def test_recent_helpful_votes(self):
"""Recent helpful votes are indexed properly."""
# Create a document and verify it doesn't show up in a
# query for recent_helpful_votes__gt=0.
r = revision(is_approved=True, save=True)
self.refresh()
eq_(
DocumentMappingType.search().filter(
document_recent_helpful_votes__gt=0).count(), 0)
# Add an unhelpful vote, it still shouldn't show up.
helpful_vote(revision=r, helpful=False, save=True)
r.document.save() # Votes don't trigger a reindex.
self.refresh()
eq_(
DocumentMappingType.search().filter(
document_recent_helpful_votes__gt=0).count(), 0)
# Add an helpful vote created 31 days ago, it still shouldn't show up.
created = datetime.now() - timedelta(days=31)
helpful_vote(revision=r, helpful=True, created=created, save=True)
r.document.save() # Votes don't trigger a reindex.
self.refresh()
eq_(
DocumentMappingType.search().filter(
document_recent_helpful_votes__gt=0).count(), 0)
# Add an helpful vote created 29 days ago, it should show up now.
created = datetime.now() - timedelta(days=29)
helpful_vote(revision=r, helpful=True, created=created, save=True)
r.document.save() # Votes don't trigger a reindex.
self.refresh()
eq_(
DocumentMappingType.search().filter(
document_recent_helpful_votes__gt=0).count(), 1)
def test_add_and_delete(self):
"""Adding a doc should add it to the search index; deleting should
delete it."""
doc = document(save=True)
revision(document=doc, is_approved=True, save=True)
self.refresh()
eq_(DocumentMappingType.search().count(), 1)
doc.delete()
self.refresh()
eq_(DocumentMappingType.search().count(), 0)
def test_wiki_no_revisions(self):
"""Don't index documents without approved revisions"""
# Create a document with no revisions and make sure the
# document is not in the index.
doc = document(save=True)
self.refresh()
eq_(DocumentMappingType.search().count(), 0)
# Create a revision that's not approved and make sure the
# document is still not in the index.
revision(document=doc, is_approved=False, save=True)
self.refresh()
eq_(DocumentMappingType.search().count(), 0)
def test_wiki_keywords(self):
"""Make sure updating keywords updates the index."""
# Create a document with a revision with no keywords. It
# shouldn't show up with a document_keywords term query for
# 'wool' since it has no keywords.
doc = document(title=u'wool hats')
doc.save()
revision(document=doc, is_approved=True, save=True)
self.refresh()
eq_(
DocumentMappingType.search().query(
document_keywords='wool').count(), 0)
revision(document=doc, is_approved=True, keywords='wool', save=True)
self.refresh()
eq_(
DocumentMappingType.search().query(
document_keywords='wool').count(), 1)
def suggestions(request):
"""A simple search view that returns OpenSearch suggestions."""
mimetype = 'application/x-suggestions+json'
term = request.GET.get('q')
if not term:
return HttpResponseBadRequest(mimetype=mimetype)
site = Site.objects.get_current()
locale = locale_or_default(request.LANGUAGE_CODE)
try:
query = dict(('%s__text' % field, term)
for field in DocumentMappingType.get_query_fields())
wiki_s = (DocumentMappingType.search()
.filter(document_is_archived=False)
.filter(document_locale=locale)
.values_dict('document_title', 'url')
.query(or_=query)[:5])
query = dict(('%s__text' % field, term)
for field in QuestionMappingType.get_query_fields())
question_s = (QuestionMappingType.search()
.filter(question_has_helpful=True)
.values_dict('question_title', 'url')
.query(or_=query)[:5])
results = list(chain(question_s, wiki_s))
except ES_EXCEPTIONS:
# If we have ES problems, we just send back an empty result
# set.
results = []
urlize = lambda r: u'https://%s%s' % (site, r['url'])
titleize = lambda r: (r['document_title'] if 'document_title' in r
else r['question_title'])
data = [term,
[titleize(r) for r in results],
[],
[urlize(r) for r in results]]
return HttpResponse(json.dumps(data), mimetype=mimetype)
def test_wiki_products(self):
"""Make sure that adding products to a Document causes it to
refresh the index.
