AWS AppSync-使用GSI排序键将全局二级索引添加到DynamoDB和分页

时间：2018-10-13 08:22:54

标签： amazon-web-services amazon-dynamodb aws-appsync

我正在尝试创建一种结构，该结构列出以postId顺序排序的帖子的评论。他们的lastChangeTime降序。

以下共享模式中的模型。 

type Comment {
  id: ID!
  postId: String!
  user: String!
  lastChangeTime: String
  commentBody: String
}

它已经有一个支持DynamoDB表和通用CRUD解析器。 并且id字段是表中的主键。

我计划按以下方式构建查询：

{
  "version": "2017-02-28",
  "operation" : "Query",
  "index" : "postId-index",
  "query" : {
    "expression": "post = :postId",
    "expressionValues" : {
      ":postId" : {
        "S" : "${ctx.args.postId}"
      }
    }
  },
  "limit": $util.defaultIfNull($ctx.args.first, 20),
  "nextToken": $util.toJson($util.defaultIfNullOrEmpty($ctx.args.after, null)),
  "scanIndexForward": false
}

要使其正常运行，我应该如何在postId（即postId-index）上添加全球二级索引（GSI）？

在定义它时，我应该在lastChangeTime上添加一个排序键吗？还是lastChangeTime字段需要自己的单独索引进行排序？

1 个答案:

答案 0 :(得分：4)

很容易。您可以通过两种不同的方式来做到这一点，也可以同时使用两种方式来获得更好的灵活性。 （如果您已经解决了该问题，希望对您有所帮助）。

通过这种方式，您可以使用查询参数动态设置sortDirection。

下面给出了详细的代码。在此之前，请注意这一点。

  1. 第一点是您的评论类型-您正在使用

    type Comment {
  id: ID!
  postId: String!
  ## rest of your type definition
}

这不是设置链接到帖子的评论类型的最佳方法。

更好的方法是：

type Comment {
    postID: ID!         ## select this as Primary key in DataSource in AppSync console
    commentID: String!    ## select this as Sort key in DataSource in AppSync console
    ## rest of your type definition
}

如果执行此操作，则DynamoDB表的结构将类似于下面显示的结构（from this AWS webpage）。

（在您的情况下，UserId为PostId，GameTitle为CommentID）

enter image description here

这样，因为所有注释都将彼此相邻记录（在相同的PostId下），因此AppSync响应时间要快得多。

在AppSync docs page中，他们还使用了以下示例：

enter image description here

  1. 正如@Lisa M Shon所述，您可以在CommentTable上启动GSI，其中PostId是分区键，而addTime是排序键。如果要使用下面提供的解析器，则称其为“ postID-addedTime-index”。确保您在GSI中的addedTime上选择了“数字”。

enter image description here

然后，您可以在架构中定义以下类型：

type Comment {
    postID: ID!
    commentID: String!
    content: String!
    addedTime: Int!
}

type CommentConnection {
    items: [Comment]
    nextToken: String
}

type Post {
    id: ID!
    postContent: String!
    addedTime: Int!
    ## Option 1. Gets Post details with all related Comments. 
    ## If 'startFromTime' is provided it will fetch all Comments starting from that timestamp.
    ## If 'startFromTime' is not provided it will fetch all Comments.
    comments(
        filter: TableCommentFilterInput,
        sortDirection: SortDirection,
        startFromTime: Int,
        limit: Int,
        nextToken: String
    ): CommentConnection
}

type Query {
    ## Option 2. It will fetch Comments only for a given PostId.
    ## If 'startFromTime' is provided it will fetch all Comments starting from that timestamp.
    ## If 'startFromTime' is not provided it will fetch all Comments.
    postCommentsByAddTime(
        postID: String!,
        startFromTime: Int!,
        sortDirection: SortDirection,
        filter: TableCommentFilterInput,
        count: Int,
        nextToken: String
    ): PaginatedComments
   ## your other queries
}

