def test_append():
for ext in ["", ".gz"]: # BZ2 does NOT support append
text = "AB"
reference = text + text
filename = 'truncated.fastq' + ext
mode = 'a'
if ext != "":
mode = 'ab'
text = text.encode()
reference = text + text
text = get_compressor(filename).compress(
text) # On Py3, need to send BYTES, not unicode
print("Trying ext=%s" % ext)
with temporary_path(filename) as path:
try:
os.unlink(path)
except OSError:
pass
with open_output(path, mode) as f:
f.write(text)
print(path)
with open_output(path, mode) as f:
f.write(text)
with xopen(path, 'r') as f:
try:
reference = reference.decode("utf-8")
except AttributeError:
pass
for appended in f:
assert appended == reference
def test_append():
for ext in ["", ".gz"]: # BZ2 does NOT support append
text = "AB"
reference = text + text
filename = 'truncated.fastq' + ext
mode = 'a'
if ext != "":
mode = 'ab'
text = text.encode()
reference = text + text
text = get_compressor(filename).compress(text) # On Py3, need to send BYTES, not unicode
print("Trying ext=%s" % ext)
with temporary_path(filename) as path:
try:
os.unlink(path)
except OSError:
pass
with open_output(path, mode) as f:
f.write(text)
print(path)
with open_output(path, mode) as f:
f.write(text)
with xopen(path, 'r') as f:
try:
reference = reference.decode("utf-8")
except AttributeError:
pass
for appended in f:
assert appended == reference
def generate_fasta(outfile, summary, union=False, perinput=False):
names = summary['input']['input_names'] or repeat(None)
n_reads = summary['record_counts'][0]
fasta_format = FastaFormat()
if union:
union_records = []
if perinput:
if outfile.endswith('.fasta'):
name_prefix = outfile[:-6]
elif outfile.endswith('.fa'):
name_prefix = outfile[:-3]
else:
name_prefix = outfile
def format_match(idx, match, records):
name2 = [
"kmer_freq={}".format(match['kmer_freq']),
"kmer_freq_type={}".format(match["kmer_freq_type"])
]
if match['abundance']:
name2.append("abundance={}".format(match['abundance']))
name2.append("abundance_frac={}".format(match['abundance'] / n_reads))
if match['contaminant_to_known_match_frac']:
name2.append("contaminant_to_known_match_frac={}".format(
match["contaminant_to_known_match_frac"]))
if match['is_known']:
name = match['known_names'][0]
name3 = []
if len(match['known_names']) > 1:
name3 = ["other_names={}".format('|'.join(match['known_names'][1:]))]
if len(match['known_seqs']) > 1:
for seq in match['known_seqs']:
records.append(fasta_format.format_entry(
"{}.{} {}".format(name, idx, ";".join(name2 + name3)),
seq))
else:
records.append(fasta_format.format_entry(
"{} {}".format(name, ";".join(name2 + name3)),
match['known_seqs'][0]))
else:
records.append(fasta_format.format_entry(
"{} {}".format(idx, ";".join(name2)),
match['longest_kmer']))
for i, (name, matches) in enumerate(zip(names, summary['detect']['matches'])):
records = []
for idx, match in enumerate(matches, 1):
format_match(idx, match, records)
if union:
union_records.extend(records)
if perinput:
with open_output("{}.{}.fasta".format(name_prefix, i), 'wt') as out:
out.write("".join(records))
if union:
with open_output(outfile, 'wt') as union_out:
union_out.write("".join(union_records))
def test_write_sequence_object(self):
fmt = FastaFormat()
with open_output(self.path, "w") as fw:
fw.write(fmt.format(Sequence("name", "CCATA")))
fw.write(fmt.format(Sequence("name2", "HELLO")))
with open(self.path) as t:
assert t.read() == '>name\nCCATA\n>name2\nHELLO\n'
def test_autodetect_fastq_format(self):
path = os.path.join(self._tmpdir, 'tmp.fastq')
fmt = get_format(path)
with open_output(path, "w") as f:
for seq in simple_fastq:
f.write(fmt.format(seq))
assert list(openseq(path)) == simple_fastq
def test_twoheaders(self):
fmt = FastqFormat()
with open_output(self.path, "w") as fw:
fw.write(fmt.format(Sequence("name", "CCATA", "!#!#!", name2="name")))
fw.write(fmt.format(Sequence("name2", "HELLO", "&&&!&", name2="name2")))
with open(self.path) as t:
assert t.read() == '@name\nCCATA\n+name\n!#!#!\n@name2\nHELLO\n+name2\n&&&!&\n'
def test_write_sequence_object(self):
fmt = FastaFormat()
with open_output(self.path, "w") as fw:
fw.write(fmt.format(Sequence("name", "CCATA")))
fw.write(fmt.format(Sequence("name2", "HELLO")))
with open(self.path) as t:
assert t.read() == '>name\nCCATA\n>name2\nHELLO\n'
def test(self):
fmt = FastqFormat()
with open_output(self.path, "w") as fw:
fw.write(fmt.format_entry("name", "CCATA", "!#!#!"))
fw.write(fmt.format_entry("name2", "HELLO", "&&&!&&"))
with open(self.path) as t:
assert t.read() == '@name\nCCATA\n+\n!#!#!\n@name2\nHELLO\n+\n&&&!&&\n'
def test_autodetect_fastq_format(self):
path = os.path.join(self._tmpdir, 'tmp.fastq')
fmt = get_format(path)
with open_output(path, "w") as f:
for seq in simple_fastq:
f.write(fmt.format(seq))
assert list(openseq(path)) == simple_fastq
def get_writer(self, file_desc, compressed=False):
"""Create the writer for a file descriptor if it does not already
exist.
