def parse_pdf(self, pdf_url):
""" Grab a single pdf url, using cache via LinkedFile
"""
existing_count = Link.objects.filter(url=pdf_url).count()
if existing_count < 1:
link = Link(url=pdf_url)
link.save()
else:
if existing_count > 1:
print "WARNING: you have two objects with the url %s. Taking the first" % pdf_url
link = Link.objects.filter(url=pdf_url).iterator().next()
filename = None
if link.linkedfile_set.count() > 0:
files = [f for f in link.linkedfile_set.order_by("last_updated") if f.link_file.name != ""]
if len(files) > 0:
filename = files[0].link_file.path
logger.debug("reusing %s from %s" % (pdf_url, filename))
if not filename:
logger.debug("getting %s" % pdf_url)
contents = urllib2.urlopen(pdf_url).read()
link_file = LinkedFile(link=link)
saved_filename = os.path.basename(urlparse(pdf_url).path)
link_file.link_file.save(saved_filename, ContentFile(contents))
filename = link_file.link_file.path
prop = GovProposalParser(filename)
# TODO: check if parsing handles more than 1 prop in a booklet
return [{"title": prop.get_title(), "date": prop.get_date(), "bill": prop}]
def parse_pdf(self,pdf_url):
""" Grab a single pdf url, using cache via LinkedFile
"""
existing_count = Link.objects.filter(url=pdf_url).count()
if existing_count < 1:
link = Link(url=pdf_url)
link.save()
else:
if existing_count > 1:
print "WARNING: you have two objects with the url %s. Taking the first" % pdf_url
link = Link.objects.filter(url=pdf_url).iterator().next()
filename = None
if link.linkedfile_set.count() > 0:
files = [f for f in link.linkedfile_set.order_by('last_updated') if f.link_file.name != '']
if len(files) > 0:
filename = files[0].link_file.path
logger.debug('reusing %s from %s' % (pdf_url, filename))
if not filename:
logger.debug('getting %s' % pdf_url)
contents = urllib2.urlopen(pdf_url).read()
link_file = LinkedFile(link=link)
saved_filename = os.path.basename(urlparse(pdf_url).path)
link_file.link_file.save(saved_filename, ContentFile(contents))
filename = link_file.link_file.path
prop = GovProposalParser(filename)
# TODO: check if parsing handles more than 1 prop in a booklet
return [{'title':prop.get_title(),'date':prop.get_date(), 'bill':prop}]
def create_or_update_single_bill(self, proposal, pdf_link, link_file):
if not(proposal['date']) or CUTOFF_DATE and proposal['date'] < CUTOFF_DATE:
return
law_name = proposal['law']
(law, created) = Law.objects.get_or_create(title=law_name)
if created:
law.save()
if law.merged_into:
law = law.merged_into
title = u''
if proposal['correction']:
title += proposal['correction']
if proposal['comment']:
title += ' ' + proposal['comment']
if len(title)<=1:
title = u'חוק חדש'
(gp,created) = GovProposal.objects.get_or_create(booklet_number=proposal['booklet'], knesset_id=18,
source_url=proposal['link'],
title=title, law=law, date=proposal['date'])
if created:
gp.save()
logger.debug("created GovProposal id = %d" % gp.id)
bill_params = dict(law=law, title=title, stage='3', stage_date=proposal['date'])
similar_bills = Bill.objects.filter(**bill_params).order_by('id')
if len(similar_bills) >= 1:
b = similar_bills[0]
if len(similar_bills) > 1:
logger.debug("multiple bills detected")
for bill in similar_bills:
if bill.id == b.id:
logger.debug("bill being used now - %d" % bill.id)
else:
logger.debug("bill with same fields - %d" % bill.id)
else:
b = Bill(**bill_params)
b.save()
gp.bill = b
gp.save()
if link_file.link is None:
link = Link(title=pdf_link, url=pdf_link,
content_type=ContentType.objects.get_for_model(gp),
object_pk=str(gp.id))
link.save()
link_file.link = link
link_file.save()
logger.debug("check updated %s" % b.get_absolute_url())
