def __init__(self, row, column, taxon_set, max_taxa, label):
self.taxon_set = taxon_set
self.label = label
message = "Cannot add '%s': Declared number of taxa (%d) already defined: %s" % \
(label,
max_taxa,
str([("%s" % t.label) for t in taxon_set]))
DataParseError.__init__(self, message=message, row=row, column=column)
def __init__(self, row, column, taxon_set, max_taxa, label):
self.taxon_set = taxon_set
self.label = label
message = "Cannot add '%s': Declared number of taxa (%d) already defined: %s" % (
label,
max_taxa,
str([("%s" % t.label) for t in taxon_set]),
)
DataParseError.__init__(self, message=message, row=row, column=column)
def data_format_error(self, message):
"""
Returns an exception object parameterized with line and
column number values.
"""
return DataParseError(message=message,
row=self.current_line_number,
column=self.current_col_number)
def read(self, stream):
"""
Main file parsing driver.
"""
if self.exclude_chars:
return self.dataset
if self.dataset is None:
self.dataset = dataobject.DataSet()
taxon_set = self.get_default_taxon_set()
self.char_matrix = self.dataset.new_char_matrix(
char_matrix_type=self.char_matrix_type, taxon_set=taxon_set)
if isinstance(self.char_matrix, dataobject.StandardCharacterMatrix) \
and len(self.char_matrix.state_alphabets) == 0:
self.char_matrix.state_alphabets.append(
dataobject.get_state_alphabet_from_symbols("0123456789"))
self.char_matrix.default_state_alphabet = self.char_matrix.state_alphabets[
0]
if self.char_matrix.default_state_alphabet is not None:
self.symbol_state_map = self.char_matrix.default_state_alphabet.symbol_state_map(
)
elif len(self.char_matrix.state_alphabets) == 0:
raise ValueError("No state alphabets defined")
elif len(self.char_matrix.state_alphabets) > 1:
raise NotImplementedError(
"Mixed state-alphabet matrices not supported")
else:
self.symbol_state_map = self.char_matrix.state_alphabets[0]
curr_vec = None
curr_taxon = None
if self.simple_rows:
legal_chars = self.char_matrix.default_state_alphabet.get_legal_symbols_as_str(
)
for line_index, line in enumerate(stream):
s = line.strip()
if not s:
continue
if s.startswith('>'):
if self.simple_rows and curr_taxon and curr_vec:
self.char_matrix[curr_taxon] = "".join(curr_vec)
name = s[1:].strip()
curr_taxon = taxon_set.require_taxon(label=name)
if curr_taxon in self.char_matrix:
raise DataParseError(
message="Fasta error: Repeated sequence name (%s) found"
% name,
row=line_index + 1,
stream=stream)
if curr_vec is not None and len(curr_vec) == 0:
raise DataParseError(
message=
"Fasta error: Expected sequence, but found another sequence name (%s)"
% name,
row=line_index + 1,
stream=stream)
if self.simple_rows:
curr_vec = []
else:
curr_vec = dataobject.CharacterDataVector(taxon=curr_taxon)
self.char_matrix[curr_taxon] = curr_vec
elif curr_vec is None:
raise DataParseError(
message=
"Fasta error: Expecting a lines starting with > before sequences",
row=line_index + 1,
stream=stream)
else:
if self.simple_rows:
for col_ind, c in enumerate(s):
c = c.strip()
if not c:
continue
if c not in legal_chars:
DataParseError(
message='Unrecognized sequence symbol "%s"' %
c,
row=line_index + 1,
column=col_ind + 1,
stream=stream)
curr_vec.append(c)
else:
for col_ind, c in enumerate(s):
c = c.strip()
if not c:
continue
try:
state = self.symbol_state_map[c]
curr_vec.append(
dataobject.CharacterDataCell(value=state))
except:
raise DataParseError(
message='Unrecognized sequence symbol "%s"' %
c,
row=line_index + 1,
column=col_ind + 1,
stream=stream)
if self.simple_rows and curr_taxon and curr_vec:
self.char_matrix[curr_taxon] = "".join(curr_vec)
return self.dataset
def _read(self,
stream,
taxon_namespace_factory=None,
tree_list_factory=None,
char_matrix_factory=None,
state_alphabet_factory=None,
global_annotations_target=None):
taxon_namespace = taxon_namespace_factory(label=None)
if self.data_type is None:
raise TypeError("Data type must be specified for this schema")
if self.data_type == "standard" and self.default_state_alphabet is not None:
char_matrix = char_matrix_factory(
self.data_type,
label=None,
taxon_namespace=taxon_namespace,
default_state_alphabet=self.default_state_alphabet,
)
else:
char_matrix = char_matrix_factory(self.data_type,
label=None,
taxon_namespace=taxon_namespace)
symbol_state_map = char_matrix.default_state_alphabet.full_symbol_state_map
curr_vec = None
curr_taxon = None
for line_index, line in enumerate(stream):
s = line.strip()
if not s:
continue
if s.startswith('>'):
name = s[1:].strip()
curr_taxon = taxon_namespace.require_taxon(label=name)
if curr_taxon in char_matrix:
raise DataParseError(
message=
"FASTA error: Repeated sequence name ('{}') found".
format(name),
line_num=line_index + 1,
stream=stream)
if curr_vec is not None and len(curr_vec) == 0:
raise DataParseError(
message=
"FASTA error: Expected sequence, but found another sequence name ('{}')"
.format(name),
line_num=line_index + 1,
stream=stream)
curr_vec = char_matrix[curr_taxon]
elif curr_vec is None:
raise DataParseError(
message=
"FASTA error: Expecting a lines starting with > before sequences",
line_num=line_index + 1,
stream=stream)
else:
states = []
for col_ind, c in enumerate(s):
c = c.strip()
if not c:
continue
try:
state = symbol_state_map[c]
except KeyError:
raise DataParseError(
message="Unrecognized sequence symbol '{}'".format(
c),
line_num=line_index + 1,
col_num=col_ind + 1,
stream=stream)
states.append(state)
curr_vec.extend(states)
product = self.Product(taxon_namespaces=None,
tree_lists=None,
char_matrices=[char_matrix])
return product
def __init__(self, *args, **kwargs):
DataParseError.__init__(self, *args, **kwargs)
def __init__(self, *args, **kwargs):
DataParseError.__init__(self, *args, **kwargs)
