def import_posts(self):
category = self.get_category()
self.write_out(self.style.STEP("- Importing entries\n"))
for post in self.blogger_manager.get_posts(self.blogger_blog_id, self.blogger_limit):
creation_date = convert_blogger_timestamp(post.published.text)
status = DRAFT if is_draft(post) else PUBLISHED
title = post.title.text or ""
content = post.content.text or ""
excerpt = self.auto_excerpt and Truncator(strip_tags(smart_unicode(content))).words(50) or ""
slug = slugify(post.title.text or get_post_id(post))[:255]
try:
entry = Entry.objects.get(creation_date=creation_date, slug=slug)
output = self.style.NOTICE("> Skipped %s (already migrated)\n" % entry)
except Entry.DoesNotExist:
entry = Entry(
status=status, title=title, content=content, creation_date=creation_date, slug=slug, excerpt=excerpt
)
entry.tags = ",".join([slugify(cat.term) for cat in post.category])
entry.last_update = convert_blogger_timestamp(post.updated.text)
entry.save()
entry.sites.add(self.SITE)
entry.categories.add(category)
entry.authors.add(self.default_author)
try:
self.import_comments(entry, post)
except gdata_service.RequestError:
# comments not available for this post
pass
entry.comment_count = entry.comments.count()
entry.save(force_update=True)
output = self.style.ITEM("> Migrated %s + %s comments\n" % (entry.title, entry.comment_count))
self.write_out(output)
def import_from_rss(rss_as_bytes):
rss = objectify.fromstring(rss_as_bytes)
current_site = Site.objects.get_current()
for item in rss.channel.item:
blob = unicode(item['{%s}encoded' % item.nsmap['content']])
category, _ = Category.objects.get_or_create(
slug=my_slugify(unicode(item.category.text)),
title=unicode(item.category.text)
)
entry = Entry()
cms_content = Content()
cms_content.save()
process_html_blob(blob, cms_content)
entry.content_ptr = cms_content
entry.title = unicode(item.title.text)
entry.creation_date = parse_rss_date(unicode(item.pubDate))
entry.slug = my_slugify(entry.title)
entry.save()
entry.sites.add(current_site)
entry.categories.add(category)
link = unicode(item.link.text)
_url_cache[link] = entry.get_absolute_url()
_url_cache[re.sub("^https", "http", link)] = entry.get_absolute_url()
add_redirect(link, entry.get_absolute_url())
def import_posts(self):
category = self.get_category()
self.write_out(self.style.STEP('- Importing entries\n'))
for post in self.blogger_manager.get_posts(self.blogger_blog_id,
self.blogger_limit):
creation_date = convert_blogger_timestamp(post.published.text)
status = DRAFT if is_draft(post) else PUBLISHED
title = post.title.text or ''
content = post.content.text or ''
excerpt = self.auto_excerpt and Truncator(
strip_tags(content)).words(50) or ''
slug = slugify(post.title.text or get_post_id(post))[:255]
try:
entry = Entry.objects.get(creation_date=creation_date,
slug=slug)
output = self.style.NOTICE(
'> Skipped %s (already migrated)\n' % entry)
except Entry.DoesNotExist:
entry = Entry(status=status,
title=title,
content=content,
creation_date=creation_date,
slug=slug,
excerpt=excerpt)
if self.default_author:
entry.author = self.default_author
entry.tags = ','.join(
[slugify(cat.term) for cat in post.category])
entry.last_update = convert_blogger_timestamp(
post.updated.text)
entry.save()
entry.sites.add(self.SITE)
entry.categories.add(category)
entry.authors.add(self.default_author)
try:
self.import_comments(entry, post)
except gdata_service.RequestError:
# comments not available for this post
pass
entry.comment_count = entry.comments.count()
entry.save(force_update=True)
output = self.style.ITEM('> Migrated %s + %s comments\n' %
(entry.title, entry.comment_count))
self.write_out(output)
def save(self, *args, **kwargs):
if self.pk is None: # or if there is no entry? What is a good check?
