我正在研究RubyOnRails应用程序。我通过Tire
gem使用Elasticsearch。 Elasticsearch索引了3个模型。这些模型是News,NewsTag和Categorization。新闻与NewsTag有着多对多的关系,而NewsTag与分类有着多对多的关系。所有这些模型都是mongodb。
映射索引: -
mapping do
indexes :cover, analyzer: 'snowball'
indexes :title_en
indexes :title_ar
indexes :content_brief_ar
indexes :content_brief_en
indexes :body_ar
indexes :body_en
indexes :likes, type: 'nested' do
end
indexes :comments, type: 'nested' do
end
indexes :news_tags, type: 'nested' do
indexes :name_en, analyzer: 'snowball'
indexes :name_ar, analyzer: 'snowball'
indexes :categorizations, type: 'nested' do
indexes :ar_name, analyzer: 'snowball'
indexes :en_name, analyzer: 'snowball'
end
end
end
def to_indexed_json
{
cover: self.cover,
title_ar: self.title_ar,
title_en: self.title_en,
content_brief_ar: self.content_brief_ar,
content_brief_en: self.content_brief_en,
body_ar: self.body_ar,
body_en: self.body_en,
news_tags: self.news_tags,
likes: self.likes,
comments: self.comments
}.to_json
end
我如何搜索
@news = News.tire.search do
query do
nested path: 'news_tags' do
query do
boolean do
should {terms 'news_tags.categorization_ids', categorization_ids }
end
end
end
end
page = page_param.to_i
search_size = per_page_param.to_i
from (page - 1) * search_size
size search_size
end
这很好用,直到我意识到检索结果的类型不是新闻,它是Item,并且在模型(新闻)中定义了一些经常调用的方法。所以我搜索了将项目包装成新闻,我找到了一些有趣的东西。
将此行添加到config / intializers / tire.rb
Tire.configure do
wrapper ProxyObject
end
和此文件app / models / proxy_object.rb
class ProxyObject < SimpleDelegator
delegate :class, :is_a?, :to => :_proxied_object
def initialize(attrs={})
klass = attrs['_type'].camelize.classify.constantize
@_proxied_object = klass.new
_assign_attrs(attrs)
super(_proxied_object)
end
private
def _proxied_object
@_proxied_object
end
def _assign_attrs(attrs={})
attrs.each_pair do |key, value|
unless _proxied_object.respond_to?("#{key}=".to_sym)
_proxied_object.class.send(:attr_accessor, key.to_sym)
end
_proxied_object.send("#{key}=".to_sym, value)
end
end
end
这个解决方案在Elasticsearch上有一个教程项目非常成功。这个解决方案使我能够调用模型方法。但是,它对我的项目不起作用。我注意到一些可能很重要的东西。在新闻索引中。
GET localhost:9200 / news / news / 58889bbb6f6d613fff050000
{
"_index": "news",
"_type": "news",
"_id": "58889bbb6f6d613fff050000",
"_version": 1,
"found": true,
"_source": {
"cover": "new2",
"title_ar": "new2",
"title_en": "new2",
"content_brief_ar": "new2",
"content_brief_en": "new2",
"body_ar": "<p>new2</p>\r\n",
"body_en": "<p>new2</p>\r\n",
"news_tags": [
{
"_id": "56dec06769702d1578000000",
"categorization_ids": [
"5888990f6f6d613fff000000"
],
"created_at": "2016-03-08T14:07:03+02:00",
"name_ar": "صحة الطفل",
"name_en": "Baby Health",
"news_ids": [
"56dec03569702d156a010000",
"5704f92769702d4c2b010000",
"574efc7969702d370a130000",
"578515d369702d4f11000000",
"58889bbb6f6d613fff050000",
"58889c3f6f6d613fff080000"
],
"updated_at": "2016-03-08T14:07:03+02:00"
}
],
"likes": [],
"comments": []
}
}
正如您所注意到的,news_ids
嵌入了news_tags
,而news_tags
中嵌入了news
。所以,我觉得有某种递归。
我的问题是,当我在@news
对象上应用任何方法时,即使count
,从搜索中检索到我收到了错误消息
(byebug) @news.count
MOPED: 127.0.0.1:27017 COMMAND database=admin command={:ismaster=>1} runtime: 1.9747ms
MOPED: 127.0.0.1:27017 UPDATE database=nabda-net_staging collection=news_tags selector={"$and"=>[{"_id"=>{"$in"=>[]}}]} update={"$pull"=>{"news_ids"=>BSON::ObjectId('5888aaba6f6d6154d0000000')}} flags=[:multi]
COMMAND database=nabda-net_staging command={:getlasterror=>1, :w=>1} runtime: 0.6387ms
*** NoMethodError Exception: undefined method `_id' for #<Hash:0x00000004b78190>
为什么我收到此错误消息? 它甚至不具有描述性。我不知道问题出在哪里。所以,任何人都可以帮助我。
答案 0 :(得分:1)
我找到了解决方案。在搜索查询中添加load: true
将解决问题。
@news = News.tire.search load: true do
query do
nested path: 'news_tags' do
query do
boolean do
should {terms 'news_tags.categorization_ids', categorization_ids }
end
end
end
end
page = page_param.to_i
search_size = per_page_param.to_i
from (page - 1) * search_size
size search_size
end
谢谢大家。