LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows
Python ai-agentsdata-engineeringdocument-aigenerative-aiidpjson-extractionllmmcp-serverocrpdf-extractionprompt-engineeringstructured-output
from rest_framework.exceptions import APIException
from backend.exceptions import UnstractBaseException
from unstract.connectors.exceptions import ConnectorError
class InvalidConnectorMode(APIException):
status_code = 400
default_detail = "Connector mode is not Valid."
class InvalidConnectorID(APIException):
status_code = 400
default_detail = "Connector ID is not Valid."
class OAuthTimeOut(APIException):
status_code = 408
default_detail = "Timed out. Please re-authenticate."
class TestConnectorException(APIException):
status_code = 500
default_detail = "Error while testing connector."
class TestConnectorInputError(UnstractBaseException):
def __init__(self, core_err: ConnectorError) -> None:
super().__init__(detail=core_err.message, core_err=core_err)
self.default_detail = core_err.message
# Preserve status code from connector error if available, otherwise default to 400
if not core_err.status_code:
self.status_code = 400