def save(self, force_insert=False, force_update=False):
if not self.stripped_title:
self.stripped_title = remove_punctuation(self.title)
if not self.stripped_author_names:
self.stripped_author_names = remove_punctuation(self.author_names)
self.mod_date = datetime.datetime.now()
super(Document, self).save(force_insert, force_update)
def save(self, force_insert=False, force_update=False):
if not self.stripped_title:
self.stripped_title = remove_punctuation(self.title)
if not self.stripped_author_names:
self.stripped_author_names = remove_punctuation(self.author_names)
self.mod_date = datetime.datetime.now()
super(Document, self).save(force_insert, force_update)
def get_first_author(record):
first_author = None
if record.authors:
first_author, drop, remainder = record.authors.partition(' ')
first_author = first_author.rstrip().rstrip(',')
first_author = remove_punctuation(first_author)
return first_author
def get_first_author(record):
first_author = None
if record.authors:
first_author, drop, remainder = record.authors.partition(' ')
first_author = first_author.rstrip().rstrip(',')
first_author = remove_punctuation(first_author)
return first_author
def create_publications(record, document, nameorder_significant=False):
# goal: create a publication record for each author, populate record with appropriate
# institution, affiliation, name order, and employee reference when and as possible
# step 1: create a publication record for each author
publications = []
authors = []
if record.authors and ';' in record.authors:
authors = record.authors.split(';')
elif record.authors and ',' in record.authors and not ';' in record.authors:
authors = record.authors.split(',')
if len(authors) == 2 and len(authors[1]) == 1:
authors = [record.authors]
elif record.book_authors and ';' in record.book_authors:
authors = record.book_authors.split(';')
elif record.book_authors and ',' in record.book_authors and not ';' in record.authors:
authors = record.book_authors.split(',')
else:
authors.append(record.authors)
authors = map(string.strip, authors)
for name in authors:
nameorder = None
if len(authors) == 1:
nameorder, no_created = NameOrder.objects.get_or_create(order='Sole')
elif name == authors[0] and not name == authors[-1]:
nameorder, no_created = NameOrder.objects.get_or_create(order='First')
elif name == authors[-1] and not name == authors[0]:
nameorder, no_created = NameOrder.objects.get_or_create(order='Last')
else:
nameorder, no_created = NameOrder.objects.get_or_create(order='Contributor')
publication, pub_created = Publication.objects.get_or_create(author_name=name.strip(),
name_order=nameorder,
document=document)
publications.append(publication)
# step 2: handle significant nameorder for CINAHL records
if nameorder_significant and nameorder and nameorder.order in ('Sole', 'First'):
publication.affiliation = record.affiliations
address = record.affiliations
if 'MSKCC' in address or 'Sloan-Kettering' in address or '1275 York' in address or 'Sloan Kettering' in address:
publication.institution = Institution.objects.get(name='Memorial Sloan-Kettering Cancer Center')
publication.save()
# step 3: handle Correspondence Address, to catch anything missed earlier
if record.correspondence_address:
for pub in publications:
if pub.author_name in record.correspondence_address:
modified = False
if not pub.affiliation:
pub.affiliation = record.correspondence_address
modified = True
if not pub.institution:
address = record.correspondence_address
if 'MSKCC' in address or 'Sloan-Kettering' in address or '1275 York' in address or 'Sloan Kettering' in address:
pub.institution = Institution.objects.get(name='Memorial Sloan-Kettering Cancer Center')
modified = True
if modified:
pub.save()
# step 4: populate each publication with appropriate employee, as possible
for pub in publications:
# handle PSYCINFO affiliations
if record.affiliations:
for affiliation in record.affiliations.split(';'):
if pub.author_name in affiliation:
pub.affiliation = affiliation
if 'MSKCC' in affiliation or 'Sloan-Kettering' in affiliation or '1275 York' in affiliation or 'Sloan Kettering' in affiliation:
pub.institution = Institution.objects.get(name='Memorial Sloan-Kettering Cancer Center')
pub.save()
# handle SCOPUS affiliations
if len(record.affiliations.split(';')) == len(authors):
name_index = None
try:
name_index = authors.index(pub.author_name)
except ValueError: #names don't match
stripped_authors = map(remove_punctuation, authors)
name_index = stripped_authors.index(remove_punctuation(pub.author_name))
if name_index:
pub.affiliation = record.affiliations.split(';')[name_index]
if 'MSKCC' in pub.affiliation or 'Sloan-Kettering' in pub.affiliation or '1275 York' in pub.affiliation or 'Sloan Kettering' in pub.affiliation:
pub.institution = Institution.objects.get(name='Memorial Sloan-Kettering Cancer Center')
pub.save()
create_author(pub, pub.author_name)
def create_document(record, source):
