Өзгерту

Бөлісу құралы:


Create data maps for import

To import data into Microsoft Dataverse, you must provide the appropriate data maps.

You can download examples of data maps from Microsoft Downloads: DataImportMaps.zip.

You use data maps to map the data contained in the source file to the Dataverse table columns. You must map every column in the source file to an appropriate column. The data in the unmapped columns is not imported during the data import operation.

The data map is represented by the Data Map (ImportMap) table. You can create a new map by creating new records of this table. The map has a unique name that is contained in the ImportMap.Name column. You can specify the name of the import source for which this data map is created by using the ImportMap.Source column.

Column, list value, and lookup mappings

To map a column, a list value, or lookup value in the source file to a Dataverse column, use the following mappings:

Column Mapping

Maps a column in a source file to a Dataverse column. For column mapping, use the ColumnMapping table. You can use 1:1 (one-to-one) or 1:N (one-to-many) relationships between source and target columns. For example, you can map account address information to the billing and shipping addresses in an order.

List Value Mapping

Maps a list value in a source file to a Dataverse column of the OptionSetValue type. For list value mapping, use the List Value Mapping (PickListMapping) table.

If a value specified in the source file column is a list value, such as an OptionSetValue, Status, State, and Boolean, you must provide a list value mapping additionally to a column mapping. For example, map the "bill" and "ship" list values in the source file to the bill and ship values of the OptionSetValue type.

Lookup Mapping

Maps a lookup value in a source file to a Dataverse column of the EntityReference type. For lookup mapping, use the LookUpMapping table.

If the value specified in the source file references a table, you must provide a lookup mapping for this value. Use the LookupMapping.LookupSourceCode column to specify whether to search for the referenced table inside the source file or inside Dataverse. If you are using early bound types, you can use the LookupSourceType enumeration to set the lookup values. To search inside the source file, use the LookupSourceType.Source value. To search inside Dataverse, use the LookupSourceType.System value. For a list of the LookupSourceCode values, see the choice values for this table. To view the metadata for your organization, install the Metadata Browser solution described in Browse table definitions in your environment. You can also browse the reference documentation for tables in the Table Reference. You can provide multiple lookup mappings. The asynchronous transformation job processes all available mappings. It finds the referenced records and updates the parse table with the record unique identifiers. For more information, see Run Data Import.

Owner mapping

Use owner mapping to map a user specified in the source file to a user in Dataverse. For logging information, use the Dataverse user logon name. For owner mapping, use the OwnerMapping table.

Notes and attachments

Mapping for notes and attachments is handled differently from other tables. Notes and attachments are used to append additional information to a record in Dataverse. Notes are stored as text and attachments are stored as files in the Dataverse database.

To create a note in Dataverse, set the Annotation.IsDocument column in the annotation (note) table to false. To create an attachment, set IsDocument to true.

Use the following settings for mapping notes and attachments:

  • Set the ColumnMapping.SourceAttributeName column to "true" or "false". The "true" value indicates an attachment. The "false" value indicates a note.

  • Set the ColumnMapping.TargetAttributeName column to IsDocument.

  • Set the ColumnMapping.ProcessCode column to the ImportProcessCode.Internal value of the ImportProcessCode enumeration, if you are using early bound types. For a list of the ProcessCode values, see the choice values for this table.

    If the source data represents a note, map the text of the note to the Annotation.NoteText column. If you are working with Salesforce files, they are usually stored on the disk under unique identification numbers. To import an attachment, you must map a file identification number that is contained in the source file to the Annotation.DocumentBody column. The DocumentBody column stores the contents of the attachment.

    The import asynchronous job checks for mappings that have the source column name set to "true" and "false" to discover notes and attachments. If it finds an attachment mapping, it looks for the specified files on the disk and uploads the file contents as attachments into Dataverse. If a file is not found, an error is returned.

    If you do not provide mapping for an annotation (note) table, the import job generates a default mapping for the note.

Note

The maximum size of files that can be uploaded is determined by the Organization.MaxUploadFileSize property. This property is set in the Email tab of the System Settings in the Dataverse application. This setting limits the size of files that can be attached to email messages, notes, and web resources. The default setting is 5 MB. However, an attachment size cannot exceed the maximum HTTP request size (the default is 16MB).

Import and export data maps

You can export an existing data map to an XML file and import XML data mappings into Dataverse. To export a data map from Dataverse, use the ExportMappingsImportMap message using ExportMappingsImportMapRequest Class or the ExportMappingsImportMap Action . To import XML data mappings and create a data map in Dataverse, use the ImportMappingsImportMap message using the ImportMappingsImportMapRequest Class or ImportMappingsImportMap Action.

See Also

Import data
Prepare source files for import
Add transformation mappings for import
Configure data import
Run data import
Data import tables
Sample: Export and import a data map
Sample: Import data using complex data map