def convert_to_nexml(obj_dict, addindent='', newl='', use_default_root_atts=True, otu_label='ot:originalLabel'):
f, wrapper = get_utf_8_string_io_writer()
write_obj_as_nexml(obj_dict,
file_obj=wrapper,
addindent=addindent,
newl=newl,
use_default_root_atts=use_default_root_atts,
otu_label=otu_label)
flush_utf_8_writer(wrapper)
return f.getvalue()
def convert_to_nexml(obj_dict, addindent='', newl='', use_default_root_atts=True, otu_label='ot:originalLabel'):
f, wrapper = get_utf_8_string_io_writer()
write_obj_as_nexml(obj_dict,
file_obj=wrapper,
addindent=addindent,
newl=newl,
use_default_root_atts=use_default_root_atts,
otu_label=otu_label)
flush_utf_8_writer(wrapper)
return f.getvalue()
def _write_nexus_format(quoted_leaf_labels, tree_name_newick_list):
f, wrapper = get_utf_8_string_io_writer()
wrapper.write('''#NEXUS
BEGIN TAXA;
Dimensions NTax = {s};
TaxLabels {l} ;
END;
BEGIN TREES;
'''.format(s=len(quoted_leaf_labels), l=' '.join(quoted_leaf_labels)))
for name, newick in tree_name_newick_list:
wrapper.write(' Tree ')
wrapper.write(name)
wrapper.write(' = ')
wrapper.write(newick)
wrapper.write('\nEND;\n')
flush_utf_8_writer(wrapper)
return f.getvalue()
def _write_nexus_format(quoted_leaf_labels, tree_name_newick_list):
f, wrapper = get_utf_8_string_io_writer()
wrapper.write('''#NEXUS
BEGIN TAXA;
Dimensions NTax = {s};
TaxLabels {l} ;
END;
BEGIN TREES;
'''.format(s=len(quoted_leaf_labels), l=' '.join(quoted_leaf_labels)))
for name, newick in tree_name_newick_list:
wrapper.write(' Tree ')
wrapper.write(name)
wrapper.write(' = ')
wrapper.write(newick)
wrapper.write('\nEND;\n')
flush_utf_8_writer(wrapper)
return f.getvalue()
def convert_tree_to_newick(tree,
otu_group,
label_key,
leaf_labels,
needs_quotes_pattern,
subtree_id=None,
bracket_ingroup=False):
assert label_key in PhyloSchema._NEWICK_PROP_VALS #pylint: disable=W0212
unlabeled_counter = 0
ingroup_node_id = tree.get('^ot:inGroupClade')
if subtree_id:
if subtree_id == 'ingroup':
root_id = ingroup_node_id
ingroup_node_id = None # turns of the comment pre-ingroup-marker
else:
root_id = subtree_id
else:
root_id = tree['^ot:rootNodeId']
edges = tree['edgeBySourceId']
if root_id not in edges:
return None
nodes = tree['nodeById']
curr_node_id = root_id
curr_edge = None
curr_sib_list = []
curr_stack = []
sio, out = get_utf_8_string_io_writer()
going_tipward = True
while True:
if going_tipward:
outgoing_edges = edges.get(curr_node_id)
if outgoing_edges is None:
curr_node = nodes[curr_node_id]
unlabeled_counter = _write_newick_leaf_label(out,
curr_node,
otu_group,
label_key,
leaf_labels,
unlabeled_counter,
needs_quotes_pattern)
_write_newick_edge_len(out, curr_edge)
going_tipward = False
else:
te = [(i, e) for i, e in outgoing_edges.items()]
te.sort() # produce a consistent rotation... Necessary?
