def parseText(text):
"""
Wraps the parser with logic to create the appropriate data structures
"""
#splits text into paragraphs to parse, which is easier than making the parser
#more complex
separated = LINESEP.split(text)
rootSection = None
currentSection = None
while bool(separated):
current = separated.pop(0)
if current == "":
continue
try:
results = ROOT.parseString(current)
except Exception as e:
logging.info('Parser Issue with: {}'.format(current))
raise e
#parsed a header, create a section
if not bool(results):
results = "Nothing, just add the paragraph"
if isinstance(results[0], HEADER) and rootSection is None:
#a new, root section
currentSection = Section(results[0].title, results[0].level)
rootSection = currentSection
for res in results[1:]:
if isinstance(res, TAG):
currentSection.add_tag(res.text)
elif isinstance(results[0], HEADER):
#a subsection
if results[0].level > currentSection._level:
currentSection = currentSection.add_subsection(results[0].title,
results[0].level)
elif results[0].level <= currentSection._level:
#a subsection of an ancestor
while results[0].level <= currentSection._level:
currentSection = currentSection.get_parent()
currentSection = currentSection.add_subsection(results[0].title,
results[0].level)
for res in results[1:]:
if isinstance(res, TAG):
currentSection.add_tag(res.text)
else:
#Paragraphs of a section
currentParagraph = currentSection.add_paragraph(current)
for res in results:
if isinstance(res, TAG):
currentParagraph['tags'].add(res.text)
elif isinstance(res, CITATION):
currentParagraph['citations'].add(res.text)
return rootSection
def test_subsection(self):
aSection = Section('aTitle', 1)
newSection = aSection.add_subsection('other', 2)
self.assertEqual(len(aSection._ordered_subsections), 1)
retrieved = aSection.section('other')
self.assertIsNotNone(retrieved)
self.assertIsInstance(retrieved, Section)
def test_deep_tags_sections(self):
aSection = Section('aTitle', 1)
aSubSection = aSection.add_subsection('subsection', 2)
aSubSection.add_tag('blah')
self.assertTrue(aSection.has_tag('blah'))
