Exemple #1
0
    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)
Exemple #2
0
    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)
Exemple #3
0
    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)
Exemple #4
0
    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)
Exemple #5
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)
Exemple #6
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)
Exemple #7
0
def opensearch_suggestions(request):
    """A simple search view that returns OpenSearch suggestions."""
    content_type = 'application/x-suggestions+json'

    term = request.GET.get('q')
    if not term:
        return HttpResponseBadRequest(content_type=content_type)

    locale = locale_or_default(request.LANGUAGE_CODE)

    # FIXME: Rewrite this using the simple search search business
    # logic. This currently returns templates (amongst other things)
    # which is totally wrong.
    try:
        query = dict(('%s__match' % field, term)
                     for field in DocumentMappingType.get_query_fields())
        # Upgrade the query to an analyzer-aware one.
        query = es_utils.es_query_with_analyzer(query, locale)

        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__match' % 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 = []

    def urlize(r):
        return u'%s://%s%s' % ('https' if request.is_secure() else 'http',
                               request.get_host(), r['url'][0])

    def titleize(r):
        # NB: Elasticsearch returns an array of strings as the value,
        # so we mimic that and then pull out the first (and only)
        # string.
        return r.get('document_title', r.get('question_title',
                                             [_('No title')]))[0]

    data = [
        term, [titleize(r) for r in results], [], [urlize(r) for r in results]
    ]
    return HttpResponse(json.dumps(data), content_type=content_type)
Exemple #8
0
def opensearch_suggestions(request):
    """A simple search view that returns OpenSearch suggestions."""
    content_type = "application/x-suggestions+json"

    term = request.GET.get("q")
    if not term:
        return HttpResponseBadRequest(content_type=content_type)

    locale = locale_or_default(request.LANGUAGE_CODE)

    # FIXME: Rewrite this using the simple search search business
    # logic. This currently returns templates (amongst other things)
    # which is totally wrong.
    try:
        query = dict(("%s__match" % field, term) for field in DocumentMappingType.get_query_fields())
        # Upgrade the query to an analyzer-aware one.
        query = es_utils.es_query_with_analyzer(query, locale)

        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__match" % 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 = []

    def urlize(r):
        return u"%s://%s%s" % ("https" if request.is_secure() else "http", request.get_host(), r["url"][0])

    def titleize(r):
        # NB: Elasticsearch returns an array of strings as the value,
        # so we mimic that and then pull out the first (and only)
        # string.
        return r.get("document_title", r.get("question_title", [_("No title")]))[0]

    data = [term, [titleize(r) for r in results], [], [urlize(r) for r in results]]
    return HttpResponse(json.dumps(data), content_type=content_type)
Exemple #9
0
    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 = DocumentFactory(title=u'wool hats')
        RevisionFactory(document=doc, is_approved=True)
        self.refresh()
        eq_(DocumentMappingType.search().query(document_title__match='wool').count(), 1)

        # Now create a revision that is a redirect and make sure the
        # document is removed from the index.
        RedirectRevisionFactory(document=doc)
        self.refresh()
        eq_(DocumentMappingType.search().query(document_title__match='wool').count(), 0)
Exemple #10
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 = DocumentFactory(title=u'wool hats')
        RevisionFactory(document=doc, is_approved=True)
        self.refresh()
        eq_(DocumentMappingType.search().query(
            document_keywords='wool').count(), 0)

        RevisionFactory(document=doc, is_approved=True, keywords='wool')
        self.refresh()

        eq_(DocumentMappingType.search().query(document_keywords='wool').count(), 1)
Exemple #11
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)

    results = s.order_by('document_display_order', '-document_recent_helpful_votes')[:100]
    results = DocumentMappingType.reshape(results)
    return results
Exemple #12
0
def suggestions(request):
    """A simple search view that returns OpenSearch suggestions."""
    content_type = 'application/x-suggestions+json'

    term = request.GET.get('q')
    if not term:
        return HttpResponseBadRequest(content_type=content_type)

    site = Site.objects.get_current()
    locale = locale_or_default(request.LANGUAGE_CODE)
    try:
        query = dict(('{0!s}__match'.format(field), term)
                     for field in DocumentMappingType.get_query_fields())
        # Upgrade the query to an analyzer-aware one.
        query = es_utils.es_query_with_analyzer(query, locale)

        wiki_s = (DocumentMappingType.search()
                  .filter(document_is_archived=False)
                  .filter(document_locale=locale)
                  .values_dict('document_title', 'url')
                  .query(or_=query)[:5])

        query = dict(('{0!s}__match'.format(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 = []

    def urlize(r):
        return u'https://{0!s}{1!s}'.format(site, r['url'])

    def titleize(r):
        return r.get('document_title', r.get('document_title'))

    data = [term,
            [titleize(r) for r in results],
            [],
            [urlize(r) for r in results]]
    return HttpResponse(json.dumps(data), content_type=content_type)
Exemple #13
0
def suggestions(request):
    """A simple search view that returns OpenSearch suggestions."""
    content_type = 'application/x-suggestions+json'

    term = request.GET.get('q')
    if not term:
        return HttpResponseBadRequest(content_type=content_type)

    site = Site.objects.get_current()
    locale = locale_or_default(request.LANGUAGE_CODE)
    try:
        query = dict(('%s__match' % field, term)
                     for field in DocumentMappingType.get_query_fields())
        # Upgrade the query to an analyzer-aware one.
        query = es_utils.es_query_with_analyzer(query, locale)

        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__match' % 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 = []

    def urlize(r):
        return u'https://%s%s' % (site, r['url'])

    def titleize(r):
        return r.get('document_title', r.get('document_title'))

    data = [term,
            [titleize(r) for r in results],
            [],
            [urlize(r) for r in results]]
    return HttpResponse(json.dumps(data), content_type=content_type)
Exemple #14
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)

    results = s.order_by('-document_recent_helpful_votes')[:100]
    results = DocumentMappingType.reshape(results)
    return results
Exemple #15
0
    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 = DocumentFactory(title='wool hats')
        RevisionFactory(document=doc, is_approved=True)
        self.refresh()
        eq_(
            DocumentMappingType.search().query(
                document_title__match='wool').count(), 1)

        # Now create a revision that is a redirect and make sure the
        # document is removed from the index.
        RedirectRevisionFactory(document=doc)
        self.refresh()
        eq_(
            DocumentMappingType.search().query(
                document_title__match='wool').count(), 0)
Exemple #16
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 = DocumentFactory(title='wool hats')
        RevisionFactory(document=doc, is_approved=True)
        self.refresh()
        eq_(
            DocumentMappingType.search().query(
                document_keywords='wool').count(), 0)

        RevisionFactory(document=doc, is_approved=True, keywords='wool')
        self.refresh()

        eq_(
            DocumentMappingType.search().query(
                document_keywords='wool').count(), 1)
Exemple #17
0
    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__match='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__match='wool').count(), 0)
Exemple #18
0
def troubleshooting_view(request):
    # Build a list of the most recently indexed 50 wiki documents.
    last_50_indexed = list(_fix_results(DocumentMappingType.search().order_by("-indexed_on")[:50]))

    last_50_reviewed = list(
        Document.objects.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})
Exemple #19
0
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)
Exemple #20
0
    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__match='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__match='wool').count(), 0)
Exemple #21
0
    def test_translations_get_parent_tags(self):
        t1 = TopicFactory(display_order=1)
        t2 = TopicFactory(display_order=2)
        p = ProductFactory()
        doc1 = DocumentFactory(
            title=u'Audio too loud',
            products=[p],
            topics=[t1, t2])
        RevisionFactory(document=doc1, is_approved=True)

        doc2 = DocumentFactory(title=u'Audio too loud bork bork', parent=doc1, tags=[u'badtag'])
        RevisionFactory(document=doc2, is_approved=True)

