Originally posted on August 28 2020 at 21:22 UTC
1. What is user import inline preprocessor?
User import inline preprocessor generates user attributes that requires some computation based on business requirements when you do not have direct one to one mapping between HR User attribute to Saviynt user attribute. Cleaning the identity data ensures that clean and complete data is entered into Identity Repository as part of user imports.
2. What are the advantages of using user import inline preprocessor?
a)Can be leveraged to generate user attributes that require computation logic and cannot be achieved using out of the box generation rules. Inline preprocessor allows to write the computation logic as part of the user import job itself and thereby eliminating the need to execute any separate job for it
b) Replaces database functions, stored procedure and custom queries to compute user attributes that may lead to database deadlocks
3. Can you do inline preprocessing in full and incremental user imports?
Yes
4. Where to configure inline preprocessor for user imports?
Connector based imports
MODIFYUSERDATAJSON in Connection based HR user imports
File based imports
User Pre-processor config in UI Upload - Select File to Upload User pop up
MODIFYUSERDATAJSON in Schema based import sav file
5. How does inline preprocessor work?
Inline processor starts only if MODIFYUSERDATAJSON is populated in user import process
MODIFYUSERDATAJSON has 3 configurations that should be defined
| Element Name | Purpose |
| ADDITIONALTABLES | This configuration is used to define the dataset required for execution of queries defined in “PREPROCESSQUERIES” section.
|
| COMPUTEDCOLUMNS | This is the list of Identity attributes which needs to be computed as part of Identity import process.
|
| PREPROCESSQUERIES | This configuration should contain a MySQL query (with desired logic) for each of the computed column identified in the “COMPUTEDCOLUMNS” section. This configuration is where you can invoke a preconfigured custom jar for any complex logic to build “COMPUTEDCOLUMNS” |
6. Some samples of MODIFYUSERDATAJSON
Sample 1
#To save Azure AccountID of user in user customproperty5
{
"ADDITIONALTABLES" :
{ "USERS" : "SELECT userkey,username FROM USERS" ,
"ACCOUNTS" : "SELECT accountid,accountkey FROM ACCOUNTS where endpointkey=10",
"USER_ACCOUNTS" : "SELECT userkey,accountkey FROM USER_ACCOUNTS where accountkey in (select distinct accountkey from accounts where endpointkey=10"
},
"COMPUTEDCOLUMNS" :
[ "customproperty5" ],
"PREPROCESSQUERIES" : [
"update NEWUSERDATA set customproperty5=(select ac.accountid from currentuser_accounts uac inner join currentusers u on uac.userkey=u.userkey innerjoin currentaccounts ac on ac.accountkey=uac.accountkey where ac.endpointkey=10)"
]
}
Sample 2
#To compute employeeclass from employeetype
{
"ADDITIONALTABLES" :
{ "USERS" : " SELECT employeetype FROM USERS " },
"COMPUTEDCOLUMNS" :
[ " employeeclass" ],
"PREPROCESSQUERIES" : [
"update NEWUSERDATA set employeeclass= Employee where employeetype like %ownemployee%", "update NEWUSERDATA set employeeclass= Contractor where employeetype like %contractor%", "update NEWUSERDATA set employeeclass= Partner where employeetype like %partner%", "update NEWUSERDATA set employeeclass= Intern where employeetype like %intern%",
]
}
