def normalize_alias_tuple(ns, nid):
ns = clean_id(ns)
ns = ns.lower()
if ns == "biblio":
return (ns, nid)
nid = clean_id(nid)
from totalimpact.providers import crossref
from totalimpact.providers import pubmed
from totalimpact.providers import arxiv
from totalimpact.providers import webpage
from totalimpact import importer
clean_nid = None
if ns=="doi" or importer.is_doi(nid):
ns = "doi"
clean_nid = crossref.clean_doi(nid)
elif ns=="pmid" or importer.is_pmid(nid):
ns = "pmid"
clean_nid = pubmed.clean_pmid(nid)
elif ns=="arxiv" or importer.is_arxiv(nid):
ns = "arxiv"
clean_nid = arxiv.clean_arxiv_id(nid)
elif ns=="url" or importer.is_url(nid):
ns = "url"
clean_nid = webpage.clean_url(nid)
elif ns not in ["doi", "pmid", "arxiv", "url"]:
clean_nid = nid
if not clean_nid:
return None
return (ns, clean_nid)
def _extract_members(self, page, query_string=None):
if 'orcid-profile' not in page:
raise ProviderContentMalformedError(
"Content does not contain expected text")
data = provider._load_json(page)
members = []
try:
orcid_works = data["orcid-profile"]["orcid-activities"][
"orcid-works"]["orcid-work"]
except KeyError:
return []
for work in orcid_works:
new_member = None
try:
ids = work["work-external-identifiers"][
"work-external-identifier"]
for myid in ids:
if myid['work-external-identifier-type'] == "DOI":
doi = myid['work-external-identifier-id']['value']
doi = crossref.clean_doi(doi)
if doi:
new_member = ("doi", doi)
if myid['work-external-identifier-type'] == "PMID":
new_member = (
"pmid",
myid['work-external-identifier-id']['value'])
except KeyError:
pass
if not new_member:
logger.info(
u"no external identifiers, try saving whole citation for {orcid}"
.format(orcid=query_string))
biblio = self._parse_orcid_work(work)
new_member = ("biblio", biblio)
if new_member:
members += [new_member]
if not members:
raise ProviderItemNotFoundError
logger.info(u"returning {n} members for {orcid}".format(
n=len(members), orcid=query_string))
return (members)
def _extract_members(self, page, query_string=None):
if 'orcid-profile' not in page:
raise ProviderContentMalformedError("Content does not contain expected text")
data = provider._load_json(page)
members = []
try:
orcid_works = data["orcid-profile"]["orcid-activities"]["orcid-works"]["orcid-work"]
except KeyError:
return []
for work in orcid_works:
new_member = None
try:
ids = work["work-external-identifiers"]["work-external-identifier"]
for myid in ids:
if myid['work-external-identifier-type'] == "DOI":
doi = myid['work-external-identifier-id']['value']
doi = crossref.clean_doi(doi)
if doi:
new_member = ("doi", doi)
if myid['work-external-identifier-type'] == "PMID":
new_member = ("pmid", myid['work-external-identifier-id']['value'])
except KeyError:
pass
if not new_member:
logger.info(u"no external identifiers, try saving whole citation for {orcid}".format(
orcid=query_string))
biblio = self._parse_orcid_work(work)
new_member = ("biblio", biblio)
if new_member:
members += [new_member]
if not members:
raise ProviderItemNotFoundError
logger.info(u"returning {n} members for {orcid}".format(
n=len(members), orcid=query_string))
return(members)