    ## rest of your schema definition

您可以同时使用-选项1和选项2，并同时使用。

完整的架构代码在此处（展开下面的代码段）： 

type Comment {
	postID: ID!
	commentID: String!
	content: String!
	addedTime: Int!
}

type CommentConnection {
	items: [Comment]
	nextToken: String
}

input CreateCommentInput {
	postID: ID!
	commentID: String!
	content: String!
	addedTime: Int!
}

input CreatePostInput {
	postContent: String!
	addedTime: Int!
}

input DeleteCommentInput {
	postID: ID!
	commentID: String!
}

input DeletePostInput {
	id: ID!
}

type Mutation {
	createComment(input: CreateCommentInput!): Comment
	updateComment(input: UpdateCommentInput!): Comment
	deleteComment(input: DeleteCommentInput!): Comment
	createPost(input: CreatePostInput!): Post
	updatePost(input: UpdatePostInput!): Post
	deletePost(input: DeletePostInput!): Post
}

type PaginatedComments {
	items: [Comment!]!
	nextToken: String
}

type Post {
	id: ID!
	postContent: String!
	addedTime: Int!
	comments(
		filter: TableCommentFilterInput,
		sortDirection: SortDirection,
		startFromTime: Int,
		limit: Int,
		nextToken: String
	): CommentConnection
}

type PostConnection {
	items: [Post]
	nextToken: String
}

type Query {
	getComment(postID: ID!, commentID: String!): Comment
	listComments(filter: TableCommentFilterInput, limit: Int, nextToken: String): CommentConnection
	getPost(id: ID!): Post
	listPosts(filter: TablePostFilterInput, limit: Int, nextToken: String): PostConnection
	postCommentsByAddTime(
		postID: String!,
		startFromTime: Int!,
		sortDirection: SortDirection,
		filter: TableCommentFilterInput,
		count: Int,
		nextToken: String
	): PaginatedComments
}

enum SortDirection {
	ASC
	DESC
}

type Subscription {
	onCreateComment(
		postID: ID,
		commentID: String,
		content: String,
		addedTime: Int
	): Comment
		@aws_subscribe(mutations: ["createComment"])
	onUpdateComment(
		postID: ID,
		commentID: String,
		content: String,
		addedTime: Int
	): Comment
		@aws_subscribe(mutations: ["updateComment"])
	onDeleteComment(
		postID: ID,
		commentID: String,
		content: String,
		addedTime: Int
	): Comment
		@aws_subscribe(mutations: ["deleteComment"])
	onCreatePost(id: ID, postContent: String, addedTime: Int): Post
		@aws_subscribe(mutations: ["createPost"])
	onUpdatePost(id: ID, postContent: String, addedTime: Int): Post
		@aws_subscribe(mutations: ["updatePost"])
	onDeletePost(id: ID, postContent: String, addedTime: Int): Post
		@aws_subscribe(mutations: ["deletePost"])
}

input TableBooleanFilterInput {
	ne: Boolean
	eq: Boolean
}

input TableCommentFilterInput {
	postID: TableIDFilterInput
	commentID: TableStringFilterInput
	content: TableStringFilterInput
	addedTime: TableIntFilterInput
}

input TableFloatFilterInput {
	ne: Float
	eq: Float
	le: Float
	lt: Float
	ge: Float
	gt: Float
	contains: Float
	notContains: Float
	between: [Float]
}

input TableIDFilterInput {
	ne: ID
	eq: ID
	le: ID
	lt: ID
	ge: ID
	gt: ID
	contains: ID
	notContains: ID
	between: [ID]
	beginsWith: ID
}

input TableIntFilterInput {
	ne: Int
	eq: Int
	le: Int
	lt: Int
	ge: Int
	gt: Int
	contains: Int
	notContains: Int
	between: [Int]
}

input TablePostFilterInput {
	id: TableIDFilterInput
	postContent: TableStringFilterInput
	addedTime: TableIntFilterInput
}

