I am getting a error message like
" All sort keys are not in a single top to bottom segment path"
Let me explain more..
I am preparing a dynamic report wherein a person can slelect columns at random from a given list and generate a report based on some where criteria.When I select one of these columns and try to generate the report I get this error. Also This is the only column that can be selected from that paricular table..
Any idea why this is happening?
The reason is because of the relationships that have been developed previously, particularly the Sort Criteria. (BY fields). Let's say that you have a host (root) record that contains an employee id, last name, first name, and some other 1-1 information . Now either through JOIN TO ALL, you have connected that employee with his/her education information. Another JOIN TO ALL connects the employee up with JOB HISTORY. Now, the way you have built those JOINS implies that education and job history have nothing to do with each other. Therefore, you cannot PRINT JOBINFO BY EMPLOYEE BY EDUCATION as you have said they are unrelated. That is the meaning of the error message. The alternatives are only to use BY fields that are common to all of the paths, in my example, fields from the root segment, or fields that are in a 1-1 relationship with that root (where you did a JOIN TO, not JOIN TO ALL.
If the files are of such a relationship that you can create a vertical join structure as I call it, you can avoid the message.
I call a horizontal join something like as follows:
JOIN KEY IN A TO KEY (OR ALL KEY) IN B AS J1
JOIN KEY IN A TO KEY (OR ALL KEY) IN C AS J2
A vertical join:
JOIN KEY IN IN A TO KEY IN B AS J1
JOIN B.KEY IN A TO KEY IN C AS J2
This only works if B has a key for every key in A. C could be missing which is 'short path' and I use SET ALL ON for those if you want data to show whether C info is present or not.
|Powered by Social Strata|