Hi,
being new to Clover (and ETL in general) I have perhaps a basic question.
I'm trying something out in the graphical interface, in order to be able to use this with Oracle Identity Analytics.
What I try to do:
I have as input a flat text file with 2 types of entries immediately after one another:
U;usernameA;field2;...field8
U;usernameB;field2;...field8
R;usernameA;field2;...;field5
R;usernameA;field2;...;field5
R;usernameB;field2;...;field5
The U records are user names, the R ones responsibilities attributed to these users.
What I need to get in the end are User records, with the responsibilities attributed to them. What I try to do is first isolate the U records and the R ones and then combine them.
When I try to filter the data with metadata assigned on the correct edge, I get an error when I run the graph because the number of records is different.
I then tried to have a separate metadata especially for the filter (with only the first character filtered out), and then after the filter try to apply the full metadata I get an error that I the metadata is different. Well, of course it is.
How could this be done?
Thanks in advance,
Rik
being new to Clover (and ETL in general) I have perhaps a basic question.
I'm trying something out in the graphical interface, in order to be able to use this with Oracle Identity Analytics.
What I try to do:
I have as input a flat text file with 2 types of entries immediately after one another:
U;usernameA;field2;...field8
U;usernameB;field2;...field8
R;usernameA;field2;...;field5
R;usernameA;field2;...;field5
R;usernameB;field2;...;field5
The U records are user names, the R ones responsibilities attributed to these users.
What I need to get in the end are User records, with the responsibilities attributed to them. What I try to do is first isolate the U records and the R ones and then combine them.
When I try to filter the data with metadata assigned on the correct edge, I get an error when I run the graph because the number of records is different.
I then tried to have a separate metadata especially for the filter (with only the first character filtered out), and then after the filter try to apply the full metadata I get an error that I the metadata is different. Well, of course it is.
How could this be done?
Thanks in advance,
Rik
-
If I understand right you have a single file with two different data records (user information and resposibility information).
For this purpose we have special reader component - Complex Data Reader http://www.cloveretl.com/documentation/UserGuide/topic/com.cloveretl.gui.docs/docs/complexdatareader.html
After few minutes I was able to prepare simple graph which reads your desired data format. Check out the attached project.
Please sign in to leave a comment.
Comments 1