.create ingestion mapping command
Applies to: ✅ Microsoft Fabric ✅ Azure Data Explorer
Creates an ingestion mapping that can be associated with a specific format and a specific table or database.
If a mapping with same name in the given scope already exists, .create
will fail. Use .create-or-alter
instead.
Permissions
At least Database Ingestor permissions are required to create a database ingestion mapping, and at least Table Ingestor permissions are required to create a table ingestion mapping.
Syntax
.create
table
TableName ingestion
MappingKind mapping
MappingName MappingFormattedAsJson
.create
database
DatabaseName ingestion
MappingKind mapping
MappingName MappingFormattedAsJson
Learn more about syntax conventions.
Parameters
Name | Type | Required | Description |
---|---|---|---|
TableName | string |
✔️ | The name of the table. |
DatabaseName | string |
✔️ | The name of the database. |
MappingKind | string |
✔️ | The type of mapping. Valid values are csv , json , avro , parquet , and orc . |
MappingName | string |
✔️ | The name of the mapping. |
MappingFormattedAsJson | string |
✔️ | The ingestion mapping definition formatted as a JSON value. |
Note
- Once created, the mapping can be referenced by its name in ingestion commands, instead of specifying the complete mapping as part of the command.
- If a mapping with the same name is created in both the table scope and the database scope, the mapping in the table scope will have a higher priority.
- When ingesting into a table and referencing a mapping whose schema does not match the ingested table schema, the ingest operation will fail.
Examples
.create table MyTable ingestion csv mapping "Mapping1"
'['
' { "column" : "rownumber", "DataType":"int", "Properties":{"Ordinal":"0"}},'
' { "column" : "rowguid", "DataType":"string", "Properties":{"Ordinal":"1"}}'
']'
.create database MyDatabase ingestion csv mapping "Mapping2"
'['
' { "column" : "rownumber", "DataType":"int", "Properties":{"Ordinal":"0"}},'
' { "column" : "rowguid", "DataType":"string", "Properties":{"Ordinal":"1"}}'
']'
Output
Name | Kind | Mapping | Database | Table |
---|---|---|---|---|
mapping1 | CSV | [{"Name":"rownumber","DataType":"int","CsvDataType":null,"Ordinal":0,"ConstValue":null},{"Name":"rowguid","DataType":"string","CsvDataType":null,"Ordinal":1,"ConstValue":null}] |
MyDatabase | MyTable |
mapping2 | CSV | [{"Name":"rownumber","DataType":"int","CsvDataType":null,"Ordinal":0,"ConstValue":null},{"Name":"rowguid","DataType":"string","CsvDataType":null,"Ordinal":1,"ConstValue":null}] |
MyDatabase |
Example: .create ingestion mapping with escape characters
.create table test_table ingestion json mapping "test_mapping_name"
'['
'{"column":"timeStamp","path":"$[\'timeStamp\']","datatype":"","transform":null},{"column":"name","path":"$[\'name\']","datatype":"","transform":null},{"column":"x-opt-partition-key","path":"$[\'x-opt-partition-key\']","datatype":"","transform":null}'
']'
Related content
- For detailed descriptions of various ingestion mapping formats such as CSV, JSON, Avro, Parquet, and Orc, see Data mappings.