        # Verify the parent has the right tags.
        doc_dict = DocumentMappingType.extract_document(doc1.id)
        eq_(sorted(doc_dict['topic']), sorted([t1.slug, t2.slug]))
        eq_(doc_dict['product'], [p.slug])

        # Verify the translation has the parent's tags.
        doc_dict = DocumentMappingType.extract_document(doc2.id)
        eq_(sorted(doc_dict['topic']), sorted([t1.slug, t2.slug]))
        eq_(doc_dict['product'], [p.slug])
Exemple #22
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'])
Exemple #23
0
    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)
Exemple #24
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)
Exemple #25
0
    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)
Exemple #26
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)
Exemple #27
0
    def test_translations_get_parent_tags(self):
        t1 = TopicFactory(display_order=1)
        t2 = TopicFactory(display_order=2)
        p = ProductFactory()
        doc1 = DocumentFactory(title='Audio too loud',
                               products=[p],
                               topics=[t1, t2])
        RevisionFactory(document=doc1, is_approved=True)

        doc2 = DocumentFactory(title='Audio too loud bork bork',
                               parent=doc1,
                               tags=['badtag'])
        RevisionFactory(document=doc2, is_approved=True)

        # Verify the parent has the right tags.
        doc_dict = DocumentMappingType.extract_document(doc1.id)
        eq_(sorted(doc_dict['topic']), sorted([t1.slug, t2.slug]))
        eq_(doc_dict['product'], [p.slug])

        # Verify the translation has the parent's tags.
        doc_dict = DocumentMappingType.extract_document(doc2.id)
        eq_(sorted(doc_dict['topic']), sorted([t1.slug, t2.slug]))
        eq_(doc_dict['product'], [p.slug])
Exemple #28
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'])
Exemple #29
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])
Exemple #30
0
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,
    })
Exemple #31
0
def troubleshooting_view(request):
    # Build a list of the most recently indexed 50 wiki documents.
    last_50_indexed = list(_fix_results(DocumentMappingType.search()
                                        .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,
         })
Exemple #32
0
 def handle(self, **options):
     index_task.delay(to_class_path(DocumentMappingType),
                      DocumentMappingType.get_indexable())
Exemple #33
0
def reindex_kb():
    """Reindex wiki_document."""
    index_task.delay(DocumentMappingType, DocumentMappingType.get_indexable())
Exemple #34
0
def reindex_kb():
    """Reindex wiki_document."""
    index_task.delay(DocumentMappingType, DocumentMappingType.get_indexable())
Exemple #35
0
def advanced_search(request, template=None):
    """ES-specific Advanced search view"""

    # JSON-specific variables
    is_json = request.GET.get("format") == "json"
    callback = request.GET.get("callback", "").strip()
    content_type = "application/x-javascript" if callback else "application/json"

    # Check callback is valid
    if is_json and callback and not jsonp_is_valid(callback):
        return HttpResponse(
            json.dumps({"error": _("Invalid callback function.")}), content_type=content_type, status=400
        )

    language = locale_or_default(request.GET.get("language", request.LANGUAGE_CODE))
    r = request.GET.copy()
    # TODO: Figure out how to get rid of 'a' and do it.
    # It basically is used to switch between showing the form or results.
    a = request.GET.get("a", "2")
    # TODO: This is so the 'a=1' stays in the URL for pagination.
    r["a"] = 1

    # 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)

    r["language"] = language

    search_form = AdvancedSearchForm(r, auto_id=False)
    search_form.set_allowed_forums(request.user)

    # This is all we use a for now I think.
    if not search_form.is_valid() or a == "2":
        if is_json:
            return HttpResponse(json.dumps({"error": _("Invalid search data.")}), content_type=content_type, status=400)

        t = template if request.MOBILE else "search/form.html"
        search_ = render(request, t, {"advanced": True, "request": request, "search_form": search_form})
        cache_period = settings.SEARCH_CACHE_PERIOD
        search_["Cache-Control"] = "max-age=%s" % (cache_period * 60)
        search_["Expires"] = (datetime.utcnow() + timedelta(minutes=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_DICT.get(lang):
        lang_name = settings.LANGUAGES_DICT[lang]
    else:
        lang_name = ""

    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = AnalyzerS().es(urls=settings.ES_URLS).indexes(es_utils.read_index("default"))

    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 not cleaned["include_archived"]:
            wiki_f &= F(document_is_archived=False)

    # End - wiki filters

    # Start - support questions filters

    if cleaned["w"] & constants.WHERE_SUPPORT:
        # These filters are ternary, they can be either YES, NO, or OFF
        ternary_filters = ("is_locked", "is_solved", "has_answers", "has_helpful", "is_archived")
        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)

        valid_forum_ids = [f.id for f in Forum.authorized_forums_for_user(request.user)]

        forum_ids = None
        if cleaned["forum"]:
            forum_ids = [f for f in cleaned["forum"] if f in valid_forum_ids]

        # If we removed all the forums they wanted to look at or if
        # they didn't specify, then we filter on the list of all
        # forums they're authorized to look at.
        if not forum_ids:
            forum_ids = valid_forum_ids

        discussion_f &= F(post_forum_id__in=forum_ids)

    # 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)

    # 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. Show the entire field highlighted.
        searcher = searcher.highlight(
            "question_content",  # support forum
            "document_summary",  # kb
            "post_content",  # contributor forum
            pre_tags=["<b>"],
            post_tags=["</b>"],
            number_of_fragments=0,
        )

        # 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__match_phrase=10.0,
            document_content__match_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 [DocumentMappingType, ThreadMappingType, QuestionMappingType]]
            )
            query = {}
            # Create a simple_query_search query for every field
            # we want to search.
            for field in query_fields:
                query["%s__sqs" % field] = cleaned_q

            # Transform the query to use locale aware analyzers.
            query = es_utils.es_query_with_analyzer(query, language)

            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)

        # 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[bounds[0] : bounds[1]]

        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, first_only=True)
                result = {"title": doc["post_title"], "type": "thread"}

            result["url"] = doc["url"]
            result["object"] = doc
            result["search_summary"] = summary
            result["rank"] = rank
            result["score"] = doc.es_meta.score
            result["explanation"] = escape(format_explanation(doc.es_meta.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")}), content_type=content_type, 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))

        log.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"))

    fallback_results = None
    if num_results == 0:
        fallback_results = _fallback_results(language, cleaned["product"])

    product = Product.objects.filter(slug__in=cleaned["product"])
    if product:
        product_titles = [_(p.title, "DB: products.Product.title") for p in product]
    else:
        product_titles = [_("All Products")]

    product_titles = ", ".join(product_titles)

    data = {
        "num_results": num_results,
        "results": results,
        "fallback_results": fallback_results,
        "product_titles": product_titles,
        "q": cleaned["q"],
        "w": cleaned["w"],
        "lang_name": lang_name,
        "advanced": True,
    }

    if is_json:
        # Models are not json serializable.
        for r in data["results"]:
            del r["object"]
        data["total"] = len(data["results"])

        data["products"] = [{"slug": p.slug, "title": p.title} for p in Product.objects.filter(visible=True)]

        if product:
            data["product"] = product[0].slug

        pages = Paginator(pages)
        data["pagination"] = dict(
            number=pages.pager.number,
            num_pages=pages.pager.paginator.num_pages,
            has_next=pages.pager.has_next(),
            has_previous=pages.pager.has_previous(),
            max=pages.max,
            span=pages.span,
            dotted_upper=pages.pager.dotted_upper,
            dotted_lower=pages.pager.dotted_lower,
            page_range=pages.pager.page_range,
            url=pages.pager.url,
        )
        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, content_type=content_type)

