tests / +28 assertions).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-25 02:33:21 +02:00
d06ea01b09 feat(form-builder): FormFieldConditionalLogicService + cycle detection + legacy backfill + snapshot
WS-5c commit 2 of 4 — the service layer, backfill migration, and
read-path switch. Per addendum Q3, conditional_logic applies to
FormField only — no library mirror and no copyLogic on
FormFieldService::insertFromLibrary.

FormFieldConditionalLogicService owns every write:
  - logicFor(field): depth-limited eager-load of the tree
  - replaceLogic(field, tree): transactional structure + operator +
    field_slug validation + cycle check + activity-log emit
    (field.conditional_logic_replaced)
  - toJsonShape(root): reconstructs the canonical ARCH §8
    `{show_when: {...}}` shape — single source of truth for the
    snapshot writer + API resources
  - assertSpecsValid(tree): public boundary guard for the FormRequest
    strict validator (WS-5c commit 3 wires this up)
  - assertNoCycles(field, tree): contract preserved from
    FormFieldService::assertNoConditionalCycle, implementation now
    reads the relational adjacency.

Backfill migration translates pre-WS-5c conditional_logic JSON to
rows. Strict dispatch: unknown operators / unknown top-level keys /
malformed groups FAIL the migration — Phase A seed-scan confirmed
the catalogue parity, so any drift is a data bug to fix at source,
not silently absorb. Rollback rebuilds canonical JSON and clears
the relational tree.

FormFieldService.create/update route `conditional_logic` through
the new service (matching the extract-and-delegate pattern from
WS-5a bindings and WS-5b validation rules). Snapshot writer + both
resources (FormFieldResource, PublicFormSchemaResource) read via
`toJsonShape(rootConditionalLogicGroup())` — byte-for-byte parity
with the pre-WS-5c JSON contract.

InvalidConditionalLogicSpecException handled in FormFieldController
as 422, same as FrozenSchemaException / CyclicDependencyException.

Tests: 20 new under tests/Feature/FormBuilder/ConditionalLogic/
(service, cycle detection, backfill forward+rollback+failure cases,
snapshot + resource parity). FormFieldApiTest cyclic rejection test
rewritten to use the new factory state. Rollback step counts in
WS-5a/b migration tests bumped +1 for the new backfill migration.
Baseline 1122 → 1142 green (3032 → 3085 assertions).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-24 23:56:39 +02:00
d494478c08 feat(form-builder): form_field_configs relational table + non-validation key split + drop validation_rules JSON columns 2026-04-24 22:42:35 +02:00
800b1b6c01 feat(form-builder): FormFieldValidationRuleService + legacy backfill + snapshot + library row-copy 2026-04-24 22:12:08 +02:00
b688ec26f0 feat(scope): declarative FK-chain strategy for OrganisationScope, register on 14 models per addendum Q2 + D-03/D-04
Refactors OrganisationScope to support a declarative, recursive FK-chain
resolver and registers the scope on 14 models that previously relied on
caller-discipline for tenant isolation.

Scope resolver (app/Models/Scopes/OrganisationScope.php):
Models now declare their strategy via:

    public static function tenantScopeStrategy(): array
    {
        return ['column' => 'organisation_id'];           // terminal
        // OR
        return ['via' => FormSchema::class, 'fk' => 'form_schema_id'];
    }

The apply() path walks the chain recursively, building whereIn subqueries
against parent models until it hits a column-based strategy. Max 3 hops;
deeper chains raise App\Exceptions\TenantScopeResolutionException. The
walker accepts BOTH the new tenantScopeStrategy() and the legacy
$organisationScopeColumn property at every hop — so PersonIdentityMatch
can chain via Person, which still uses the legacy event_id bridge, without
requiring Person/Event/Shift/FestivalSection/TimeSlot to migrate to the
new convention in this work package. That migration is a separate
backlog ticket — explicitly scope-controlled per the addendum.

Fourteen newly-scoped models:

  Form-builder child models (D-03):
    FormSchemaSection             via FormSchema                    (1 hop)
    FormField                     via FormSchema                    (1 hop)
    FormSubmission                column organisation_id (Commit 2)
    FormValue                     via FormSubmission                (1 hop)
    FormValueOption               via FormValue -> FormSubmission   (2 hops)
    FormSubmissionSectionStatus   via FormSubmission                (1 hop)
    FormSubmissionDelegation      via FormSubmission                (1 hop)
    FormSchemaWebhook             via FormSchema                    (1 hop)
    FormWebhookDelivery           via FormSubmission                (1 hop)

  Event-data models (D-04 event-data subset):
    ShiftAssignment               via Shift (legacy festival_section_id)
    ShiftWaitlist                 via Shift
    VolunteerAvailability         via TimeSlot (legacy event_id)
    PersonSectionPreference       via FestivalSection (legacy event_id)
    PersonIdentityMatch           via Person (legacy event_id)

Note — task directive specified VolunteerAvailability "via: Event, fk: event_id",
but the table has no event_id column (only person_id + time_slot_id).
Rerouted via TimeSlot, which carries the legacy event_id bridge; same
end result, correct FK.

Security-relevant callers made explicit:
  PublicFormSchemaResource::toArray() now eagerly loads fields + sections
  with withoutGlobalScope(OrganisationScope::class). Prior to this commit
  the public form endpoint silently relied on those relations being
  unscoped. The PublicFormCrossOrgScopeTest pre-existing assertions still
  pass — behaviour unchanged, intent now explicit.

