Hi folks,
I have a report request that I am working on but I am trying to wrap my mind around the best way to approach things, and I'd appreciate input.
"They" are trying to get a feel for all (ALL) the fields in our Remedy helpdesk app that aren't being used. The way *I* proposed to do this was through three nested SQL queries...
1. List all (views) in remedy database
2. List all fields in #1.
3. Counts for specific "not used" values for #2.
In my fevered mind, I would get the top 'n' values in #3 and append their counts, because we have several "non-null" values that still equate to "not used" -- blank (empty strings), 0's, "X", etc.
The man requesting the data says that he doesn't need that level of breakdown, so I will instead come up with conditional test and then just sum those.
The output app will be Crystal Reports, probably 9.0, but we currently have 8.0 fielded. The fact that this is a "run-once" report gives us leeway to use 9.0 if required. Heck, I don't care if the output is in Excel. Once we get it to work it will probably be run against diff't sites for limited time periods, so technically, it's not a "run-once", just "run-once-in-a-blue-moon".
This thing will probably run like a three-legged dog (or two-legged... it might not even hop).
Does anybody see any un-obvious flaws in this or have other input/suggestions?
Is there a better way? Are there factors I'm not taking into account? Anybody ever had to do something similar?
--
Marc Visconte
CSC
Lead RMS Developer
Crystal Reports
I have a report request that I am working on but I am trying to wrap my mind around the best way to approach things, and I'd appreciate input.
"They" are trying to get a feel for all (ALL) the fields in our Remedy helpdesk app that aren't being used. The way *I* proposed to do this was through three nested SQL queries...
1. List all (views) in remedy database
2. List all fields in #1.
3. Counts for specific "not used" values for #2.
In my fevered mind, I would get the top 'n' values in #3 and append their counts, because we have several "non-null" values that still equate to "not used" -- blank (empty strings), 0's, "X", etc.
The man requesting the data says that he doesn't need that level of breakdown, so I will instead come up with conditional test and then just sum those.
The output app will be Crystal Reports, probably 9.0, but we currently have 8.0 fielded. The fact that this is a "run-once" report gives us leeway to use 9.0 if required. Heck, I don't care if the output is in Excel. Once we get it to work it will probably be run against diff't sites for limited time periods, so technically, it's not a "run-once", just "run-once-in-a-blue-moon".
This thing will probably run like a three-legged dog (or two-legged... it might not even hop).
Does anybody see any un-obvious flaws in this or have other input/suggestions?
Is there a better way? Are there factors I'm not taking into account? Anybody ever had to do something similar?
--
Marc Visconte
CSC
Lead RMS Developer
Crystal Reports