def _parse_quality_scores(fh, seq_len, variant, phred_offset, prev):
phred_scores = []
qual_len = 0
for chunk in _line_generator(fh, skip_blanks=False):
if chunk:
if chunk.startswith('@') and qual_len == seq_len:
return np.hstack(phred_scores), chunk
else:
if not prev:
_blank_error("after '+' or within quality scores")
qual_len += len(chunk)
if qual_len > seq_len:
raise FASTQFormatError(
"Found more quality score characters than sequence "
"characters. Extra quality score characters: %r" %
chunk[-(qual_len - seq_len):])
phred_scores.append(
_decode_qual_to_phred(chunk,
variant=variant,
phred_offset=phred_offset))
prev = chunk
if qual_len != seq_len:
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
return np.hstack(phred_scores), None
def _parse_sequence_data(fh, prev):
seq_chunks = []
for chunk in _line_generator(fh, skip_blanks=False):
if chunk.startswith('+'):
if not prev:
_blank_error("before '+'")
if not seq_chunks:
raise FASTQFormatError(
"Found FASTQ record without sequence data.")
return ''.join(seq_chunks), chunk
elif chunk.startswith('@'):
raise FASTQFormatError(
"Found FASTQ record that is missing a quality (+) header line "
"after sequence data.")
else:
if not prev:
_blank_error("after header or within sequence")
if _whitespace_regex.search(chunk):
raise FASTQFormatError(
"Found whitespace in sequence data: %r" % str(chunk))
seq_chunks.append(chunk)
prev = chunk
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
def _fastq_to_generator(fh,
variant=None,
phred_offset=None,
constructor=Sequence,
**kwargs):
# Skip any blank or whitespace-only lines at beginning of file
seq_header = next(_line_generator(fh, skip_blanks=True))
if not seq_header.startswith('@'):
raise FASTQFormatError(
"Expected sequence (@) header line at start of file: %r" %
str(seq_header))
while seq_header is not None:
id_, desc = _parse_fasta_like_header(seq_header)
seq, qual_header = _parse_sequence_data(fh, seq_header)
if qual_header != '+' and qual_header[1:] != seq_header[1:]:
raise FASTQFormatError(
"Sequence (@) and quality (+) header lines do not match: "
"%r != %r" % (str(seq_header[1:]), str(qual_header[1:])))
phred_scores, seq_header = _parse_quality_scores(
fh, len(seq), variant, phred_offset, qual_header)
yield constructor(seq,
metadata={
'id': id_,
'description': desc
},
positional_metadata={'quality': phred_scores},
**kwargs)
def _parse_sequence_data(fh):
seq_chunks = []
for chunk in _line_generator(fh):
if chunk.startswith('+'):
if not seq_chunks:
raise FASTQFormatError(
"Found FASTQ record without sequence data.")
return ''.join(seq_chunks), chunk
elif chunk.startswith('@'):
raise FASTQFormatError(
"Found FASTQ record that is missing a quality (+) header line "
"after sequence data.")
else:
if _whitespace_regex.search(chunk):
raise FASTQFormatError(
"Found whitespace in sequence data: %r" % chunk)
seq_chunks.append(chunk)
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
def _fastq_to_generator(fh,
variant=None,
phred_offset=None,
constructor=BiologicalSequence):
seq_header = next(_line_generator(fh))
if not seq_header.startswith('@'):
raise FASTQFormatError(
"Expected sequence (@) header line at start of file: %r" %
seq_header)
while seq_header is not None:
id_, desc = _parse_fasta_like_header(seq_header)
seq, qual_header = _parse_sequence_data(fh)
if qual_header != '+' and qual_header[1:] != seq_header[1:]:
raise FASTQFormatError(
"Sequence (@) and quality (+) header lines do not match: "
"%r != %r" % (seq_header[1:], qual_header[1:]))
phred_scores, seq_header = _parse_quality_scores(
fh, len(seq), variant, phred_offset)
yield constructor(seq, id=id_, description=desc, quality=phred_scores)
def _parse_quality_scores(fh, seq_len, variant, phred_offset):
phred_scores = []
qual_len = 0
for chunk in _line_generator(fh):
if chunk.startswith('@') and qual_len == seq_len:
return phred_scores, chunk
else:
qual_len += len(chunk)
if qual_len > seq_len:
raise FASTQFormatError(
"Found more quality score characters than sequence "
"characters. Extra quality score characters: %r" %
chunk[-(qual_len - seq_len):])
phred_scores.extend(
_decode_qual_to_phred(chunk,
variant=variant,
phred_offset=phred_offset))
if qual_len != seq_len:
raise FASTQFormatError(
"Found incomplete/truncated FASTQ record at end of file.")
return phred_scores, None
def _line_generator(fh):
for line in fh:
line = line.rstrip('\n')
if not line:
raise FASTQFormatError("Found blank line in FASTQ-formatted file.")
yield line
def _blank_error(unique_text):
error_string = ("Found blank or whitespace-only line {} in "
"FASTQ file").format(unique_text)
raise FASTQFormatError(error_string)