# take action on first save
# NB! Try to make sure that the slug is unique so we can have separate links to separate messages
entry = Entry(
title=self.create_entry_title(),
content=self.create_entry_content(),
status=HIDDEN,
)
entry.save()
entry.slug = self.create_entry_slug() + '-' + str(entry.pk)
entry.sites.add(Site.objects.get(pk=settings.SITE_ID))
entry.authorized_users.add(self.user)
entry.content_template += '.%s' % (self.__class__.__name__.lower())
entry.detail_template += '.%s' % (self.__class__.__name__.lower())
entry.save()
# Should we perhaps save somewhere this Site.objects.get(pk=settings.SITE_ID) so we don't have to query it every time? Though, I guess, it is not such a big deal here, as we are rarely saving things that have associated blog posts. We are not generating blog posts (entries) every second...
self.entry = entry
else:
if self.entry is None:
pass # QUESTION. Should we still post? Better late than never? Though, with transactions, we'd never get here.
super(PrivatelyPublishedModelMixin, self).save(*args, **kwargs)
def import_posts(self):
category = self.get_category()
self.write_out(self.style.STEP('- Importing entries\n'))
for post in self.blogger_manager.get_posts(self.blogger_blog_id):
creation_date = convert_blogger_timestamp(post.published.text)
status = DRAFT if is_draft(post) else PUBLISHED
title = post.title.text or ''
content = post.content.text or ''
slug = slugify(post.title.text or get_post_id(post))[:255]
try:
entry = Entry.objects.get(creation_date=creation_date,
slug=slug)
output = self.style.NOTICE('> Skipped %s (already migrated)\n'
% entry)
except Entry.DoesNotExist:
entry = Entry(status=status, title=title, content=content,
creation_date=creation_date, slug=slug)
if self.default_author:
entry.author = self.default_author
entry.tags = ','.join([slugify(cat.term) for
cat in post.category])
entry.last_update = convert_blogger_timestamp(
post.updated.text)
entry.save()
entry.sites.add(self.SITE)
entry.categories.add(category)
entry.authors.add(self.default_author)
try:
self.import_comments(entry, post)
except gdata_service.RequestError:
# comments not available for this post
pass
output = self.style.ITEM('> Migrated %s + %s comments\n'
% (entry.title, len(Comment.objects.for_model(entry))))
self.write_out(output)
def import_entries(self, feed_entries):
"""Import entries"""
for feed_entry in feed_entries:
self.write_out('> %s... ' % feed_entry.title)
if feed_entry.get('publised_parsed'):
creation_date = datetime(*feed_entry.published_parsed[:6])
if settings.USE_TZ:
creation_date = timezone.make_aware(
creation_date, timezone.utc)
else:
creation_date = timezone.now()
slug = slugify(feed_entry.title)[:255]
if Entry.objects.filter(creation_date__year=creation_date.year,
creation_date__month=creation_date.month,
creation_date__day=creation_date.day,
slug=slug):
self.write_out(self.style.NOTICE(
'SKIPPED (already imported)\n'))
continue
categories = self.import_categories(feed_entry)
entry_dict = {'title': feed_entry.title[:255],
'content': feed_entry.description,
'excerpt': feed_entry.get('summary'),
'status': PUBLISHED,
'creation_date': creation_date,
'start_publication': creation_date,
'last_update': timezone.now(),
'slug': slug}
if not entry_dict['excerpt'] and self.auto_excerpt:
entry_dict['excerpt'] = Truncator(
strip_tags(feed_entry.description)).words(50)
if self.tags:
entry_dict['tags'] = self.import_tags(categories)
entry = Entry(**entry_dict)
entry.save()
entry.categories.add(*categories)
entry.sites.add(self.SITE)
if self.image_enclosure:
for enclosure in feed_entry.enclosures:
if 'image' in enclosure.get('type') \
and enclosure.get('href'):
img_tmp = NamedTemporaryFile(delete=True)
img_tmp.write(urlopen(enclosure['href']).read())
img_tmp.flush()
entry.image.save(os.path.basename(enclosure['href']),
File(img_tmp))
break
if self.default_author:
entry.authors.add(self.default_author)
elif feed_entry.get('author_detail'):
try:
author = Author.objects.create_user(
slugify(feed_entry.author_detail.get('name')),
feed_entry.author_detail.get('email', ''))
except IntegrityError:
author = Author.objects.get(
username=slugify(feed_entry.author_detail.get('name')))
entry.authors.add(author)
self.write_out(self.style.ITEM('OK\n'))
def test_restricted_access(self):
client = Client()
user_g = create_test_user("Guido van Rossum", "*****@*****.**",
"3jh3j")
user_h = create_test_user("Hilda von Varden", "*****@*****.**",
"777777777")
