def test_paragraphs2(self):
"""Test paragraphs are correctly split with no closing element."""
r = HtmlReader()
d = r.parse('<p>First para<p>Second Para')
self.assertEqual(len(d.elements), 2)
self.assertEqual(d.elements[0].text, 'First para')
self.assertEqual(d.elements[1].text, 'Second Para')
for el in d.elements:
self.assertIsInstance(el, Paragraph)
def test_linebreak2(self):
"""Test br splits paragraph."""
r = HtmlReader()
d = r.parse('<span>First line</span><br/><span>Second line</span>')
self.assertEqual(len(d.elements), 2)
self.assertEqual(d.elements[0].text, 'First line')
self.assertEqual(d.elements[1].text, 'Second line')
for el in d.elements:
self.assertIsInstance(el, Paragraph)
def read_html_paper(paper_path):
"""Opens a HTML paper and stores it as a chemdataextractor Document"""
f = open(paper_path, 'rb')
doc = Document.from_file(f, readers=[HtmlReader()])
return doc
def extract_sentences(paper_path, para_yes):
"""extracts sentences from a paper into two lists, given that para_yes contains
a list of document element numbers corresponding to paragraphs manually identified
as those containing synthesis information"""
f = open(paper_path, 'rb')
doc = Document.from_file(f, readers=[HtmlReader()])
sen_yes_arr = list()
sen_no_arr = list()
elem_all = np.arange(0, len(doc))
para_no = np.delete(elem_all, para_yes)
for i in para_no:
if type(doc.elements[i]) == chemdataextractor.doc.text.Paragraph:
for sentence in doc.elements[i]:
sen_no_arr.append(sentence)
for i in para_yes:
if type(doc.elements[i]) == chemdataextractor.doc.text.Paragraph:
for sentence in doc.elements[i]:
sen_yes_arr.append(sentence)
return sen_yes_arr, sen_no_arr