fix(wire-format): flip sistent consumer reads of meshery wire to canonical camelCase (Phase 7c)

[leecalcote]

Summary

Sistent components consume meshery / meshery-cloud wire payloads. Per the ecosystem identifier-naming contract (TypeScript property: camelCase everywhere, see meshery/schemas docs/identifier-naming-contributor-guide.md), consumer reads must match the canonical camelCase wire keys that meshery-cloud emits post Phase 4. This change is a surgical field-rename sweep — no behavioural or structural changes.

This is the sistent half of Phase 7c.

Flipped (45 sites across 20 files)

Catalog / design wire (Pattern, MesheryPattern, CatalogContent):

• catalog_data → catalogData
• content_class → contentClass
• published_version → publishedVersion
• pattern_file → patternFile

Pagination wire (MesheryViewPage / TeamsPage / EnvironmentsPage / DesignsPage / UsersPage / EventsPage):

• total_count → totalCount

Role wire (Role):

• role_name → roleName — response read in InviteUserModal only. The order: 'role_name asc' request parameter is left as-is until the cloud server adds dual-accept on the sort key.

Component / model wire (Component):

• display_name → displayName

Files changed

• src/custom/CustomCatalog/CustomCard.tsx — Pattern interface and reads
• src/custom/CustomCatalog/Helper.ts — getVersion() reads
• src/custom/CatalogDetail/OverviewSection.tsx
• src/custom/CatalogDetail/RelatedDesigns.tsx
• src/custom/CatalogDetail/LeftPanel.tsx
• src/custom/CatalogDetail/CaveatsSection.tsx
• src/custom/CatalogDetail/ActionButton.tsx
• src/custom/CatalogCard/CatalogCard.tsx
• src/custom/PerformersSection/PerformersSection.tsx
• src/custom/Workspaces/WorkspaceTeamsTable.tsx
• src/custom/Workspaces/WorkspaceViewsTable.tsx
• src/custom/Workspaces/EnvironmentTable.tsx
• src/custom/Workspaces/WorkspaceRecentActivityModal.tsx — also flipped local EventsResponse type
• src/custom/Workspaces/DesignTable.tsx
• src/custom/Workspaces/hooks/useDesignAssignment.tsx
• src/custom/Workspaces/hooks/useEnvironmentAssignment.tsx
• src/custom/Workspaces/hooks/useTeamAssignment.tsx
• src/custom/Workspaces/hooks/useViewsAssignment.tsx
• src/custom/UsersTable/UsersTable.tsx — totalCount only
• src/custom/DashboardWidgets/GettingStartedWidget/InviteUserModal.tsx

Intentionally retained as sistent-internal contract (not meshery wire)

• src/hooks/useRoomActivity.ts — provider_url, data.user_map are sistent's collaboration WebSocket signaling protocol identifiers, not meshery wire fields. The signaling server (meshery-cloud collaboration endpoint) speaks this protocol with sistent and is independent of the REST API wire-format contract.
• src/custom/CollaboratorAvatarGroup/CollaboratorAvatarGroup.tsx — User.border_color is part of sistent's exported Users prop shape (the sistent API surface that consumers pass in). Flipping this would be a public API break, not a wire-format alignment.

Already retained-as-deferred per prior phases (no change in this PR)

• src/custom/UsersTable/UsersTable.tsx — fullUser.organization_with_user_roles?.role_names deferred until the containing response type is migrated upstream (existing inline TODO).
• src/custom/TeamTable/TeamTableConfiguration.tsx — team_names / team_name deferred until meshery-cloud bulk-delete handler lands dual-accept on the request body (existing decision from PR fix(schemas): Phase 2.K — align Sistent with canonical camelCase wire format #1431 / cascade-10).

Test plan

• npx tsc --noEmit — no new errors vs baseline of 105 pre-existing errors (all unrelated: missing @types/jest, @types/node, @types/js-yaml, @sistent/mui-datatables declaration, etc.).
• npm run lint — clean.
• npm test — 21/21 tests passing across 7 suites.
• npm run build — tsup build succeeds (CJS + ESM + DTS).
• Final grep sweep confirms no remaining snake_case wire-format reads in scope. The four retained reads (useRoomActivity user_map, CollaboratorAvatarGroup border_color, UsersTable organization_with_user_roles.role_names, TeamTableConfiguration team_names/team_name) are documented above with rationale.