document = None
q = Q()
if record.doi:
documents = Document.objects.filter(doi__exact=record.doi)
if len(documents) == 1:
document = documents[0]
if record.authors and record.pages and not document:
# here's the first de-punctuated comparison point
first_author = get_first_author(record) # includes de-punctuation
first_page = record.pages.split('-')[0].strip()
if first_page and not first_page.isspace():
documents = Document.objects.filter(page_range__isnull=False).filter(stripped_author_names__istartswith=first_author,
page_range__istartswith=first_page)
if len(documents) == 1:
document = documents[0]
if record.authors and record.title and not document:
# here's the second de-punctuated comparison point
first_author = get_first_author(record) # includes de-punctuation
stripped_title = remove_punctuation(record.title)
documents = Document.objects.filter(stripped_author_names__istartswith=first_author,
stripped_title__iexact=stripped_title)
if len(documents) == 1:
document = documents[0]
if not document:
document = Document.objects.create(title=record.title,
author_names=record.authors,
doi=record.doi)
modified = False
if record.doi and not document.doi:
document.doi = record.doi
modified = True
if record.title and not document.title:
document.title = record.title
modified = True
if document.title and not document.stripped_title:
document.stripped_title = remove_punctuation(document.title)
modified = True
if record.authors and not document.author_names:
document.author_names = record.authors
modified = True
if record.authors and not document.stripped_author_names:
# here's where we set the de-punctuated author
document.stripped_author_names = remove_punctuation(record.authors)
modified = True
if record.source and not document.source:
document.source = source
modified = True
if record.abstract and not document.abstract:
document.abstract = record.abstract
modified = True
if record.volume and not document.volume:
document.volume = record.volume
modified = True
if record.issue and not document.issue:
document.issue = record.issue
modified = True
if record.pages and not document.page_range:
document.page_range = record.pages
modified = True
if record.language and not document.language:
document.language = record.language
modified = True
if record.publish_year and not document.publish_year:
document.publish_year = record.publish_year
modified = True
if record.publish_date and not document.publish_date:
document.publish_date = record.publish_date
modified = True
if record.document_type and not document.document_type:
document.document_type = map_document_type(record.document_type)
modified = True
if record.document_subtype and not document.document_subtype:
document.document_subtype = record.document_subtype
modified = True
if record.dmt and not document.dmt:
document.dmt = DiseaseManagementTeam.objects.get(id=record.dmt)
modified = True
if record.affiliations and not document.affiliations:
document.affiliations = record.affiliations
modified = True
if modified:
document.save()
return document
def create_publications(record, document, nameorder_significant=False):
# goal: create a publication record for each author, populate record with appropriate
# institution, affiliation, name order, and employee reference when and as possible
# step 1: create a publication record for each author
publications = []
authors = []
if record.authors and ';' in record.authors:
authors = record.authors.split(';')
elif record.authors and ',' in record.authors and not ';' in record.authors:
authors = record.authors.split(',')
if len(authors) == 2 and len(authors[1]) == 1:
authors = [record.authors]
elif record.book_authors and ';' in record.book_authors:
authors = record.book_authors.split(';')
elif record.book_authors and ',' in record.book_authors and not ';' in record.authors:
authors = record.book_authors.split(',')
else:
authors.append(record.authors)
authors = map(string.strip, authors)
for name in authors:
nameorder = None
if len(authors) == 1:
nameorder, no_created = NameOrder.objects.get_or_create(
order='Sole')
elif name == authors[0] and not name == authors[-1]:
nameorder, no_created = NameOrder.objects.get_or_create(
order='First')
elif name == authors[-1] and not name == authors[0]:
nameorder, no_created = NameOrder.objects.get_or_create(
order='Last')
else:
nameorder, no_created = NameOrder.objects.get_or_create(
order='Contributor')
publication, pub_created = Publication.objects.get_or_create(
author_name=name.strip(), name_order=nameorder, document=document)
publications.append(publication)
# step 2: handle significant nameorder for CINAHL records
if nameorder_significant and nameorder and nameorder.order in (
'Sole', 'First'):
publication.affiliation = record.affiliations
address = record.affiliations
if 'MSKCC' in address or 'Sloan-Kettering' in address or '1275 York' in address or 'Sloan Kettering' in address:
publication.institution = Institution.objects.get(
