用JsonResponse分页

Question

我想在我的JsonResponse中添加分页。

我目前正在使用django.http.JsonResponse从弹性搜索API中生成json。我想添加一个分页功能，以便包括在内。我的代码如下：

class ResultQueryView(View):
    def get(self, request):
        resource_meta = request.GET.getlist("resource_meta")
        locations = request.GET.getlist("location")
        page = request.GET.get("page")
        logger.info("Got search query where resource_meta: {} and locations: {}".format(resource_meta, locations))
        results = resource_query(resource_meta, locations)
        resource_ids = [r["_id"] for r in results['hits']['hits']]
        resources = get_enriched_resources(request.user, Resource.objects.filter(internal_id__in=resource_ids))
        serialized = ResourceSerializer(resources, many=True)
        return JsonResponse({"resources": serialized.data})

Answer 1

使用Django的Paginator。

from django.core.paginator import EmptyPage, PageNotAnInteger, Paginator

class ViewPaginatorMixin(object):
    min_limit = 1
    max_limit = 10

    def paginate(self, object_list, page=1, limit=10, **kwargs):
        try:
            page = int(page)
            if page < 1:
                page = 1
        except (TypeError, ValueError):
            page = 1

        try:
            limit = int(limit)
            if limit < self.min_limit:
                limit = self.min_limit
            if limit > self.max_limit:
                limit = self.max_limit
        except (ValueError, TypeError):
            limit = self.max_limit

        paginator = Paginator(object_list, limit)
        try:
            objects = paginator.page(page)
        except PageNotAnInteger:
            objects = paginator.page(1)
        except EmptyPage:
            objects = paginator.page(paginator.num_pages)
        data = {
            'previous_page': objects.has_previous() and objects.previous_page_number() or None,
            'next_page': objects.has_next() and objects.next_page_number() or None,
            'data': list(objects)
        }
        return data

现在，使用ViewPaginatorMixin支持View分页。

class ResultQueryView(ViewPaginatorMixin, View):
    def get(self, request):
       // code
       serialized = ResourceSerializer(resources, many=True)
       return JsonResponse({"resources": self.paginate(serialized.data, page, limit)})

Answer 2

一个简单的解决方案可以是在构建serialized.data之前对JsonResponse进行切片(甚至可以用预期页面的总n来注释结果，即math.ceil(len(serialized.data) / PAGE_SIZE))：

PAGE_SIZE = 10

start = page * PAGE_SIZE
stop = min(start + PAGE_SIZE, len(serialized.data))
#return JsonResponse({"resources": serialized.data})
return JsonResponse({"resources": serialized.data[start:stop]})

测试：

class FakeSerialized(object):
    def __init__(self):
        self.data = list(range(0,35))

serialized = FakeSerialized()
print('All data:', serialized.data)
PAGE_SIZE = 10

for page in range(0, 5):

    start = page * PAGE_SIZE
    stop = min(start + PAGE_SIZE, len(serialized.data))
    data = serialized.data[start:stop]

    print('Page %d:' % page, data)

结果：

$ python3 ./paginate.py
All data: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34]
Page 0: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
Page 1: [10, 11, 12, 13, 14, 15, 16, 17, 18, 19]
Page 2: [20, 21, 22, 23, 24, 25, 26, 27, 28, 29]
Page 3: [30, 31, 32, 33, 34]
Page 4: []

Answer 3

可以重写get_queryset()的ListAPIView，以返回根据需要构造的查询集，然后使用视图的内建分页：

from rest_framework.generics import ListAPIView
from rest_framework.pagination import PageNumberPagination


class ResultQueryView(ListAPIView):
    serializer_class = ResourceSerializer
    # configure this according to your needs
    pagination_class = PageNumberPagination

    def get_queryset(self):
        resource_meta = self.request.GET.getlist("resource_meta")
        locations = self.request.GET.getlist("location")
        results = resource_query(resource_meta, locations)
        resource_ids = [r["_id"] for r in results['hits']['hits']]
        return Resource.objects.filter(internal_id__in=resource_ids))