openapi: 3.0.0 info: title: OCR Invoice Job API description: API for managing individual OCR invoice jobs version: 1.0.0 contact: name: Repzo url: https://repzo.com servers: - url: https://sv.api.repzo.me description: Production server - url: https://staging.sv.api.repzo.me description: Staging server tags: - name: OCR Invoice Job description: Operations related to individual OCR invoice job management paths: /ocr-invoice-job: get: tags: - OCR Invoice Job summary: Get all OCR invoice jobs description: Retrieve a list of all OCR invoice jobs with optional filtering and pagination parameters: - name: limit in: query description: Maximum number of records to return required: false schema: type: integer minimum: 1 maximum: 100 default: 50 - name: offset in: query description: Number of records to skip for pagination required: false schema: type: integer minimum: 0 default: 0 - name: sort in: query description: Field to sort by required: false schema: type: string - name: order in: query description: Sort order (asc or desc) required: false schema: type: string enum: [asc, desc] default: desc - name: filter in: query description: Filter conditions in JSON format required: false schema: type: string - name: search in: query description: Search term for text fields required: false schema: type: string responses: "200": description: List of OCR invoice jobs retrieved successfully content: application/json: schema: type: object properties: success: type: boolean example: true data: type: array items: $ref: "#/components/schemas/OcrInvoiceJob" total: type: integer description: Total number of records "400": description: Bad request "401": description: Unauthorized "500": description: Internal server error post: tags: - OCR Invoice Job summary: Create a new OCR invoice job description: Create a new OCR invoice job with the provided data requestBody: required: true content: application/json: schema: $ref: "#/components/schemas/OcrInvoiceJobInput" responses: "201": description: OCR invoice job created successfully content: application/json: schema: type: object properties: success: type: boolean example: true data: $ref: "#/components/schemas/OcrInvoiceJob" "400": description: Bad request "401": description: Unauthorized "500": description: Internal server error /ocr-invoice-job/{id}: get: tags: - OCR Invoice Job summary: Get OCR invoice job by ID description: Retrieve a specific OCR invoice job by its ID parameters: - name: id in: path required: true description: OCR invoice job ID schema: type: string responses: "200": description: OCR invoice job retrieved successfully content: application/json: schema: type: object properties: success: type: boolean example: true data: $ref: "#/components/schemas/OcrInvoiceJob" "404": description: OCR invoice job not found "401": description: Unauthorized "500": description: Internal server error put: tags: - OCR Invoice Job summary: Update OCR invoice job description: Update an existing OCR invoice job with new data parameters: - name: id in: path required: true description: OCR invoice job ID schema: type: string requestBody: required: true content: application/json: schema: $ref: "#/components/schemas/OcrInvoiceJobInput" responses: "200": description: OCR invoice job updated successfully content: application/json: schema: type: object properties: success: type: boolean example: true data: $ref: "#/components/schemas/OcrInvoiceJob" "400": description: Bad request "404": description: OCR invoice job not found "401": description: Unauthorized "500": description: Internal server error delete: tags: - OCR Invoice Job summary: Delete OCR invoice job description: Delete an existing OCR invoice job parameters: - name: id in: path required: true description: OCR invoice job ID schema: type: string responses: "200": description: OCR invoice job deleted successfully content: application/json: schema: type: object properties: success: type: boolean example: true message: type: string example: OCR invoice job deleted successfully "404": description: OCR invoice job not found "401": description: Unauthorized "500": description: Internal server error /ocr-invoice-job/{id}/process: post: tags: - OCR Invoice Job summary: Process OCR invoice job description: Start processing an OCR invoice job parameters: - name: id in: path required: true description: OCR invoice job ID schema: type: string responses: "200": description: OCR invoice job processing started content: application/json: schema: type: object properties: success: type: boolean example: true data: type: object properties: jobId: type: string status: type: string enum: [queued, processing] "404": description: OCR invoice job not found "401": description: Unauthorized "500": description: Internal server error /ocr-invoice-job/{id}/results: get: tags: - OCR Invoice Job summary: Get OCR results description: Retrieve the OCR processing results for a job parameters: - name: id in: path required: true description: OCR invoice job ID schema: type: string responses: "200": description: OCR results retrieved successfully content: application/json: schema: type: object properties: success: type: boolean example: true data: $ref: "#/components/schemas/OcrResults" "404": description: OCR invoice job not found "401": description: Unauthorized "500": description: Internal server error components: schemas: OcrInvoiceJob: type: object properties: _id: type: string description: Unique identifier for the OCR invoice job example: "60f7b1b3e4b0e8b3f8b3f8b3" jobNumber: type: string description: Unique job number example: "OCR-2023-001" documentName: type: string description: Name of the document being processed example: "invoice_2023_001.pdf" documentUrl: type: string format: uri description: URL of the document to process example: "https://storage.example.com/invoices/invoice_2023_001.pdf" documentType: type: string enum: [pdf, png, jpg, jpeg, tiff, bmp] description: Type of document example: "pdf" template: type: string description: OCR template ID to use example: "60f7b1b3e4b0e8b3f8b3f8b4" group: type: string description: OCR job group ID if part of a batch example: "60f7b1b3e4b0e8b3f8b3f8b5" status: type: string enum: [created, queued, processing, completed, failed, cancelled] default: created description: Processing status priority: type: string enum: [low, medium, high, urgent] default: medium description: Processing priority progress: type: object properties: percentage: type: number format: double minimum: 0 maximum: 100 description: Processing progress percentage currentStep: type: string description: Current processing step totalSteps: type: integer description: Total number of processing steps startedAt: type: string format: date-time description: When processing started example: "2023-01-15T08:00:00Z" completedAt: type: string format: date-time description: When processing completed example: "2023-01-15T08:05:00Z" processingTime: type: number format: double description: Processing time in seconds example: 45.6 documentInfo: type: object properties: pages: type: integer description: Number of pages in document size: type: integer description: File size in bytes format: type: string description: Document format resolution: type: string description: Document resolution language: type: string description: Detected language extractedData: type: object properties: invoice: type: object properties: number: type: string description: Invoice number date: type: string format: date description: Invoice date dueDate: type: string format: date description: Due date supplier: type: object properties: name: type: string address: type: string taxId: type: string customer: type: object properties: name: type: string address: type: string amounts: type: object properties: subtotal: type: number format: double tax: type: number format: double total: type: number format: double currency: type: string lineItems: type: array items: type: object properties: description: type: string quantity: type: number format: double unitPrice: type: number format: double amount: type: number format: double confidence: type: object properties: overall: type: number format: double minimum: 0 maximum: 1 description: Overall confidence score fields: type: object additionalProperties: type: number format: double description: Confidence per field validationResults: type: object properties: isValid: type: boolean description: Whether validation passed errors: type: array items: type: object properties: field: type: string message: type: string severity: type: string enum: [error, warning, info] cost: type: number format: double minimum: 0 description: Processing cost currency: type: string description: Currency code example: "USD" errorLog: type: string description: Error log if processing failed retryCount: type: integer minimum: 0 description: Number of retry attempts createdBy: type: string description: User who created the job example: "60f7b1b3e4b0e8b3f8b3f8b6" metadata: type: object additionalProperties: true description: Additional metadata createdAt: type: string format: date-time description: Record creation timestamp example: "2023-01-15T08:00:00Z" updatedAt: type: string format: date-time description: Record last update timestamp example: "2023-01-15T10:30:00Z" required: - jobNumber - documentName - documentUrl - template OcrInvoiceJobInput: type: object properties: jobNumber: type: string description: Unique job number example: "OCR-2023-001" documentName: type: string description: Name of the document being processed example: "invoice_2023_001.pdf" documentUrl: type: string format: uri description: URL of the document to process example: "https://storage.example.com/invoices/invoice_2023_001.pdf" documentType: type: string enum: [pdf, png, jpg, jpeg, tiff, bmp] description: Type of document example: "pdf" template: type: string description: OCR template ID to use example: "60f7b1b3e4b0e8b3f8b3f8b4" group: type: string description: OCR job group ID if part of a batch example: "60f7b1b3e4b0e8b3f8b3f8b5" priority: type: string enum: [low, medium, high, urgent] default: medium description: Processing priority createdBy: type: string description: User who created the job example: "60f7b1b3e4b0e8b3f8b3f8b6" metadata: type: object additionalProperties: true description: Additional metadata required: - jobNumber - documentName - documentUrl - template OcrResults: type: object properties: extractedData: type: object description: All extracted data from the document confidence: type: object description: Confidence scores for extracted data validationResults: type: object description: Validation results processingTime: type: number format: double description: Total processing time securitySchemes: ApiKeyAuth: type: apiKey in: header name: api-key security: - ApiKeyAuth: []