Python containers.OrderedDict类代码示例

def bravais_lattice_to_space_groups(chiral_only=True):
  from cctbx import sgtbx
  from cctbx.sgtbx import bravais_types
  from libtbx.containers import OrderedDict
  bravais_lattice_to_sg = OrderedDict()
  for sgn in range(230):
    sg = sgtbx.space_group_info(number=sgn+1).group()
    if (not chiral_only) or (sg.is_chiral()):
      bravais_lattice = bravais_types.bravais_lattice(group=sg)
      bravais_lattice_to_sg.setdefault(str(bravais_lattice), [])
      bravais_lattice_to_sg[str(bravais_lattice)].append(sg)
  return bravais_lattice_to_sg

 def unique_beams(self):
   ''' Iterate through unique beams. '''
   from dxtbx.imageset import ImageSweep
   from libtbx.containers import OrderedDict
   obj = OrderedDict()
   for iset in self._imagesets:
     if isinstance(iset, ImageSweep):
       obj[iset.get_beam()] = None
     else:
       for i in range(len(iset)):
         obj[iset.get_beam(i)] = None
   return obj.keys()

class cctbx_data_structures_from_cif(object):
  def __init__(self,
               file_object=None,
               file_path=None,
               cif_model=None,
               data_structure_builder=None,
               data_block_name=None,
               base_array_info=None,
               **kwds):
    assert file_object is None or cif_model is None
    if data_structure_builder is None:
      data_structure_builders = (
        builders.miller_array_builder, builders.crystal_structure_builder)
    else:
      assert data_structure_builder in (
        builders.miller_array_builder, builders.crystal_structure_builder)
      data_structure_builders = (data_structure_builder,)

    self.xray_structures = OrderedDict()
    self.miller_arrays = OrderedDict()
    if cif_model is None:
      cif_model = reader(file_path=file_path, file_object=file_object).model()
    if not len(cif_model):
      raise Sorry("No data block found in CIF")
    if data_block_name is not None and not data_block_name in cif_model:
      if (file_path is None):
        msg = 'Unknown CIF data block name: "%s"' % data_block_name
      else:
        msg = 'Unknown CIF data block name "%s" in file: "%s"' % (
          data_block_name, file_path)
      raise RuntimeError(msg)
    errors = []
    wavelengths = {}
    for key, block in cif_model.items():
      if data_block_name is not None and key != data_block_name: continue
      for builder in data_structure_builders:
        if builder == builders.crystal_structure_builder:
          if '_atom_site_fract_x' in block or '_atom_site_Cartn_x' in block:
            self.xray_structures.setdefault(key, builder(block).structure)
        elif builder == builders.miller_array_builder:
          block_wavelengths = builders.get_wavelengths(block)
          if (block_wavelengths is not None) :
            wavelengths = block_wavelengths
          if base_array_info is not None:
            base_array_info = base_array_info.customized_copy(labels=[key])
          if ( '_refln_index_h' in block or '_refln.index_h' in block or
               '_diffrn_refln' in block
               ):
            self.miller_arrays.setdefault(
              key, builder(block, base_array_info=base_array_info,
                wavelengths=wavelengths).arrays())

    class image_data_cache(object):
      def __init__(self, imageset, size=10):
        self.imageset = imageset
        self.size = size
        self._image_data = OrderedDict()

      def __getitem__(self, i):
        image_data = self._image_data.get(i)
        if image_data is None:
          image_data = self.imageset.get_raw_data(i)
          if len(self._image_data) >= self.size:
            # remove the oldest entry in the cache
            del self._image_data[self._image_data.keys()[0]]
          self._image_data[i] = image_data
        return image_data

 def _unique_detectors_dict(self):
   ''' Returns an ordered dictionary of detector objects. '''
   from dxtbx.imageset import ImageSweep
   from libtbx.containers import OrderedDict
   obj = OrderedDict()
   for iset in self._imagesets:
     if isinstance(iset, ImageSweep):
       obj[iset.get_detector()] = None
     else:
       for i in range(len(iset)):
         obj[iset.get_detector(i)] = None
   detector_id = 0
   for detector in obj.keys():
     obj[detector] = detector_id
     detector_id = detector_id + 1
   return obj

 def unique_scans(self):
   ''' Iterate through unique scans. '''
   from dxtbx.imageset import ImageSweep
   from libtbx.containers import OrderedDict
   obj = OrderedDict()
   for iset in self._imagesets:
     if isinstance(iset, ImageSweep):
       obj[iset.get_scan()] = None
     else:
       for i in range(len(iset)):
         try:
           model = iset.get_scan(i)
           if model is not None:
             obj[model] = None
         except Exception:
           pass
   return obj.keys()