    data.update(
        {
            "product": product,
            "products": Product.objects.filter(visible=True),
            "pages": pages,
            "search_form": search_form,
        }
    )
    results_ = render(request, template, data)
    cache_period = settings.SEARCH_CACHE_PERIOD
    results_["Cache-Control"] = "max-age=%s" % (cache_period * 60)
    results_["Expires"] = (datetime.utcnow() + timedelta(minutes=cache_period)).strftime(EXPIRES_FMT)
    results_.set_cookie(settings.LAST_SEARCH_COOKIE, urlquote(cleaned["q"]), max_age=3600, secure=False, httponly=False)

    return results_
Exemple #36
0
def advanced_search(request, template=None):
    """Elasticsearch-specific Advanced search view"""

    to_json = JSONRenderer().render

    # 1. Prep request.
    r = request.GET.copy()
    # TODO: Figure out how to get rid of 'a' and do it.
    # It basically is used to switch between showing the form or results.
    a = request.GET.get('a', '2')
    # TODO: This is so the 'a=1' stays in the URL for pagination.
    r['a'] = 1

    language = locale_or_default(request.GET.get('language', request.LANGUAGE_CODE))
    r['language'] = language
    lang = language.lower()
    lang_name = settings.LANGUAGES_DICT.get(lang) or ''

    # 2. Build form.
    search_form = AdvancedSearchForm(r, auto_id=False)
    search_form.set_allowed_forums(request.user)

    # 3. Validate request.
    # Note: a == 2 means "show the form"--that's all we use it for now.
    if a == '2' or not search_form.is_valid():
        if request.IS_JSON:
            return HttpResponse(
                json.dumps({'error': _('Invalid search data.')}),
                content_type=request.CONTENT_TYPE,
                status=400)

        t = template if request.MOBILE else 'search/form.html'
        data = {'advanced': True,
                'request': request,
                'search_form': search_form}
        # get value for search input from last search term.
        last_search = request.COOKIES.get(settings.LAST_SEARCH_COOKIE)
        # If there is any cached input from last search, pass it to template
        if last_search and 'q' not in r:
            cached_field = urlquote(last_search)
            data.update({'cached_field': cached_field})

        return cache_control(
            render(request, t, data),
            settings.SEARCH_CACHE_PERIOD)

    # 4. Generate search.
    cleaned = search_form.cleaned_data

    # On mobile, we default to just wiki results.
    if request.MOBILE and cleaned['w'] == constants.WHERE_BASIC:
        cleaned['w'] = constants.WHERE_WIKI

    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = (AnalyzerS().es(urls=settings.ES_URLS)
                .indexes(es_utils.read_index('default')))

    doctypes = []
    final_filter = F()
    unix_now = int(time.time())
    interval_filters = (
        ('created', cleaned['created'], cleaned['created_date']),
        ('updated', cleaned['updated'], cleaned['updated_date'])
    )

    # Start - wiki search configuration

    if cleaned['w'] & constants.WHERE_WIKI:
        wiki_f = F(model='wiki_document')

        # 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 not cleaned['include_archived']:
            wiki_f &= F(document_is_archived=False)

        # Apply sortby
        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

        doctypes.append(DocumentMappingType.get_mapping_type_name())
        final_filter |= wiki_f

    # End - wiki search configuration

    # Start - support questions configuration

    if cleaned['w'] & constants.WHERE_SUPPORT:
        question_f = F(model='questions_question')

        # These filters are ternary, they can be either YES, NO, or OFF
        ternary_filters = ('is_locked', 'is_solved', 'has_answers',
                           'has_helpful', 'is_archived')
        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)

        # 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'])

        # Apply sortby
        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 created and updated filters
        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)}

                question_f &= F(**before)

            elif filter_option == constants.INTERVAL_AFTER:
                after = {filter_name + '__gte': min(filter_date, unix_now),
                         filter_name + '__lte': unix_now}

                question_f &= F(**after)

        doctypes.append(QuestionMappingType.get_mapping_type_name())
        final_filter |= question_f

    # End - support questions configuration

    # Start - discussion forum configuration

    if cleaned['w'] & constants.WHERE_DISCUSSION:
        discussion_f = F(model='forums_thread')

        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)

        valid_forum_ids = [f.id for f in Forum.authorized_forums_for_user(request.user)]

        forum_ids = None
        if cleaned['forum']:
            forum_ids = [f for f in cleaned['forum'] if f in valid_forum_ids]

        # If we removed all the forums they wanted to look at or if
        # they didn't specify, then we filter on the list of all
        # forums they're authorized to look at.
        if not forum_ids:
            forum_ids = valid_forum_ids

        discussion_f &= F(post_forum_id__in=forum_ids)

        # Apply created and updated filters
        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)

            elif filter_option == constants.INTERVAL_AFTER:
                after = {filter_name + '__gte': min(filter_date, unix_now),
                         filter_name + '__lte': unix_now}

                discussion_f &= F(**after)

        doctypes.append(ThreadMappingType.get_mapping_type_name())
        final_filter |= discussion_f

    # End - discussion forum configuration

    # Done with all the filtery stuff--time  to generate results

    searcher = searcher.doctypes(*doctypes)
    searcher = searcher.filter(final_filter)

    if 'explain' in request.GET and request.GET['explain'] == '1':
        searcher = searcher.explain()

    cleaned_q = cleaned['q']

    # Set up the highlights. Show the entire field highlighted.
    searcher = searcher.highlight(
        'question_content',  # support forum
        'document_summary',  # kb
        'post_content',  # contributor forum
        pre_tags=['<b>'],
        post_tags=['</b>'],
        number_of_fragments=0)

    searcher = apply_boosts(searcher)

    # Build the query
    if cleaned_q:
        query_fields = chain(*[
            cls.get_query_fields() for cls in [
                DocumentMappingType,
                ThreadMappingType,
                QuestionMappingType
            ]
        ])
        query = {}
        # Create a simple_query_search query for every field we want to search.
        for field in query_fields:
            query['%s__sqs' % field] = cleaned_q

        # Transform the query to use locale aware analyzers.
        query = es_utils.es_query_with_analyzer(query, language)

        searcher = searcher.query(should=True, **query)

    searcher = searcher[:settings.SEARCH_MAX_RESULTS]

    # 5. Generate output
    pages = paginate(request, searcher, settings.SEARCH_RESULTS_PER_PAGE)

    if pages.paginator.count == 0:
        # If we know there aren't any results, show fallback_results.
        fallback_results = _fallback_results(language, cleaned['product'])
        results = []
    else:
        fallback_results = None
        results = build_results_list(pages, request.IS_JSON)

    items = [(k, v) for k in search_form.fields for
             v in r.getlist(k) if v and k != 'a']
    items.append(('a', '2'))

    product = Product.objects.filter(slug__in=cleaned['product'])
    if product:
        product_titles = [pgettext('DB: products.Product.title', p.title) for p in product]
    else:
        product_titles = [_('All Products')]