Test fix: FormSchemaApiTest::test_publish_sets_is_published_true was
flaky (factory randomly picked EVENT_REGISTRATION which requires
bindings). Pinned to USER_PROFILE for determinism; PurposeSchemaLifecycleTest
covers the binding-enforcement path.

Test flip: MultiTenancyTest::test_form_schema_webhook_is_not_globally_scoped
renamed to is_scoped_via_fk_chain and asserts the new behaviour: scope
filters by route org, withoutGlobalScope() still exposes cross-org rows.
The test's original purpose ("pin current behaviour so a future refactor
is intentional") is now satisfied by Commit 3 being that intentional
refactor.

Docs:
  SCHEMA.md §3.5.11 Rule 5 — tenantScopeStrategy() convention documented;
    the 14 newly-scoped models enumerated; link to addendum Q2.
  ARCH-FORM-BUILDER.md §4.14 — new section "Multi-tenancy scope chain"
    with the hop-count table for all 14 chains and the withoutGlobalScope
    pattern for cross-org callers.

Tests: tests/Feature/MultiTenancy/ScopeLeakageTest.php — two orgs with
fully-populated record chains down to each of the 14 leaf models; asserts
scoped queries never cross, withoutGlobalScope still does. Plus: three-
hop chain (FormValueOption) explicitly exercised, legacy-column bridge
verified, over-deep chain raises TenantScopeResolutionException. 16 tests /
31 new assertions. Full suite: 1000 passed (2706 assertions).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-24 17:08:33 +02:00
6ba921442c fix(form-builder): explicit OrganisationScope bypass on every public-form query
Five models that the public form endpoints touch carry a global
OrganisationScope: FormSchema, Event, TimeSlot, FestivalSection,
PersonTag. The initial S2c implementation relied on the scope no-opping
because /public/forms/* has no `{organisation}` route parameter and
OrganisationScope::resolveOrganisationId returns null in that case.

That's accidentally-correct. Any middleware that sets an implicit org
context later (route model binding for platform admin, impersonation,
default-org fallback on an authed Sanctum session) would start
filtering public schema resolution by the wrong org.

- PublicFormTokenResolver: both FormSchema::query() calls now pass
  withoutGlobalScope(OrganisationScope::class). public_token is
  globally unique so this is safe.
- PublicFormController::timeSlots() / sections() / festivalEventIds():
  Event, TimeSlot, FestivalSection queries all explicit now, including
  the eager-loaded event relation on time-slots.
- PublicFormController::ownerEvent(): narrowed from
  Event::withoutGlobalScopes() to withoutGlobalScope(OrganisationScope)
  so future scopes (soft-delete, archived) aren't accidentally
  stripped.
- PublicFormSchemaResource::availableTagsByCategory: same narrowing on
  the PersonTag query.

PublicFormCrossOrgScopeTest pins the expectation — 4 cases hit every
public endpoint under a stashed foreign-org route parameter and assert
the owner-org data still surfaces. Verified the tests fail when the
fix is reverted (all 4 return `SCHEMA_NOT_FOUND` with the bypass
absent).

Full suite 893 → 897 green.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-17 23:16:22 +02:00
e4294702c5 feat(form-builder): public resources — TAG_PICKER tags, schema version, submission shape
S2c D1, D5, D7.

PublicFormSchemaResource:
- Carries available_tags on every TAG_PICKER field, respecting
  validation_rules.tag_categories when set. Tags prefetched once per
  org to avoid N+1.
- Surfaces the schema version and an opened_at timestamp so the portal
  can detect drift between GET-time and submit-time.

PublicFormSubmissionResource (new): the submission envelope the portal
sees on every lifecycle endpoint. No PII echo — public_submitter_name,
public_submitter_email, public_submitter_ip never appear. Admin
metadata (review_status, reviewed_by, schema_snapshot, submitter user
id) is filtered out. Includes schema_drift computed from
schema.version vs schema_version_at_submit, and identity_match read
from the §31.1 listener's column with a Dutch message per state
(pending|matched|none).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-17 22:55:54 +02:00
4b7e66b83f feat(form-builder): API resources with FieldAccessService filtering
Phase 4 of S2b. Nine resources that shape the universal form builder
responses. FieldAccessService::filterVisibleFields gates every field
array — the primary defence tested by FormResourceSecurityTest (§22.9).

- FormSchemaResource: includes fields_count, submissions_count,
  has_submissions, is_locked (derived from edit_lock_*), public_form_url
  when public_token is set, and filtered fields collection.
- FormSchemaSummaryResource: lean list-endpoint variant.
- FormFieldResource: effective_label / help_text / options resolved via
  FormLocaleResolver + translations JSON, plus TAG_PICKER available_tags
  filtered by validation_rules.tag_categories.
- FormSubmissionResource: values keyed by field slug with FieldAccessService
  filtering, section_statuses, active delegations, review_info,
  submitted_in_locale, submission_duration_seconds.
- FormSubmissionSummaryResource: lean list variant.
- FormTemplateResource, FormFieldLibraryResource.
- PublicFormSchemaResource: strictly limited per §10 — only
  is_portal_visible=true AND is_admin_only=false fields, no PII hints,
  no role_restrictions, no submissions_count.
- FormSchemaWebhookResource: url/secret never returned; only url_host +
  has_secret boolean.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-17 21:13:40 +02:00