for tag in tags:
tag = tag.title()
try:
tag_m, created = Tag.objects.get_or_create(name=tag)
except Exception, e:
tag_m = Tag.objects.get(name=tag)
url_m.tags.add(tag_m)
tagstring = url_parser.tagstring or ""
url_m.raw_tags = tagstring
url_m.save()
lsum = LinkSum(
tags=tagstring, summary=url_parser.summary, url=url_m,
title=url_parser.title, link=url_parser.url, collection_id=default_collection.pk,
read=False, recommanded=1, source=source,
user_id=user_id,
)
if isinstance(lsum.summary, basestring) and len(lsum.summary) == 0:
lsum.summary = ""
if url_parser.image and not url_m.image:
url_m.image = url_parser.image
url_m.save()
try:
filtr = url_parser.find_collection(lsum, filters)
if filtr and filtr.xpath is not None and len(filtr.xpath.strip()) == 0:
filtr.xpath = None
filtr.save()
def index_url(link, user_id, link_at, author_name, source_name):
"""Entry point to store our link & linksum into the database"""
if not isinstance(user_id, (list, tuple)):
user_id = [int(user_id)]
urls = find_urls(link)
if not urls:
logger.info(u"No link provided")
return
#loads our filters
filters = Filter.objects.filter(Q(user__pk__in=user_id) | Q(user__isnull=True))
try:
source = Source.objects.get(name__iexact=source_name)
except Source.DoesNotExist:
logger.info(u"source %s unknown" % source_name)
return -1
try:
author = Author.objects.get(name=author_name)
except Author.DoesNotExist:
logger.info(u"Created author '{0}'".format(author_name))
author, created = Author.objects.get_or_create(name=author_name, source=source)
for url in urls:
link_extractor = LinkExtractor(url)
try:
link_extractor.fetch_url_content()
except index_exc.FetchException:
logger.exception(u"Can't fetch link")
continue
try:
link_extractor.extract()
except Exception, exc:
logger.warning(u"Can't extract url_content from %s" % url, exc_info=True)
continue
try:
url_instance = create_url(link_extractor)
except index_exc.UrlCreationException:
logger.warning(u"can't create url for link {0}".format(url), exc_info=True)
continue
users_lsums = dict([(user, None) for user in user_id])
for user in user_id:
try:
links_numb = LinkSum.objects.get(url=url_instance, user__id=user)
links_numb.recommanded += 1
links_numb.save()
links_numb.authors.add(author)
del users_lsums[user]
logger.info(u"linksum already in database")
except LinkSum.DoesNotExist:
pass
lsum = LinkSum(url=url_instance, collection_id=default_collection.pk,
read=False)
try:
find_collection(lsum, filters)
except index_exc.DeleteLinkException:
logger.info(u"Link not saved, filtered")
continue
for user in users_lsums:
try:
lsum.user_id = user
lsum.save()
lsum.authors.add(author)
lsum.sources.add(source)
for tags in url_instance.tags.all():
lsum.tags.add(tags)
except Exception, exc:
logger.exception(exc)
logger.info(u"Added new link for user {0}".format(user_id))