    # FIXME: This is probably bad l10n.
    product_titles = ', '.join(product_titles)

    data = {
        'num_results': pages.paginator.count,
        'results': results,
        'fallback_results': fallback_results,
        'product_titles': product_titles,
        'q': cleaned['q'],
        'w': cleaned['w'],
        'lang_name': lang_name,
        'advanced': True,
        'products': Product.objects.filter(visible=True)
    }

    if request.IS_JSON:
        data['total'] = len(data['results'])
        data['products'] = [{'slug': p.slug, 'title': p.title}
                            for p in data['products']]

        if product:
            data['product'] = product[0].slug

        pages = Paginator(pages)
        data['pagination'] = dict(
            number=pages.pager.number,
            num_pages=pages.pager.paginator.num_pages,
            has_next=pages.pager.has_next(),
            has_previous=pages.pager.has_previous(),
            max=pages.max,
            span=pages.span,
            dotted_upper=pages.pager.dotted_upper,
            dotted_lower=pages.pager.dotted_lower,
            page_range=pages.pager.page_range,
            url=pages.pager.url,
        )
        if not results:
            data['message'] = _('No pages matched the search criteria')
        json_data = to_json(data)
        if request.JSON_CALLBACK:
            json_data = request.JSON_CALLBACK + '(' + json_data + ');'
        return HttpResponse(json_data, content_type=request.CONTENT_TYPE)

    data.update({
        'product': product,
        'pages': pages,
        'search_form': search_form
    })
    resp = cache_control(render(request, template, data), settings.SEARCH_CACHE_PERIOD)
    resp.set_cookie(settings.LAST_SEARCH_COOKIE, urlquote(cleaned['q']),
                    max_age=3600, secure=False, httponly=False)
    return resp
Exemple #37
0
 def get_query_fields(self):
     return DocumentMappingType.get_query_fields()
Exemple #38
0
def simple_search(request, template=None):
    """ES-specific simple search view.

    This view is for end user searching of the Knowledge Base and
    Support Forum. Filtering options are limited to:
    * product (`product=firefox`, for example, for only Firefox results)
    * document type (`w=2`, for esample, for Support Forum questions only)
    """

    # Redirect to old Advanced Search URLs (?a={1,2}) to the new URL.
    a = request.GET.get("a")
    if a in ["1", "2"]:
        new_url = reverse("search.advanced") + "?" + request.GET.urlencode()
        return HttpResponseRedirect(new_url)

    # JSON-specific variables
    is_json = request.GET.get("format") == "json"
    callback = request.GET.get("callback", "").strip()
    content_type = "application/x-javascript" if callback else "application/json"

    # Check callback is valid
    if is_json and callback and not jsonp_is_valid(callback):
        return HttpResponse(
            json.dumps({"error": _("Invalid callback function.")}), content_type=content_type, status=400
        )

    language = locale_or_default(request.GET.get("language", request.LANGUAGE_CODE))
    r = request.GET.copy()

    # TODO: Do we really need to add this to the URL if it isn't already there?
    r["w"] = r.get("w", constants.WHERE_BASIC)

    # TODO: Break out a separate simple search form.
    search_form = SimpleSearchForm(r, auto_id=False)

    if not search_form.is_valid():
        if is_json:
            return HttpResponse(json.dumps({"error": _("Invalid search data.")}), content_type=content_type, status=400)

        t = template if request.MOBILE else "search/form.html"
        search_ = render(request, t, {"advanced": False, "request": request, "search_form": search_form})
        cache_period = settings.SEARCH_CACHE_PERIOD
        search_["Cache-Control"] = "max-age=%s" % (cache_period * 60)
        search_["Expires"] = (datetime.utcnow() + timedelta(minutes=cache_period)).strftime(EXPIRES_FMT)
        return search_

    cleaned = search_form.cleaned_data

    # On mobile, we default to just wiki results.
    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_DICT.get(lang):
        lang_name = settings.LANGUAGES_DICT[lang]
    else:
        lang_name = ""

    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = AnalyzerS().es(urls=settings.ES_URLS).indexes(es_utils.read_index("default"))

    wiki_f = F(model="wiki_document")
    question_f = F(model="questions_question")

    cleaned_q = cleaned["q"]
    products = cleaned["product"]

    if not products and "all_products" not in request.GET:
        lowered_q = cleaned_q.lower()

        if "thunderbird" in lowered_q:
            products.append("thunderbird")
        elif "android" in lowered_q:
            products.append("mobile")
        elif "ios" in lowered_q or "ipad" in lowered_q or "ipod" in lowered_q or "iphone" in lowered_q:
            products.append("ios")
        elif "firefox os" in lowered_q:
            products.append("firefox-os")
        elif "firefox" in lowered_q:
            products.append("firefox")

    # Start - wiki filters

    if cleaned["w"] & constants.WHERE_WIKI:
        # Category filter
        wiki_f &= F(document_category__in=settings.SEARCH_DEFAULT_CATEGORIES)

        # Locale filter
        wiki_f &= F(document_locale=language)

        # Product filter
        for p in products:
            wiki_f &= F(product=p)

        # Archived bit
        wiki_f &= F(document_is_archived=False)

    # End - wiki filters

    # Start - support questions filters

    if cleaned["w"] & constants.WHERE_SUPPORT:
        # Has helpful answers is set by default if using basic search
        cleaned["has_helpful"] = constants.TERNARY_YES

        # No archived questions in default search.
        cleaned["is_archived"] = constants.TERNARY_NO

        # These filters are ternary, they can be either YES, NO, or OFF
        ternary_filters = ("has_helpful", "is_archived")
        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)

        # Product filter
        for p in products:
            question_f &= F(product=p)

    # End - support questions filters

    # 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

    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:
        # Set up the highlights. Show the entire field highlighted.
        searcher = searcher.highlight(
            "question_content",  # support forum
            "document_summary",  # kb
            pre_tags=["<b>"],
            post_tags=["</b>"],
            number_of_fragments=0,
        )

        # Set up boosts
        searcher = searcher.boost(
            question_title=4.0,
            question_content=3.0,
            question_answer_content=3.0,
            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__match_phrase=10.0,
            document_content__match_phrase=8.0,
        )

        # Build the query
        query_fields = chain(*[cls.get_query_fields() for cls in [DocumentMappingType, QuestionMappingType]])
        query = {}
        # Create match and match_phrase queries for every field
        # we want to search.
        for field in query_fields:
            for query_type in ["match", "match_phrase"]:
                query["%s__%s" % (field, query_type)] = cleaned_q

        # Transform the query to use locale aware analyzers.
        query = es_utils.es_query_with_analyzer(query, language)

        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)

        # 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[bounds[0] : bounds[1]]

        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"],
                }

            result["url"] = doc["url"]
            result["object"] = doc
            result["search_summary"] = summary
            result["rank"] = rank
            result["score"] = doc.es_meta.score
            result["explanation"] = escape(format_explanation(doc.es_meta.explanation))
            result["id"] = doc["id"]
            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")}), content_type=content_type, 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))

        log.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"))

    fallback_results = None
    if num_results == 0:
        fallback_results = _fallback_results(language, cleaned["product"])

    product = Product.objects.filter(slug__in=cleaned["product"])
    if product:
        product_titles = [_(p.title, "DB: products.Product.title") for p in product]
    else:
        product_titles = [_("All Products")]

    product_titles = ", ".join(product_titles)

    data = {
        "num_results": num_results,
        "results": results,
        "fallback_results": fallback_results,
        "product_titles": product_titles,
        "q": cleaned["q"],
        "w": cleaned["w"],
        "lang_name": lang_name,
    }

    if is_json:
        # Models are not json serializable.
        for r in data["results"]:
            del r["object"]
        data["total"] = len(data["results"])

        data["products"] = [{"slug": p.slug, "title": p.title} for p in Product.objects.filter(visible=True)]

        if product:
            data["product"] = product[0].slug

        pages = Paginator(pages)
        data["pagination"] = dict(
            number=pages.pager.number,
            num_pages=pages.pager.paginator.num_pages,
            has_next=pages.pager.has_next(),
            has_previous=pages.pager.has_previous(),
            max=pages.max,
            span=pages.span,
            dotted_upper=pages.pager.dotted_upper,
            dotted_lower=pages.pager.dotted_lower,
            page_range=pages.pager.page_range,
            url=pages.pager.url,
        )
        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, content_type=content_type)

    data.update(
        {
            "product": product,
            "products": Product.objects.filter(visible=True),
            "pages": pages,
            "search_form": search_form,
            "advanced": False,
        }
    )
    results_ = render(request, template, data)
    cache_period = settings.SEARCH_CACHE_PERIOD
    results_["Cache-Control"] = "max-age=%s" % (cache_period * 60)
    results_["Expires"] = (datetime.utcnow() + timedelta(minutes=cache_period)).strftime(EXPIRES_FMT)
    results_.set_cookie(settings.LAST_SEARCH_COOKIE, urlquote(cleaned["q"]), max_age=3600, secure=False, httponly=False)

    return results_
Exemple #39
0
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)
    search_form.set_allowed_forums(request.user)

