I tried doing a bulk upload on Oracle with the following file:
first_names, last_name, email, id, type, access_level, guest
Gabriel, Marklund, gabriel@marklund.com, gabriel, student, full, f
Mohan, Pakkurti, mohan@pakk.com, mohan, staff, full, f
Lars, Pind, lars@pind.com, lars, professor, limited, f
Eva, Marklund, eva@marklund.com, eva, external, full, t
If I omit the id field the upload will break since no default value is
set for that field:
can't read "row(id)": no such element in array
while executing
"dotlrn::user_add -user_id $user_id -id $row(id) -type $row(type)
-can_browse"
("full" arm line 2)
invoked from within
"switch -exact $row(access_level) {
limited {
That seems easy enough to fix. However when with the id field included
I get (after having made the proc oacs::csv_foreach not catch its
error messages so that I can see the stack trace, do we really need
that catch?):
Object type "dotlrn_staff_profile_rel" does not exist
while executing
"error "Object type "$object_type" does not exist""
(procedure "package_instantiate_object" line 15)
invoked from within
"package_instantiate_object -creation_user $creation_user
-creation_ip $creation_ip -start_with "relationship" -form_id
$form_id -extra_vars $extr..."
invoked from within
In the database I only have the similar sounding object types:
dotlrn_professor_profile_rel
dotlrn_student_profile_rel
dotlrn_admin_profile_rel
dotlrn_external_profile_rel
dotlrn_user_profile_rel
It seems the staff user class is not yet implemented in the database?
Indeed, changing staff to student in my CSV file the upload succeeds.
It's good that you are able to add the users to a group (i.e. a class)
directly after the bulk upload. You can only add the uploaded users to
one group though it seems.
