def _parse_table(self, config_file):
self.alignments = OrderedDict()
cfg_stream, close = fileio.process_file_arg(config_file)
try:
table_started = False
table_finished = False
row_num = 0
for i, l in enumerate(cfg_stream):
line = l.strip()
if self._end_pattern.match(line):
if not table_started:
raise errors.SampleTableError(
'hit end of sample table before beginning')
if len(self.alignments) < 1:
raise errors.SampleTableError(
'no rows found in sample table')
table_finished = True
break
if self._begin_pattern.match(line):
table_started = True
continue
if not table_started:
continue
if (line == '') or (line.startswith('#')):
continue
row_num += 1
try:
al = AlignmentConfig(line)
except errors.SampleTableRowError as e:
_LOG.error('sample table row {0} is invalid'.format(
row_num))
raise e
if not al.taxon_name in self.alignments:
self.alignments[al.taxon_name] = OrderedDict()
self.alignments[al.taxon_name][al.locus_name] = al
self._ordering.append((al.taxon_name, al.locus_name))
continue
if al.locus_name in self.alignments[al.taxon_name]:
raise errors.SampleTableError('locus {0} found twice '
'for taxon {1} at row {2} of sample '
'table'.format(al.locus_name, al.taxon_name,
row_num))
self.alignments[al.taxon_name][al.locus_name] = al
self._ordering.append((al.taxon_name, al.locus_name))
if not table_started:
raise errors.SampleTableError('no sample table found')
if not table_finished:
raise errors.SampleTableError('no end of table found')
finally:
if close:
cfg_stream.close()
class SampleTable(object):
_begin_pattern = re.compile(r'^begin\s*sample_tbl$', re.IGNORECASE)
_end_pattern = re.compile(r'^end\s*sample_tbl$', re.IGNORECASE)
def __init__(self, config_file):
self.alignments = None
self._ordering = []
self._parse_table(config_file)
def _parse_table(self, config_file):
self.alignments = OrderedDict()
cfg_stream, close = fileio.process_file_arg(config_file)
try:
table_started = False
table_finished = False
row_num = 0
for i, l in enumerate(cfg_stream):
line = l.strip()
if self._end_pattern.match(line):
if not table_started:
raise errors.SampleTableError(
'hit end of sample table before beginning')
if len(self.alignments) < 1:
raise errors.SampleTableError(
'no rows found in sample table')
table_finished = True
break
if self._begin_pattern.match(line):
table_started = True
continue
if not table_started:
continue
if (line == '') or (line.startswith('#')):
continue
row_num += 1
try:
al = AlignmentConfig(line)
except errors.SampleTableRowError as e:
_LOG.error('sample table row {0} is invalid'.format(
row_num))
raise e
if not al.taxon_name in self.alignments:
self.alignments[al.taxon_name] = OrderedDict()
self.alignments[al.taxon_name][al.locus_name] = al
self._ordering.append((al.taxon_name, al.locus_name))
continue
if al.locus_name in self.alignments[al.taxon_name]:
raise errors.SampleTableError('locus {0} found twice '
'for taxon {1} at row {2} of sample '
'table'.format(al.locus_name, al.taxon_name,
row_num))
self.alignments[al.taxon_name][al.locus_name] = al
self._ordering.append((al.taxon_name, al.locus_name))
if not table_started:
raise errors.SampleTableError('no sample table found')
if not table_finished:
raise errors.SampleTableError('no end of table found')
finally:
if close:
cfg_stream.close()
def _get_taxa(self):
return self.alignments.keys()
taxa = property(_get_taxa)
def _get_loci(self):
l = []
for t, d in self.alignments.iteritems():
for locus in d.iterkeys():
if not locus in l:
l.append(locus)
return l
loci = property(_get_loci)
def _get_number_of_taxa(self):
return len(self.taxa)
npairs = property(_get_number_of_taxa)
def get_sample_table_string(self):
return '\n'.join(('BEGIN SAMPLE_TBL',
'\n'.join((str(self.alignments[t][l]) for t, l in self._ordering)),
'END SAMPLE_TBL'))
def __str__(self):
return self.get_sample_table_string()
def equals(self, other):
if not isinstance(other, SampleTable):
return False
if len(self.alignments) != len(other.alignments):
return False
for i1, i2 in zip(self.alignments.items(), other.alignments.items()):
if i1[0] != i2[0]:
return False
if len(i1[1]) != len(i2[1]):
return False
for t1, t2 in zip(i1[1].items(), i2[1].items()):
if t1[0] != t2[0]:
return False
if not t1[1].equals(t2[1]):
return False
if self._ordering != self._ordering:
return False
return True