In UC4/Automic Application Manager 9.4.3 (not Workflow), I have a process flow shared among several user teams which first uploads a CSV file into the database via SQLLDR. But, the users often typo the filename - which of course must be exact. In that case, the SQLLDR job component aborts, and IT (my team) is notified to clear it. We then have to contact the user and let them know that they have a typo in the file name.
We have other process flows which use the File data type which allows the user to select a file from a defined directory, which works great for those flows. The problem I have using the File data type for this process flow is that each user team uses a different directory on the shared drive which is samba-mounted to a directory on the server, where UC4 sees it as 'dataload'. I can set the File data type directory to the parent folder, but the user team folders are in subdirectories (e.g. /dataload/stu/Dual_Enroll for the K-12 team, /dataload/stu/Admissions for the Admissions team, etc.). If the file data type is looking at /dataload/stu, the subfolders are not presented to be selected or drilled down into.
I am considering writing a bash shell script, but am not clear how this could interact with the user to select files, although if it could do so, it could of course traverse the file structure.
I have also considered having several prompts which would be attached to different directories, and hopefully security could control which prompt each team sees, so that only one directory could be viewed and only one file selected. I suppose in that case, there would need to conditions on the component to set the directory based on which prompt actually contains a value.
I have tried using a File data type prompt set to the parent folder /dataload/stu, but it only presents files at that level, and does not allow the user to drill down into their team's own folder.
Has anyone done this, and if so how? Or, does anyone have suggestions for how I might perform this? I am loathe to create a new process flow for each team, just to have a different directory for each team to search. I am hopeful that is not the only way to accomplish having the different teams share the process flow, without being alerted when the SQLLDR job aborts due to a mistyped file name.