def test_filter_chain(ctx):
from melk.util.dibject import Dibject, dibjectify
from melkman.filters import NewsItemFilterFactory, ACCEPT_ITEM, REJECT_ITEM
filter_factory = NewsItemFilterFactory(ctx.component_manager)
chain = [
{'op': 'match_author',
'config': {'values': ['fred']},
'action': 'reject'},
{'op': 'and',
'config': {
'filters': [
{'op': 'match_author',
'config': {'values': ['barney']}},
{'op': 'match_field',
'config': {'field': 'foo',
'values': ['bar']}}]},
'action': 'accept'},
{'op': 'match_field',
'config': {'field': 'foo',
'values': ['bar']},
'action': 'reject'},
{'op': 'match_all',
'config': {},
'action': 'accept'}
]
chain = [dibjectify(x) for x in chain]
chain = filter_factory.create_chain(chain)
assert chain(dummy_news_item({'author': 'fred'})) == REJECT_ITEM
assert chain(dummy_news_item({})) == ACCEPT_ITEM
assert chain(dummy_news_item({'details': {'foo': 'bar'}})) == REJECT_ITEM
assert chain(dummy_news_item({'author': 'barney', 'details': {'foo': 'bar'}})) == ACCEPT_ITEM
def dummy_news_item(d):
di = DummyItem(dibjectify(d))
di.setdefault('author', 'Whoever T. Merriweather')
di.setdefault('item_id', random_id())
di.setdefault('timestamp', datetime.utcnow())
di.setdefault('title', 'The News Title')
di.setdefault('link', 'http://example.org/blagosphere?id=12')
di.setdefault('source_title', 'The Blags')
di.setdefault('source_url', 'http://example.org/blagosphere')
di.setdefault('summary', 'abaraljsrs sjrkja rsj klrjewori ew rwa riojweroiwer iowr wre')
di.setdefault('details', Dibject())
return di
def __init__(self, config):
self.config = dibjectify(config)
self._local = green_local()
find_plugins_by_entry_point(MELKMAN_PLUGIN_ENTRY_POINT)
self._broker = None
def parse_feed(content, feed_url):
fake_headers = {
'content-location': feed_url,
'content-type': 'text/xml; charset=utf-8',
}
ff = feedparser.parse(content, header_defaults=fake_headers)
# make a clean copy composed of built-in types
ff = dibjectify(ff)
if ff is None or not 'feed' in ff:
raise InvalidFeedError()
#
# perform some cleanup...
#
source_url = canonical_url(feed_url)
# make sure the feed has an id...
if not 'id' in ff.feed:
ff.feed['id'] = source_url.lower()
# make sure the feed has a self referential link
has_self_ref = False
ff.feed.setdefault('links', [])
for link in ff.feed.links:
if link.rel == 'self':
has_self_ref = True
break
if not has_self_ref:
ff.feed.links.append(Dibject(rel='self', href=source_url, title=''))
# create a structure holding the appropriate source information
# from the feed. This will be copied into each entry.
source_info = Dibject()
for k in ['id', 'title', 'title_detail', 'link', 'links', 'icon']:
try:
source_info[k] = deepcopy(ff.feed[k])
except KeyError:
pass
out_entries = []
for e in ff.get('entries', []):
# make sure it has an id
eid = e.get('id', None)
if eid is None:
eid = find_best_entry_id(e)
if eid is None:
# throw this entry out, it has no
# id, title, summary or content
# that is recognizable...
continue
e['id'] = eid
# assign a guid based on the id given and the source url
e['melk_id'] = melk_id(eid, source_url.lower())
# build a 'source' entry for each entry which points
# back to this feed. if there is already a source
# specified in the entry, we move it aside to
# original_source.
if 'source' in e:
e['original_source'] = e.source
e.source = deepcopy(source_info)
out_entries.append(e)
ff['entries'] = out_entries
return ff
