def replace_markerless(self, stack, node, depth):
"""Assign a unique index to all of the MARKERLESS paragraphs"""
if node.label[-1] == mtypes.MARKERLESS:
keyterm = KeyTerms.get_keyterm(node, ignore_definitions=False)
if keyterm:
p_num = keyterm_to_int(keyterm)
else:
# len(n.label[-1]) < 6 filters out keyterm nodes
p_num = sum(n.is_markerless() and len(n.label[-1]) < 6
for n in stack.peek_level(depth)) + 1
node.label[-1] = 'p{}'.format(p_num)
par_list[1] = match.section
elif match.appendix:
par_list[1] = "Appendix:" + match.appendix
# Set paragraph depths
for p in match_list[2:]:
par_list[match_list.index(p)] = p
par = tokens.Paragraph(par_list)
return [par]
_keyterm_label_part = (
Suppress(Marker("keyterm")) +
QuotedString(quoteChar='(', endQuoteChar=')')
).setParseAction(lambda m: "p{}".format(keyterm_to_int(m[0])))
_simple_label_part = Word(string.ascii_lowercase + string.ascii_uppercase +
string.digits)
_label_part = _keyterm_label_part | _simple_label_part
override_label = (
Suppress("[") +
Marker("label") + Suppress(":") +
atomic.part +
Suppress("-") +
(atomic.section | atomic.appendix) +
ZeroOrMore(Suppress("-") + _label_part) +
Suppress("]")
).setParseAction(tokenize_override_ps)
# Looks like: [subject-group(Some text Goes Here)]
def test_keyterm_to_int(self):
"""keyterm_to_int should standardize the keyterm"""
self.assertEqual(keyterm_to_int('Abc 123 More.'),
keyterm_to_int(' abc123 mOrE'))
self.assertTrue(keyterm_to_int('a term') > 10000)