"""
p = product(slug=u'desktop', save=True)
eq_(DocumentMappingType.search().filter(product=p.slug).count(), 0)
doc = document(save=True)
revision(document=doc, is_approved=True, save=True)
self.refresh()
eq_(DocumentMappingType.search().filter(product=p.slug).count(), 0)
doc.products.add(p)
self.refresh()
eq_(DocumentMappingType.search().filter(product=p.slug).count(), 1)
doc.products.remove(p)
self.refresh()
# Make sure the document itself is still there and that we didn't
# accidentally delete it through screwed up signal handling:
eq_(DocumentMappingType.search().filter().count(), 1)
eq_(DocumentMappingType.search().filter(product=p.slug).count(), 0)
def test_wiki_topics(self):
"""Make sure that adding topics to a Document causes it to
refresh the index.
"""
t = topic(slug=u'hiphop', save=True)
eq_(DocumentMappingType.search().filter(topic=t.slug).count(), 0)
doc = document(save=True)
revision(document=doc, is_approved=True, save=True)
self.refresh()
eq_(DocumentMappingType.search().filter(topic=t.slug).count(), 0)
doc.topics.add(t)
self.refresh()
eq_(DocumentMappingType.search().filter(topic=t.slug).count(), 1)
doc.topics.clear()
self.refresh()
# Make sure the document itself is still there and that we didn't
# accidentally delete it through screwed up signal handling:
eq_(DocumentMappingType.search().filter().count(), 1)
eq_(DocumentMappingType.search().filter(topic=t.slug).count(), 0)
def test_translations_get_parent_tags(self):
doc1 = document(title=u'Audio too loud')
doc1.save()
revision(document=doc1, is_approved=True, save=True)
doc1.topics.add(topic(slug='cookies', save=True))
doc1.topics.add(topic(slug='general', save=True))
doc1.products.add(product(slug='desktop', save=True))
doc2 = document(title=u'Audio too loud bork bork', parent=doc1)
doc2.save()
revision(document=doc2, is_approved=True, save=True)
doc2.tags.add(u'badtag')
# Verify the parent has the right tags.
doc_dict = DocumentMappingType.extract_document(doc1.id)
eq_(doc_dict['topic'], [u'cookies', u'general'])
eq_(doc_dict['product'], [u'desktop'])
# Verify the translation has the parent's tags.
doc_dict = DocumentMappingType.extract_document(doc2.id)
eq_(doc_dict['topic'], [u'cookies', u'general'])
eq_(doc_dict['product'], [u'desktop'])
def test_wiki_redirects(self):
"""Make sure we don't index redirects"""
# First create a revision that doesn't have a redirect and
# make sure it's in the index.
doc = document(title=u'wool hats')
doc.save()
revision(document=doc, is_approved=True, save=True)
self.refresh()
eq_(
DocumentMappingType.search().query(
document_title__text='wool').count(), 1)
# Now create a revision that is a redirect and make sure the
# document is removed from the index.
revision(document=doc,
content=REDIRECT_CONTENT,
is_approved=True,
save=True)
self.refresh()
eq_(
DocumentMappingType.search().query(
document_title__text='wool').count(), 0)
def _es_documents_for(locale, topics=None, products=None):
"""ES implementation of documents_for."""
s = (DocumentMappingType.search()
.values_dict('id', 'document_title', 'url', 'document_parent_id',
'document_summary')
.filter(document_locale=locale, document_is_archived=False,
document_category__in=settings.IA_DEFAULT_CATEGORIES))
for topic in topics or []:
s = s.filter(topic=topic.slug)
for product in products or []:
s = s.filter(product=product.slug)
return list(s.order_by('-document_recent_helpful_votes')[:100])
def troubleshooting_view(request):
# Build a list of the most recently indexed 50 wiki documents.
last_50_indexed = list(_fix_value_dicts(DocumentMappingType.search()
.values_dict()
.order_by('-indexed_on')[:50]))
last_50_reviewed = list(Document.uncached
.filter(current_revision__is_approved=True)
.order_by('-current_revision__reviewed')[:50])
diff_list = diff_it_for_realz(last_50_indexed, last_50_reviewed)
return render(
request,
'admin/search_troubleshooting.html',
{'title': 'Index Troubleshooting',
'diffs': diff_list,
})
def search(request, template=None):
"""ES-specific search view"""
# JSON-specific variables
is_json = (request.GET.get('format') == 'json')
callback = request.GET.get('callback', '').strip()
mimetype = 'application/x-javascript' if callback else 'application/json'
# Search "Expires" header format
expires_fmt = '%A, %d %B %Y %H:%M:%S GMT'
# Check callback is valid
if is_json and callback and not jsonp_is_valid(callback):
return HttpResponse(
json.dumps({'error': _('Invalid callback function.')}),
mimetype=mimetype, status=400)
language = locale_or_default(
request.GET.get('language', request.LANGUAGE_CODE))
r = request.GET.copy()
a = request.GET.get('a', '0')
# Search default values
try:
category = (map(int, r.getlist('category')) or
settings.SEARCH_DEFAULT_CATEGORIES)
except ValueError:
category = settings.SEARCH_DEFAULT_CATEGORIES
r.setlist('category', category)
# Basic form
if a == '0':
r['w'] = r.get('w', constants.WHERE_BASIC)
# Advanced form
if a == '2':
r['language'] = language
r['a'] = '1'
