范围 - 查询嵌套文档的性能影响？

I want to mirror the data of an external database through their API.

我想通过他们的API镜像外部数据库的数据。

However, I also want to store a few extra fields on each document. I figured the structure would be easier to reason about, if there was a clear distinction between the data supplied by the API and the extra fields I have attached.

但是,我还想在每个文档上存储一些额外的字段。我认为如果API提供的数据与我附加的额外字段之间有明显的区别,结构将更容易推理。

Example:

Schema({
  youtube: {
    type: Object
  },
  'youtube.etag': {
    type: String
  },
  'youtube.commentThreadId': {
    type: String,
    index: 1
  },
  'youtube.channelId': {
    type: String,
    index: 1
  },
  'youtube.videoId': {
    type: String,
    index: 1
  },
  'youtube.canReply': {
    type: Boolean
  },
  'youtube.totalReplyCount': {
    type: Number
  },
  'youtube.isPublic': {
    type: Boolean
  },
  lastSync: {
    type: Date,
  },
  updatedAt: {
    type: Date
  },
  createdAt: {
    type: Date
  },
});

In my case, scoping the API data on 'youtube' as an embedded document seemed to do the trick, but I would like to know if their is any performance hit by doing this, compares to having the fields at the outermost level of the document.

在我的情况下,将'youtube'上的API数据作为嵌入式文档进行定位似乎可以解决问题,但我想知道它们是否会因此而受到任何性能影响,而不是将文档放在文档的最外层。。

1 个解决方案

#1

Organizing your data into sub-documents like this is a good idea and has no effect on performance.

将数据组织到这样的子文档中是一个好主意,对性能没有影响。

However, your schema structure is invalid because you can't have .s in your key names. It should be defined like this instead:

但是,您的架构结构无效,因为您的密钥名称中不能包含.s。它应该像这样定义:

Schema({
  youtube: {
    etag: String,
    commentThreadId: {type: String, index: 1},
    channelId: {type: String, index: 1},
    videoId: {type: String, index: 1},
    canReply: Boolean,
    totalReplyCount: Number,
    isPublic: Boolean
  },
  lastSync: {
    type: Date,
  },
  updatedAt: {
    type: Date
  },
  createdAt: {
    type: Date
  },
});

#1