Mapping configurations¶

This topic describes the mapping configurations when importing data into NebulaGraph Database Developer Edition, which include node mapping configurations and edge mapping configurations.

Node mapping configurations¶

The node mapping configurations are specified in the section sources.nodetypes of the configuration file.

Sample configuration¶

sources: [
    {
//        type:"hdfs"
//        format:"csv"
//        header:false
//        separator:","
//        path: "hdfs://127.0.0.1:9000/import_movie/sub_movie.csv"
        nodetypes: [
            {
                name: "movie"
                sourceFields:[movieid, name]
                nebulaFields:[id, name]
                batchSize: 1000
                enableRealTimeQPSLog: true
                mode: "INSERT IGNORE"
            }
        ]
        edgetypes: []
    }
]

Configuration options¶

Option	Description	Type	Required	Default value
`name`	The name of the node type in the graph to which data is written.	String	Yes	N/A
`sourceFields`	The list of names of fields in the data source. The values of these fields are written as values of the properties of the node type. When importing data from NebulaGraph 3.x, set `sourceFields` to the list of names of properties from which data is retrieved. Besides, you can also specify `_vertexId` to retrieve vertex IDs; specify `_srcId`, `_dstId`, and `_rank` to retrieve source vertex IDs, destination vertex IDs, and ranks of edges respectively.	List	Yes	N/A
`nebulaFields`	The list of names of properties of the node type. The order of the property names must correspond to the order of the field names in `sourceFields`.	List	Yes	N/A

For details on the supported data types of the properties in nebulaFields and the format requirements on fields in sourceFields by special data types, see Supported data types.

Common write configurations¶

The following configuration options can also be specified as global settings in the section writeCommon of the configuration file: batchSize, enableRealTimeQPSLog, and mode. Note that the configurations on specific node types override the global settings.

For more information about their descriptions, see Common write configurations.

Edge mapping configurations¶

The edge mapping configurations are specified in the section sources.edgetypes of the configuration file.

Sample configuration¶

sources: [
    {
//        type:"hdfs"
//        format:"csv"
//        header:false
//        separator:","
//        path: "hdfs://127.0.0.1:9000/import_movie/sub_user_watched_movies.csv"
        nodetypes: [ ]
        edgetypes: [
            {
                name: "watch"
                sourceKeys: [userid]
                nebulaSrcKeys: [id]
                targetKeys: [movieid]
                nebulaDstKeys: [id]
                sourceFields:[rate]
                nebulaFields:[rate]
                batchSize: 1000
                enableRealTimeQPSLog: true
                mode: "INSERT IGNORE"
                ignoreHangingEdge: false
            }
        ]
    }
]

Configuration options¶

Option	Description	Type	Required	Default value
`name`	The name of the edge type in the graph to which data is written.	String	Yes	N/A
`sourceKeys`	The list of fields in the data source mapped as the primary key of the source node type of the edge type.	List	Yes	N/A
`nebulaSrcKeys`	The list of names of properties that serve as the primary key of the source node type of the edge type. The order of the property names must correspond to the order of the field names in `sourceKeys`. This option is optional if the primary key of the source node type consists of only one property.	List	No	N/A
`targetKeys`	The list of fields in the data source mapped as the primary key of the destination node type of the edge type.	List	Yes	N/A
`nebulaDstKeys`	The list of names of properties that serve as the primary key of the destination node type of the edge type. The order of the property names must correspond to the order of the field names in `targetKeys`. This option is optional if the primary key of the destination node type consists of only one property.	List	No	N/A
`sourceFields`	The list of names of fields in the data source. The values of these fields are written as values of the properties of the edge type. When importing data from NebulaGraph 3.x, set `sourceFields` to the list of names of properties from which data is retrieved. Besides, you can also specify `_vertexId` to retrieve vertex IDs; specify `_srcId`, `_dstId`, and `_rank` to retrieve source vertex IDs, destination vertex IDs, and ranks of edges respectively.	List	No	N/A
`nebulaFields`	The list of names of properties of the edge type. The order of the property names must correspond to the order of the field names in `sourceFields`.	List	No	N/A
`ignoreHangingEdge`	Whether to ignore edges that do not connect to nodes on either or both sides, known as dangling edges. Options: `true`: No error is reported. The write operation for a dangling edge is ignored, but it is still counted as a successful write. `false`: An error is reported. The write operation for a dangling edge fails.	Bool	No	`false`

For details on the supported data types of the properties in nebulaSrcKeys,nebulaDstKeys, and nebulaFields, as well as the format requirements on fields in sourceKeys, targetKeys and sourceFields by special data types, see Supported data types.

Common write configurations¶

The following configuration options can also be specified as global settings in the section writeCommon of the configuration file: batchSize, enableRealTimeQPSLog, mode, and ignoreHangingEdge. Note that the configurations on specific edge types override the global settings.

For more information about their descriptions, see Common write configurations.

Supported data types¶

NebulaGraph Exchange supports mapping source data to properties of the following data types:

Data type	Description
Numeric	Includes `INT`, `DOUBLE`, `FLOAT`, and `DECIMAL`.
String	Includes `STRING`.
Boolean	Includes `BOOL`.
Date and time	Includes `DATE`, `DURATION`, `LOCAL DATETIME`, `LOCAL TIME`, `ZONED DATETIME`, and `ZONED TIME`.
List	`LIST<E>`, where `E` can be `INT`, `DOUBLE`, `FLOAT`, `STRING`, or `BOOL`. Import format: Starts with `[` and ends with `]`, without the `LIST` prefix. Elements are separated by commas. For example: `[1, 2, 3]` or `["1", "2", "3"]`.
Set	`SET<E>`, where `E` can be any predefined data type or constructed data type other than records and paths. Import format: Starts with `{` and ends with `}`, without the `SET` prefix. Elements are separated by commas. For example: `{1, 2, 3}`.
Map	`MAP<key_type, value_type>`, where keys and values can be any predefined data type or constructed data type other than records and paths. Import format: Starts with `{` and ends with `}`, without the `MAP` prefix. Uses `key:value` pairs separated by commas. For example: `{'a':123,'b':true}`.
Vector	`VECTOR<n, FLOAT>`, where `n` is the number of dimensions, ranging from 1 to 2,147,483,647. Import format: Starts with `[` and ends with `]`, without the `VECTOR` prefix. Elements must be numeric values without quotes, and the number of elements must be consistent across rows. For example: `[1.0, 2.0, 3.0]`.
Geospatial	Includes `POINT`, `LINESTRING`, and `POLYGON`. Values must be in WKT format. For example: `POINT(37.7749 -12.4194)`, `LINESTRING(1 2, 2 4, 4 8)`, `POLYGON((0 0, 0 10, 10 10, 10 0, 0 0))`.