    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 = (AnalyzerS().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)

        valid_forum_ids = [
            f.id for f in Forum.authorized_forums_for_user(request.user)]

        forum_ids = None
        if cleaned['forum']:
            forum_ids = [f for f in cleaned['forum'] if f in valid_forum_ids]

        # If we removed all the forums they wanted to look at or if
        # they didn't specify, then we filter on the list of all
        # forums they're authorized to look at.
        if not forum_ids:
            forum_ids = valid_forum_ids

        discussion_f &= F(post_forum_id__in=forum_ids)

    # 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. Show the entire field highlighted.
        searcher = searcher.highlight(
            'question_content',  # support forum
            'document_summary',  # kb
            'post_content',  # contributor forum
            pre_tags=['<b>'],
            post_tags=['</b>'],
            number_of_fragments=0)

        # 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

            # Transform the query to use locale aware analyzers.
            query = es_utils.es_query_with_analyzer(query, language)

            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)

        # 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]]

        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, first_only=True)
                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': Product.objects.filter(slug__in=cleaned['product']),
        'products': Product.objects.filter(visible=True),
        '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_
Exemple #40
0
 def get_doctypes(self):
     return [DocumentMappingType.get_mapping_type_name()]
Exemple #41
0
def generate_simple_search(search_form, language, with_highlights=False):
    """Generates an S given a form

    :arg search_form: a validated SimpleSearch form
    :arg language: the language code
    :arg with_highlights: whether or not to ask for highlights

    :returns: a fully formed S

    """
    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = (es_utils.AnalyzerS().es(
        urls=settings.ES_URLS,
        timeout=settings.ES_TIMEOUT,
        use_ssl=settings.ES_USE_SSL,
        http_auth=settings.ES_HTTP_AUTH,
        connection_class=RequestsHttpConnection,
    ).indexes(es_utils.read_index("default")))

    cleaned = search_form.cleaned_data

    doctypes = []
    final_filter = es_utils.F()
    cleaned_q = cleaned["q"]
    products = cleaned["product"]

    # Handle wiki filters
    if cleaned["w"] & constants.WHERE_WIKI:
        wiki_f = es_utils.F(
            model="wiki_document",
            document_category__in=settings.SEARCH_DEFAULT_CATEGORIES,
            document_locale=language,
            document_is_archived=False,
        )

        for p in products:
            wiki_f &= es_utils.F(product=p)

        doctypes.append(DocumentMappingType.get_mapping_type_name())
        final_filter |= wiki_f

    # Handle question filters
    if cleaned["w"] & constants.WHERE_SUPPORT:
        question_f = es_utils.F(model="questions_question",
                                question_is_archived=False,
                                question_has_helpful=True)

        for p in products:
            question_f &= es_utils.F(product=p)

        doctypes.append(QuestionMappingType.get_mapping_type_name())
        final_filter |= question_f

    # Build a filter for those filters and add the other bits to
    # finish the search
    searcher = searcher.doctypes(*doctypes)
    searcher = searcher.filter(final_filter)

    if cleaned["explain"]:
        searcher = searcher.explain()

    if with_highlights:
        # Set up the highlights. Show the entire field highlighted.
        searcher = searcher.highlight(
            "question_content",  # support forum
            "document_summary",  # kb
            pre_tags=["<b>"],
            post_tags=["</b>"],
            number_of_fragments=0,
        )

    searcher = apply_boosts(searcher)

    # Build the query
    query_fields = chain(*[
        cls.get_query_fields()
        for cls in [DocumentMappingType, QuestionMappingType]
    ])
    query = {}
    # Create match and match_phrase queries for every field
    # we want to search.
    for field in query_fields:
        for query_type in ["match", "match_phrase"]:
            query["%s__%s" % (field, query_type)] = cleaned_q

    # Transform the query to use locale aware analyzers.
    query = es_utils.es_query_with_analyzer(query, language)

    searcher = searcher.query(should=True, **query)
    return searcher
Exemple #42
0
 def get_query_fields(self):
     return DocumentMappingType.get_query_fields()
Exemple #43
0
 def handle(self, **options):
     index_task.delay(DocumentMappingType,
                      DocumentMappingType.get_indexable())
Exemple #44
0
 def get_doctypes(self):
     return [DocumentMappingType.get_mapping_type_name()]
Exemple #45
0
def simple_search(request, template=None):
    """ES-specific simple search view.

    This view is for end user searching of the Knowledge Base and
    Support Forum. Filtering options are limited to:
    * product (`product=firefox`, for example, for only Firefox results)
    * document type (`w=2`, for esample, for Support Forum questions only)
    """

    # Redirect to old Advanced Search URLs (?a={1,2}) to the new URL.
    a = request.GET.get('a')
    if a in ['1', '2']:
        new_url = reverse('search.advanced') + '?' + request.GET.urlencode()
        return HttpResponseRedirect(new_url)

    # JSON-specific variables
    is_json = (request.GET.get('format') == 'json')
    callback = request.GET.get('callback', '').strip()
    content_type = (
        'application/x-javascript' if callback else 'application/json')

    # Check callback is valid
    if is_json and callback and not jsonp_is_valid(callback):
        return HttpResponse(
            json.dumps({'error': _('Invalid callback function.')}),
            content_type=content_type, status=400)

    language = locale_or_default(
        request.GET.get('language', request.LANGUAGE_CODE))
    r = request.GET.copy()

    # TODO: Do we really need to add this to the URL if it isn't already there?
    r['w'] = r.get('w', constants.WHERE_BASIC)

    # TODO: Break out a separate simple search form.
    search_form = SimpleSearchForm(r, auto_id=False)

    if not search_form.is_valid():
        if is_json:
            return HttpResponse(
                json.dumps({'error': _('Invalid search data.')}),
                content_type=content_type,
                status=400)

        t = template if request.MOBILE else 'search/form.html'
        search_ = render(request, t, {
            'advanced': False, 'request': request,
            'search_form': search_form})
        cache_period = settings.SEARCH_CACHE_PERIOD
        search_['Cache-Control'] = 'max-age={0!s}'.format((cache_period * 60))
        search_['Expires'] = (
            (datetime.utcnow() + timedelta(minutes=cache_period))
            .strftime(EXPIRES_FMT))
        return search_

    cleaned = search_form.cleaned_data

    # On mobile, we default to just wiki results.
    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_DICT.get(lang):
        lang_name = settings.LANGUAGES_DICT[lang]
    else:
        lang_name = ''

