def test_extract_text():
pmc_id = '4322985'
xml_str = pmc_client.get_xml(pmc_id)
text = pmc_client.extract_text(xml_str)
assert text is not None
assert 'RAS VS BRAF ONCOGENES AND TARGETED THERAPIES' in text
assert unicode_strs(text)
def preprocess_nxml_file(self, filename, pmid, extra_annotations):
"""Preprocess an NXML file for use with the ISI reader.
Preprocessing is done by extracting plain text from NXML and then
creating a text file with one sentence per line.
Parameters
----------
filename : str
Filename (more specifically the file path) of an nxml file to
process
pmid : str
The PMID from which it comes, or None if not specified
extra_annotations : dict
Extra annotations to be added to each statement, possibly including
metadata about the source (annotations with the key "interaction"
will be overridden)
"""
with open(filename, 'r') as fh:
txt_content = extract_text(fh.read())
# We need to remove some common LaTEX commands from the converted text
# or the reader will get confused
cmd1 = r'[^ \{\}]+\{[^\{\}]+\}\{[^\{\}]+\}'
cmd2 = r'[^ \{\}]+\{[^\{\}]+\}'
txt_content = re.sub(cmd1, '', txt_content)
txt_content = re.sub(cmd2, '', txt_content)
# Prepocess text extracted from nxml
self.preprocess_plain_text_string(txt_content, pmid, extra_annotations)
def prep_input(self, content_iter):
logger.info('Prepping input.')
for content in content_iter:
# If it's an NXML, we get the raw text and save it as new content
if content.is_format('nxml'):
txt = extract_text(content.get_text())
content = \
Content.from_string(str(content.get_id()),
'txt', txt)
quality_issue = self._check_content(content.get_text())
if quality_issue is not None:
logger.warning('Skipping %s due to: %s' %
(content.get_id(), quality_issue))
continue
new_fpath = content.copy_to(self.input_dir)
self.num_input += 1
logger.debug('%s saved for reading by Eidos.' % new_fpath)
return
def test_extract_text2():
xml_str = '<article><body><p><p>some text</p>a</p></body></article>'
text = pmc_client.extract_text(xml_str)
assert text == 'a\nsome text\n'
assert unicode_strs(text)
def test_extract_text():
pmc_id = '4322985'
xml_str = pmc_client.get_xml(pmc_id)
text = pmc_client.extract_text(xml_str)
assert text is not None
assert unicode_strs(text)
def test_extract_text():
pmc_id = '4322985'
xml_str = pmc_client.get_xml(pmc_id)
text = pmc_client.extract_text(xml_str)
assert text is not None
assert unicode_strs(text)