Refs

• meshery/schemas docs/identifier-naming-contributor-guide.md — the 26-row canonical naming directory
• Phase 2.K cascade (fix(schemas): Phase 2.K cascade — flip deferred 10 wire keys #1435) — prior wire-format alignment in sistent

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

[leecalcote]

Phase 8a Post-Release Integration Validation Results — 2026-04-28

Ran an end-to-end validation against the v1.0.14 / sistent v0.20.1 / meshkit v1.0.7 / schemas v1.2.6 ecosystem. Overall: GREEN with two follow-up findings, neither blocking.

What passed

• Builds: meshery-cloud go build ./server/..., meshery go build ./..., meshkit go build ./..., sistent npm run build — all clean.
• Go server unit tests: meshery (22 packages), meshkit (all packages) — all pass. meshery-cloud all packages pass except one DAO test (Finding [chore] Migrate Storybook from Meshery to Sistent repo #2 below).
• UI tests: meshery-cloud/ui jest — 92 suites / 650 tests pass; meshery/ui vitest — 5 suites / 34 tests pass; sistent jest — 21 suites / 63 tests pass.
• Type checks: meshery-cloud/ui & meshery/ui pass clean. sistent has 10 pre-existing type errors in theme/palette.ts and tsup.config.ts (last touched 2025-09, unrelated to this overhaul).
• Schema audit: make validate-schemas and consumer-audit both report zero blocking violations — the canonical-casing migration is clean across all consumers.

Two findings (both follow-up, neither blocking)

Finding #1 — Schemas pin drift (HIGH-COORDINATION): All three Go modules (meshery-cloud, meshery, meshkit) pin github.com/meshery/schemas v1.2.2 while the published tag is v1.2.6. The cloud integration test TestPublishedDocsOpenAPIVersionNotAhead correctly catches this:

docs openapi.yml version v1.2.6 is ahead of repo schemas pin v1.2.2

Runtime is fine — the canonical-casing migration was implemented in consumers via local-flip PRs and v1.2.3..v1.2.6 doesn't break the v1.2.2 ABI — but the schema-driven contract demands consumers track the latest. Suggested hotfix: one PR per Go repo bumping schemas v1.2.2 → v1.2.6 + go mod tidy, dispatched in order schemas → meshkit → meshery-cloud → meshery.

Finding #2 — Seeder-bypass NULL joined_at (MEDIUM-MAINTENANCE): make db-user-init invokes the legacy SQL function public.create_and_assign_my_team(uuid) (defined in the 2019 initial-schema migration) which inserts into users_teams_mappings without the new joined_at column. The Go DAO path (AddUserToTeam, CreateTeamAndAssignToUser) correctly stamps joined_at, but the SQL bypass leaves 7 NULL rows on every fresh seed. The migration 20260424000002_add_users_teams_mappings_joined_at backfills idempotently on next server boot, so production self-heals — but the DAO test TestUsersTeamsMappingsJoinedAtBackfilled fails on freshly-seeded test DBs until a server start sweeps the NULL rows. Suggested hotfix: CREATE OR REPLACE FUNCTION public.create_and_assign_my_team(...) in a new migration adding joined_at to the INSERT column list and now() to the values list.

Out-of-scope follow-ups (not regressions)

• Sistent module-augmentation type errors in palette.ts (pre-existing since 2025-09).
• make db-init does not create/migrate the kratos & hydra Postgres DBs — every new developer hits this on integration-test runs. Worth wiring into db-init or documenting in docs/development.md.
• 60+ duplicate code 'NNNN' - skipping export warnings from make error (pre-existing, benign).

Verdict

Ship v1.0.14 as-is. Plan a v1.0.15 hotfix bundling the schemas-pin bump and the SQL-function fix. Both fixes total <20 lines each. The canonical-casing migration is clean per the consumer-audit — the most important contract gate the user-facing audit could check.

Full report: /tmp/phase8a-validation-report.md on the validation host.