Skip to content

data-model.md

Latest commit

 

History

History
82 lines (64 loc) · 3.59 KB

data-model.md

File metadata and controls

82 lines (64 loc) · 3.59 KB
Raw

Data Model

vtex.search-resolver does not own a persistent data model — it composes GraphQL resolvers over the catalog, messages, rewriter, and Intelligent Search services. The schema-of-record is vtex.search-graphql/graphql/schema.graphql.

Resolver composition

// node/resolvers/index.ts
import { fieldResolvers as benefitsFieldResolvers } from './benefits'
import {
  fieldResolvers as searchFieldResolvers,
  queries as searchQueries,
} from './search'
import { queries as statsQueries } from './stats'

export const resolvers = {
  ...searchFieldResolvers,
  ...benefitsFieldResolvers,
  Query: {
    ...searchQueries,
    ...statsQueries,
  },
}
Module Owns Key files
node/resolvers/search/ All catalog / search / autocomplete / product / brand / category / facets fields index.ts, product.ts, brand.ts, category.ts, autocomplete.ts, productPriceRange.ts, offer.ts, discount.ts, assemblyOption.ts, itemMetadata*.ts, newURLs.ts, plus modules/
node/resolvers/benefits/ Product benefits (promotions, discount lists) field resolvers —
node/resolvers/stats/ searchURLsCount (no-op since VBase removal, always returns []); topSearches lives in resolvers/search/, not here —
node/directives/ Schema directives (@cacheControl, @withSegment, etc.) —
node/commons/ Shared helpers (caching keys, segment parsing) —
node/services/ Higher-level service objects composed in resolvers —
node/utils/ Pure utilities (searchState, query parsing, sluggers) —

Clients (outbound)

Client File Talks to Notes
search node/clients/search.ts Legacy VTEX search backend Cached (searchCache, capacity 3000, 6s timeout)
intelligentSearchApi node/clients/intelligent-search-api.ts vtex.intelligent-search-api (the IO app) ExternalClient; routed through the IO mesh
intsch node/clients/intsch/ Intelligent Search platform service directly Lower-level; used when going through the IO app is not desirable
checkout node/clients/checkout.ts Checkout APIs —
rewriter node/clients/rewriter.ts vtex.rewriter URL/slug resolution

LRU caches (set in node/index.ts)

Cache Capacity Backs
segmentCache 1000 Segment client
searchCache 3000 Legacy search client
messagesCache 3000 messagesGraphQL (translations)
appsCache 1500 apps client (@vtex/api)
intschCache 3000 Direct IS client

Settings (per-account behavior)

Relevant manifest.json:settingsSchema flags:

Flag Default Effect
slugifyLinks false When true, links are slugified via slugify; when false, the default catalog slug is used.
shouldUseNewPDPEndpoint false Routes PDP queries through the new IS endpoint instead of the legacy search backend.
enableHybridSearch false When true, sends semanticRatio on intsch product-search requests.
enableDeliveryPromisePreview false When true, sends dpPreview on intsch product-search/facets requests.

PLP (productSearch / products / productSuggestions / facets) always uses intsch; there is no PLP app-setting toggle.

These are read from VTEX app settings at runtime, not from FeatureHub.

In-code feature flags

node/featureFlags.ts currently exports:

export const flags = {
  VTEX_ASSETS_URL: true,
}

Code-level toggles for behaviors not yet exposed as app settings. Treat as transitional.