def test_parsers_error(self):
"""DelimitedRecordFinder should raise RecordError if trailing data"""
good = [
" \t abc \n",
"\t def\n",
"// \t\n",
"\t\n",
"\t efg \n",
"\t\t//\n",
]
blank = ["", " ", "\t \t\n\n"]
bad = ["abc"]
result = [["abc", "def", "//"], ["efg", "//"]]
r = DelimitedRecordFinder("//")
self.assertEqual(list(r(good)), result)
self.assertEqual(list(r(good + blank)), result)
try:
list(r(good + bad))
except RecordError:
pass
else:
raise AssertionError("Parser failed to raise error on bad data")
r = DelimitedRecordFinder("//", strict=False)
self.assertEqual(list(r(good + bad)), result + [["abc"]])
def test_parsers(self):
"""DelimitedRecordFinder should split records into lines correctly"""
lines = "abc\ndef\n//\nefg\n//".split()
self.assertEqual(
list(DelimitedRecordFinder("//")(lines)),
[["abc", "def", "//"], ["efg", "//"]],
)
self.assertEqual(
list(DelimitedRecordFinder("//", keep_delimiter=False)(lines)),
[["abc", "def"], ["efg"]],
)
def MinimalGreengenesParser(lines,
LineDelim="=",
RecStart="BEGIN",
RecEnd="END"):
"""Parses raw Greengeens 16S rRNA Gene records
lines : open records file
LineDelim : individual line delimiter, eg foo=bar
RecStart : start identifier for a record
RecEnd : end identifier for a record
"""
line_parser = DefaultDelimitedSplitter(delimiter=LineDelim)
# parse what the ending record looks like so it can match after being split
RecordDelim = line_parser(RecEnd)
# make sure to ignore the starting record
ignore = make_ignore_f(RecStart)
parser = DelimitedRecordFinder(RecordDelim,
constructor=line_parser,
keep_delimiter=False,
ignore=ignore)
for record in parser(lines):
yield GenericRecord(record)
def test_parsers_strip(self):
"""DelimitedRecordFinder should trim each line correctly"""
lines = " \t abc \n \t def\n // \t\n\t\t efg \n//".split("\n")
self.assertEqual(
list(DelimitedRecordFinder("//")(lines)),
[["abc", "def", "//"], ["efg", "//"]],
)
def test_parsers_ignore(self):
"""DelimitedRecordFinder should skip lines to ignore."""
def never(line):
return False
def ignore_labels(line):
return (not line) or line.isspace() or line.startswith("#")
lines = [">abc", "\n", "1", "$$", ">def", "#ignore", "2", "$$"]
self.assertEqual(
list(DelimitedRecordFinder("$$")(lines)),
[[">abc", "1", "$$"], [">def", "#ignore", "2", "$$"]],
)
self.assertEqual(
list(DelimitedRecordFinder("$$", ignore=never)(lines)),
[[">abc", "", "1", "$$"], [">def", "#ignore", "2", "$$"]],
)
self.assertEqual(
list(DelimitedRecordFinder("$$", ignore=ignore_labels)(lines)),
[[">abc", "1", "$$"], [">def", "2", "$$"]],
)
all_chars = maketrans("", "")
dna_lc = "utacgrywsmkbdhvn"
dna_lc_cmp = "aatgcyrwskmvhdbn"
dna_trans = maketrans(dna_lc + dna_lc.upper(), dna_lc_cmp + dna_lc_cmp.upper())
rna_lc = "utacgrywsmkbdhvn"
rna_lc_cmp = "aaugcyrwskmvhdbn"
rna_trans = maketrans(rna_lc + rna_lc.upper(), rna_lc_cmp + rna_lc_cmp.upper())
locus_fields = [
None, "locus", "length", None, "mol_type", "topology", "db", "date"
]
_locus_parser = FieldWrapper(locus_fields)
# need to turn off line stripping, because whitespace is significant
GbFinder = DelimitedRecordFinder("//", constructor=rstrip)
class PartialRecordError(Exception):
pass
def parse_locus(line):
"""Parses a locus line, including conversion of Length to an int.
WARNING: Gives incorrect results on legacy records that omit the topology.
All records spot-checked on 8/30/05 had been updated to include the topology
even when prior versions omitted it.
"""
result = _locus_parser(line)
try:
def test_parsers_empty(self):
"""DelimitedRecordFinder should return empty list on empty lines"""
self.assertEqual(list(DelimitedRecordFinder("//")([" ", "\n"])), [])
self.assertEqual(list(DelimitedRecordFinder("//")([])), [])
def QMEPsiBlast9(lines):
"""Returns successive query, match, e-value from lines of Psi-Blast run.
Assumes tabular output. Uses last iteration from each query.
WARNING: Allows duplicates in result
"""
result = []
for query in PsiBlastQueryFinder(lines):
for iteration in PsiBlastFinder(query):
pass
result.extend(QMEBlast9(iteration))
return result
fastacmd_taxonomy_splitter = DelimitedRecordFinder(delimiter="", ignore=never_ignore)
fasta_field_map = {
"NCBI sequence id": "seq_id",
"NCBI taxonomy id": "tax_id",
"Common name": "common_name",
"Scientific name": "scientific_name",
}
def FastacmdTaxonomyParser(lines):
"""Yields successive records from the results of fastacmd -T.
Format is four lines separated by newline:
NCBI sequence
NCBI taxonomy
Common name
__author__ = "Sandra Smit"
__copyright__ = "Copyright 2007-2020, The Cogent Project"
__credits__ = ["Sandra Smit", "Rob Knight"]
__license__ = "BSD-3"
__version__ = "2020.6.30a"
__maintainer__ = "Sandra Smit"
__email__ = "*****@*****.**"
__status__ = "Development"
strip = str.strip
maketrans = str.maketrans
RdbFinder = DelimitedRecordFinder("//")
_field_names = {
"acc": "rRNA",
"src": "Source",
"str": "Strain",
"ta1": "Taxonomy1",
"ta2": "Taxonomy2",
"ta3": "Taxonomy3",
"ta4": "Taxonomy4",
"chg": "Changes",
"rem": "Remarks",
"aut": "Authors",
"ttl": "Title",
"jou": "Journal",
"dat": "JournalYear",
