Regenerate gql types - #2064
Conversation
The GraphQL Schema Check (codegen:check) introspects the live schema at api.estuary.dev and diffs it against the committed src/gql-types. Production added DataPlane.id, an IdFilter input, and a DataPlanesFilter.id field, so the checked-in schema.graphql and graphql.ts went stale and the check failed on every open PR. Regenerated with npm run codegen.
⚪ Code HealthNo change to the dead-code surface. 48 Unused files
66 Unused exports
30 Unused exported types
14 Unused exported enum members
5 Unused dependencies
3 Unused devDependencies
|
GregorShear
changed the title
…ash, stale e2e assertion - useStorageMappings now walks every page via useAllPages instead of taking the server's silent 50-row default page, so the edit dialog can resolve any clicked row and PrefixCard's duplicate/coverage validators see the complete list. A dedupe by catalogPrefix guards against a graphcache invalidation re-appending the refetched page. - Gate the edit dialog's notFound self-close on the storage mappings loading flag (deep-link race) and on query errors (notFound previously closed the dialog before the error state could render). - Drop refetchMappings from create/update: graphcache already invalidates every storageMappings query on those mutations, and a first-page-only network refetch is misleading now that the query paginates. - Guard row.spec.data_planes in the settings table and type it optional: the server omits the field from the spec JSON when the list is empty (serde skip_serializing_if), which crashed the whole Settings section on such a mapping. - Remove the Playwright assertion on the removed "Last Updated" column. The GraphQL StorageMapping type carries no timestamp field, so the column cannot come back client-side. - gql-types regenerated for the query change only; the DataPlane.id/IdFilter schema drift stays in #2064.
2 participants
No description provided.