def _parse_tree_statement(self, taxon_set=None):
"""
Processes a TREE command. Assumes that the file reader is
positioned right after the "TREE" token in a TREE command.
Calls on the NewickStatementParser of the trees module.
"""
token = self.stream_tokenizer.read_next_token()
if token == '*':
token = self.stream_tokenizer.read_next_token()
tree_name = token
token = self.stream_tokenizer.read_next_token()
if token != '=':
raise self.data_format_error("Expecting '=' in definition of Tree '%s' but found '%s'" % (tree_name, token))
tree_comments = self.stream_tokenizer.comments
tree = nexustokenizer.tree_from_token_stream(stream_tokenizer=self.stream_tokenizer,
taxon_set=taxon_set,
translate_dict=self.tree_translate_dict,
encode_splits=self.encode_splits,
rooting_interpreter=self.rooting_interpreter,
finish_node_func=self.finish_node_func,
extract_comment_metadata=self.extract_comment_metadata,
store_tree_weights=self.store_tree_weights,
preserve_underscores=self.preserve_underscores,
suppress_internal_node_taxa=self.suppress_internal_node_taxa,
edge_len_type=self.edge_len_type,
case_sensitive_taxon_labels=self.case_sensitive_taxon_labels)
tree.label = tree_name
if tree_comments is not None and len(tree_comments) > 0:
tree.comments.extend(tree_comments)
if self.stream_tokenizer.current_token != ';':
self.stream_tokenizer.skip_to_semicolon()
return tree
def _parse_tree_statement(self, taxon_set=None):
"""
Processes a TREE command. Assumes that the file reader is
positioned right after the "TREE" token in a TREE command.
Calls on the NewickStatementParser of the trees module.
"""
token = self.stream_tokenizer.read_next_token()
if token == '*':
token = self.stream_tokenizer.read_next_token()
tree_name = token
token = self.stream_tokenizer.read_next_token()
if token != '=':
raise self.data_format_error(
"Expecting '=' in definition of Tree '%s' but found '%s'" %
(tree_name, token))
tree_comments = self.stream_tokenizer.comments
tree = nexustokenizer.tree_from_token_stream(
stream_tokenizer=self.stream_tokenizer,
taxon_set=taxon_set,
translate_dict=self.tree_translate_dict,
encode_splits=self.encode_splits,
rooting_interpreter=self.rooting_interpreter,
finish_node_func=self.finish_node_func,
extract_comment_metadata=self.extract_comment_metadata,
store_tree_weights=self.store_tree_weights,
preserve_underscores=self.preserve_underscores,
suppress_internal_node_taxa=self.suppress_internal_node_taxa,
edge_len_type=self.edge_len_type,
case_sensitive_taxon_labels=self.case_sensitive_taxon_labels)
tree.label = tree_name
if tree_comments is not None and len(tree_comments) > 0:
tree.comments.extend(tree_comments)
if self.stream_tokenizer.current_token != ';':
self.stream_tokenizer.skip_to_semicolon()
return tree
def _ncl_tree_tokens_to_native_tree(self,
ncl_tb,
taxa_block,
tree_tokens,
rooted_flag=None):
if not tree_tokens:
return None
iid = ncl_tb.GetInstanceIdentifierString()
if taxa_block is None:
taxa_block = self._ncl_taxa_block_to_native(ncl_tb)
# self.taxa_block = taxa_block
lti = ListOfTokenIterator(tree_tokens)
lti.tree_rooted = rooted_flag
if iid not in self.tree_translate_dicts:
self.tree_translate_dicts[ncl_tb] = {}
for n, t in enumerate(taxa_block):
self.tree_translate_dicts[ncl_tb][str(n + 1)] = t
if self.encode_splits:
t.clade_mask = (1 << n)
return nexustokenizer.tree_from_token_stream(
lti,
taxon_set=taxa_block,
translate_dict=self.tree_translate_dicts[ncl_tb],
encode_splits=self.encode_splits,
rooting_interpreter=self.rooting_interpreter,
finish_node_func=self.finish_node_func)
