def fields_of(memo_type: type) -> FieldNamespace

Return a namespace of first-class field references for a dbzero memo type.

Field references identify memo fields for APIs such as db0.index_of(...). They do not read or write instance values, and they do not replace normal Python class attributes.

fields = db0.fields_of(Task)
priority_field = fields.priority
priority_index = db0.index_of(priority_field)

Parameters

• memo_type type
  A dbzero memo class.

Returns

A FieldNamespace bound to the memo type.

Attribute Lookup

Use attribute lookup for known field declarations. This is the preferred syntax for indexed fields:

priority = db0.index_of(db0.fields_of(Task).priority)

Attribute lookup is strict. Unknown names raise AttributeError, which helps catch typos early.

For dataclasses, fields_of preserves normal class-attribute behavior:

from dataclasses import dataclass
 
@db0.memo
@dataclass
class Task:
    title: str
    priority: int = 0
 
assert Task.priority == 0
assert db0.fields_of(Task).priority is db0.fields_of(Task).priority

Dynamic Field Names

Use subscription when the field name is dynamic or not a valid Python attribute:

field_name = "priority"
priority_field = db0.fields_of(Task)[field_name]
 
keyword_field = db0.fields_of(Task)["from"]

Subscription accepts persistent field names and returns a FieldRef. The consuming API, such as db0.index_of(...), validates whether that field is valid for the requested operation.

Notes

• fields_of accepts memo types, not instances.
• Field references are process-local objects, not durable values.
• A FieldRef can be compared and hashed, but it is not a descriptor and is not installed on the memo type.
• fields_of does not open a prefix or materialize a field by itself.