from __future__ import absolute_import, division, print_function
from libtbx.utils import Sorry

class application:
  def __init__(self,param):

    self.param = param
    self.application_level_validation()

  def application_level_validation(self):

    if self.param.merging.reverse_lookup is not None:
      if self.param.data_reindex_op != "h,k,l":
        raise Sorry("The data reindex operator "+self.param.data_reindex_op+
        """ cannot be given in combination with the reverse lookup file
       """+self.param.merging.reverse_lookup+""".  The reverse lookup table itself contains
       a reindexing operator for each image in the data set.""")

class samosa:
  def __init__(self,param,allow_model=False):

    self.param = param
    self.allow_model = allow_model
    self.application_level_validation()

  def application_level_validation(self):

    if self.allow_model==False and self.param.model is not None:
      raise Sorry("""For samosa, no PDB structural model is used for frame-to-frame
      scaling, therefore the model phil parameter must be set to None.""")
    if self.param.target_unit_cell is None or self.param.target_space_group is None:
      raise Sorry("""For samosa, target_unit_cell and space_group must be given""")
    if not self.param.scaling.algorithm in ['mark1','levmar']:
      raise Sorry("""Must specify either mark1 or levmar algorithm for scaling.
      (Both algorithms have the same effect within samosa.""")
    #if self.param.significance_filter.apply is True:
    #  raise Sorry("""No significance filter for samosa.  Variance weighting is
    #  used to downweight weak data.""")
    if self.param.raw_data.sdfac_auto or self.param.raw_data.sdfac_refine:
      raise Sorry("""SDFAC manipulation not implemented for samosa.""")
    if self.param.postrefinement.enable:
      raise Sorry("""Postrefinement not currently available in samosa.""")
    if not self.param.include_negatives:
      raise Sorry("""Normally negative values are included in samosa.""")
    if not self.param.scaling.report_ML:
      raise Sorry("""Must have 'report_ML' estimates of mosaic model from integration program""")
    if "PartialityDeff" in self.param.levmar.parameter_flags and \
       "PartialityEtaDeff" in self.param.levmar.parameter_flags:
      raise Sorry("""Only one partiality model""")
    if self.param.memory.shared_array_allocation is None:
      raise Sorry("""For samosa.join, the expected number of measurements must be given
      in the memory.shared_array_allocation phil parameter""")
    if self.param.backend != "Flex":
      raise Sorry("""For samosa, we are only using the Flex backend at present""")

    # samosa.scale requires mtz_file to be set & file to be readable in order to produce isomorphous stats
    if self.param.scaling.mtz_file is not None:
      from iotbx import mtz
      mtz_object = mtz.object(file_name=self.param.scaling.mtz_file)
      comparison_column_found = False
      obs_labels = []
      for array in mtz_object.as_miller_arrays():
        this_label = array.info().label_string().lower()
        if array.observation_type() is not None:
          obs_labels.append(this_label.split(',')[0])
        if this_label.find(self.param.scaling.mtz_column_F)==0:
          comparison_column_found = True
          break
      if not comparison_column_found:
        raise Sorry(self.param.scaling.mtz_file +
                  " has labels [" +
                  ", ".join(obs_labels) +
                  "].  Please set scaling.mtz_column_F to one of these.")