if bracket_ingroup and (ingroup_node_id == curr_node_id):
out.write('[pre-ingroup-marker]')
out.write('(')
next_p = te.pop(0)
curr_stack.append((curr_edge, curr_node_id, curr_sib_list))
curr_edge, curr_sib_list = next_p[1], te
curr_node_id = curr_edge['@target']
if not going_tipward:
next_up_edge_id = None
while True:
if curr_sib_list:
out.write(',')
next_up_edge_id, next_up_edge = curr_sib_list.pop(0)
break
if curr_stack:
curr_edge, curr_node_id, curr_sib_list = curr_stack.pop(-1)
curr_node = nodes[curr_node_id]
out.write(')')
_write_newick_internal_label(out,
curr_node,
otu_group,
label_key,
needs_quotes_pattern)
_write_newick_edge_len(out, curr_edge)
if bracket_ingroup and (ingroup_node_id == curr_node_id):
out.write('[post-ingroup-marker]')
else:
break
if next_up_edge_id is None:
break
curr_edge = next_up_edge
curr_node_id = curr_edge['@target']
going_tipward = True
out.write(';')
flush_utf_8_writer(out)
return sio.getvalue()
def convert(self, src, serialize=None, output_dest=None, src_schema=None):
if src_schema is None:
src_format = PhyloSchema.NEXSON
current_format = None
else:
src_format = src_schema.format_code
current_format = src_schema.version
if not self.can_convert_from():
m = 'Conversion of {c} to {d} is not supported'.format(c=self.content, d=self.description)
raise NotImplementedError(m)
if src_format != PhyloSchema.NEXSON:
raise NotImplementedError('Only conversion from NexSON is currently supported')
if self.format_code == PhyloSchema.NEXSON:
d = src
if self.content == 'study':
d = convert_nexson_format(src,
out_nexson_format=self.version,
current_format=current_format,
remove_old_structs=True,
pristine_if_invalid=False,
sort_arbitrary=False)
elif self.content in ('tree', 'subtree'):
if self.content == 'tree' and self.cull_nonmatching:
d = cull_nonmatching_trees(d, self.content_id, current_format)
d = convert_nexson_format(d,
out_nexson_format=self.version,
current_format=current_format,
remove_old_structs=True,
pristine_if_invalid=False,
sort_arbitrary=False)
else:
i_t_o_list = extract_tree_nexson(d, self.content_id, current_format)
d = {}
for ito_tup in i_t_o_list:
i, t = ito_tup[0], ito_tup[1]
d[i] = t
elif self.content == 'meta':
strip_to_meta_only(d, current_format)
elif self.content == 'otus':
d = extract_otus_nexson(d, self.content_id, current_format)
elif self.content == 'otu':
d = extract_otu_nexson(d, self.content_id, current_format)
elif self.content == 'otumap':
if self.content_id is None:
r = extract_otu_nexson(d, None, current_format)
else:
p = extract_otus_nexson(d, self.content_id, current_format)
if p is None:
r = extract_otu_nexson(d, self.content_id, current_format)
else:
r = {}
for v in p.values():
r.update(v.get('otuById', {}))
if not r:
return None
d = _otu_dict_to_otumap(r)
elif self.content == 'treelist':
i_t_o_list = extract_tree_nexson(d,
self.content_id,
current_format)
d = [i[0] for i in i_t_o_list]
if d is None:
return None
if serialize:
if output_dest:
write_as_json(d, output_dest)
return None
else:
f, wrapper = get_utf_8_string_io_writer()
write_as_json(d, wrapper)
flush_utf_8_writer(wrapper)
return f.getvalue()
else:
return d
# Non-NexSON types go here...
if (serialize is not None) and (not serialize):
raise ValueError('Conversion without serialization is only supported for the NexSON format')
if output_dest:
if is_str_type(output_dest):
output_dest = codecs.open(output_dest, 'w', encoding='utf-8')
if self.format_code == PhyloSchema.NEXML:
if output_dest:
write_obj_as_nexml(src, output_dest, addindent=' ', newl='\n', otu_label=self.otu_label_prop)
return
return convert_to_nexml(src, addindent=' ', newl='\n', otu_label=self.otu_label_prop)
elif self.format_code in [PhyloSchema.NEXUS, PhyloSchema.NEWICK]:
if self.content in ('tree', 'subtree'):
if isinstance(self.content_id, list) or isinstance(self.content_id, tuple):
ci, subtree_id = self.content_id
else:
ci, subtree_id = self.content_id, None
else:
ci, subtree_id = None, None
response = extract_tree(src, ci, self, subtree_id=subtree_id)
# these formats are always serialized...
if output_dest:
output_dest.write(response)
output_dest.write('\n')
return response
assert False
def convert_tree_to_newick(tree,
otu_group,
label_key,
leaf_labels,
needs_quotes_pattern,
subtree_id=None,
bracket_ingroup=False):
assert label_key in PhyloSchema._NEWICK_PROP_VALS #pylint: disable=W0212
unlabeled_counter = 0
ingroup_node_id = tree.get('^ot:inGroupClade')
if subtree_id:
if subtree_id == 'ingroup':
root_id = ingroup_node_id
ingroup_node_id = None # turns of the comment pre-ingroup-marker
else:
root_id = subtree_id
else:
root_id = tree['^ot:rootNodeId']
edges = tree['edgeBySourceId']
if root_id not in edges:
return None
nodes = tree['nodeById']
curr_node_id = root_id
curr_edge = None
curr_sib_list = []
curr_stack = []
sio, out = get_utf_8_string_io_writer()
going_tipward = True
while True:
if going_tipward:
outgoing_edges = edges.get(curr_node_id)
if outgoing_edges is None:
curr_node = nodes[curr_node_id]
unlabeled_counter = _write_newick_leaf_label(out,
curr_node,
otu_group,
label_key,
leaf_labels,
unlabeled_counter,
needs_quotes_pattern)
_write_newick_edge_len(out, curr_edge)
going_tipward = False
else:
te = [(i, e) for i, e in outgoing_edges.items()]
te.sort() # produce a consistent rotation... Necessary?