def _ncl_tree_tokens_to_native_tree(self, ncl_tb, taxa_block, tree_tokens, rooted_flag=None):
if not tree_tokens:
return None
iid = ncl_tb.GetInstanceIdentifierString()
if taxa_block is None:
taxa_block = self._ncl_taxa_block_to_native(ncl_tb)
# self.taxa_block = taxa_block
lti = ListOfTokenIterator(tree_tokens)
lti.tree_rooted = rooted_flag
if iid not in self.tree_translate_dicts:
self.tree_translate_dicts[ncl_tb] = {}
for n, t in enumerate(taxa_block):
self.tree_translate_dicts[ncl_tb][str(n + 1)] = t
if self.encode_splits:
t.clade_mask = (1 << n)
return nexustokenizer.tree_from_token_stream(lti,
taxon_set=taxa_block,
translate_dict=self.tree_translate_dicts[ncl_tb],
encode_splits=self.encode_splits,
rooting_interpreter=self.rooting_interpreter,
finish_node_func=self.finish_node_func)
def tree_source_iter(stream, **kwargs):
"""
Iterates over a NEWICK-formatted source of trees given by file-like object
`stream`
Note that if `encode_splits` is True, then a `taxon_set` has to be given.
This is because adding Taxon objects to a taxon set may invalidate split
bitmasks. Because NEWICK tree taxa are added to a TaxonSet as they are found
on a tree, there is a strong possibility that all split bitmasks get
invalidated in the middle of parsing a tree. To avoid this, and, more
importantly to avoid errors downstream in client code due to this, we
force specification of a `taxon_set` if `encode_splits` is requested.
The following optional keyword arguments are also recognized:
`taxon_set`
TaxonSet object to use when reading data.
`as_rooted=True` (or `as_unrooted=False`)
Unconditionally interprets all trees as rooted.
`as_unrooted=True` (or `as_rooted=False`)
Unconditionally interprets all trees as unrooted.
`default_as_rooted=True` (or `default_as_unrooted=False`)
Interprets all trees as rooted if rooting not given by `[&R]`
or `[&U]` comments.
`default_as_unrooted=True` (or `default_as_rooted=False`)
Interprets all trees as rooted if rooting not given by `[&R]`
or `[&U]` comments.
`edge_len_type`
Specifies the type of the edge lengths (int or float).
`extract_comment_metadata`
If True, any comments that begin with '&' or '&&' associated with
items will be processed and stored as part of the annotation set of
the object (`annotations`) If False, this will be skipped. Defaults
to False.
`store_tree_weights`
If True, process the tree weight ("[&W 1/2]") comment
associated with each tree, if any.
`encode_splits`
Specifies whether or not split bitmasks will be calculated and
attached to the edges.
`finish_node_func`
Is a function that will be applied to each node after it has
been constructed.
`case_sensitive_taxon_labels`
If True, then taxon labels are case sensitive (different cases
= different taxa); defaults to False.
`allow_duplicate_taxon_labels`
if True, allow duplicate labels on trees
`preserve_underscores`
If True, unquoted underscores in labels will *not* converted to
spaces. Defaults to False: all underscores not protected by
quotes will be converted to spaces.
`suppress_internal_node_taxa`
If False, internal node labels will be instantantiatd into Taxon
objects. Defaults to True: internal node labels will *not* be
treated as taxa.
`allow_duplicate_taxon_labels`
If True, then multiple identical taxon labels will be allowed.
Defaults to False: treat multiple identical taxon labels as an
error.
`hyphens_as_tokens`
If True, hyphens will be treated as special punctuation
characters. Defaults to False, hyphens not treated as special
punctuation characters.
