Profound Logic Forums

Is it possible to combine the Login and change password screens into one screen? The request is to merge them into one screen. As I understand the request, the log-in screen would display initially with a button to select change password. If the button is clicked, the log-in fields would disappear and the fields and button to change password would appear. A back button would be available to both screens.

I have the DDS for PUICHGPWD and PUISCREED in my library and merging the screens is simple, but the RPG is harder to locate.

The screens are driven from internal processes and cannot be changed/combined in that way. You can't make the sign on program call the change password program, for example, unless it does so automatically for expired password or password about to expire.

That said, you can add a call to the change password screen from your own application screens -- call program PUI0001203.

Use this prototype:
For user, pass the "current user profile" field (358-367) from the SDS. The other parameters are not required and can be omitted.

Thanks, this helps me stop chasing squirrels.

Can I bypass the existing logon and change password screens with a custom program that uses the APIs to handle these functions?

Yes. Profound UI supports running "anonymous" sessions, meaning you are not required to sign on.

For security, it has a database of programs that are allowed to be called in this way. This is maintained under Launch->Maintain Anonymous Programs. DB file where they are stored is PUI0001200.

Any program listed in there can be called up using:

/profoundui/start?pgm=mylib/mypgm

The program could then present its own "sign on" screen and do its own thing.

We took your responses and modified our requirements and everything is working, to a point. We modified the login and change password screens, but our QA tester discovered that if she logs in and immediately clicks on change password that the login screen is still visible underneath the change password screen.

My theory is that it is because the screens were changed to use images as the background rather than a panel. Can you tell me if there is any validity to this before I spend (billable) time testing my theory? You have my consent to refer to me as that lazy s.o.b. in Atlanta.

Using images vs. panels should have nothing to do with this. It is likely related to overlay/assume properties on the record format. I would check there.