> ## Documentation Index > Fetch the complete documentation index at: https://docs.doofinder.com/llms.txt > Use this file to discover all available pages before exploring further. # Update in bulk from temporal > Updates a list of items in the temporal index in a single bulk operation. ## OpenAPI ````yaml /openapi/management-v2.yaml patch /api/v2/search_engines/{hashid}/indices/{name}/temp/items/_bulk openapi: 3.0.0 info: title: Management API contact: name: Management API Support url: https://docs.doofinder.com/ version: '2.0' x-logo: url: https://admin.doofinder.com/images/logo.svg backgroundColor: '#33268C' altText: Management API Search Engine description: > # Introduction Management API allows you to perform some of the administrative tasks you can do on your search engines using the Admin control panel, directly from your code. # Basics ## Endpoint All requests should be done via `https` to the right endpoint: ``` https://{search-zone}-api.doofinder.com ``` Where `{search-zone}` is the code of the datacenter where your search engines are located. For instance: ``` https://eu1-api.doofinder.com https://us1-api.doofinder.com ``` ## Authentication To authenticate you need an API key. If you don't have one you can generate it in the Admin by going to your Account and then to API Keys, [here](https://admin.doofinder.com/admin/account/api). A valid API key looks like this: ``` ab46030xza33960aac71a10248489b6c26172f07 ``` ### API Token You can authenticate with the previous API key. The correct way to authenticate is to send a HTTP Header with the name `Authorization` and the value `Token {api-key}`. I.e.: ``` Authorization: Token ab46030xza33960aac71a10248489b6c26172f07 ``` ### JWT Token If you prefer you can authenticate with a [JSON Web Token](https://jwt.io). The token must be signed with an API management key and there are some claims required in the JWT payload. These claims are: * `iat` (issued at): Creation datetime timestamp, i.e. the moment when the JWT was created. * `exp` (expiration time): Expiration datetime timestamp, i.e. the moment when the JWT is going to expire and will no longer be valid. The time span between issued and expiration dates must be shorter than a week. * `name`: Your user code. It is your unique identifier as doofinder user. You can find this code in your profile page in the administration panel. To authenticate using JWT you must send a HTTP header with the name `Authorization` and the value `Bearer {JWT-token}`. I.e.: ``` Authorization: Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJuYW1lIjoidGVzdCIsImlhdCI6MTUxNjIzOTAyMn0.QX_3HF-T2-vlvzGDbAzZyc1Cd-J9qROSes3bxlgB4uk ``` ## Conventions Along most of the code samples you will find placeholders for some common variable values. They are: * `{hashid}`: The search engine's unique id. e.g.: d8fdeab7fce96a19d3fc7b0ca7a1e98b. * `{index}`: When storing items, they're always stored under a certain _index_. e.g.: product. * `{token}`: Your personal authentication token obtained in the control panel. ## Api limits ### Requests rate The API has a limit of 20 requests per second. If you do more than that, the next requests will get a `429 Too Many Requests` error. There is no penalty associated to that error, but you must wait one second before attempting more requests. ### Body size The body size for requests with data sending (eg: POST or PUT requests) has a limit of 8MB. If you do a request with a request body bigger than 8 MB, the response will be `413 Payload Too Large`. This is important mostly for batch calls like batch create items. ### Batch size Bulk item operations (create, update, delete) accept a maximum of **100 items per request**. Requests with more than 100 items will be rejected with a `422 Unprocessable Entity` error. Split larger payloads into batches of up to 100 items. # Objects ## Search Engines A search engine is a collection of indices where data is stored. It has some properties like name, language, and some other configuration options that are shared among all of its indices. Each search engine is uniquely identified by its `hashid` which is an auto-generated string. Search engines have a _"process"_ operation which takes items from all its data sources and insert them into their indices. This operation has a zero downtime, you can still search into indices while processing. ## Indices Indices are collections of items, the same way a search engine is a collection of indices. An index also has a set of data sources which usually are urls where the items for indexing are taken from. If you need to reindex a whole index you can create a temporary index. The mission of a temporary index is indexing items while you can still search the main index, having zero downtime. ## Data Sources Data sources are indices items sources. When you process a search engine, we take the items from its data sources and "insert" them into the indices. You do not need any data source if you index items only by API. ## Items Items are objects stored in an index. Those are the elements you search for when doing query to a search engine. They