# TODO: Rewrite so SearchForm is unbound initially and we can use
# `initial` on the form fields.
if 'include_archived' not in r:
r['include_archived'] = False
search_form = SearchForm(r)
if not search_form.is_valid() or a == '2':
if is_json:
return HttpResponse(
json.dumps({'error': _('Invalid search data.')}),
mimetype=mimetype,
status=400)
t = template if request.MOBILE else 'search/form.html'
search_ = render(request, t, {
'advanced': a, 'request': request,
'search_form': search_form})
search_['Cache-Control'] = 'max-age=%s' % \
(settings.SEARCH_CACHE_PERIOD * 60)
search_['Expires'] = (datetime.utcnow() +
timedelta(
minutes=settings.SEARCH_CACHE_PERIOD)) \
.strftime(expires_fmt)
return search_
cleaned = search_form.cleaned_data
if request.MOBILE and cleaned['w'] == constants.WHERE_BASIC:
cleaned['w'] = constants.WHERE_WIKI
page = max(smart_int(request.GET.get('page')), 1)
offset = (page - 1) * settings.SEARCH_RESULTS_PER_PAGE
lang = language.lower()
if settings.LANGUAGES.get(lang):
lang_name = settings.LANGUAGES[lang]
else:
lang_name = ''
# We use a regular S here because we want to search across
# multiple doctypes.
searcher = (UntypedS().es(urls=settings.ES_URLS)
.indexes(es_utils.READ_INDEX))
wiki_f = F(model='wiki_document')
question_f = F(model='questions_question')
discussion_f = F(model='forums_thread')
# Start - wiki filters
if cleaned['w'] & constants.WHERE_WIKI:
# Category filter
if cleaned['category']:
wiki_f &= F(document_category__in=cleaned['category'])
# Locale filter
wiki_f &= F(document_locale=language)
# Product filter
products = cleaned['product']
for p in products:
wiki_f &= F(product=p)
# Topics filter
topics = cleaned['topics']
for t in topics:
wiki_f &= F(topic=t)
# Archived bit
if a == '0' and not cleaned['include_archived']:
# Default to NO for basic search:
cleaned['include_archived'] = False
if not cleaned['include_archived']:
wiki_f &= F(document_is_archived=False)
# End - wiki filters
# Start - support questions filters
if cleaned['w'] & constants.WHERE_SUPPORT:
# Solved is set by default if using basic search
if a == '0' and not cleaned['has_helpful']:
cleaned['has_helpful'] = constants.TERNARY_YES
# These filters are ternary, they can be either YES, NO, or OFF
ternary_filters = ('is_locked', 'is_solved', 'has_answers',
'has_helpful')
d = dict(('question_%s' % filter_name,
_ternary_filter(cleaned[filter_name]))
for filter_name in ternary_filters if cleaned[filter_name])
if d:
question_f &= F(**d)
if cleaned['asked_by']:
question_f &= F(question_creator=cleaned['asked_by'])
if cleaned['answered_by']:
question_f &= F(question_answer_creator=cleaned['answered_by'])
q_tags = [t.strip() for t in cleaned['q_tags'].split(',')]
for t in q_tags:
if t:
question_f &= F(question_tag=t)
# Product filter
products = cleaned['product']
for p in products:
question_f &= F(product=p)
# Topics filter
topics = cleaned['topics']
for t in topics:
question_f &= F(topic=t)
# End - support questions filters
# Start - discussion forum filters
if cleaned['w'] & constants.WHERE_DISCUSSION:
if cleaned['author']:
discussion_f &= F(post_author_ord=cleaned['author'])
if cleaned['thread_type']:
if constants.DISCUSSION_STICKY in cleaned['thread_type']:
discussion_f &= F(post_is_sticky=1)
if constants.DISCUSSION_LOCKED in cleaned['thread_type']:
discussion_f &= F(post_is_locked=1)
if cleaned['forum']:
discussion_f &= F(post_forum_id__in=cleaned['forum'])
# End - discussion forum filters
# Created filter
unix_now = int(time.time())
interval_filters = (
('created', cleaned['created'], cleaned['created_date']),
('updated', cleaned['updated'], cleaned['updated_date']))
for filter_name, filter_option, filter_date in interval_filters:
if filter_option == constants.INTERVAL_BEFORE:
before = {filter_name + '__gte': 0,
filter_name + '__lte': max(filter_date, 0)}
discussion_f &= F(**before)
question_f &= F(**before)
elif filter_option == constants.INTERVAL_AFTER:
after = {filter_name + '__gte': min(filter_date, unix_now),
filter_name + '__lte': unix_now}
discussion_f &= F(**after)
question_f &= F(**after)