def exercise_odict():
  from libtbx.containers import OrderedDict as odict
  d = odict([('banana',3), ('apple',4), ('pear',1)])
  d.setdefault('orange', 2)
  assert d.has_key('orange')
  assert d['orange'] == 2
  assert d.keys() == ['banana', 'apple', 'pear', 'orange']
  assert d.values() == [3, 4, 1, 2]
  d = odict.fromkeys(('b','c','a'))
  assert d.keys() == ['b', 'c', 'a']

 def __init__(self, header=None, data=None):
   self._columns = OrderedDict()
   self.keys_lower = {}
   if header is not None:
     for key in header:
       self.setdefault(key, flex.std_string())
     if data is not None:
       # the number of data items must be an exact multiple of the number of headers
       assert len(data) % len(header) == 0, "Wrong number of data items for loop"
       n_rows = len(data)//len(header)
       n_columns = len(header)
       for i in range(n_rows):
         self.add_row([data[i*n_columns+j] for j in range(n_columns)])
   elif header is None and data is not None:
     assert isinstance(data, dict) or isinstance(data, OrderedDict)
     self.add_columns(data)
     self.keys_lower = dict(
       [(key.lower(), key) for key in self._columns.keys()])

  def __init__(self, unmerged_intensities, batches_all, n_bins=20, d_min=None,
               id_to_batches=None):

    sel = unmerged_intensities.sigmas() > 0
    unmerged_intensities = unmerged_intensities.select(sel)
    batches_all = batches_all.select(sel)

    unmerged_intensities.setup_binner(n_bins=n_bins)
    unmerged_intensities.show_summary()
    self.unmerged_intensities = unmerged_intensities
    self.merged_intensities = unmerged_intensities.merge_equivalents().array()

    separate = separate_unmerged(
      unmerged_intensities, batches_all, id_to_batches=id_to_batches)
    self.intensities = separate.intensities
    self.batches = separate.batches
    run_id_to_batch_id = separate.run_id_to_batch_id
    self.individual_merged_intensities = OrderedDict()
    for k in self.intensities.keys():
      self.intensities[k] = self.intensities[k].resolution_filter(d_min=d_min)
      self.batches[k] = self.batches[k].resolution_filter(d_min=d_min)
      self.individual_merged_intensities[k] = self.intensities[k].merge_equivalents().array()

    if run_id_to_batch_id is not None:
      labels = run_id_to_batch_id.values()
    else:
      labels = None
    racc = self.relative_anomalous_cc()
    if racc is not None:
      self.plot_relative_anomalous_cc(racc, labels=labels)
    correlation_matrix, linkage_matrix = self.compute_correlation_coefficient_matrix()

    self._cluster_dict = self.to_dict(correlation_matrix, linkage_matrix)

    self.plot_cc_matrix(correlation_matrix, linkage_matrix, labels=labels)

    self.write_output()

  def __init__(self, xinfo_file, sweep_ids=None, sweep_ranges=None):
    '''Initialise myself from an input .xinfo file.'''

    # first initialise all of the data structures which will hold the
    # information...

    self._project = None
    self._crystals = OrderedDict()

    if sweep_ids is not None:
      sweep_ids = [s.lower() for s in sweep_ids]
    if sweep_ranges is not None:
      assert sweep_ids is not None
      assert len(sweep_ids) == len(sweep_ranges)
    self._sweep_ids = sweep_ids
    self._sweep_ranges = sweep_ranges

    # read the contents of the xinfo file

    self._parse_project(xinfo_file)

    self._validate()

    return

 def sort(self, recursive=False, key=None, reverse=False):
   self.blocks = OrderedDict(sorted(self.blocks.items(), key=key, reverse=reverse))
   if recursive:
     for b in self.blocks.values():
       b.sort(recursive=recursive, reverse=reverse)

class loop(DictMixin):
  def __init__(self, header=None, data=None):
    self._columns = OrderedDict()
    self.keys_lower = {}
    if header is not None:
      for key in header:
        self.setdefault(key, flex.std_string())
      if data is not None:
        # the number of data items must be an exact multiple of the number of headers
        assert len(data) % len(header) == 0, "Wrong number of data items for loop"
        n_rows = len(data)//len(header)
        n_columns = len(header)
        for i in range(n_rows):
          self.add_row([data[i*n_columns+j] for j in range(n_columns)])
    elif header is None and data is not None:
      assert isinstance(data, dict) or isinstance(data, OrderedDict)
      self.add_columns(data)
      self.keys_lower = dict(
        [(key.lower(), key) for key in self._columns.keys()])