# must specify slug. Otherwise get_aboslute_url is broken and will get NoReverseMatch.
entry = Entry(title="Django's wonderful permalink decorator 12321",
slug="django",
status=PUBLISHED,
start_publication=datetime.datetime(
2011, 8, 15, 8, 15, 12, 0, pytz.UTC))
private_entry_g = Entry(
title="Secret communication of Guido with space travellers",
slug="secret-never-see",
status=HIDDEN)
private_entry_h = Entry(
title="Hilda spent her vacation in New Zealand",
slug="-k-a-n-g-a-r-o-o-",
status=HIDDEN)
### NB!
### This will fail with an in-memory sqlite3 database, but works fine with an on-disk sqlite3 database. With in-memory database it fails with 'no such table zinnia_category'. If we omit status-PUBLISHED, it works just fine. This is really weird, because the code in a previous test manually queries the database (as sqlite3) and confirms that various Zinnia tables exist, including 'zinnia_category'. Perhaps, Zinnia has some extra functionality when saving PUBLISHED entries, and it does something that breaks with an in-memory database...
entry.save()
private_entry_g.save()
private_entry_h.save()
# Adding the respective users as authorized. We can only do it after first saving the models.
private_entry_g.authorized_users.add(user_g)
private_entry_g.save()
private_entry_h.authorized_users.add(user_h)
private_entry_h.save()
# Logging in user_g
client.login(username='******', password='******')
# Now, we did not set the site for the new entries (see django.contrib.sites).
# Thus, we do not expect them to show up when we follow a direct link.
# By the way, they still shows up in the dashboard; this is probably a Zinnia issue.
self.assertEqual(client.get(entry.get_absolute_url()).status_code, 404)
self.assertEqual(
client.get(private_entry_g.get_absolute_url()).status_code, 404)
self.assertEqual(
client.get(private_entry_h.get_absolute_url()).status_code, 404)
# django.contrib.sites
# Adding the current site to the Entry.
# Otherwise it will not be displayed in the detailed view.
# For some weird reason it is still displayed in the dashboard (that is, zinnia archive of entries). Created an issue for Zinnia on github...
self.site = Site.objects.get(pk=settings.SITE_ID)
entry.sites.add(self.site)
entry.save()
private_entry_g.sites.add(self.site)
private_entry_g.save()
private_entry_h.sites.add(self.site)
private_entry_h.save()
# Test a direct link to the entries. We should see the public entry and the private entry by one of the users
self.assertContains(client.get(entry.get_absolute_url()),
'wonderful permalink decorator 12321',
status_code=200)
# self.assertContains(client.get(private_entry_g.get_absolute_url()), 'Secret communication of Guido with', status_code=200)
# self.assertEquals(client.get(private_entry_h.get_absolute_url()).status_code, 404)
# Also, there should be no link to the hidden entry from the detail view of other entries!
self.assertNotContains(client.get(entry.get_absolute_url()),
private_entry_h.get_absolute_url())
self.assertNotContains(client.get(private_entry_g.get_absolute_url()),
private_entry_h.get_absolute_url())
# Test the dashboard
response_dash = client.get(reverse('dashboard'), follow=True)
self.assertContains(response_dash,
'wonderful permalink decorator 12321',
status_code=200)
self.assertContains(response_dash,
'Secret communication of Guido with',
status_code=200)
self.assertNotContains(response_dash,
'spent her vacation',
status_code=200)
# The dashboard should contain links "Continue Reading" to the visible posts
self.assertContains(response_dash, entry.get_absolute_url())
# self.assertContains(response_dash, private_entry_g.get_absolute_url())
# self.assertNotContains(response_dash, private_entry_h.get_absolute_url())
client.logout()
