# -----------------------------------------------------------------------------
# Subsampled_Grid wraps a set of Grid_Data objects.  One of the wrapped grid
# objects is the primary data and the others are subsamples.
#

from VolumeData import Grid_Data

# -----------------------------------------------------------------------------
#
class Subsampled_Grid(Grid_Data):

  def __init__(self, primary_grid_data):

    pg = primary_grid_data

    self.available_subsamplings = {(1,1,1): pg}

    Grid_Data.__init__(self, pg.size, pg.value_type, pg.origin, pg.step,
                       pg.cell_angles, pg.rotation, pg.symmetries,
                       name = pg.name, path = pg.path, file_type = pg.file_type,
                       grid_id = pg.grid_id, default_color = pg.rgba)
    self.data_cache = None      # Caching done by underlying grid objects.

  # ---------------------------------------------------------------------------
  # It is the caller's responsibility to verify that the passed in subsampled
  # data has a matching set of components and a valid subsampling size.
  #
  def add_subsamples(self, grid_data, cell_size):

    csize = tuple(cell_size)
    if self.available_subsamplings.has_key(csize):
      del self.available_subsamplings[csize]
    self.available_subsamplings[csize] = grid_data
    
  # ---------------------------------------------------------------------------
  #
  def matrix(self, ijk_origin = (0,0,0), ijk_size = None,
             ijk_step = (1,1,1), progress = None, from_cache_only = False,
             subsampling = (1,1,1)):

    if ijk_size == None:
      ijk_size = self.size

    d = self.available_subsamplings[tuple(subsampling)]
    m = d.matrix(ijk_origin, ijk_size, ijk_step, progress, from_cache_only)
    return m

  # ---------------------------------------------------------------------------
  #
  def read_matrix(self, ijk_origin = (0,0,0), ijk_size = None,
                  ijk_step = (1,1,1), progress = None):

    m = self.matrix(ijk_origin, ijk_size, ijk_step, progress)
    return m
        
  # ---------------------------------------------------------------------------
  #
  def cached_data(self, ijk_origin, ijk_size, ijk_step):

    m = self.matrix(ijk_origin, ijk_size, ijk_step, from_cache_only = True)
    return m

  # ---------------------------------------------------------------------------
  #
  def clear_cache(self):

    for d in self.available_subsamplings.values():
      d.clear_cache()