Python search.TrigramSimilarity方法代码示例

# 需要导入模块: from django.contrib.postgres import search [as 别名]
# 或者: from django.contrib.postgres.search import TrigramSimilarity [as 别名]
def post_search(request):
    form = SearchForm()
    query = None
    results = []
    if 'query' in request.GET:
        form = SearchForm(request.GET)
        if form.is_valid():
            query = form.cleaned_data['query']
            results = Post.published.annotate(
                similarity=TrigramSimilarity('title', query),
            ).filter(similarity__gt=0.1).order_by('-similarity')
    return render(request,
                  'blog/post/search.html',
                  {'form': form,
                   'query': query,
                   'results': results}) 

# 需要导入模块: from django.contrib.postgres import search [as 别名]
# 或者: from django.contrib.postgres.search import TrigramSimilarity [as 别名]
def post_search(request):
    form = SearchForm()
    query = None
    results = []
    if 'query' in request.GET:
        form = SearchForm(request.GET)
        if form.is_valid():
            query = form.cleaned_data['query']
            results = Post.objects.annotate(
                similarity=TrigramSimilarity('title', query),
            ).filter(similarity__gt=0.3).order_by('-similarity')
    return render(request,
                  'blog/post/search.html',
                  {'form': form,
                   'query': query,
                   'results': results}) 

# 需要导入模块: from django.contrib.postgres import search [as 别名]
# 或者: from django.contrib.postgres.search import TrigramSimilarity [as 别名]
def _find_keyword_or_split(self, text: str):
        """Attempts to find a keyword for the full string, in case of failure,
        attempts to chip off the last word and repeats. Similarity depends on
        the length of the string, hence edit (Levenshtein) distance filter was
        introduced so that 'amerikkalainen jalkapallo' is matched to one keyword
        and 'nuorisotalo jalkapallo' is matched to two.
        The function allows 2 edits per every 5 symbols, which helps both with
        simple plurals and true typos.

        TODO: when a two-word keyword is preceded by something as in
        'nuorisotalo amerikkalainen jalkapallo' it is mapped to 'nuorisotalo'
        'jalkapallo', and 'amerikkalaiset', should be 'nuorisotalo' and
        'amerikkalainen jalkapallo'.

        TODO: correct language treatment, so that 'body condition' is searched
        for in the English, not the Finnish keywords.
        """

        text = text.strip(" ")
        match = (
            Keyword.objects.annotate(similarity=TrigramSimilarity("name_fi", text))
            .order_by("similarity")
            .last()
        )

        if match and distance(match.name_fi, text) / len(text) < 0.2:
            return set([match])

        if " " not in text:
            return set()

        split = text.rsplit(" ", 1)
        keywords = set()
        for i in split:
            keywords |= self._find_keyword_or_split(i)

        return keywords 

# 需要导入模块: from django.contrib.postgres import search [as 别名]
# 或者: from django.contrib.postgres.search import TrigramSimilarity [as 别名]
def search_taxonomy_node(request, short_name):
    if not request.user.is_authenticated:
        return HttpResponse('Unauthorized', status=401)
    dataset = get_object_or_404(Dataset, short_name=short_name)
    if not dataset.user_is_maintainer(request.user):
        raise HttpResponseNotAllowed
    taxonomy = dataset.taxonomy
    query = request.GET.get('q', '')

    # vector = SearchVector('name', weight='A') + SearchVector('description', weight='C')
    # query = SearchQuery(query)
    # qs_results = TaxonomyNode.objects.filter(taxonomy__dataset=dataset)\
    #                                  .annotate(rank=SearchRank(vector, query)).filter(rank__gte=0.3)\
    #                                  .order_by('rank')

    qs_results = TaxonomyNode.objects.filter(taxonomy__dataset=dataset)\
                                     .annotate(similarity=TrigramSimilarity('name', query) +
                                                          TrigramSimilarity('description', query),)\
                                     .filter(similarity__gte=0.2)\
                                     .order_by('-similarity')

    results = [{'name': node.name,
                'node_id': node.node_id,
                'path': ' > '.join([TaxonomyNode.objects.get(node_id=node_id).name for node_id in path_list]),
                'big_id': ','.join(path_list),
                'omitted': node.omitted_curation_task}
               for node in qs_results
               for path_list in taxonomy.get_hierarchy_paths(node.node_id)]

    return JsonResponse(results[:10], safe=False)


#############################
# CONTRIBUTE TO DATASET VIEWS
############################# 

# 需要导入模块: from django.contrib.postgres import search [as 别名]
# 或者: from django.contrib.postgres.search import TrigramSimilarity [as 别名]
def test_trigram_similarity(self):
        search = 'Bat sat on cat.'
        self.assertQuerysetEqual(
            self.Model.objects.filter(
                field__trigram_similar=search,
            ).annotate(similarity=TrigramSimilarity('field', search)).order_by('-similarity'),
            [('Cat sat on mat.', 0.625), ('Dog sat on rug.', 0.333333)],
            transform=lambda instance: (instance.field, instance.similarity),
            ordered=True,
        )