Storage API

The storage API owns canonical memory records, serialization, versions, and evidence metadata. It writes only inside resolved workspace data paths.

.. py:module:: memoryledger.storage

.. py:function:: default_config_text(project_name, memoryledger_dir)

module:

memoryledger.storage

Render legacy-compatible configuration text for initialization or migration.

type project_name:

sphinx_autodoc_typehints_type:

\:py\:class\:\str``

param project_name:

type memoryledger_dir:

:sphinx_autodoc_typehints_type:\:py\:class\:\str``

param memoryledger_dir:

rtype:

sphinx_autodoc_typehints_type:

\:py\:class\:\str``

.. py:function:: find_config(start=None)

module:

memoryledger.storage

Find a supported project configuration walking upward from a path.

type start:

sphinx_autodoc_typehints_type:

\:py\:class\:\~pathlib.Path` | :py:obj:`None``

param start:

rtype:

sphinx_autodoc_typehints_type:

\:py\:class\:\~pathlib.Path` | :py:obj:`None``

.. py:class:: Store(config)

module:

memoryledger.storage

Bases: :py:class:object

Read and mutate durable memory records in a resolved workspace.

type config:

sphinx_autodoc_typehints_type:

\:py\:class\:\~memoryledger.models.Config``

param config:

.. py:method:: Store.ensure_initialized() :module: memoryledger.storage

Create legacy durable state explicitly before a mutation.

:rtype: :sphinx_autodoc_typehints_type:`\:py\:obj\:\`None\``

.. py:module:: memoryledger.guardrails

from memoryledger.storage import Store

Storage methods can mutate durable records and require a validated workspace. Guardrail exceptions should be surfaced to callers rather than bypassed.