may have different schemas depending on their index **preset**. For instance, a `product` item will have a price, category, etc. In case you want to group items as a product with variants, you have to add these two fields: * `group_id`: All the items with the same group id will be members of the group. Could be the parent id or just an independent id. * `group_leader`: Boolean indicating if the item is the leader of the variant group, which will be the item shown among its group in the search results. It is not mandatory if there is no item leader. You could index the items in any order because the grouping feature is made in search phase. An example of request payload would be: ``` [ { "title": "Item Variant 1", "link": "http://www.example.com/img/variant1.png", "id": "1", "df_manual_boost": 1, "categories": [...], "group_id": "99", "group_leader": true }, { "title": "Item Variant 2", "link": "http://www.example.com/img/variant2.png", "id": "2", "df_manual_boost": 1, "categories": [...], "group_id": "99", "group_leader": false }, ] ``` ### Items transformations When your Index has `product` preset, some defaults are added to your item at creation. For instance * `categories` field if not provided, would be a empty list at response `[]`. * `best_price` would get the min value between `price` and `sale_price` item fields if present, if not, would be `null`. Fields `price` and `sale_price` should be a number like: 15.56. If you want to update `best_price`, `price` and `sale_price` should be present. When your Index has `generic` preset, this defaults will not be applied. termsOfService: https://www.doofinder.com/en/terms-and-conditions servers: - url: https://{search_zone}-api.doofinder.com variables: search_zone: enum: - eu1 - us1 - ap1 default: eu1 - url: / security: - api_token: [] - jwt_token: [] tags: - name: Search Engines description: >- All search engines CRUD operations, including handling data sources processing. - name: Indices description: All indices and temporary indices CRUD operations. - name: Items description: >- Handling content of indices. Allows to retrieve and update the items of an index. - name: Stats description: Provides statistical information about your search engines. externalDocs: url: https://docs.doofinder.com/api/management/v2/stats/ paths: /api/v2/search_engines/{hashid}/indices/{name}/temp/items/_bulk: patch: tags: - Items summary: Update in bulk from temporal description: >- Updates a list of items in the temporal index in a single bulk operation. operationId: items_temp_bulk_update parameters: - $ref: '#/components/parameters/searchengine_hashid' - $ref: '#/components/parameters/index_name' requestBody: content: application/json: schema: $ref: '#/components/schemas/BulkRequest' required: true responses: '200': $ref: '#/components/responses/BulkResponse' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/Forbidden' '404': $ref: '#/components/responses/NotFound' '408': $ref: '#/components/responses/Timeout' '409': $ref: '#/components/responses/Conflict' '413': $ref: '#/components/responses/TooManyItemsBody' '422': $ref: '#/components/responses/TooManyItemsBulk' '429': $ref: '#/components/responses/TooManyRequests' '500': $ref: '#/components/responses/InternalServerError' components: parameters: searchengine_hashid: name: hashid in: path description: Unique id of a search engine. required: true schema: type: string pattern: ^[a-f0-9]{32}$ index_name: name: name in: path description: Name of an index. required: true schema: type: string pattern: ^[a-z][a-z0-9_]*$ schemas: BulkRequest: title: Bulk Request type: array description: >- A list of items data for a bulk operation with a minimum of 1 item and a maximum of 100. items: $ref: '#/components/schemas/IndexableItem' minItems: 1 maxItems: 100 IndexableItem: title: Item description: >- A "searchable" element. Items are the objects that form the search results. type: object additionalProperties: true properties: id: type: string description: Item id group_id: description: >- This field indicates the group to which this item belongs to. All items with the same `group_id` will be collapsed into one in search results, returning the most relevant one or the group leader if they all have the same score. type: string nullable: true group_leader: description: >- This field indicates the item chosen as the default among its group. It will be returned in search results if there is no other item with a higher score. type: boolean nullable: true df_group_leader: description: >- (Deprecated) This field indicates the item chosen as the default among its group. It will be returned in search results if there is no other item with a higher score. type: boolean nullable: true deprecated: true df_manual_boost: description: >- A numeric score boosting. It multiplies the natural score of the item for a search. For instance, if boost is greater than `1.0` the item will appear higher in the results. If it is lower than `1.0`, it will appear lower. The minimum value is 0.0. type: number format: double minimum: 0 nullable: true categories: description: > This field has special behaviour when Indice has `product` preset, categories shall be specified as: ``` { ... categories: [ 'parent category > inner category > leaft category', 'other parent > other inner > other leaft' ] } ``` This way, when using categories field in facets, it will work as expected. When Index has `generic` preset, this works as normal field but it should be string. oneOf: - type: array items: type: string - type: string best_price: description: > Auto created field that gets the min value between `price` or `sale_price` fields, if added in the document. It gets null if doesn't find any of these fields. It is not necessary to send this field to index it. type: number nullable: true image_link: type: string format: uri nullable: true link: type: string format: uri nullable: true title: type: string nullable: true example: title: Adidas Originals Trefoil Hoodie link: >- https://woocommerce.doofinder.com/product/adidas-originals-trefoil-hoodie/ image_link: >- https://woocommerce.doofinder.com/wp-content/uploads/2024/02/suda-6-scaled.jpg id: '1234' color: blue price: 15.99 sale_price: 11.99 categories: - Clothes > Hoodie - Men > Hoodie BulkResult: title: Bulk Result description: Result for bulk operations. type: object properties: results: description: List of results of each bulk operation. type: array items: type: object properties: result: description: Result of the operation. type: string enum: - created - updated - deleted id: description: Item unique identifier. type: string error: description: Result error of the operation. type: string nullable: true errors: description: Indicates if there is any error in the results. type: boolean example: results: - result: updated id: '1234' errors: false ManagementError: type: object description: Standard error wrapper for all Management API error responses. properties: error: type: object properties: code: type: string description: Machine-readable error code. example: not_found message: type: string description: Human-readable error description. example: Not found details: type: array description: >- Optional list of per-field validation details (present on bad_params and errors_in_fields). items: $ref: '#/components/schemas/ManagementErrorDetail' required: - code - message required: - error ManagementErrorDetail: type: object description: Details about a specific validation failure on a field. properties: name: type: string description: Field name. example: language fields: type: array items: type: string description: Affected field names. example: - language reason: type: string description: Validation failure reason. example: is invalid responses: BulkResponse: description: Bulk operation successful. content: application/json: schema: $ref: '#/components/schemas/BulkResult' BadRequest: description: Bad Request — invalid or missing parameters. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: bad_params message: Bad parameters error details: - name: language fields: - language reason: is invalid Unauthorized: description: Unauthorized — no valid API token provided. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: not_authenticated message: Authentication required Forbidden: description: Forbidden — token is valid but does not have access to this resource. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: access_denied message: Access denied NotFound: description: Not Found — the requested resource does not exist. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: not_found message: Not found Timeout: description: >- Request Timeout — the search engine or index is busy; retry after a short delay. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: searchengine_timeout message: Timeout connecting to backend Conflict: description: >- Conflict — the search engine is locked for processing or in an inactive state. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: searchengine_locked message: Search engine is being processed TooManyItemsBody: description: Request Entity Too Large — the bulk payload exceeds 8 MB. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: too_many_items message: Body is too large (max 8 MB) TooManyItemsBulk: description: Unprocessable Entity — the bulk array exceeds 100 items. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: too_many_items message: Too many items in bulk request (max 100) TooManyRequests: description: Too Many Requests — rate limit exceeded; back off and retry. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: too_many_requests message: Too many requests InternalServerError: description: Internal Server Error — unexpected backend failure. content: application/json: schema: $ref: '#/components/schemas/ManagementError' example: error: code: internal_error message: Internal server error securitySchemes: api_token: type: apiKey description: | Doofinder API key. Pass it as: Authorization: Token name: Authorization in: header jwt_token: type: apiKey description: | JWT Bearer token. Sign a JWT with your API key (HS256) and pass it as: Authorization: Bearer name: Authorization in: header ````