# In basic search, we limit questions from the last
# SEARCH_DEFAULT_MAX_QUESTION_AGE seconds.
if a == '0':
start_date = unix_now - settings.SEARCH_DEFAULT_MAX_QUESTION_AGE
question_f &= F(created__gte=start_date)
# Note: num_voted (with a d) is a different field than num_votes
# (with an s). The former is a dropdown and the latter is an
# integer value.
if cleaned['num_voted'] == constants.INTERVAL_BEFORE:
question_f &= F(question_num_votes__lte=max(cleaned['num_votes'], 0))
elif cleaned['num_voted'] == constants.INTERVAL_AFTER:
question_f &= F(question_num_votes__gte=cleaned['num_votes'])
# Done with all the filtery stuff--time to generate results
# Combine all the filters and add to the searcher
doctypes = []
final_filter = F()
if cleaned['w'] & constants.WHERE_WIKI:
doctypes.append(DocumentMappingType.get_mapping_type_name())
final_filter |= wiki_f
if cleaned['w'] & constants.WHERE_SUPPORT:
doctypes.append(QuestionMappingType.get_mapping_type_name())
final_filter |= question_f
if cleaned['w'] & constants.WHERE_DISCUSSION:
doctypes.append(ThreadMappingType.get_mapping_type_name())
final_filter |= discussion_f
searcher = searcher.doctypes(*doctypes)
searcher = searcher.filter(final_filter)
if 'explain' in request.GET and request.GET['explain'] == '1':
searcher = searcher.explain()
documents = ComposedList()
try:
cleaned_q = cleaned['q']
# Set up the highlights
# First 500 characters of content in one big fragment
searcher = searcher.highlight(
'question_content', 'discussion_content', 'document_summary',
pre_tags=['<b>'],
post_tags=['</b>'],
number_of_fragments=0,
fragment_size=500)
# Set up boosts
searcher = searcher.boost(
question_title=4.0,
question_content=3.0,
question_answer_content=3.0,
post_title=2.0,
post_content=1.0,
document_title=6.0,
document_content=1.0,
document_keywords=8.0,
document_summary=2.0,
# Text phrases in document titles and content get an extra
# boost.
document_title__text_phrase=10.0,
document_content__text_phrase=8.0)
# Apply sortby for advanced search of questions
if cleaned['w'] == constants.WHERE_SUPPORT:
sortby = cleaned['sortby']
try:
searcher = searcher.order_by(
*constants.SORT_QUESTIONS[sortby])
except IndexError:
# Skip index errors because they imply the user is
# sending us sortby values that aren't valid.
pass
# Apply sortby for advanced search of kb documents
if cleaned['w'] == constants.WHERE_WIKI:
sortby = cleaned['sortby_documents']
try:
searcher = searcher.order_by(
*constants.SORT_DOCUMENTS[sortby])
except IndexError:
# Skip index errors because they imply the user is
# sending us sortby values that aren't valid.
pass
# Build the query
if cleaned_q:
query_fields = chain(*[cls.get_query_fields()
for cls in get_mapping_types()])
query = {}
# Create text and text_phrase queries for every field
# we want to search.
for field in query_fields:
for query_type in ['text', 'text_phrase']:
query['%s__%s' % (field, query_type)] = cleaned_q
searcher = searcher.query(should=True, **query)
num_results = min(searcher.count(), settings.SEARCH_MAX_RESULTS)
