def get_cols(self, lines):
"""Initialize Column objects from a multi-line ASCII header
Parameters
----------
lines : `list`
List of table lines
"""
re_name_def = re.compile("^\s*%\s+(?P<colname>\w+)")
self.names = []
for line in lines:
if not line.startswith('%'):
break # End of header lines
else:
match = re_name_def.search(line)
if match:
self.names.append(match.group('colname'))
if not self.names:
raise core.InconsistentTableError(
'No column names found in Omega header')
self.cols = []
for n in self.names:
col = core.Column(name=n)
self.cols.append(col)
def process_lines(self, lines):
"""
Convert the given input into a list of SoupString rows
for further processing.
"""
try:
from bs4 import BeautifulSoup
except ImportError:
raise core.OptionalTableImportError(
'BeautifulSoup must be '
'installed to read HTML tables')
if 'parser' not in self.html:
soup = BeautifulSoup('\n'.join(lines))
else: # use a custom backend parser
soup = BeautifulSoup('\n'.join(lines), self.html['parser'])
tables = soup.find_all('table')
for i, possible_table in enumerate(tables):
if html.identify_table(possible_table, self.html, i + 1):
table = possible_table # Find the correct table
break
else:
if isinstance(self.html['table_id'], int):
err_descr = 'number {0}'.format(self.html['table_id'])
else:
err_descr = "id '{0}'".format(self.html['table_id'])
raise core.InconsistentTableError(
'ERROR: HTML table {0} not found'.format(err_descr))
self.html['attrs'] = table.attrs
# Get all table rows
soup_list = [html.SoupString(x) for x in table.find_all('tr')]
return soup_list
def get_cols(self, lines):
"""Initialize Column objects from a multi-line ASCII header
Parameters
----------
lines : `list`
List of table lines
"""
re_name_def = re.compile(r'^\s*%\s+(?P<colname>\w+)')
self.names = []
for line in lines:
if not line: # ignore empty lines in header (windows)
continue
if not line.startswith('%'): # end of header lines
break
match = re_name_def.search(line)
if match:
self.names.append(match.group('colname'))
if not self.names:
raise core.InconsistentTableError(
'No column names found in Omega header')
self.cols = [] # pylint: disable=attribute-defined-outside-init
for name in self.names:
col = core.Column(name=name)
self.cols.append(col)
def __call__(self, lines):
for line in lines:
if not isinstance(line, html.SoupString):
raise TypeError('HTML lines should be of type SoupString')
soup = line.soup
# If header is duplicated, don't return it as data!
header_elements = soup.find_all('th')
if header_elements:
continue
data_elements = soup.find_all('td')
if data_elements:
# Return multirows as a couple for HTMLWithGroupsData handling
yield [(el.text.strip(), int(el['rowspan']))
if el.has_attr('rowspan') else el.text.strip()
for el in data_elements]
if len(lines) == 0:
raise core.InconsistentTableError('HTML tables must contain data '
'in a <table> tag')
def get_cols(self, lines):
"""Initialize Column objects from a multi-line ASCII header
Parameters
----------
lines : `list`
List of table lines
"""
re_name_def = re.compile(
r'^\s*#\s+' # whitespace and comment marker
r'(?P<colnumber>[0-9]+)\s+-\s+' # number of column
r'(?P<colname>(.*))'
)
self.names = []
include_cuts = False
for line in lines:
if not line: # ignore empty lines in header (windows)
continue
if not line.startswith('# '): # end of header lines
break
if line.startswith('# -/+'):
include_cuts = True
else:
match = re_name_def.search(line)
if match:
self.names.append(match.group('colname').rstrip())
if not self.names:
raise core.InconsistentTableError(
'No column names found in cWB header')
if include_cuts:
self.cols = [ # pylint: disable=attribute-defined-outside-init
core.Column(name='selection cut 1'),
core.Column(name='selection cut 2'),
]
else:
self.cols = [] # pylint: disable=attribute-defined-outside-init
for name in self.names:
col = core.Column(name=name)
self.cols.append(col)
def get_cols(self, lines):
"""Initialize Column objects from a multi-line ASCII header
Parameters
----------
lines : `list`
List of table lines
"""
re_name_def = re.compile(
"^\s*#\s+" # whitespace and comment marker
"(?P<colnumber>[0-9]+)\s+-\s+" # number of column
"(?P<colname>(.*))"
)
self.names = []
for line in lines:
if not line.startswith('# '):
break # End of header lines
elif line.startswith('# -/+'):
include_cuts = True
else:
match = re_name_def.search(line)
if match:
self.names.append(match.group('colname').rstrip())
if not self.names:
raise core.InconsistentTableError(
'No column names found in cWB header')
if include_cuts:
self.cols = [
core.Column(name='selection cut 1'),
core.Column(name='selection cut 2'),
]
else:
self.cols = []
for n in self.names:
col = core.Column(name=n)
self.cols.append(col)