# page end
page_end = row[7]
# reprint
try:
reprint = bool(int(row[10]))
except:
reprint = False
# lookup number for pdf
lookup = row[11]
paper = Paper(author=author,
date=date,
title=title,
journal=journal,
volume = volume,
page_start=page_start,
page_end=page_end,
reprint=reprint,
lookup=lookup)
paper.save()
# keywords
kwords = [x.strip() for x in row[8].split(',')]
for kword in kwords:
obj, created = Keyword.objects.get_or_create(keyword=kword)
paper.keywords.add(obj)
def retrieve(self, request, pk=None):
"""
Logic for parsing paper info from PubMed using efetch
"""
try:
paper = Paper.objects.get(pmid=pk)
except Paper.DoesNotExist:
try:
handle = Entrez.efetch("pubmed", id=str(pk), rettype="medline")
attr = {}
kw = ['DP', 'TI', 'TA', 'VI', 'VI', 'PG', 'AB', 'PMC']
attr = attr.fromkeys(kw, None)
AU = []
MH = []
buff = ''
for line in handle:
if line.strip():
if line.startswith(' '):
buff = buff + ' ' + line.strip()
elif not buff:
buff = line.strip()
else:
key, val = buff.split('-', 1)
if key.strip() in kw:
attr[key.strip()] = val.strip()
elif key.strip() == 'AU':
AU.append(val.strip())
elif key.strip() == 'MH':
MH.append(val.strip())
buff = line.strip()
attr['AU'] = ', '.join(AU)
attr['MH'] = ', '.join(MH)
attr['DP'] = re.sub("[^0-9]", "", attr['DP'])
if attr['PMC']:
attr['PMC'] = re.sub("[^0-9]", "", attr['PMC'])
if attr['TI']:
attr['TI'] = attr['TI'].strip('.')
if attr['DP']:
attr['DP'] = attr['DP'][:4]
paper = Paper(
pmid=pk,
author=attr.get('AU', None),
pub_year=attr.get('DP', None),
title=attr.get('TI', None),
journal=attr.get('TA', None),
volume=attr.get('VI', None),
pages=attr.get('PG', None),
abstract=attr.get('AB', None),
keywords=attr.get('MH', None),
pmc=attr.get('PMC', None)
)
handle.close()
paper.save()
except:
return JsonResponse(status=404, data={'success':'false','message':'PMID:%s not found' % pk})
serializer = PaperSerializer(paper)
return Response(serializer.data)
