Search Configuration for Embedded Search
This section describes the required configuration settings for using Embedded Search as a search tool in Personnel & Organization
.
For the Personnel & Organization
component, several Embedded Search templates (models) are delivered which you can extend and reassemble to your specific requirements. The Embedded Search models use connectors to interact actively with the TREX server.
For more background information, see Embedded Search.
The business function HCM, Personnel & Organization 03 (HCM_PAO_CI_3
) is activated. Only when this business function is active can you access the required Customizing settings.
For the Embedded Search, you need the product version TREX 710 Revision 55
or higher.
Embedded Search is available for Personnel & Organization
and offers the following features:
Extensive text search in object attributes and descriptions
You can define this in the Modeler for Search and Analytics
in Customizing for Personnel & Organization
under .
Fuzzy search for texts
Ranking of hit list (for example, hits found in personnel numbers and employee names are displayed before hits in other attributes)
Individual advanced search criteria based on user authorizations
For example, if a user is only allowed to see employees of personnel area US01 and US02, he or she would not see any other personnel areas when using the input help.
Default values for filters and advanced search criteria
To use Embedded Search, make settings in Customizing for Personnel & Organization
under .
To use the standard models for Embedded Search delivered for Personnel & Organization
, Employee Self-Service
, or Manager Self-Service
, perform the following steps to configure Embedded Search as search tool:
You have assigned the search category group SAP_STANDARD_ES (for Personnel & Organization
) or SAP_PEOPLE_ES
(for Employee Self-Service
) to the personalization object key Search Category Group
(HRPAO_SEARCH_CATEGORY_GROUP
) of the respective composite role assigned to your users.
Activate the BAdI implementations for infotype updates. For more information, see Customizing for Personnel & Organization
under .
In the Modeler for Search and Analytics
, you have to add the basic authorization objects to the authorization path of the main models (as described in section Implementation for Customer-Specific Search Categories
under 5. Define Authorizations for Search Models
).
You have to set up the TREX connection and create connectors for the models (for the models HRPAO_EMPLOYEE
, HRPAO_EMPLOYEE_VS
, HRPAO_POSITION_VS
, and HRPAO_ORG_UNIT_VS
).
For more background information, see Setting Up Embedded Search.
You also have to start the initial indexing of data and register the connectors for real-time indexing. For more information, see Customizing for Personnel & Organization
under .
If you have copied one of the main Embedded Search models (for object types employee, organizational unit, and position) to your own namespace or if you have added a new request to an existing main search model, make the following Customizing settings:
Define Search Categories
You define a search category in Customizing for Personnel & Organization
under in view V_T77PAOSCAT
. To define a new search category, copy one of the delivered search categories and maintain the text, object type and provider type. For Embedded Search, enter the provider type SAP_ES
.
Assign Embedded Search Models to Search Categories
You assign your Embedded Search model to the new search category in Customizing for Personnel & Organization
under in view V_T77PAOSSPES
. Here, you have to add a new entry for your search category (defined in the first step) with Search Request Type
= Search Request. Assign the Embedded Search model, the Root Node
name and the Request
.
Note
If you have created a virtual model to enable the “value suggest” functionality, assign the virtual model to the search category. The Search Request Type
is Value Suggest Request in this case.
Group Search Categories
You create a new search category group and assign the newly created search category to this category group in Customizing for Personnel & Organization
under in view cluster VC_T77PAOSCATGRP
:
For the Web Dynpro Application ‘PAOM Master Data Application’, you assign your newly created search category with the Search Context
= Master Data Application.
For the UI5 Search Lane, you assign your newly created search category with the Search Context
= Full Search Page.
The following fields are only relevant for the Search Lane on the landing page:
No Link: With this indicator you can enable or disable the link on the search result objects in the Search lane
The following fields are used for launching profile pages from the Search lane:
Role: In this field you specify the launchpad role of the application to be launched
Instance: In this field you specify the launchpad instance of the application to be launched
Application Alias: In this field you specify the application alias of the application to be launched, based on the role and instance entered above.
Assign Search Category Group to PFCG Role
You have to assign the newly created search category group to the personalization object HRPAO_SEARCH_CATEGORY_GROUP
. This personalization object key is contained in the standard composite PFCG role HR Professional.
Define Authorizations for Search Models
The authorization concept for Embedded Search models is designed for ABAP basic authorization objects and HR structural authorizations. The SAP standard models for Personnel & Organization
include structural authorizations, but not ABAP basic authorization objects since authorization concepts differ considerably from customer to customer.
Structural Authorizations
The structural authorizations are assigned with the models HRPAO_STRUCAUTH_USR
and HRPAO_STRUCAUTH_OBJ
to the main models for each object type. In the standard delivery no grouping is done except for users to whom no structural authorization profile is assigned. It is highly recommended to group users with similar authorizations, especially for the SAP_PEOPLE_ES
category. Otherwise, the performance of the indexing and the size of the indices is problematic. The grouping has to be implemented with an implementation for the Business Add-In (BAdI) HRPAO_B_SEARCH_ES_HRVIEW
.
Basic Authorizations
Depending on your authorization concept, the basic authorization objects such as P_ORGIN
for employees have to be maintained for the Embedded Search models. For the object types position, organizational unit and job, the authorization object PLOG
is already checked (for infotype Object
(1000)) as part of the structural authorization indexing.
To assign basic ABAP authorizations, proceed as follows:
Start the Modeler for Search and Analytics
by using transaction ESH_MODELER
.
Select the model for which you would like to add ABAP authorization objects and choose Edit Model
.
On the Node Response
tab, select the node.
Go to the Authorization
tab and choose ABAP Auth. Objects
.
On the ABAP Auth. Objects
tab, add the required ABAP authorization object and define the fields of the authorization object. The authorization field values can be defined as fixed values by selecting. AUTHC
: R (reading authorization). If the value is only known at runtime, you can define a path (by using the Select Path
button) to refer to a target node attribute (for example for the fields from infotype Organizational Assignment
(0001) of authorization object P_ORGIN
).
Note
You have to import the authorization object first by using the Import
button.
Define the logical conjunction for the Check ID of the authorization object.
Choose Save
.
Note
If you use a high level of grouping, you could also implement the basic authorization check in the BAdI implementation of the model HRPAO_STRUCAUTH_OBJ
. This is not recommended if a high number of checks are executed because the users are not grouped which would cause performance issues. Keep in mind that the user can only see new objects after the delta indexing for the structural authorization model has run.
If you assign the fields of the authorization object (such as P_ORGIN
) to the node ORG_ASSIGNMENT
, extended periods of responsibility are not taken into account. If you have this requirement and you are not able to group the users to a degree that allows basic authorization checks during the indexing of the structural authorization model, a solution is to have a separate node for infotype 0001 where the fields BEGDA
and ENDDA
are recalculated during indexing (BEGDA
= BEGDA
+ ADAYS
, ENDDA
= ENDDA
+ ADAYS
). Then, you assign the fields of the basic authorization object to the fields of this node instead of the node ORG_ASSIGNMENT
. As in the node ORG_ASSIGNMENT
, the fields BEGDA
and ENDDA
have to have the semantics Valid from
and Valid to
. The new node would only be used for authorization purposes.