> ## Documentation Index > Fetch the complete documentation index at: https://docs.datalinks.com/llms.txt > Use this file to discover all available pages before exploring further. # Ingest data > Load data into the specified dataset. ## OpenAPI ````yaml /api-reference/openapi.yml post /ingest/{namespace}/{datasetName} openapi: 3.0.3 info: title: DataLinks version: 2.2.3 servers: - url: https://api.datalinks.com/api/v1 description: production server security: - bearerAuth: [] - openId: - implicit paths: /ingest/{namespace}/{datasetName}: post: tags: - ingest summary: Ingest data description: Load data into the specified dataset. operationId: ingestData parameters: - name: namespace in: path description: Namespace for the dataset. required: true schema: type: string example: cinematography - name: datasetName in: path description: Name of the dataset. required: true schema: type: string examples: movies: summary: Movies value: movies actors: summary: Actors value: actors requestBody: content: application/json: schema: $ref: '#/components/schemas/IngestRequest' examples: movies: $ref: '#/components/examples/moviesExample' actors: $ref: '#/components/examples/actorsExample' unstructured: $ref: '#/components/examples/unstructuredExample' required: true responses: '200': description: Returns the ingestion result. '401': $ref: '#/components/responses/Unauthorized' components: schemas: IngestRequest: required: - data type: object properties: data: type: array items: type: object additionalProperties: type: string infer: $ref: '#/components/schemas/IngestRequestInfer' link: $ref: '#/components/schemas/IngestLink' curate: type: boolean nullable: true dataDescription: $ref: '#/components/schemas/DataDescription' schemaDefinition: $ref: '#/components/schemas/SchemaDefinition' additionalInstructions: type: string description: Additional instructions to guide the AI system during ingestion. IngestRequestInfer: type: object properties: steps: type: array items: $ref: '#/components/schemas/IngestRequestInferStep' IngestLink: type: object properties: ExactMatch: nullable: true anyOf: - $ref: '#/components/schemas/ExactMatchConfig' - $ref: '#/components/schemas/ResolutionParameters' GeoMatch: nullable: true anyOf: - $ref: '#/components/schemas/GeoMatchConfig' - $ref: '#/components/schemas/ResolutionParameters' DataDescription: maxLength: 10000 type: string description: >- A description of the dataset to help the AI system understand the context. example: Employee dataset containing demographics and compensation. SchemaDefinition: type: object additionalProperties: type: string description: >- A schema definition mapping field names to descriptions, to guide the AI system in structuring the extracted data. example: id: unique employee identifier name: full employee name age: employee age in years department: work department salary: annual salary in USD ErrorResponse: required: - error_code - message type: object properties: error_code: type: string example: UNAUTHORIZED error_message: type: string example: '' message: type: string example: Missing authentication token sub_errors: type: array items: required: - code - message type: object properties: code: type: string message: type: string IngestRequestInferStep: type: object discriminator: propertyName: type mapping: table: $ref: '#/components/schemas/TableStep' rows: $ref: '#/components/schemas/RowsStep' normalise: $ref: '#/components/schemas/NormaliseStep' validate: $ref: '#/components/schemas/ValidateStep' reverseGeo: $ref: '#/components/schemas/ReverseGeoStep' namedEntities: $ref: '#/components/schemas/NamedEntitiesStep' anyOf: - $ref: '#/components/schemas/TableStep' - $ref: '#/components/schemas/RowsStep' - $ref: '#/components/schemas/NormaliseStep' - $ref: '#/components/schemas/ValidateStep' - $ref: '#/components/schemas/ReverseGeoStep' - $ref: '#/components/schemas/NamedEntitiesStep' ExactMatchConfig: type: object properties: minDistinct: type: number minVariation: type: number nullable: true ResolutionParameters: type: object properties: bidirectional: type: boolean targetColumns: type: array items: type: string ignorePrivateColumns: type: boolean GeoMatchConfig: type: object properties: distance: type: number distanceUnit: type: string nullable: true TableStep: required: - deriveFrom - type type: object properties: type: type: string enum: - table deriveFrom: type: string description: Name of the source column to extract from. additionalInstructions: type: string description: >- Additional instructions to guide the AI system for higher accuracy during extraction. example: >- This text contains a table comma separated, but instead of line breaks we are using a ';'. model: type: string description: The LLM model to use for this step. provider: type: string description: The LLM provider to use for this step. replaceOriginalColumn: type: boolean description: >- When