使用elasticsearch dsl索引数据时出错

Question

我有以下两种模式：

class PostUser(models.Model):

    user_id = models.CharField(max_length=1000,blank=True,null=True)
    reputation = models.CharField(max_length = 1000 , blank = True , null = True)

    def __unicode__(self):
        return self.user_id

    def indexing(self):
        obj = PostUserIndex(
            meta = {'id': self.id},
            user_id = self.user_id,
            reputation = self.reputation,
        )
        obj.save(index = 'post-user-index')
        return obj.to_dict(include_meta=True)

class Posts(models.Model):

    user_post_id = models.CharField(max_length = 1000 , blank = True , null = True)
    score = models.CharField(max_length = 1000 , blank = True , null = True)
    owner_user_id = models.ForeignKey(PostUser,default="-100")


    def __unicode__(self):

        return self.user_post_id

    def indexing(self):
        obj = PostsIndex(
            meta = {'id': self.id},
            user_post_id = self.user_post_id,
            score = self.score,
            owner_user_id = self.owner_user_id,
        )
        obj.save(index = 'newuserposts-index')
        return obj.to_dict(include_meta=True)

我试图对数据进行索引的方式如下：

class PostUserIndex(DocType):
    user_id = Text()
    reputation = Text()


class PostsIndex(DocType):
    user_post_id = Text()
    score = Text()
    owner_user_id = Nested(PostUserIndex)

然后，我尝试运行以下方法来索引数据：

def posts_indexing():
    PostsIndex.init(index='newuserposts-index')
    es = Elasticsearch()
    bulk(client=es, actions=(b.indexing() for b in models.Posts.objects.all().iterator()))

我尝试了不同的方法，手动输入嵌套属性，并从doctype更改为PostUser的内部文档，但我还是遇到了奇怪的错误。

错误：

AttributeError：“PostUser”对象没有属性“复制”

Traceback (most recent call last):
  File "<console>", line 1, in <module>
  File "/Users/ammarkhan/Desktop/danny/src/dataquerying/datatoindex.py", line 74, in new_user_posts_indexing
    bulk(client=es, actions=(b.indexing() for b in models.Posts.objects.all().iterator()))
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch/helpers/__init__.py", line 257, in bulk
    for ok, item in streaming_bulk(client, actions, **kwargs):
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch/helpers/__init__.py", line 180, in streaming_bulk
    client.transport.serializer):
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch/helpers/__init__.py", line 58, in _chunk_actions
    for action, data in actions:
  File "/Users/ammarkhan/Desktop/danny/src/dataquerying/datatoindex.py", line 74, in <genexpr>
    bulk(client=es, actions=(b.indexing() for b in models.Posts.objects.all().iterator()))
  File "/Users/ammarkhan/Desktop/danny/src/dataquerying/models.py", line 167, in indexing
    obj.save(index = 'newuserposts-index')
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/document.py", line 405, in save
    self.full_clean()
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/utils.py", line 417, in full_clean
    self.clean_fields()
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/utils.py", line 403, in clean_fields
    data = field.clean(data)
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/field.py", line 179, in clean
    data = super(Object, self).clean(data)
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/field.py", line 90, in clean
    data = self.deserialize(data)
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/field.py", line 86, in deserialize
    return self._deserialize(data)
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/field.py", line 166, in _deserialize
    return self._wrap(data)
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/field.py", line 142, in _wrap
    return self._doc_class.from_es(data)
  File "/Users/ammarkhan/Desktop/danny/lib/python2.7/site-packages/elasticsearch_dsl/utils.py", line 342, in from_es
    meta = hit.copy()
AttributeError: ‘PostUser' object has no attribute 'copy'

Answer 1

在您的.save方法中调用indexing，这将将文档保存到elasticsearch，然后将其传递给bulk以完成相同的任务，save是额外的。

您还将PostUser的一个实例分配给owner_user_id，而不是通过调用它上的indexing方法来正确地序列化它(其中没有save )：

  def indexing(self):
    obj = PostsIndex(
        meta = {'id': self.id, 'index': 'newuserposts-index'},
        user_post_id = self.user_post_id,
        score = self.score,
        owner_user_id = self.owner_user_id.indexing(),
    )
    return obj.to_dict(include_meta=True)