    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = (AnalyzerS().es(urls=settings.ES_URLS)
                .indexes(es_utils.read_index('default')))

    wiki_f = F(model='wiki_document')
    question_f = F(model='questions_question')

    cleaned_q = cleaned['q']
    products = cleaned['product']

    if not products and 'all_products' not in request.GET:
        lowered_q = cleaned_q.lower()

        if 'thunderbird' in lowered_q:
            products.append('thunderbird')
        elif 'android' in lowered_q:
            products.append('mobile')
        elif ('ios' in lowered_q or 'ipad' in lowered_q or 'ipod' in lowered_q or
              'iphone' in lowered_q):
            products.append('ios')
        elif 'firefox os' in lowered_q:
            products.append('firefox-os')
        elif 'firefox' in lowered_q:
            products.append('firefox')

    # Start - wiki filters

    if cleaned['w'] & constants.WHERE_WIKI:
        # Category filter
        wiki_f &= F(document_category__in=settings.SEARCH_DEFAULT_CATEGORIES)

        # Locale filter
        wiki_f &= F(document_locale=language)

        # Product filter
        for p in products:
            wiki_f &= F(product=p)

        # Archived bit
        wiki_f &= F(document_is_archived=False)

    # End - wiki filters

    # Start - support questions filters

    if cleaned['w'] & constants.WHERE_SUPPORT:
        # Has helpful answers is set by default if using basic search
        cleaned['has_helpful'] = constants.TERNARY_YES

        # No archived questions in default search.
        cleaned['is_archived'] = constants.TERNARY_NO

        # These filters are ternary, they can be either YES, NO, or OFF
        ternary_filters = ('has_helpful', 'is_archived')
        d = dict(('question_{0!s}'.format(filter_name),
                  _ternary_filter(cleaned[filter_name]))
                 for filter_name in ternary_filters if cleaned[filter_name])
        if d:
            question_f &= F(**d)

        # Product filter
        for p in products:
            question_f &= F(product=p)

    # End - support questions filters

    # 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

    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:
        # Set up the highlights. Show the entire field highlighted.
        searcher = searcher.highlight(
            'question_content',  # support forum
            'document_summary',  # kb
            pre_tags=['<b>'],
            post_tags=['</b>'],
            number_of_fragments=0)

        # Set up boosts
        searcher = searcher.boost(
            question_title=4.0,
            question_content=3.0,
            question_answer_content=3.0,
            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__match_phrase=10.0,
            document_content__match_phrase=8.0)

        # Build the query
        query_fields = chain(*[
            cls.get_query_fields() for cls in [
                DocumentMappingType,
                QuestionMappingType
            ]
        ])
        query = {}
        # Create match and match_phrase queries for every field
        # we want to search.
        for field in query_fields:
            for query_type in ['match', 'match_phrase']:
                query['{0!s}__{1!s}'.format(field, query_type)] = cleaned_q

        # Transform the query to use locale aware analyzers.
        query = es_utils.es_query_with_analyzer(query, language)

        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)

        # 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[bounds[0]:bounds[1]]

        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']}

            result['url'] = doc['url']
            result['object'] = doc
            result['search_summary'] = summary
            result['rank'] = rank
            result['score'] = doc.es_meta.score
            result['explanation'] = escape(format_explanation(
                doc.es_meta.explanation))
            result['id'] = doc['id']
            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')}),
                                content_type=content_type, 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))

        log.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'))

    fallback_results = None
    if num_results == 0:
        fallback_results = _fallback_results(language, cleaned['product'])

    product = Product.objects.filter(slug__in=cleaned['product'])
    if product:
        product_titles = [_(p.title, 'DB: products.Product.title')
                          for p in product]
    else:
        product_titles = [_('All Products')]

    product_titles = ', '.join(product_titles)

    data = {
        'num_results': num_results,
        'results': results,
        'fallback_results': fallback_results,
        'product_titles': product_titles,
        'q': cleaned['q'],
        'w': cleaned['w'],
        'lang_name': lang_name, }

    if is_json:
        # Models are not json serializable.
        for r in data['results']:
            del r['object']
        data['total'] = len(data['results'])

        data['products'] = ([{'slug': p.slug, 'title': p.title}
                             for p in Product.objects.filter(visible=True)])

        if product:
            data['product'] = product[0].slug

        pages = Paginator(pages)
        data['pagination'] = dict(
            number=pages.pager.number,
            num_pages=pages.pager.paginator.num_pages,
            has_next=pages.pager.has_next(),
            has_previous=pages.pager.has_previous(),
            max=pages.max,
            span=pages.span,
            dotted_upper=pages.pager.dotted_upper,
            dotted_lower=pages.pager.dotted_lower,
            page_range=pages.pager.page_range,
            url=pages.pager.url,
        )
        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, content_type=content_type)

    data.update({
        'product': product,
        'products': Product.objects.filter(visible=True),
        'pages': pages,
        'search_form': search_form,
        'advanced': False,
    })
    results_ = render(request, template, data)
    cache_period = settings.SEARCH_CACHE_PERIOD
    results_['Cache-Control'] = 'max-age={0!s}'.format((cache_period * 60))
    results_['Expires'] = (
        (datetime.utcnow() + timedelta(minutes=cache_period))
        .strftime(EXPIRES_FMT))
    results_.set_cookie(settings.LAST_SEARCH_COOKIE, urlquote(cleaned['q']),
                        max_age=3600, secure=False, httponly=False)

    return results_
Exemple #46
0
def simple_search(request, template=None):
    """ES-specific simple search view.

    This view is for end user searching of the Knowledge Base and
    Support Forum. Filtering options are limited to:
    * product (`product=firefox`, for example, for only Firefox results)
    * document type (`w=2`, for esample, for Support Forum questions only)
    """

    # Redirect to old Advanced Search URLs (?a={1,2}) to the new URL.
    a = request.GET.get('a')
    if a in ['1', '2']:
        new_url = reverse('search.advanced') + '?' + request.GET.urlencode()
        return HttpResponseRedirect(new_url)

    # JSON-specific variables
    is_json = (request.GET.get('format') == 'json')
    callback = request.GET.get('callback', '').strip()
    content_type = ('application/x-javascript'
                    if callback else 'application/json')

    # Check callback is valid
    if is_json and callback and not jsonp_is_valid(callback):
        return HttpResponse(json.dumps(
            {'error': _('Invalid callback function.')}),
                            content_type=content_type,
                            status=400)

    language = locale_or_default(
        request.GET.get('language', request.LANGUAGE_CODE))
    r = request.GET.copy()

    # TODO: Do we really need to add this to the URL if it isn't already there?
    r['w'] = r.get('w', constants.WHERE_BASIC)

    # TODO: Break out a separate simple search form.
    search_form = SimpleSearchForm(r, auto_id=False)

    if not search_form.is_valid():
        if is_json:
            return HttpResponse(json.dumps(
                {'error': _('Invalid search data.')}),
                                content_type=content_type,
                                status=400)

        t = template if request.MOBILE else 'search/form.html'
        search_ = render(request, t, {
            'advanced': False,
            'request': request,
            'search_form': search_form
        })
        cache_period = settings.SEARCH_CACHE_PERIOD
        search_['Cache-Control'] = 'max-age=%s' % (cache_period * 60)
        search_['Expires'] = (
            (datetime.utcnow() +
             timedelta(minutes=cache_period)).strftime(EXPIRES_FMT))
        return search_

    cleaned = search_form.cleaned_data

    # On mobile, we default to just wiki results.
    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_DICT.get(lang):
        lang_name = settings.LANGUAGES_DICT[lang]
    else:
        lang_name = ''

    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = (AnalyzerS().es(urls=settings.ES_URLS).indexes(
        es_utils.read_index('default')))

    wiki_f = F(model='wiki_document')
    question_f = F(model='questions_question')