input TableStringFilterInput {
	ne: String
	eq: String
	le: String
	lt: String
	ge: String
	gt: String
	contains: String
	notContains: String
	between: [String]
	beginsWith: String
}

input UpdateCommentInput {
	postID: ID!
	commentID: String!
	content: String
	addedTime: Int
}

input UpdatePostInput {
	id: ID!
	postContent: String
	addedTime: Int
}

schema {
	query: Query
	mutation: Mutation
	subscription: Subscription
}

  1. （与选项1相关）。在右侧面板的AppSync控制台的“架构”页面上，找到“帖子”并与“ comments（...）：CommentConnection”相对，单击“附加”，添加“ CommentTable”作为源，并在VTL中添加以下解析器代码：

在请求映射模板中：

    #set( $startFromTime = $util.defaultIfNull($context.args.startFromTime, 0) )
    {
        "version" : "2017-02-28",
        "operation" : "Query",
        "index" : "postID-addedTime-index",
        "query" : {
          "expression": "postID = :postID and addedTime > :startFrom",
            "expressionValues" : {
              ":postID" : { "S" : "$context.source.id" },
              ":startFrom" : { "N" : "$startFromTime" }
            }
        },
        "scanIndexForward":   #if( $context.args.sortDirection )
          #if( $context.args.sortDirection == "ASC" )
              true
          #else
              false
          #end
        #else
            true
        #end,

        #if( ${context.arguments.count} )
            ,"limit": ${context.arguments.count}
        #end
        #if( ${context.arguments.nextToken} )
            ,"nextToken": "${context.arguments.nextToken}"
        #end
    }

在响应映射模板中：

{
    "items": $utils.toJson($context.result.items)
    #if( ${context.result.nextToken} )
        ,"nextToken": "${context.result.nextToken}"
    #end
}
  1. （与选项2相关）。在右侧面板的AppSync控制台的“架构”页面上，找到“查询”并与“ postCommentsByAddTime（...）：PaginatedComments”相对，单击“附加”，添加CommentTable作为数据源，并在VTL中添加以下解析器代码：

在请求映射模板中：

{
    "version" : "2017-02-28",
    "operation" : "Query",
    "index" : "postID-addedTime-index",
    "query" : {
      "expression": "postID = :postID and addedTime > :startFrom",
        "expressionValues" : {
          ":postID" : { "S" : "${context.arguments.postID}" },
          ":startFrom" : { "N" : "${context.arguments.startFromTime}" }
        }
    }
    #if( ${context.arguments.count} )
        ,"limit": ${context.arguments.count}
    #end
    #if( ${context.arguments.nextToken} )
        ,"nextToken": "${context.arguments.nextToken}"
    #end
}

在响应映射模板中：

{
    "items": $utils.toJson($context.result.items)
    #if( ${context.result.nextToken} )
        ,"nextToken": "${context.result.nextToken}"
    #end
}

就是这样。

现在您可以使用以下所有查询：

query ListPosts {
  listPosts{
    items {
      id
      postContent
      ## all below arguments are nullable
      comments(startFromTime: 121111112222, count: 4
      ## default sortDirection is ASC, you can change it this way
      ## sortDirection: DESC
    ) {
        items {
          postID
          commentID
          content
          addedTime
        }
      }
    }
  }
}

query GetPost {
  getPost(id: "6548e596-d1ed-4203-a32f-52cfab8c9b20") {
    id
    comments (
    ## you can also add all three or any or none of these
    ## sortDirection: DESC,
    ## startFromTime: 189283212122
    ## count: 5
    ) {
        items {
        postID
        commentID
        content
        addedTime
      }
  }
  }
}

query GetCommentsByTime {
  postCommentsByAddTime(postID: "6548e596-d1ed-4203-a32f-52cfab8c9b20", startFromTime: 12423455352342, count: 2) {
    items {
      postID
      commentID
      content
      addedTime
    }
  }
}
相关问题
最新问题