true, the source column specified in deriveFrom is removed from the output after extraction. default: false RowsStep: required: - deriveFrom - type type: object properties: type: type: string enum: - rows deriveFrom: type: string description: Name of the column that contains the JSON content. NormaliseStep: required: - mode - type type: object properties: type: type: string enum: - normalise - normalize mode: type: string description: Normalisation mode. enum: - all-in-one - field-by-field - embeddings targetCols: type: object additionalProperties: type: string description: Target columns with their descriptions. dataDescription: $ref: '#/components/schemas/DataDescription' model: type: string description: The LLM model to use for this step. provider: type: string description: The LLM provider to use for this step. ValidateStep: required: - columns - mode - type type: object properties: type: type: string enum: - validate mode: type: string description: Validation mode. enum: - regex - rows - fields columns: type: array description: Columns to validate. items: type: string model: type: string description: The LLM model to use for this step. provider: type: string description: The LLM provider to use for this step. ReverseGeoStep: required: - type type: object properties: type: type: string enum: - reverseGeo deriveFrom: type: string description: Name of the location column. NamedEntitiesStep: required: - type type: object properties: type: type: string enum: - namedEntities examples: moviesExample: summary: Movies value: data: - name: Braveheart year: 1995 budget: 72000000 director: Mel Gibson - name: Argo year: 2012 budget: 44500000 director: Ben Affleck - name: A Quiet Place year: 2018 budget: 17000000 director: John Krasinski - name: Dances with Wolves year: 1990 budget: 22000000 director: Kevin Costner - name: Unforgiven year: 1992 budget: 14000000 director: Clint Eastwood - name: The Postman year: 1997 budget: 80000000 director: Kevin Costner link: ExactMatch: null actorsExample: summary: Actors value: data: - name: Mel Gibson age: 68 nationality: American - name: Sophie Marceau age: 57 nationality: French - name: Ben Affleck age: 52 nationality: American - name: Bryan Cranston age: 68 nationality: American - name: John Krasinski age: 45 nationality: American - name: Emily Blunt age: 41 nationality: British - name: Kevin Costner age: 69 nationality: American - name: Mary McDonnell age: 72 nationality: American - name: Clint Eastwood age: 94 nationality: American - name: Morgan Freeman age: 86 nationality: American - name: Kevin Costner age: 69 nationality: American - name: Will Patton age: 70 nationality: American link: ExactMatch: null unstructuredExample: summary: Example of extracting rows from unstructured data value: data: - text: >- Meryl Streep, known for her incredible acting, has starred in unforgettable films like The Devil Wears Prada and Sophie's Choice. She's won three Oscars over her illustrious career. Anne Hathaway shared the screen with her in The Devil Wears Prada and later shone in Les Misérables, earning her own Oscar. Leonardo DiCaprio, a household name, is celebrated for roles in Titanic, Inception, and The Revenant. His long-awaited Oscar win came for The Revenant. His Titanic co-star Kate Winslet, who also has an Oscar to her name, delivered powerful performances in The Reader and of course, Titanic. Tom Hanks, the quintessential leading man, gave us unforgettable performances in Forrest Gump, Cast Away, and Saving Private Ryan. He's a two-time Oscar winner. Robin Wright, who starred alongside him in Forrest Gump, is also well-remembered for her role in The Princess Bride. Cate Blanchett, a two-time Oscar winner, mesmerized audiences in films like Blue Jasmine, Carol, and Elizabeth. Rooney Mara, her co-star in Carol, also made a mark with her performance in The Girl with the Dragon Tattoo. Denzel Washington, one of the most respected actors in Hollywood, has won two Oscars for his work in Training Day and Glory. He also delivered a heartfelt performance in Fences, where he starred alongside Viola Davis. Davis herself is a powerhouse of talent, with an Oscar for her role in Fences and another standout performance in The Help. infer: steps: - deriveFrom: text additionalInstructions: >- If you find a numeric field use only the value and omit the rest. type: table responses: Unauthorized: description: >- Unauthorized. This error occurs if the authentication token is missing or invalid. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' securitySchemes: bearerAuth: type: http description: > Use a Bearer token for authentication. Submit the token using the `Authorization` header: `Authorization: Bearer `. scheme: bearer openId: type: oauth2 flows: implicit: authorizationUrl: >- https://login.datalinks.com/realms/datalinks-realm/protocol/openid-connect/auth scopes: openid: openid ````