def index():
for row in _data:
issns = []
if row[_ISSN]: issns.append(row[_ISSN])
if row[_EISSN]: issns.append(row[_EISSN])
jnames = []
if row[_TITLE]: jnames.append(row[_TITLE])
if row[_TITLE_ALT] and row[
_TITLE_ALT] not in jnames: jnames.append(row[_TITLE_ALT])
eissn = []
if row[_EISSN]: eissn.append(row[_EISSN])
publisher = []
if row[_PUBLISHER]: publisher.append(row[_PUBLISHER])
Journal.add(issn=issns, electronic_issn=eissn,
journal_title=jnames, publisher_name=publisher)
def extract_data(filepath, pmidfile):
fn = os.path.split(filepath)[1]
# extract the xml from the zip
z = zipfile.ZipFile(filepath)
inf = z.filelist[0]
f = z.open(inf)
# parse the xml
tree = cetree.parse(f)
root = tree.getroot()
pmids = []
records = []
for mlc in root.findall("MedlineCitation"):
record = {"electronic_issn": [], "print_issn": [], "issn": [], "journal_title": [], "journal_abbr": []}
has_data = False
# record any pubmed ids that we encounter
mypmid = None
pmidels = mlc.findall("PMID")
for e in pmidels:
mypmid = e.text
if e.text not in pmids:
pmids.append(e.text)
article = mlc.find("Article")
journal = article.find("Journal")
# record any ISSNs in their appropriate type field
issnels = journal.findall("ISSN")
for e in issnels:
has_data = True
if e.get("IssnType") == "Electronic":
if e.text not in record["electronic_issn"]:
record["electronic_issn"].append(e.text)
if e.text in record["issn"]:
record["issn"].remove(e.text)
elif e.get("IssnType") == "Print":
if e.text not in record["print_issn"]:
record["print_issn"].append(e.text)
if e.text in record["issn"]:
record["issn"].remove(e.text)
else:
if e.text not in record["issn"]:
record["issn"].append(e.text)
# add journal titles
titlels = journal.findall("Title")
for e in titlels:
has_data = True
if e.text not in record["journal_title"]:
record["journal_title"].append(e.text)
# add journal abbreviations
isoels = journal.findall("ISOAbbreviation")
for e in isoels:
has_data = True
if e.text not in record["journal_abbr"]:
record["journal_abbr"].append(e.text)
info = mlc.find("MedlineJournalInfo")
# record medline's version of the name (which may well be the iso abbreviation)
taels = info.findall("MedlineTA")
for e in taels:
has_data = True
if e.text not in record["journal_title"] and e.text not in record["journal_abbr"]:
record["journal_abbr"].append(e.text)
# record the medline knowledge of this other issn - we don't necessarily know what type it is
links = info.findall("ISSNLinking")
for e in links:
has_data = True
if (
e.text not in record["electronic_issn"]
and e.text not in record["print_issn"]
and e.text not in record["issn"]
):
record["issn"].append(e.text)
# now write it to the ACAT via catflap
if has_data:
jnames = record.get("journal_title", [])
for a in record.get("journal_abbr", []):
if a not in jnames:
jnames.append(a)
source = "medline " + fn
if mypmid is not None:
source += " " + mypmid
Journal.add(
issn=record.get("issn", []) + record.get("electronic_issn", []) + record.get("print_issn", []),
journal_title=jnames,
electronic_issn=record.get("electronic_issn", []),
print_issn=record.get("print_issn", []),
journal_abbreviation=record.get("journal_abbr", []),
)
# save the pmids found in this file
save_pmids(pmidfile, pmids)