read.bulk.product_paths

Every product this store publishes, reduced to what a link is made of: its handle and the ltree category_path of its canonical category (null when the product is in no category). It is read.products with everything a URL does not need removed — the same published set, about 1/100th of the bytes — for a caller that walks the whole catalog to build addresses (a sitemap, a shopping feed, a channel export). Paginated; at most 1000 per page, and over the ceiling is an error, never a truncated page. Served on the BULK face: the same published answer as the public read, charged to the caller’s own bulk budget instead of the shopper’s, and never to be reached by a browser.

• Surface: GET /v1/read/bulk/product_paths
• Auth: ★ Bulk read (machine) — a TENANT credential is REQUIRED (no credential is unauthorized), and store must be a store of THAT tenant (another tenant’s is forbidden). The ANSWER is the public read’s, byte for byte: this store’s published assortment, no drafts, no actor. What the credential buys is the BUDGET — this face has its own RateLimit-* bucket, so walking a whole catalogue here takes nothing from the ceiling a shopper’s browser shares.

Params (JSON Schema)

{
  "properties": {
    "limit": {
      "default": 1000,
      "maximum": 1000,
      "minimum": 1,
      "type": "integer"
    },
    "page": {
      "default": 1,
      "maximum": 9007199254740991,
      "minimum": 1,
      "type": "integer"
    },
    "store": {
      "minLength": 1,
      "type": "string"
    }
  },
  "required": [
    "store"
  ],
  "type": "object"
}

Result (200)

Open (unknown) — this capability does not declare an output schema.

Errors

invalid_params (400) · not_found (404) · unauthorized (401) · forbidden (403)