:py:mod:`gnome.utilities.cache`
===============================

.. py:module:: gnome.utilities.cache

.. autoapi-nested-parse::

   cache system for caching element data on disk for
   accessing again for output, etc.



Module Contents
---------------

Classes
~~~~~~~

.. autoapisummary::

   gnome.utilities.cache.ElementCache



Functions
~~~~~~~~~

.. autoapisummary::

   gnome.utilities.cache.clean_up_cache



Attributes
~~~~~~~~~~

.. autoapisummary::

   gnome.utilities.cache._cache_dir


.. py:data:: _cache_dir

   

.. py:exception:: CacheError


   Bases: :py:obj:`Exception`

   Here so we can be sure the user knows the error is coming from here

   Initialize self.  See help(type(self)) for accurate signature.


.. py:function:: clean_up_cache(dir_name=_cache_dir)

   Deletes a cache dir.

   Designed to be called at program exit. and/or when
   individual cache objects are deleted

   raises a warning if there is problem deleting a particular directory


.. py:class:: ElementCache(cache_dir=None, enabled=True)


   Bases: :py:obj:`object`

   Cache for element data -- i.e. the data associated with the particles.
   This caches UncertainSpillContainerPair
   The cache can be accessed to re-draw the LE movies, etc.

   TODO: This is a really fragile module in terms of handling multiple
         instances.  The __del__() method of previous instances can clear
         the _cache_dir at the whim of the GC.
         We may want to manage this differently.

   initialize a new cache object

   :param cache_dir=None: full path to the directory where the cache
                          should be stored.
                          If not provided, a temp dir will be created by
                          the python tempfile module

   .. py:method:: __del__()

      Clear out the cache when this object is deleted


   .. py:method:: _make_filename(step_num, uncertain=False)

      Returns a filename of the temp file generated from step_num

      :param step_num: the model step number that is saved/reloaded

      This here so that loading and saving use the same code


   .. py:method:: create_new_dir(cache_dir=None)


   .. py:method:: save_timestep(step_num, spill_container_pair)

      add a time step of data to the cache

      :param step_num: the step number of the data
      :param spill_container: the spill container at this step


   .. py:method:: load_timestep(step_num)

      Returns a SpillContainer with the data arrays cached on disk

      :param step_num: the step number you want to load.


   .. py:method:: _set_weathering_data(sc, data)

      add mass balance data to arrays


   .. py:method:: _get_weathering_data(data_arrays)


   .. py:method:: rewind()

      Rewinds the cache -- clearing out everything