def test_parsers_ignore(self):
"""LabeledRecordFinder should skip lines to ignore."""
def never(line):
return False
def ignore_labels(line):
return (not line) or line.isspace() or line.startswith("#")
def is_start(line):
return line.startswith(">")
lines = [">abc", "\n", "1", ">def", "#ignore", "2"]
self.assertEqual(
list(LabeledRecordFinder(is_start)(lines)),
[[">abc", "1"], [">def", "#ignore", "2"]],
)
self.assertEqual(
list(LabeledRecordFinder(is_start, ignore=never)(lines)),
[[">abc", "", "1"], [">def", "#ignore", "2"]],
)
self.assertEqual(
list(LabeledRecordFinder(is_start, ignore=ignore_labels)(lines)),
[[">abc", "1"], [">def", "2"]],
)
# separate by semicolons
# get rid of leading/trailing spaces
taxa = list(map(strip, taxonomy.split(";")))
# delete trailing period if present
last = taxa[-1]
if last.endswith("."):
taxa[-1] = last[:-1]
return species, taxa
def is_feature_component_start(line):
"""Checks if a line starts with '/', ignoring whitespace."""
return line.lstrip().startswith("/")
feature_component_iterator = LabeledRecordFinder(is_feature_component_start)
_join_with_empty = dict.fromkeys(["translation"])
_leave_as_lines = {}
def parse_feature(lines):
"""Parses a feature. Doesn't handle subfeatures.
Returns dict containing:
'type': source, gene, CDS, etc.
'location': unparsed location string
...then, key-value pairs for each annotation,
e.g. '/gene="MNBH"' -> {'gene':['MNBH']} (i.e. quotes stripped)
All relations are assumed 'to many', and order will be preserved.
"""
def is_gde_label(x):
"""Checks if x looks like a GDE label line."""
return x and x[0] in "%#"
def is_blank_or_comment(x):
"""Checks if x is blank or a FASTA comment line."""
return (not x) or x.startswith("#") or x.isspace()
def is_blank(x):
"""Checks if x is blank."""
return (not x) or x.isspace()
FastaFinder = LabeledRecordFinder(is_fasta_label, ignore=is_blank_or_comment)
def MinimalFastaParser(infile,
strict=True,
label_to_name=str,
finder=FastaFinder,
label_characters=">"):
"""Yields successive sequences from infile as (label, seq) tuples.
If strict is True (default), raises RecordError when label or seq missing.
"""
try:
infile = open_(infile)
close_at_end = True
except (TypeError, AttributeError):
def setUp(self):
"""Define a standard LabeledRecordFinder"""
self.FastaLike = LabeledRecordFinder(lambda x: x.startswith(">"))
WARNING: Only maps the data type if the key is in label_constructors above.
"""
if not line.startswith("#"):
raise ValueError("Labels must start with a # symbol.")
if line.find(":") == -1:
raise ValueError("Labels must contain a : symbol.")
key, value = list(map(strip, line[1:].split(":", 1)))
key = key.upper()
if key in label_constructors:
value = label_constructors[key](value)
return key, value
BlatFinder = LabeledRecordFinder(query_finder, constructor=strip, ignore=is_blat_junk)
BlastFinder = LabeledRecordFinder(query_finder, constructor=strip, ignore=is_blast_junk)
PsiBlastFinder = LabeledRecordFinder(
iter_finder, constructor=strip, ignore=is_blast_junk
)
PsiBlastQueryFinder = LabeledRecordFinder(
iteration_set_finder, constructor=strip, ignore=is_blast_junk
)
def GenericBlastParser9(lines, finder, make_col_headers=False):
"""Yields successive records from lines (props, data list)
__version__ = "2019.10.24a"
__maintainer__ = "Rob Knight"
__email__ = "*****@*****.**"
__status__ = "Development"
maketrans = str.maketrans
strip = str.strip
rstrip = str.rstrip
def ll_start(line):
"""Returns True if line looks like the start of a LocusLink record."""
return line.startswith(">>")
LLFinder = LabeledRecordFinder(ll_start)
pipes = DelimitedSplitter("|", None)
first_pipe = DelimitedSplitter("|")
commas = DelimitedSplitter(",", None)
first_colon = DelimitedSplitter(":", 1)
accession_wrapper = FieldWrapper(["Accession", "Gi", "Strain"], pipes)
def _read_accession(line):
"""Reads accession lines: format is Accession | Gi | Strain."""
return MappedRecord(accession_wrapper(line))
rell_wrapper = FieldWrapper(["Description", "Id", "IdType", "Printable"], pipes)