name='Memorial Sloan-Kettering Cancer Center')
publication.save()
# step 3: handle Correspondence Address, to catch anything missed earlier
if record.correspondence_address:
for pub in publications:
if pub.author_name in record.correspondence_address:
modified = False
if not pub.affiliation:
pub.affiliation = record.correspondence_address
modified = True
if not pub.institution:
address = record.correspondence_address
if 'MSKCC' in address or 'Sloan-Kettering' in address or '1275 York' in address or 'Sloan Kettering' in address:
pub.institution = Institution.objects.get(
name='Memorial Sloan-Kettering Cancer Center')
modified = True
if modified:
pub.save()
# step 4: populate each publication with appropriate employee, as possible
for pub in publications:
# handle PSYCINFO affiliations
if record.affiliations:
for affiliation in record.affiliations.split(';'):
if pub.author_name in affiliation:
pub.affiliation = affiliation
if 'MSKCC' in affiliation or 'Sloan-Kettering' in affiliation or '1275 York' in affiliation or 'Sloan Kettering' in affiliation:
pub.institution = Institution.objects.get(
name='Memorial Sloan-Kettering Cancer Center')
pub.save()
# handle SCOPUS affiliations
if len(record.affiliations.split(';')) == len(authors):
name_index = None
try:
name_index = authors.index(pub.author_name)
except ValueError: #names don't match
stripped_authors = map(remove_punctuation, authors)
name_index = stripped_authors.index(
remove_punctuation(pub.author_name))
if name_index:
pub.affiliation = record.affiliations.split(
';')[name_index]
if 'MSKCC' in pub.affiliation or 'Sloan-Kettering' in pub.affiliation or '1275 York' in pub.affiliation or 'Sloan Kettering' in pub.affiliation:
pub.institution = Institution.objects.get(
name='Memorial Sloan-Kettering Cancer Center')
pub.save()
create_author(pub, pub.author_name)
def create_document(record, source):
document = None
q = Q()
if record.doi:
documents = Document.objects.filter(doi__exact=record.doi)
if len(documents) == 1:
document = documents[0]
if record.authors and record.pages and not document:
# here's the first de-punctuated comparison point
first_author = get_first_author(record) # includes de-punctuation
first_page = record.pages.split('-')[0].strip()
if first_page and not first_page.isspace():
documents = Document.objects.filter(
page_range__isnull=False).filter(
stripped_author_names__istartswith=first_author,
page_range__istartswith=first_page)
if len(documents) == 1:
document = documents[0]
if record.authors and record.title and not document:
# here's the second de-punctuated comparison point
first_author = get_first_author(record) # includes de-punctuation
stripped_title = remove_punctuation(record.title)
documents = Document.objects.filter(
stripped_author_names__istartswith=first_author,
stripped_title__iexact=stripped_title)
if len(documents) == 1:
document = documents[0]
if not document:
document = Document.objects.create(title=record.title,
author_names=record.authors,
doi=record.doi)
modified = False
if record.doi and not document.doi:
document.doi = record.doi
modified = True
if record.title and not document.title:
document.title = record.title
modified = True
if document.title and not document.stripped_title:
document.stripped_title = remove_punctuation(document.title)
modified = True
if record.authors and not document.author_names:
document.author_names = record.authors
modified = True
if record.authors and not document.stripped_author_names:
# here's where we set the de-punctuated author
document.stripped_author_names = remove_punctuation(record.authors)
modified = True
if record.source and not document.source:
document.source = source
modified = True
if record.abstract and not document.abstract:
document.abstract = record.abstract
modified = True
if record.volume and not document.volume:
document.volume = record.volume
modified = True
if record.issue and not document.issue:
document.issue = record.issue
modified = True
if record.pages and not document.page_range:
document.page_range = record.pages
modified = True
if record.language and not document.language:
document.language = record.language
modified = True
if record.publish_year and not document.publish_year:
document.publish_year = record.publish_year
modified = True
if record.publish_date and not document.publish_date:
document.publish_date = record.publish_date
modified = True
if record.document_type and not document.document_type:
document.document_type = map_document_type(record.document_type)
modified = True
if record.document_subtype and not document.document_subtype:
document.document_subtype = record.document_subtype
modified = True
if record.dmt and not document.dmt:
document.dmt = DiseaseManagementTeam.objects.get(id=record.dmt)
modified = True
if record.affiliations and not document.affiliations:
document.affiliations = record.affiliations
modified = True
if modified:
document.save()
return document