"""
if "taxon_set" in kwargs:
taxon_set = kwargs["taxon_set"]
del(kwargs["taxon_set"])
else:
taxon_set = None
if "encode_splits" in kwargs and taxon_set is None:
raise Exception('When encoding splits on trees, a pre-populated TaxonSet instance ' \
+ "must be provided using the 'taxon_set' keyword to avoid taxon/split bitmask values "\
+ "changing as new Taxon objects are added to the set.")
preserve_underscores = kwargs.get('preserve_underscores', False)
hyphens_as_tokens = kwargs.get('hyphens_as_tokens', nexustokenizer.DEFAULT_HYPHENS_AS_TOKENS)
extract_comment_metadata = kwargs.get("extract_comment_metadata", False)
newick_stream = nexustokenizer.NexusTokenizer(stream,
preserve_underscores=preserve_underscores,
hyphens_as_tokens=hyphens_as_tokens,
extract_comment_metadata=extract_comment_metadata,
case_sensitive_taxon_labels=kwargs.get('case_sensitive_taxon_labels', False))
while not newick_stream.eof:
t = nexustokenizer.tree_from_token_stream(newick_stream, taxon_set=taxon_set, **kwargs)
if t is not None:
yield t
else:
raise StopIteration()
def tree_source_iter(stream, **kwargs):
"""
Iterates over a NEWICK-formatted source of trees given by file-like object
`stream`
Note that if `encode_splits` is True, then a `taxon_set` has to be given.
This is because adding Taxon objects to a taxon set may invalidate split
bitmasks. Because NEWICK tree taxa are added to a TaxonSet as they are found
on a tree, there is a strong possibility that all split bitmasks get
invalidated in the middle of parsing a tree. To avoid this, and, more
importantly to avoid errors downstream in client code due to this, we
force specification of a `taxon_set` if `encode_splits` is requested.
The following optional keyword arguments are also recognized:
`taxon_set`
TaxonSet object to use when reading data.
`as_rooted=True` (or `as_unrooted=False`)
Unconditionally interprets all trees as rooted.
`as_unrooted=True` (or `as_rooted=False`)
Unconditionally interprets all trees as unrooted.
`default_as_rooted=True` (or `default_as_unrooted=False`)
Interprets all trees as rooted if rooting not given by `[&R]`
or `[&U]` comments.
`default_as_unrooted=True` (or `default_as_rooted=False`)
Interprets all trees as rooted if rooting not given by `[&R]`
or `[&U]` comments.
`edge_len_type`
Specifies the type of the edge lengths (int or float).
`extract_comment_metadata`
If True, any comments that begin with '&' or '&&' associated with
items will be processed and stored as part of the annotation set of
the object (`annotations`) If False, this will be skipped. Defaults
to False.
`store_tree_weights`
If True, process the tree weight ("[&W 1/2]") comment
associated with each tree, if any.
`encode_splits`
Specifies whether or not split bitmasks will be calculated and
attached to the edges.
`finish_node_func`
Is a function that will be applied to each node after it has
been constructed.
`case_sensitive_taxon_labels`
If True, then taxon labels are case sensitive (different cases
= different taxa); defaults to False.
`allow_duplicate_taxon_labels`
if True, allow duplicate labels on trees
`preserve_underscores`
If True, unquoted underscores in labels will *not* converted to
spaces. Defaults to False: all underscores not protected by
quotes will be converted to spaces.
`suppress_internal_node_taxa`
If False, internal node labels will be instantantiatd into Taxon
objects. Defaults to True: internal node labels will *not* be
treated as taxa.
`allow_duplicate_taxon_labels`
If True, then multiple identical taxon labels will be allowed.
Defaults to False: treat multiple identical taxon labels as an
error.
`hyphens_as_tokens`
If True, hyphens will be treated as special punctuation
characters. Defaults to False, hyphens not treated as special
punctuation characters.