if bracket_ingroup and (ingroup_node_id == curr_node_id):
out.write('[pre-ingroup-marker]')
out.write('(')
next_p = te.pop(0)
curr_stack.append((curr_edge, curr_node_id, curr_sib_list))
curr_edge, curr_sib_list = next_p[1], te
curr_node_id = curr_edge['@target']
if not going_tipward:
next_up_edge_id = None
while True:
if curr_sib_list:
out.write(',')
next_up_edge_id, next_up_edge = curr_sib_list.pop(0)
break
if curr_stack:
curr_edge, curr_node_id, curr_sib_list = curr_stack.pop(-1)
curr_node = nodes[curr_node_id]
out.write(')')
_write_newick_internal_label(out,
curr_node,
otu_group,
label_key,
needs_quotes_pattern)
_write_newick_edge_len(out, curr_edge)
if bracket_ingroup and (ingroup_node_id == curr_node_id):
out.write('[post-ingroup-marker]')
else:
break
if next_up_edge_id is None:
break
curr_edge = next_up_edge
curr_node_id = curr_edge['@target']
going_tipward = True
out.write(';')
flush_utf_8_writer(out)
return sio.getvalue()
def convert(self, src, serialize=None, output_dest=None, src_schema=None):
if src_schema is None:
src_format = PhyloSchema.NEXSON
current_format = None
else:
src_format = src_schema.format_code
current_format = src_schema.version
if not self.can_convert_from():
m = 'Conversion of {c} to {d} is not supported'.format(c=self.content, d=self.description)
raise NotImplementedError(m)
if src_format != PhyloSchema.NEXSON:
raise NotImplementedError('Only conversion from NexSON is currently supported')
if self.format_code == PhyloSchema.NEXSON:
d = src
if self.content == 'study':
d = convert_nexson_format(src,
out_nexson_format=self.version,
current_format=current_format,
remove_old_structs=True,
pristine_if_invalid=False,
sort_arbitrary=False)
elif self.content in ('tree', 'subtree'):
if self.content == 'tree' and self.cull_nonmatching:
d = cull_nonmatching_trees(d, self.content_id, current_format)
d = convert_nexson_format(d,
out_nexson_format=self.version,
current_format=current_format,
remove_old_structs=True,
pristine_if_invalid=False,
sort_arbitrary=False)
else:
i_t_o_list = extract_tree_nexson(d, self.content_id, current_format)
d = {}
for ito_tup in i_t_o_list:
i, t = ito_tup[0], ito_tup[1]
d[i] = t
elif self.content == 'meta':
strip_to_meta_only(d, current_format)
elif self.content == 'otus':
d = extract_otus_nexson(d, self.content_id, current_format)
elif self.content == 'otu':
d = extract_otu_nexson(d, self.content_id, current_format)
elif self.content == 'otumap':
if self.content_id is None:
r = extract_otu_nexson(d, None, current_format)
else:
p = extract_otus_nexson(d, self.content_id, current_format)
if p is None:
r = extract_otu_nexson(d, self.content_id, current_format)
else:
r = {}
for v in p.values():
r.update(v.get('otuById', {}))
if not r:
return None
d = _otu_dict_to_otumap(r)
elif self.content == 'treelist':
i_t_o_list = extract_tree_nexson(d,
self.content_id,
current_format)
d = [i[0] for i in i_t_o_list]
if d is None:
return None
if serialize:
if output_dest:
write_as_json(d, output_dest)
return None
else:
f, wrapper = get_utf_8_string_io_writer()
write_as_json(d, wrapper)
flush_utf_8_writer(wrapper)
return f.getvalue()
else:
return d
# Non-NexSON types go here...
if (serialize is not None) and (not serialize):
raise ValueError('Conversion without serialization is only supported for the NexSON format')
if output_dest:
if is_str_type(output_dest):
output_dest = codecs.open(output_dest, 'w', encoding='utf-8')
if self.format_code == PhyloSchema.NEXML:
if output_dest:
write_obj_as_nexml(src, output_dest, addindent=' ', newl='\n', otu_label=self.otu_label_prop)
return
return convert_to_nexml(src, addindent=' ', newl='\n', otu_label=self.otu_label_prop)
elif self.format_code in [PhyloSchema.NEXUS, PhyloSchema.NEWICK]:
if self.content in ('tree', 'subtree'):
if isinstance(self.content_id, list) or isinstance(self.content_id, tuple):
ci, subtree_id = self.content_id
else:
ci, subtree_id = self.content_id, None
else:
ci, subtree_id = None, None
response = extract_tree(src, ci, self, subtree_id=subtree_id)
# these formats are always serialized...
if output_dest:
output_dest.write(response)
output_dest.write('\n')
return response
assert False