  def __setitem__(self, key, value):
    if not re.match(tag_re, key):
      raise Sorry("%s is not a valid data name" %key)
    if len(self) > 0:
      assert len(value) == self.size()
    if not isinstance(value, flex.std_string):
      for flex_numeric_type in (flex.int, flex.double):
        if isinstance(value, flex_numeric_type):
          value = value.as_string()
        else:
          try:
            value = flex_numeric_type(value).as_string()
          except TypeError:
            continue
          else:
            break
      if not isinstance(value, flex.std_string):
        value = flex.std_string(value)
    # value must be a mutable type
    assert hasattr(value, '__setitem__')
    self._columns[key] = value
    self.keys_lower[key.lower()] = key

  def __getitem__(self, key):
    return self._columns[self.keys_lower[key.lower()]]

  def __delitem__(self, key):
    del self._columns[self.keys_lower[key.lower()]]
    del self.keys_lower[key.lower()]

  def keys(self):
    return self._columns.keys()

  def __repr__(self):
    return repr(OrderedDict(self.iteritems()))

  def name(self):
    return common_substring(self.keys()).rstrip('_').rstrip('.')

  def size(self):
    size = 0
    for column in self.values():
      size = max(size, len(column))
    return size

  def n_rows(self):
    size = 0
    for column in self.values():
      size = max(size, len(column))
    return size

  def n_columns(self):
    return len(self.keys())

  def add_row(self, row, default_value="?"):
    if isinstance(row, dict):
      for key in self:
        if key in row:
          self[key].append(str(row[key]))
        else:
          self[key].append(default_value)
    else:
      assert len(row) == len(self)
      for i, key in enumerate(self):
        self[key].append(str(row[i]))

  def add_column(self, key, values):
    if self.size() != 0:
      assert len(values) == self.size()
    self[key] = values
    self.keys_lower[key.lower()] = key

  def add_columns(self, columns):
    assert isinstance(columns, dict) or isinstance(columns, OrderedDict)
    for key, value in columns.iteritems():
      self.add_column(key, value)

  def update_column(self, key, values):
    assert type(key)==type(""), "first argument is column key string"
    if self.size() != 0:
#.........这里部分代码省略.........

class XInfo(object):
  '''A class to represent all of the input to the xia2dpa system, with
  enough information to allow structure solution, as parsed from a
  .xinfo file, an example of which is in the source code.'''

  def __init__(self, xinfo_file, sweep_ids=None, sweep_ranges=None):
    '''Initialise myself from an input .xinfo file.'''

    # first initialise all of the data structures which will hold the
    # information...

    self._project = None
    self._crystals = OrderedDict()

    if sweep_ids is not None:
      sweep_ids = [s.lower() for s in sweep_ids]
    if sweep_ranges is not None:
      assert sweep_ids is not None
      assert len(sweep_ids) == len(sweep_ranges)
    self._sweep_ids = sweep_ids
    self._sweep_ranges = sweep_ranges

    # read the contents of the xinfo file

    self._parse_project(xinfo_file)

    self._validate()

    return

  def get_output(self):
    '''Generate a string representation of the project.'''

    text = 'Project %s\n' % self._project
    for crystal in self._crystals.keys():
      text += 'Crystal %s\n' % crystal
      text += '%s\n' % self._crystals[crystal].get_output()

    # remove a trailing newline...

    return text[:-1]

  def get_project(self):
    return self._project

  def get_crystals(self):
    return self._crystals

  def _validate(self):
    '''Validate the structure of this object, ensuring that
    everything looks right... raise exception if I find something
    wrong.'''

    return True

  def _parse_project(self, xinfo_file):
    '''Parse & validate the contents of the .xinfo file. This parses the
    project element (i.e. the whole thing..)'''

    project_records = []

    for r in open(xinfo_file, 'r').readlines():
      record = r.strip()
      if not record:
        pass
      elif record[0] == '!' or record[0] == '#':
        pass
      else :
        # then it may contain something useful...
        project_records.append(record)

    # so now we have loaded the whole file into memory stripping
    # out the crud... let's look for something useful

    for i in range(len(project_records)):
      record = project_records[i]
      if 'BEGIN PROJECT' in record:
        self._project = record.replace('BEGIN PROJECT', '').strip()
      if 'END PROJECT' in record:
        if not self._project == record.replace(
            'END PROJECT', '').strip():
          raise RuntimeError, 'error parsing END PROJECT record'