"""
if "taxon_set" in kwargs:
taxon_set = kwargs["taxon_set"]
del (kwargs["taxon_set"])
else:
taxon_set = None
if "encode_splits" in kwargs and taxon_set is None:
raise Exception('When encoding splits on trees, a pre-populated TaxonSet instance ' \
+ "must be provided using the 'taxon_set' keyword to avoid taxon/split bitmask values "\
+ "changing as new Taxon objects are added to the set.")
preserve_underscores = kwargs.get('preserve_underscores', False)
hyphens_as_tokens = kwargs.get('hyphens_as_tokens',
nexustokenizer.DEFAULT_HYPHENS_AS_TOKENS)
extract_comment_metadata = kwargs.get("extract_comment_metadata", False)
newick_stream = nexustokenizer.NexusTokenizer(
stream,
preserve_underscores=preserve_underscores,
hyphens_as_tokens=hyphens_as_tokens,
extract_comment_metadata=extract_comment_metadata,
case_sensitive_taxon_labels=kwargs.get('case_sensitive_taxon_labels',
False))
while not newick_stream.eof:
t = nexustokenizer.tree_from_token_stream(newick_stream,
taxon_set=taxon_set,
**kwargs)
if t is not None:
yield t
else:
raise StopIteration()
def tree_source_iter(stream, **kwargs):
"""
Iterates over a NEWICK-formatted source of trees given by file-like object
`stream`
Note that if `encode_splits` is True, then a `taxon_set` has to be given.
This is because adding Taxon objects to a taxon set may invalidate split
bitmasks. Because NEWICK tree taxa are added to a TaxonSet as they are found
on a tree, there is a strong possibility that all split bitmasks get
invalidated in the middle of parsing a tree. To avoid this, and, more
importantly to avoid errors downstream in client code due to this, we
force specification of a `taxon_set` if `encode_splits` is requested.
The following optional keyword arguments are also recognized:
- `taxon_set`: TaxonSet object to use when reading data
- `as_rooted=True` (or `as_unrooted=False`): interprets trees as rooted
- `as_unrooted=True` (or `as_rooted=False`): interprets trees as unrooted
- `default_as_rooted=True` (or `default_as_unrooted=False`): interprets
all trees as rooted if rooting not given by `[&R]` or `[&U]` comments
- `default_as_unrooted=True` (or `default_as_rooted=False`): interprets
all trees as rooted if rooting not given by `[&R]` or `[&U]` comments
- `edge_len_type`: specifies the type of the edge lengths (int or float)
- `encode_splits`: specifies whether or not split bitmasks will be
calculated and attached to the edges.
- `extract_comment_metadata`: if True, any 'hot comments' (i.e.,
comments that begin with '&') or NHX comments associated with
items will be processed and stored as a dictionary attribute of the
object: "comment_metadata".
- `store_tree_weights`: if True, process the tree weight ("[&W 1/2]")
comment associated with each tree, if any.
- `finish_node_func`: is a function that will be applied to each node
after it has been constructed.
- `case_insensitive_taxon_labels`: If False, then taxon labels are
case sensitive (different cases = different taxa); defaults to True
"""
if "taxon_set" in kwargs:
taxon_set = kwargs["taxon_set"]
del (kwargs["taxon_set"])
else:
taxon_set = None
if "encode_splits" in kwargs and taxon_set is None:
raise Exception(
"When encoding splits on trees, a pre-populated TaxonSet instance "
+ "must be provided using the 'taxon_set' keyword to avoid taxon/split bitmask values "
+ "changing as new Taxon objects are added to the set."
)
preserve_underscores = kwargs.get("preserve_underscores", False)
hyphens_as_tokens = kwargs.get("hyphens_as_tokens", nexustokenizer.DEFAULT_HYPHENS_AS_TOKENS)
extract_comment_metadata = kwargs.get("extract_comment_metadata", False)
newick_stream = nexustokenizer.NexusTokenizer(
stream,
preserve_underscores=preserve_underscores,
hyphens_as_tokens=hyphens_as_tokens,
extract_comment_metadata=extract_comment_metadata,
case_insensitive_taxon_labels=kwargs.get("case_insensitive_taxon_labels", True),
)
while not newick_stream.eof:
t = nexustokenizer.tree_from_token_stream(newick_stream, taxon_set=taxon_set, **kwargs)
if t is not None:
yield t
else:
raise StopIteration()