# TODO - Can ditch the ComposedList here, but we need
# something that paginate can use to figure out the paging.
documents = ComposedList()
documents.set_count(('results', searcher), num_results)
results_per_page = settings.SEARCH_RESULTS_PER_PAGE
pages = paginate(request, documents, results_per_page)
# Facets
product_facets = {}
# If we know there aren't any results, let's cheat and in
# doing that, not hit ES again.
if num_results == 0:
searcher = []
else:
# Get the documents we want to show and add them to
# docs_for_page
documents = documents[offset:offset + results_per_page]
if len(documents) == 0:
# If the user requested a page that's beyond the
# pagination, then documents is an empty list and
# there are no results to show.
searcher = []
else:
bounds = documents[0][1]
searcher = searcher.values_dict()[bounds[0]:bounds[1]]
# If we are doing basic search, we show product facets.
if a == '0':
pfc = searcher.facet(
'product', filtered=True).facet_counts()
product_facets = dict(
[(p['term'], p['count']) for p in pfc['product']])
results = []
for i, doc in enumerate(searcher):
rank = i + offset
if doc['model'] == 'wiki_document':
summary = _build_es_excerpt(doc)
if not summary:
summary = doc['document_summary']
result = {
'title': doc['document_title'],
'type': 'document'}
elif doc['model'] == 'questions_question':
summary = _build_es_excerpt(doc)
if not summary:
# We're excerpting only question_content, so if
# the query matched question_title or
# question_answer_content, then there won't be any
# question_content excerpts. In that case, just
# show the question--but only the first 500
# characters.
summary = bleach.clean(
doc['question_content'], strip=True)[:500]
result = {
'title': doc['question_title'],
'type': 'question',
'is_solved': doc['question_is_solved'],
'num_answers': doc['question_num_answers'],
'num_votes': doc['question_num_votes'],
'num_votes_past_week': doc['question_num_votes_past_week']}
else:
summary = _build_es_excerpt(doc)
result = {
'title': doc['post_title'],
'type': 'thread'}
result['url'] = doc['url']
result['object'] = ObjectDict(doc)
result['search_summary'] = summary
result['rank'] = rank
result['score'] = doc._score
result['explanation'] = escape(format_explanation(
doc._explanation))
results.append(result)
except ES_EXCEPTIONS as exc:
# Handle timeout and all those other transient errors with a
# "Search Unavailable" rather than a Django error page.
if is_json:
return HttpResponse(json.dumps({'error':
_('Search Unavailable')}),
mimetype=mimetype, status=503)
# Cheating here: Convert from 'Timeout()' to 'timeout' so
# we have less code, but still have good stats.
exc_bucket = repr(exc).lower().strip('()')
statsd.incr('search.esunified.{0}'.format(exc_bucket))
import logging
logging.exception(exc)
t = 'search/mobile/down.html' if request.MOBILE else 'search/down.html'
return render(request, t, {'q': cleaned['q']}, status=503)
items = [(k, v) for k in search_form.fields for
v in r.getlist(k) if v and k != 'a']
items.append(('a', '2'))
if is_json:
# Models are not json serializable.
for r in results:
del r['object']
data = {}
data['results'] = results
data['total'] = len(results)
data['query'] = cleaned['q']
if not results:
data['message'] = _('No pages matched the search criteria')
json_data = json.dumps(data)
if callback:
json_data = callback + '(' + json_data + ');'
return HttpResponse(json_data, mimetype=mimetype)
fallback_results = None
if num_results == 0:
fallback_results = _fallback_results(language, cleaned['product'])
results_ = render(request, template, {
'num_results': num_results,
'results': results,
'fallback_results': fallback_results,
'q': cleaned['q'],
'w': cleaned['w'],
'product': cleaned['product'],
'products': Product.objects.filter(visible=True),
'product_facets': product_facets,
'pages': pages,
'search_form': search_form,
'lang_name': lang_name, })
results_['Cache-Control'] = 'max-age=%s' % \
(settings.SEARCH_CACHE_PERIOD * 60)
results_['Expires'] = (datetime.utcnow() +
timedelta(minutes=settings.SEARCH_CACHE_PERIOD)) \
.strftime(expires_fmt)
results_.set_cookie(settings.LAST_SEARCH_COOKIE, urlquote(cleaned['q']),
max_age=3600, secure=False, httponly=False)
return results_
def reindex_kb():
"""Reindex wiki_document."""
index_task.delay(DocumentMappingType, DocumentMappingType.get_indexable())