      # next look for crystals
      if 'BEGIN CRYSTAL ' in record:
        crystal_records = [record]
        while True:
          i += 1
          record = project_records[i]
          crystal_records.append(record)
          if 'END CRYSTAL ' in record:
            break

        self._parse_crystal(crystal_records)

      # that's everything, because parse_crystal handles
      # the rest...

    return

#.........这里部分代码省略.........

 def __init__(self, imageset, size=10):
   self.imageset = imageset
   self.size = size
   self._image_data = OrderedDict()

class multi_crystal_analysis(object):

  def __init__(self, unmerged_intensities, batches_all, n_bins=20, d_min=None,
               id_to_batches=None):

    sel = unmerged_intensities.sigmas() > 0
    unmerged_intensities = unmerged_intensities.select(sel)
    batches_all = batches_all.select(sel)

    unmerged_intensities.setup_binner(n_bins=n_bins)
    unmerged_intensities.show_summary()
    self.unmerged_intensities = unmerged_intensities
    self.merged_intensities = unmerged_intensities.merge_equivalents().array()

    separate = separate_unmerged(
      unmerged_intensities, batches_all, id_to_batches=id_to_batches)
    self.intensities = separate.intensities
    self.batches = separate.batches
    run_id_to_batch_id = separate.run_id_to_batch_id
    self.individual_merged_intensities = OrderedDict()
    for k in self.intensities.keys():
      self.intensities[k] = self.intensities[k].resolution_filter(d_min=d_min)
      self.batches[k] = self.batches[k].resolution_filter(d_min=d_min)
      self.individual_merged_intensities[k] = self.intensities[k].merge_equivalents().array()

    if run_id_to_batch_id is not None:
      labels = run_id_to_batch_id.values()
    else:
      labels = None
    racc = self.relative_anomalous_cc()
    if racc is not None:
      self.plot_relative_anomalous_cc(racc, labels=labels)
    correlation_matrix, linkage_matrix = self.compute_correlation_coefficient_matrix()

    self._cluster_dict = self.to_dict(correlation_matrix, linkage_matrix)

    self.plot_cc_matrix(correlation_matrix, linkage_matrix, labels=labels)

    self.write_output()

  def to_dict(self, correlation_matrix, linkage_matrix):

    from scipy.cluster import hierarchy
    tree = hierarchy.to_tree(linkage_matrix, rd=False)
    leaves_list = hierarchy.leaves_list(linkage_matrix)

    d = {}

    # http://w3facility.org/question/scipy-dendrogram-to-json-for-d3-js-tree-visualisation/
    # https://gist.github.com/mdml/7537455

    def add_node(node):
      if node.is_leaf(): return
      cluster_id = node.get_id() - len(linkage_matrix) - 1
      row = linkage_matrix[cluster_id]
      d[cluster_id+1] = {
        'datasets': [i+1 for i in sorted(node.pre_order())],
        'height': row[2],
      }

      # Recursively add the current node's children
      if node.left: add_node(node.left)
      if node.right: add_node(node.right)

    add_node(tree)

    return d

  def relative_anomalous_cc(self):
    if self.unmerged_intensities.anomalous_flag():
      d_min = min([ma.d_min() for ma in self.intensities.values()])
      racc = flex.double()
      full_set_anom_diffs = self.merged_intensities.anomalous_differences()
      for i_wedge in self.individual_merged_intensities.keys():
        ma_i = self.individual_merged_intensities[i_wedge].resolution_filter(d_min=d_min)
        anom_i = ma_i.anomalous_differences()
        anom_cc = anom_i.correlation(full_set_anom_diffs, assert_is_similar_symmetry=False).coefficient()
        racc.append(anom_cc)
      return racc

  def plot_relative_anomalous_cc(self, racc, labels=None):
    perm = flex.sort_permutation(racc)
    fig = pyplot.figure(dpi=1200, figsize=(16,12))
    pyplot.bar(range(len(racc)), list(racc.select(perm)))
    if labels is None:
      labels = ["%.0f" %(j+1) for j in perm]
    assert len(labels) == len(racc)
    pyplot.xticks([i+0.5 for i in range(len(racc))], labels)
    locs, labels = pyplot.xticks()
    pyplot.setp(labels, rotation=70)
    pyplot.xlabel("Dataset")
    pyplot.ylabel("Relative anomalous correlation coefficient")
    fig.savefig("racc.png")

  def compute_correlation_coefficient_matrix(self):
    from scipy.cluster import hierarchy
    import scipy.spatial.distance as ssd

    correlation_matrix = flex.double(
      flex.grid(len(self.intensities), len(self.intensities)))
#.........这里部分代码省略.........