    # Start - wiki filters

    if cleaned['w'] & constants.WHERE_WIKI:
        # Category filter
        wiki_f &= F(document_category__in=settings.SEARCH_DEFAULT_CATEGORIES)

        # Locale filter
        wiki_f &= F(document_locale=language)

        # Product filter
        products = cleaned['product']
        for p in products:
            wiki_f &= F(product=p)

        # Archived bit
        wiki_f &= F(document_is_archived=False)

    # End - wiki filters

    # Start - support questions filters

    if cleaned['w'] & constants.WHERE_SUPPORT:
        # Has helpful answers is set by default if using basic search
        cleaned['has_helpful'] = constants.TERNARY_YES

        # No archived questions in default search.
        cleaned['is_archived'] = constants.TERNARY_NO

        # These filters are ternary, they can be either YES, NO, or OFF
        ternary_filters = ('has_helpful', 'is_archived')
        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)

        # Product filter
        products = cleaned['product']
        for p in products:
            question_f &= F(product=p)

    # End - support questions filters

    # 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

    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. Show the entire field highlighted.
        searcher = searcher.highlight(
            'question_content',  # support forum
            'document_summary',  # kb
            pre_tags=['<b>'],
            post_tags=['</b>'],
            number_of_fragments=0)

        # Set up boosts
        searcher = searcher.boost(
            question_title=4.0,
            question_content=3.0,
            question_answer_content=3.0,
            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__match_phrase=10.0,
            document_content__match_phrase=8.0)

        # Build the query
        query_fields = chain(*[
            cls.get_query_fields()
            for cls in [DocumentMappingType, QuestionMappingType]
        ])
        query = {}
        # Create match and match_phrase queries for every field
        # we want to search.
        for field in query_fields:
            for query_type in ['match', 'match_phrase']:
                query['%s__%s' % (field, query_type)] = cleaned_q

        # Transform the query to use locale aware analyzers.
        query = es_utils.es_query_with_analyzer(query, language)

        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)

        # 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[bounds[0]:bounds[1]]

        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']
                }

            result['url'] = doc['url']
            result['object'] = doc
            result['search_summary'] = summary
            result['rank'] = rank
            result['score'] = doc.es_meta.score
            result['explanation'] = escape(
                format_explanation(doc.es_meta.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')}),
                                content_type=content_type,
                                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))

        log.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'))

    fallback_results = None
    if num_results == 0:
        fallback_results = _fallback_results(language, cleaned['product'])

    product = Product.objects.filter(slug__in=cleaned['product'])
    if product:
        product_titles = [
            _(p.title, 'DB: products.Product.title') for p in product
        ]
    else:
        product_titles = [_('All Products')]

    product_titles = ', '.join(product_titles)

    data = {
        'num_results': num_results,
        'results': results,
        'fallback_results': fallback_results,
        'product_titles': product_titles,
        'q': cleaned['q'],
        'w': cleaned['w'],
        'lang_name': lang_name,
    }

    if is_json:
        # Models are not json serializable.
        for r in data['results']:
            del r['object']
        data['total'] = len(data['results'])

        data['products'] = ([{
            'slug': p.slug,
            'title': p.title
        } for p in Product.objects.filter(visible=True)])

        if product:
            data['product'] = product[0].slug

        pages = Paginator(pages)
        data['pagination'] = dict(
            number=pages.pager.number,
            num_pages=pages.pager.paginator.num_pages,
            has_next=pages.pager.has_next(),
            has_previous=pages.pager.has_previous(),
            max=pages.max,
            span=pages.span,
            dotted_upper=pages.pager.dotted_upper,
            dotted_lower=pages.pager.dotted_lower,
            page_range=pages.pager.page_range,
            url=pages.pager.url,
        )
        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, content_type=content_type)

    data.update({
        'product': product,
        'products': Product.objects.filter(visible=True),
        'pages': pages,
        'search_form': search_form,
        'advanced': False,
    })
    results_ = render(request, template, data)
    cache_period = settings.SEARCH_CACHE_PERIOD
    results_['Cache-Control'] = 'max-age=%s' % (cache_period * 60)
    results_['Expires'] = (
        (datetime.utcnow() +
         timedelta(minutes=cache_period)).strftime(EXPIRES_FMT))
    results_.set_cookie(settings.LAST_SEARCH_COOKIE,
                        urlquote(cleaned['q']),
                        max_age=3600,
                        secure=False,
                        httponly=False)

    return results_
Exemple #47
0
def generate_simple_search(search_form, language, with_highlights=False):
    """Generates an S given a form

    :arg search_form: a validated SimpleSearch form
    :arg language: the language code
    :arg with_highlights: whether or not to ask for highlights

    :returns: a fully formed S

    """
    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = (
        es_utils.AnalyzerS().es(
            urls=settings.ES_URLS,
            timeout=settings.ES_TIMEOUT,
            use_ssl=settings.ES_USE_SSL,
            http_auth=settings.ES_HTTP_AUTH,
        )
        .indexes(es_utils.read_index('default'))
    )

    cleaned = search_form.cleaned_data

    doctypes = []
    final_filter = es_utils.F()
    cleaned_q = cleaned['q']
    products = cleaned['product']

    # Handle wiki filters
    if cleaned['w'] & constants.WHERE_WIKI:
        wiki_f = es_utils.F(model='wiki_document',
                            document_category__in=settings.SEARCH_DEFAULT_CATEGORIES,
                            document_locale=language,
                            document_is_archived=False)

        for p in products:
            wiki_f &= es_utils.F(product=p)

        doctypes.append(DocumentMappingType.get_mapping_type_name())
        final_filter |= wiki_f

    # Handle question filters
    if cleaned['w'] & constants.WHERE_SUPPORT:
        question_f = es_utils.F(model='questions_question',
                                question_is_archived=False,
                                question_has_helpful=True)

        for p in products:
            question_f &= es_utils.F(product=p)

        doctypes.append(QuestionMappingType.get_mapping_type_name())
        final_filter |= question_f

    # Build a filter for those filters and add the other bits to
    # finish the search
    searcher = searcher.doctypes(*doctypes)
    searcher = searcher.filter(final_filter)

    if cleaned['explain']:
        searcher = searcher.explain()

    if with_highlights:
        # Set up the highlights. Show the entire field highlighted.
        searcher = searcher.highlight(
            'question_content',  # support forum
            'document_summary',  # kb
            pre_tags=['<b>'],
            post_tags=['</b>'],
            number_of_fragments=0
        )

    searcher = apply_boosts(searcher)

    # Build the query
    query_fields = chain(*[
        cls.get_query_fields() for cls in [
            DocumentMappingType,
            QuestionMappingType
        ]
    ])
    query = {}
    # Create match and match_phrase queries for every field
    # we want to search.
    for field in query_fields:
        for query_type in ['match', 'match_phrase']:
            query['%s__%s' % (field, query_type)] = cleaned_q

    # Transform the query to use locale aware analyzers.
    query = es_utils.es_query_with_analyzer(query, language)

    searcher = searcher.query(should=True, **query)
    return searcher
Exemple #48
0
def advanced_search(request):
    """Elasticsearch-specific Advanced search view"""

    to_json = JSONRenderer().render
    template = 'search/results.html'

    # 1. Prep request.
    r = request.GET.copy()
    # TODO: Figure out how to get rid of 'a' and do it.
    # It basically is used to switch between showing the form or results.
    a = request.GET.get('a', '2')
    # TODO: This is so the 'a=1' stays in the URL for pagination.
    r['a'] = 1

    language = locale_or_default(
        request.GET.get('language', request.LANGUAGE_CODE))
    r['language'] = language
    lang = language.lower()
    lang_name = settings.LANGUAGES_DICT.get(lang) or ''

