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 import serializers
from backend.constants import FieldLengthConstants as FLC
from connector_processor.constants import ConnectorKeys
class TestConnectorSerializer(serializers.Serializer):
connector_id = serializers.CharField(max_length=FLC.CONNECTOR_ID_LENGTH)
connector_metadata = serializers.JSONField()
class ConnectorSchemaQuerySerializer(serializers.Serializer):
"""Serializer for validating connector schema query parameters."""
id = serializers.CharField(
required=True,
allow_blank=False,
error_messages={"required": "ID is mandatory.", "blank": "ID cannot be empty."},
)
class SupportedConnectorsQuerySerializer(serializers.Serializer):
"""Serializer for validating supported connectors query parameters."""
type = serializers.ChoiceField(
choices=[ConnectorKeys.INPUT, ConnectorKeys.OUTPUT],
required=False,
help_text="Optional filter by connector type. If not provided, returns all connectors.",
)
connector_mode = serializers.CharField(
required=False, allow_blank=True, help_text="Optional connector mode filter"
)