def ocr_url(identifier):
"""Generate URL to get OCR data from
"""
identifier_parts = identifier.split('?')
if len(identifier_parts) > 1:
return ocr_base_url + identifier_parts[1]
else:
log.error('Could not generate OCR link for identifier ' + identifier)
return ''
def article_ocr(identifier):
"""Retrieve ocr full text for article with identifier
"""
url = DelpherAPI.ocr_url(identifier)
try:
response = request.get(url)
except:
log.exception('Could not get OCR data for url {url}.'.format(**locals()))
return '<failed to load>'
if response is None:
log.error('Did not get OCR data for url {url}.'.format(**locals()))
return '<failed to load>'
else:
# Each paragraph is one item in the list
return "\n\n".join([response[key] for key in sorted(response.keys()) if key != 'title'])