Schema Idea

ums/utils/__init__.py

@@ -38,3 +38,8 @@ from ums.utils.types import (
 from ums.utils.request import ManagementRequest
 
 from ums.utils.functions import list_shared_data, list_shared_schema
+
+from ums.utils.schema import (
+	ExtractionSchema,
+	ExtractedData
+)

ums/utils/schema.py

@@ -13,15 +13,73 @@
 	The types are implemented using [pydantic](https://docs.pydantic.dev/).
 	It provides validation, allow JSON serialization and works well with [FastAPI](https://fastapi.tiangolo.com/) which is used internally for the http request between the agents and the management.
 
+	**This is work in progress!**
 """
 
-from enum import Enum
-
-from typing import List, Any
+from typing import List, Any, Dict
 
 from pydantic import BaseModel
 
 class ExtractionSchema(BaseModel):
 	"""
 		This is the basic class used as superclass for all extracted information from data items.
-	"""
+
+		For all the `ExtractionSchema` is is required that the data can be serialized to json. 
+		Thus, mostly only default data types like `int, str, bool, list, dict, tuple` also including `ExtractionSchema` and `RiddleInformation` can be used here!
+	"""
+
+class ExtractedContent(ExtractionSchema):
+	"""
+		An extracted content item.
+	"""
+
+	type : str 
+	"""
+		The type, as a string, the actual string will depend on the extraction agent.
+	"""
+
+	content : str | Any
+	"""
+		The extracted content
+	"""
+
+class ExtractedPositions(ExtractionSchema):
+	"""
+		A position (like time, coordinates, ...) where something was extracted (each position should belong to a content item).
+	"""
+
+	type : str 
+	"""
+		The type, as a string, the actual string will depend on the extraction agent.
+	"""
+
+	position : str | int | Any
+	"""
+		The position, will also depend on the extraction agent.
+	"""
+
+	description : str | Any = None
+	"""
+		An optional description for more details.
+	"""
+
+class ExtractedData(ExtractionSchema):
+	"""
+		Contains the extracted items from a data file.
+	"""
+
+	contents : List[ExtractedContent] = []
+	"""
+		The extracted contents (i.e., transcriptions etc.), each item here should belong a position item at the same index.
+	"""
+
+	positions : List[ExtractedPositions] = []
+	"""
+		The positions of extracted contents, each item here should belong a content item at the same index.
+	"""
+
+	other : Dict[str, Any] = {}
+	"""
+		Possibly more data. Use a keywords (depending on agent) and store the data there.
+	"""
+