I investigated why Experts ETL was unable to generate data for a few persons who all had either R (Retired) or D (Deceased) empl_status values. My presumption was that we weren't handling those correctly, but we actually were. The actual reason seems to be that some HR person removed all the um_college values from their job data. We use those values as selection criteria in our pure_employee_eligible_job view, so now that view no longer includes data for these people. I also discovered that, while all rows in ps_dwhr_job have position_nbr values, some of them are just a single space. That was true for almost all of the data I was investigating. Turns out our ETL software handles those bogus values just fine, too. The end results are that we'll just ignore the errors for those people, since their jobs have all ended and they don't display in Pure anyway; and we now have additional tests that include all these strange values.
I investigated why Experts ETL was unable to generate data for a few persons who all had either
R(Retired) orD(Deceased)empl_statusvalues. My presumption was that we weren't handling those correctly, but we actually were. The actual reason seems to be that some HR person removed all theum_collegevalues from their job data. We use those values as selection criteria in ourpure_employee_eligible_jobview, so now that view no longer includes data for these people. I also discovered that, while all rows inps_dwhr_jobhaveposition_nbrvalues, some of them are just a single space. That was true for almost all of the data I was investigating. Turns out our ETL software handles those bogus values just fine, too. The end results are that we'll just ignore the errors for those people, since their jobs have all ended and they don't display in Pure anyway; and we now have additional tests that include all these strange values.