# LOG OUT.
# Still try a direct link to a public entry! But the private one should return 404.
self.assertContains(client.get(entry.get_absolute_url()),
'wonderful permalink decorator 12321',
status_code=200)
self.assertEquals(
client.get(private_entry_g.get_absolute_url()).status_code, 404)
self.assertEquals(
client.get(private_entry_h.get_absolute_url()).status_code, 404)
# Since we are logged out, the public entry cannot show links to either hidden entry
self.assertNotContains(client.get(entry.get_absolute_url()),
private_entry_g.get_absolute_url())
self.assertNotContains(client.get(entry.get_absolute_url()),
private_entry_h.get_absolute_url())
# Now logging in user_h
client.login(username="******", password="******")
# Test a direct link to the entries. We should see the public entry and the private entry by the other user
self.assertContains(client.get(entry.get_absolute_url()),
'wonderful permalink decorator 12321',
status_code=200)
self.assertEquals(
client.get(private_entry_g.get_absolute_url()).status_code, 404)
self.assertContains(client.get(private_entry_h.get_absolute_url()),
'Hilda spent her vacation in New Zealand',
status_code=200)
# Also, there should be no link to the hidden entry from the detail view of other entries!
# self.assertNotContains(client.get(entry.get_absolute_url()), private_entry_g.get_absolute_url())
# self.assertNotContains(client.get(private_entry_h.get_absolute_url()), private_entry_g.get_absolute_url())
# Test the dashboard
response_dash = client.get(reverse('dashboard'), follow=True)
self.assertContains(response_dash,
'wonderful permalink decorator 12321',
status_code=200)
self.assertNotContains(response_dash,
'Secret communication of Guido with',
status_code=200)
self.assertContains(response_dash,
'spent her vacation',
status_code=200)
# The dashboard should contain links "Continue Reading" to the visible posts
self.assertContains(response_dash, entry.get_absolute_url())
# self.assertNotContains(response_dash, private_entry_g.get_absolute_url())
# self.assertContains(response_dash, private_entry_h.get_absolute_url())
client.logout()
def import_entries(self, feed_entries):
"""
Import entries.
"""
for feed_entry in feed_entries:
self.write_out('> %s... ' % feed_entry.title)
if feed_entry.get('published_parsed'):
creation_date = datetime(*feed_entry.published_parsed[:6])
if settings.USE_TZ:
creation_date = timezone.make_aware(
creation_date, timezone.utc)
else:
creation_date = timezone.now()
slug = slugify(feed_entry.title)[:255]
if Entry.objects.filter(creation_date__year=creation_date.year,
creation_date__month=creation_date.month,
creation_date__day=creation_date.day,
slug=slug):
self.write_out(
self.style.NOTICE('SKIPPED (already imported)\n'))
continue
categories = self.import_categories(feed_entry)
entry_dict = {
'title': feed_entry.title[:255],
'content': feed_entry.description,
'excerpt': strip_tags(feed_entry.get('summary')),
'status': PUBLISHED,
'creation_date': creation_date,
'start_publication': creation_date,
'last_update': timezone.now(),
'slug': slug
}
if not entry_dict['excerpt'] and self.auto_excerpt:
entry_dict['excerpt'] = Truncator(
strip_tags(feed_entry.description)).words(50)
if self.tags:
entry_dict['tags'] = self.import_tags(categories)
entry = Entry(**entry_dict)
entry.save()
entry.categories.add(*categories)
entry.sites.add(self.SITE)
if self.image_enclosure:
for enclosure in feed_entry.enclosures:
if ('image' in enclosure.get('type')
and enclosure.get('href')):
img_tmp = NamedTemporaryFile(delete=True)
img_tmp.write(urlopen(enclosure['href']).read())
img_tmp.flush()
entry.image.save(os.path.basename(enclosure['href']),
File(img_tmp))
break
if self.default_author:
entry.authors.add(self.default_author)
elif feed_entry.get('author_detail'):
try:
author = Author.objects.create_user(
slugify(feed_entry.author_detail.get('name')),
feed_entry.author_detail.get('email', ''))
except IntegrityError:
author = Author.objects.get(
username=slugify(feed_entry.author_detail.get('name')))
entry.authors.add(author)
self.write_out(self.style.ITEM('OK\n'))