To create a Glue table

Step 1: Create a Glue table definition

Run the transaction /DVD/GL11 to open Data Dictionary.
Enter the name of a new table and click Create.
Enter the description of the table in Short Description.
Under the Attributes tab fill in the following fields based on your preference:
- Enable update (optional) – simulates the update operation for tables on storages which don't support the update operation, for more information see Enable update.
Choose, if you prefer a transparent or a binary storage for the target table:
- For a transparent storage, enter its ID in the field StorageID (Load).
- For a binary storage:
  1. Enter its ID in the field StorageID (Load).
  2. Select File storage options.
  3. Fill in the following fields:
    - Container name – directory where files generated by Glue are stored (obligatory parameter). You can also specify a path to the container by using slash sign as a separator between the containers.
    - Delimiter type – character which separates values in generated .csv files. This character can be selected from F4 help (obligatory parameter).
    - Put values into quotes – option for putting values into the quotes (use when there is a delimiter character present in data).
    - Use extending escaping (optional) – replaces escape characters, such as newline, backspace, tabulator, etc. by the 'space' character. This can be customized in the table /DVD/RL_EXT_ESC and a default set of values can be filled by transaction /DVD/RL_DEFAULT_ESC.
    - Include header (optional) – includes a header line at the beginning of the file with data.
    - One file extraction (optional) – keeps all data in one file (data from one extraction is stored in only one file).
    - Compress file(optional) – compresses the file into a gzip. This feature cannot be selected if the parameter One file extraction is enabled and vice versa.
    - Compression level - compression level of gzip file. Available values can be selected from F4 help. Default value is 5. The parameter will appear only if the Compress file option is checked.
Select Storage Management. It will navigate you to Storage Management setup where you can see and manage (if you have the rights to do so) storages.

Step 2: Insert fields manually or import them from an existing SAP DDIC table, an InfoProvider, a file or a BEx Query/View

You can either manually insert fields or import them from an existing SAP DDIC table, an InfoProvider, a file or a BEx Query/BEx Query View. You may define fields (columns) of a new table.

a. Import fields from an existing SAP table

To generate the field structure of a Glue table based on existing SAP DDIC table, an InfoProvider, a file or a BEx Query/View

Select the Fields tab.
Select Import Fields. (Field names, data elements, types, lengths and descriptions are imported from the selected source.)
1. To import fields from a table, mark From table.
2. To import fields from an InfoProvider, mark From InfoProvider.
3. To import fields from a file, mark From File.
4. To import fields from a BEx Query or a BEx Query View, mark Bex Query/View. The program then automatically finds out which type of object is specified in the corresponding parameter. It finds out whether it is a BEx Query or a BEx Query View and it translates query elements to the DDIC fields form.
  
  If you cannot import fields from BEx Query or BEx Query View due to unsupported structure (e.g. Characteristics in columns (axis X), enabled Hierarchy or multiple structures in rows (axis Y)), adjust your query in BEx Query Analyzer or Query Monitor (TCODE RSRT), save it as Query View and use it to import the fields.
You can use standard SAP field names or the "meaningful" ones. To use the meaningful fields select Use meaningful fields. For more information, see Meaningful fields.

An example of imported fields.

b. Insert fields manually

If you choose to manually insert fields

Select the Fields tab.
To insert or append a new field, click Insert or Append row.
Fill in:
- Field name – name of the field
- Key field – mark, if it is a key field

The total number of characters for the primary Key field value is restricted to 338.

- Data element
- Data Type – automatically filled according to the Data element or you can set it up manually by clicking Data Type/Element specification . You can then manually specify the Data Type, Length, Decimal Places, Short text.
  
  The following data types are not supported when you create a Glue table: LRAW, RAW, RSTR, VARC.
- Partitioning
  You may set the partitioning for fields, based on which data is distributed into blocks during the extraction. This step enables you to modify the data grouped in these blocks. By default the partitioning is set on the GLREQUEST field.
  
  If you use a Hive or an Impala storage, please keep in mind which fields you choose for Partitioning. Selecting too granular data for partitioning may decrease the performance and the stability of the whole cluster.
- Reference Fieldname – the reference field for Data enrichment, containing the name of the field with original values. For more information see Data Enrichment.
  To delete a row, highlight the particular row and click Delete Row. You may view following ALV operations:

4. Save. (You can save the table anytime during your work.) All actions you made so far are performed on the SAP side and have no effect on Hadoop system.

5. If the table was not saved before, the system will ask you to fill in the Package.

6. Activate. Only after you activate the Glue table, it is created on an external storage (for example Hadoop). The activation of a Glue table doesn't influence other Glue components.

Browser not supported