    # 2. Build form.
    search_form = AdvancedSearchForm(r, auto_id=False)
    search_form.set_allowed_forums(request.user)

    # 3. Validate request.
    # Note: a == 2 means "show the form"--that's all we use it for now.
    if a == '2' or not search_form.is_valid():
        if request.IS_JSON:
            return HttpResponse(json.dumps(
                {'error': _('Invalid search data.')}),
                                content_type=request.CONTENT_TYPE,
                                status=400)

        t = 'search/form.html'
        data = {
            'advanced': True,
            'request': request,
            'search_form': search_form
        }
        # get value for search input from last search term.
        last_search = request.COOKIES.get(settings.LAST_SEARCH_COOKIE)
        # If there is any cached input from last search, pass it to template
        if last_search and 'q' not in r:
            cached_field = urlquote(last_search)
            data.update({'cached_field': cached_field})

        return cache_control(render(request, t, data),
                             settings.SEARCH_CACHE_PERIOD)

    # 4. Generate search.
    cleaned = search_form.cleaned_data

    # We use a regular S here because we want to search across
    # multiple doctypes.
    searcher = (AnalyzerS().es(
        urls=settings.ES_URLS,
        timeout=settings.ES_TIMEOUT,
        use_ssl=settings.ES_USE_SSL,
        http_auth=settings.ES_HTTP_AUTH,
        connection_class=RequestsHttpConnection).indexes(
            es_utils.read_index('default')))

    doctypes = []
    final_filter = F()
    unix_now = int(time.time())
    interval_filters = (('created', cleaned['created'],
                         cleaned['created_date']),
                        ('updated', cleaned['updated'],
                         cleaned['updated_date']))

    # Start - wiki search configuration

    if cleaned['w'] & constants.WHERE_WIKI:
        wiki_f = F(model='wiki_document')

        # 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 not cleaned['include_archived']:
            wiki_f &= F(document_is_archived=False)

        # Apply sortby
        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

        doctypes.append(DocumentMappingType.get_mapping_type_name())
        final_filter |= wiki_f

    # End - wiki search configuration

    # Start - support questions configuration

    if cleaned['w'] & constants.WHERE_SUPPORT:
        question_f = F(model='questions_question')

        # These filters are ternary, they can be either YES, NO, or OFF
        ternary_filters = ('is_locked', 'is_solved', 'has_answers',
                           'has_helpful', 'is_archived')
        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)

        # 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'])

        # Apply sortby
        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 created and updated filters
        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)
                }

                question_f &= F(**before)

            elif filter_option == constants.INTERVAL_AFTER:
                after = {
                    filter_name + '__gte': min(filter_date, unix_now),
                    filter_name + '__lte': unix_now
                }

                question_f &= F(**after)

        doctypes.append(QuestionMappingType.get_mapping_type_name())
        final_filter |= question_f

    # End - support questions configuration

    # Start - discussion forum configuration

    if cleaned['w'] & constants.WHERE_DISCUSSION:
        discussion_f = F(model='forums_thread')

        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)

        valid_forum_ids = [
            f.id for f in Forum.authorized_forums_for_user(request.user)
        ]

        forum_ids = None
        if cleaned['forum']:
            forum_ids = [f for f in cleaned['forum'] if f in valid_forum_ids]

        # If we removed all the forums they wanted to look at or if
        # they didn't specify, then we filter on the list of all
        # forums they're authorized to look at.
        if not forum_ids:
            forum_ids = valid_forum_ids

        discussion_f &= F(post_forum_id__in=forum_ids)

        # Apply created and updated filters
        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)

            elif filter_option == constants.INTERVAL_AFTER:
                after = {
                    filter_name + '__gte': min(filter_date, unix_now),
                    filter_name + '__lte': unix_now
                }

                discussion_f &= F(**after)

        doctypes.append(ThreadMappingType.get_mapping_type_name())
        final_filter |= discussion_f

    # End - discussion forum configuration

    # Done with all the filtery stuff--time  to generate results

    searcher = searcher.doctypes(*doctypes)
    searcher = searcher.filter(final_filter)

    if 'explain' in request.GET and request.GET['explain'] == '1':
        searcher = searcher.explain()

    cleaned_q = cleaned['q']

    # Set up the highlights. Show the entire field highlighted.
    searcher = searcher.highlight(
        'question_content',  # support forum
        'document_summary',  # kb
        'post_content',  # contributor forum
        pre_tags=['<b>'],
        post_tags=['</b>'],
        number_of_fragments=0)

    searcher = apply_boosts(searcher)

    # Build the query
    if cleaned_q:
        query_fields = chain(*[
            cls.get_query_fields() for cls in
            [DocumentMappingType, ThreadMappingType, QuestionMappingType]
        ])
        query = {}
        # Create a simple_query_search query for every field we want to search.
        for field in query_fields:
            query['%s__sqs' % field] = cleaned_q

        # Transform the query to use locale aware analyzers.
        query = es_utils.es_query_with_analyzer(query, language)

        searcher = searcher.query(should=True, **query)

    searcher = searcher[:settings.SEARCH_MAX_RESULTS]

    # 5. Generate output
    pages = paginate(request, searcher, settings.SEARCH_RESULTS_PER_PAGE)

    if pages.paginator.count == 0:
        # If we know there aren't any results, show fallback_results.
        fallback_results = _fallback_results(language, cleaned['product'])
        results = []
    else:
        fallback_results = None
        results = build_results_list(pages, request.IS_JSON)

    items = [(k, v) for k in search_form.fields for v in r.getlist(k)
             if v and k != 'a']
    items.append(('a', '2'))

    product = Product.objects.filter(slug__in=cleaned['product'])
    if product:
        product_titles = [
            pgettext('DB: products.Product.title', p.title) for p in product
        ]
    else:
        product_titles = [_('All Products')]

    # FIXME: This is probably bad l10n.
    product_titles = ', '.join(product_titles)

    data = {
        'num_results': pages.paginator.count,
        'results': results,
        'fallback_results': fallback_results,
        'product_titles': product_titles,
        'q': cleaned['q'],
        'w': cleaned['w'],
        'lang_name': lang_name,
        'advanced': True,
        'products': Product.objects.filter(visible=True)
    }

    if request.IS_JSON:
        data['total'] = len(data['results'])
        data['products'] = [{
            'slug': p.slug,
            'title': p.title
        } for p in data['products']]

        if product:
            data['product'] = product[0].slug

        pages = Paginator(pages)
        data['pagination'] = dict(
            number=pages.pager.number,
            num_pages=pages.pager.paginator.num_pages,
            has_next=pages.pager.has_next(),
            has_previous=pages.pager.has_previous(),
            max=pages.max,
            span=pages.span,
            dotted_upper=pages.pager.dotted_upper,
            dotted_lower=pages.pager.dotted_lower,
            page_range=pages.pager.page_range,
            url=pages.pager.url,
        )
        if not results:
            data['message'] = _('No pages matched the search criteria')
        json_data = to_json(data)
        if request.JSON_CALLBACK:
            json_data = request.JSON_CALLBACK + '(' + json_data + ');'
        return HttpResponse(json_data, content_type=request.CONTENT_TYPE)

    data.update({
        'product': product,
        'pages': pages,
        'search_form': search_form
    })
    resp = cache_control(render(request, template, data),
                         settings.SEARCH_CACHE_PERIOD)
    resp.set_cookie(settings.LAST_SEARCH_COOKIE,
                    urlquote(cleaned['q']),
                    max_age=3600,
                    secure=False,
                    httponly=False)
    return resp
Exemple #49
0
 def handle(self, **options):
     index_task.delay(DocumentMappingType, DocumentMappingType.get_indexable())