def set_by_name(cls, subject, work=None, authority=None):
''' use this method whenever you would be creating a new subject!'''
subject = subject.strip()
# make sure it's not a ; delineated list
subjects = subject.split(';')
for additional_subject in subjects[1:]:
cls.set_by_name(additional_subject, work, authority)
subject = subjects[0]
# make sure there's no heading
headingmatch = re.match(r'^!(.+):(.+)', subject)
if headingmatch:
subject = headingmatch.group(2).strip()
authority = headingmatch.group(1).strip()
elif subject.startswith('nyt:'):
subject = subject[4:].split('=')[0].replace(
'_', ' ').strip().capitalize()
subject = 'NYT Bestseller - {}'.format(subject)
authority = 'nyt'
elif subject.startswith('award:'):
subject = subject[6:].split('=')[0].replace(
'_', ' ').strip().capitalize()
subject = 'Award Winner - {}'.format(subject)
authority = 'award'
if valid_subject(subject):
(subject_obj, created) = cls.objects.get_or_create(name=subject)
if not subject_obj.authority and authority:
subject_obj.authority = authority
subject_obj.save()
subject_obj.works.add(work)
return subject_obj
else:
return None
def get_keywords(self):
value = self.check_metas(['keywords']).strip(',;')
if value:
subjects = []
for subject in re.split(' *[;,] *', value):
if valid_subject(subject):
subjects.append(subject)
self.set('subjects', subjects)
def attach_more_doab_metadata(edition,
description,
subjects,
publication_date,
publisher_name=None,
language=None,
authors=u''):
"""
for given edition, attach description, subjects, publication date to
corresponding Edition and Work
"""
# if edition doesn't have a publication date, update it
if not edition.publication_date:
edition.publication_date = publication_date
# if edition.publisher_name is empty, set it
if not edition.publisher_name:
edition.set_publisher(publisher_name)
edition.save()
# attach description to work if it's not empty
work = edition.work
if not work.description:
work.description = description
# update subjects
for s in subjects:
if valid_subject(s):
models.Subject.set_by_name(s, work=work)
# set reading level of work if it's empty; doab is for adults.
if not work.age_level:
work.age_level = '18-'
if language:
work.language = language
work.save()
if authors and authors == authors: # test for authors != NaN
authlist = creator_list(authors)
if edition.authors.all().count() < len(authlist):
edition.authors.clear()
if authlist is not None:
for [rel, auth] in authlist:
edition.add_author(auth, rel)
return edition
def handle(self, **options):
semicolon_subjects = Subject.objects.filter(name__contains=";")
for subject in semicolon_subjects:
for work in subject.works.all():
Subject.set_by_name(subject.name, work=work)
subject.delete()
nyt_subjects = Subject.objects.filter(name__startswith="nyt:")
for subject in nyt_subjects:
for work in subject.works.all():
Subject.set_by_name(subject.name, work=work)
subject.delete()
award_subjects = Subject.objects.filter(name__startswith="award:")
for subject in award_subjects:
for work in subject.works.all():
Subject.set_by_name(subject.name, work=work)
subject.delete()
period_subjects = Subject.objects.filter(name__contains=".")
for subject in period_subjects:
if not valid_subject(subject.name):
subject.delete()