Args:
file_desc: File descriptor. If `compressed==True`, this is a tuple
(path, mode), otherwise it's only a path.
compressed: Whether data has already been compressed.
Returns:
The writer.
"""
if compressed:
path, mode = file_desc
else:
path = file_desc
if path not in self.writers:
if self.suffix:
real_path = add_suffix_to_path(path, self.suffix)
else:
real_path = path
# TODO: test whether O_NONBLOCK allows non-blocking write to NFS
if compressed:
self.writers[path] = open_output(real_path, mode)
else:
self.writers[path] = xopen(real_path, "w")
return self.writers[path]
def get_writer(self, file_desc, compressed=False):
"""Create the writer for a file descriptor if it does not already
exist.
Args:
file_desc: File descriptor. If `compressed==True`, this is a tuple
(path, mode), otherwise it's only a path.
compressed: Whether data has already been compressed.
Returns:
The writer.
"""
if compressed:
path, mode = file_desc
else:
path = file_desc
if path not in self.writers:
if self.suffix:
real_path = add_suffix_to_path(path, self.suffix)
else:
real_path = path
# TODO: test whether O_NONBLOCK allows non-blocking write to NFS
if compressed:
self.writers[path] = open_output(real_path, mode)
else:
self.writers[path] = xopen(real_path, "w")
return self.writers[path]
def test_write_qualities_to_fasta(self):
path = os.path.join(self._tmpdir, 'tmp.fasta')
fmt = get_format(path, qualities=True)
assert isinstance(fmt, FastaFormat)
with open_output(path, "w") as f:
for seq in simple_fastq:
f.write(fmt.format(seq))
assert list(openseq(path)) == simple_fasta
def generate_text_report(self, fmt, summary, outfile, **kwargs):
if fmt == 'txt':
with open_output(outfile, context_wrapper=True) as out:
generate_reports(out, summary, **kwargs)
elif fmt == 'fasta':
generate_fasta(outfile, summary, **kwargs)
else:
super().generate_from_template(fmt, summary, outfile, **kwargs)
def test_write_qualities_to_fasta(self):
path = os.path.join(self._tmpdir, 'tmp.fasta')
fmt = get_format(path, qualities=True)
assert isinstance(fmt, FastaFormat)
with open_output(path, "w") as f:
for seq in simple_fastq:
f.write(fmt.format(seq))
assert list(openseq(path)) == simple_fasta
def test(self):
fmt = FastqFormat()
with open_output(self.path, "w") as fw:
fw.write(fmt.format_entry("name", "CCATA", "!#!#!"))
fw.write(fmt.format_entry("name2", "HELLO", "&&&!&&"))
with open(self.path) as t:
assert t.read(
) == '@name\nCCATA\n+\n!#!#!\n@name2\nHELLO\n+\n&&&!&&\n'
def close(self):
"""Close all outputs.
"""
for path in self.force_create:
if path not in self.writers and path != STDOUT:
with open_output(path, "w"):
pass
for writer in self.writers.values():
if writer not in (sys.stdout, sys.stderr):
writer.close()
def test_linelength(self):
fmt = FastaFormat(line_length=3)
with open_output(self.path, "w") as fw:
fw.write(fmt.format_entry("r1", "ACG"))
fw.write(fmt.format_entry("r2", "CCAT"))
fw.write(fmt.format_entry("r3", "TACCAG"))
with open(self.path) as t:
x=t.read()
print(x)
assert x == '>r1\nACG\n>r2\nCCA\nT\n>r3\nTAC\nCAG\n'
def test_linelength(self):
fmt = FastaFormat(line_length=3)
with open_output(self.path, "w") as fw:
fw.write(fmt.format_entry("r1", "ACG"))
fw.write(fmt.format_entry("r2", "CCAT"))
fw.write(fmt.format_entry("r3", "TACCAG"))
with open(self.path) as t:
x = t.read()
print(x)
assert x == '>r1\nACG\n>r2\nCCA\nT\n>r3\nTAC\nCAG\n'
def close(self):
"""Close all outputs.
"""
for path in self.force_create:
if path not in self.writers and path != STDOUT:
with open_output(path, "w"):
pass
for writer in self.writers.values():
if writer not in (sys.stdout, sys.stderr):
writer.close()
def test_twoheaders(self):
fmt = FastqFormat()
with open_output(self.path, "w") as fw:
fw.write(
fmt.format(Sequence("name", "CCATA", "!#!#!", name2="name")))
fw.write(
fmt.format(Sequence("name2", "HELLO", "&&&!&", name2="name2")))
with open(self.path) as t:
assert t.read(
) == '@name\nCCATA\n+name\n!#!#!\n@name2\nHELLO\n+name2\n&&&!&\n'
def serialize(self, obj, fmt, mode, outfile, **kwargs):
"""Serialize a summary dict to a file.
Args:
obj: The summary dict.
fmt: The serialization format (e.g. json, yaml).
mode: The file mode (b=binary, t=text).
outfile: The output file.
kwargs: Additional arguments to pass to the `dump` method.
"""
mod = importlib.import_module(fmt)
with open_output(outfile, 'w' + mode) as stream:
mod.dump(obj, stream, **kwargs)
def serialize(self, obj, fmt, mode, outfile, **kwargs):
"""Serialize a summary dict to a file.
Args:
obj: The summary dict.
fmt: The serialization format (e.g. json, yaml).
mode: The file mode (b=binary, t=text).
outfile: The output file.
kwargs: Additional arguments to pass to the `dump` method.
"""
mod = importlib.import_module(fmt)
with open_output(outfile, 'w' + mode) as stream:
mod.dump(obj, stream, **kwargs)
