Skip to content

README.md

Latest commit

 

History

History
71 lines (47 loc) · 2.27 KB

README.md

File metadata and controls

71 lines (47 loc) · 2.27 KB

Nebuia Integration Library

This library provides an interface to interact with the Nebuia API, allowing operations for document processing, custom brain searching, and batch management.

Key Features

  • Integrator initialization with API credentials
  • Custom brain searching
  • Document management (retrieve, clear, delete)
  • Batch handling (create, get documents, add files)
  • Document type retrieval

Installation

[Specific installation instructions for each language]

Basic Usage

# Integrator initialization
integrator = Integrator(with_base='http://nebuia.instance/api/v1', key='api_key', secret='api_secret')

# Usage example: Custom brain search
results = integrator.search_in_brain(search_params=SearchParameters(batch="brain_id", param="flu", k=2, type_search="literal"))

# Usage example: Get documents by status
docs = integrator.get_documents_by_status(status=StatusDocument.ERROR_LINK)

# Usage example: Create a new batch
status, batch_id = integrator.create_batch("name_batch", batch_type=BatchType.TESTING)

API Reference

Integrator Class

Main Methods:

  • search_in_brain(search_params: SearchParameters) -> dict
  • get_documents_by_status(status: StatusDocument) -> list
  • clear_document_by_uuid(uuid: str) -> dict
  • delete_document(uuid: str) -> dict
  • get_documents_by_batch_id(batch_id: str) -> BatchDocuments
  • append_to_batch(batch_id: str, files: list[File]) -> dict
  • get_document_types() -> list[DocumentType]
  • create_batch(name: str, batch_type: BatchType) -> tuple[bool, str]

Data Structures

  • File: Represents a file to process
  • SearchParameters: Parameters for brain searching
  • StatusDocument: Enumeration of document statuses
  • BatchType: Enumeration of batch types

Error Handling

The library uses the Loguru logger to log information and errors. Make sure to configure Loguru appropriately in your application.

Implementation Notes

  • The library should handle API responses and convert them into appropriate data structures for each language.
  • Input validations should be implemented for method parameters.
  • Error handling should be consistent across all methods, using specific exceptions when appropriate.

Contributions

[Instructions for contributing to the project]

License

[License information]