Importing Data from SAP: Difference between revisions

From QPR ProcessAnalyzer Wiki
Jump to navigation Jump to search
No edit summary
Line 16: Line 16:
* OBJECT: S_TABU_DIS (= What tables we can read)  
* OBJECT: S_TABU_DIS (= What tables we can read)  
** ACTVT = 03 (Display)
** ACTVT = 03 (Display)
** DICBERCLS = * (all tables) OR individually named tables that QPR PA is allowed to read
** DICBERCLS = * (all tables), or individually named tables that QPR ProcessAnalyzer is needs to read
   
   
User created for the extractions can be a system user without any access to SAP GUI.
User created for the extractions can be a system user without any access to SAP GUI.

Revision as of 00:04, 2 March 2023

Introduction

Data can be extracted from SAP R/3 using its RFC interface. In QPR ProcessAnalyzer the extraction is done with the expression language ExtractSap function or SQL scripting --#ImportSapQuery command. In addition to the RFC interface, data can be fetched from SAP HANA with its ODBC connection using the expression language ImportODBCSecure function or SQL scripting ImportOdbcQuery command. Note that the RFC interface also available in SAP HANA.

For the RFC interface, QPR ProcessAnalyzer is using SAP NetWeaver RFC Library 7.50 provided by SAP allowing developers to connect to remote-enabled function modules in SAP. In SAP, QPR ProcessAnalyzer is calling the RFC (Remote Function Module) BBP_RFC_READ_TABLE that is available in all modern SAP systems to read individual tables. This setup requires no installations of 3rd party components into the SAP. All that is needed is a user, password and right authorizations for that user.

The connection can be made from the QPR ProcessAnalyzer server as long as all firewall rules are set up correctly to access the SAP, or via QPR ScriptLaucher which can be installed into an on-premise machine inside company external firewall where SAP can be reached. In addition, QPR ScriptLauncher needs have a connection QPR ProcessAnalyzer server. Please find instructions on how to install QPR ScriptLauncher.

Authorizations / Requirements in SAP side

If the BBP_RFC_READ_TABLE is not available in SAP, the older RFC_READ_TABLE can be used as well. If you have created custom versions of BBP_RFC_READ_TABLE those can be used as well, as long as the Remote Function Module's interface is the same as in BBP_RFC_READ_TABLE.

In order to establish a connection to the RFC BBP_RFC_READ_TABLE, a user needs to be created in the SAP with following authorization objects:

  • OBJECT: S_RFC (= What functions we can call)
    • ACTVT = 16 (execute)
    • RFC_NAME = RFC1, RFCPING, RFC_METADATA, SYST, SDTX, BBPB
    • RFC_TYPE = FGUR (Function Group)
  • OBJECT: S_TABU_DIS (= What tables we can read)
    • ACTVT = 03 (Display)
    • DICBERCLS = * (all tables), or individually named tables that QPR ProcessAnalyzer is needs to read

User created for the extractions can be a system user without any access to SAP GUI.

Note: if RFC call to the BBP_RFC_READ_TABLE can be established with other authorization settings, that is enough for QPR ProcessAnalyzer to successfully extract data.

Data conversion in RFC interface

Data extracted from SAP using the RFC interface has specific data types. Data will be converted into respective data types in QPR ProcessAnalyzer as follows:

SAP data type Expression language:
ReadSap function
SQL scripting:
--#ImportSapQuery
I (integer) long INT
F (floating point number) double FLOAT
P (fixed point number, packed number) double FLOAT
D (date) datetime DATETIME2
T (time) timespan TIME
C (text) string NVARCHAR(length) or
NVARCHAR(max) if length is greater than 8000
N (numeric text) string NVARCHAR(length) or
NVARCHAR(max) if length is greater than 8000
X (binary data) byte[] VARBINARY(length) or
VARBINARY(max) if length is greater than 8000

Conversions are determined by the ConvertDataTypes setting for each data type separately. If no conversions are explicitly set, --#ImportSapQuery (in SQL scripting) does not make conversions (to maintain backwards compatibility), and ExtractSap function (in expression language) will make all possible conversion. When there is no conversion defined, data is converted to strings as defined by the table above.

Date (D) and time (T) type of fields are usually related to the same timestamp and thus they are bound together as follows: If T type of column is defined directly after the the D type of column, the time part in the T column is added to the D column, and no new column is created for the the T column. This is done only if both D and T types are among the converted types.

In the C datatype, spaces from the beginning and end are removed (trimmed).

More information about SAP data types: https://help.sap.com/saphelp_nwpi71/helpdata/en/fc/eb2fd9358411d1829f0000e829fbfe/content.htm?no_cache=true.