Reporting on UnifaceFlow
Author: paul.chapelhow@b-plan.com (chapelhow)
Some of our customers like to use their 3rd party reporting tools on UnifaceFlow. However they have a problem with segmented filelds in the UnifaceFlow tables. We have had requests to get particular data that is stored in these fields, in particular the values of Parameters that are passed in/out of the tasks (URELDATA field of UWPRCI). These paramters usually contain customer specific data which can help them identify tasks.
As I see it there are 2 options to satisfy this request:
1. Do not use segmented fields in runtime data.
2. Publish the schema model so that we are able to access the data required before it gets into the segmented fields, though with only the 4 current runtime tables I am not sure this is a valid solution?.
Paul Chapelhow
3 Comments
Local Administrator
Hi Paul,
there are some very easy ways right out of the box:
- use the uniface /cpy combined with a /whr clause to convert these data to XLM from the commandline.
- write with Uniface a (web) service to deliver these data upon request.
- in your last step of each workflox, add a block to publish all the log data in whatever format you like (XML named by the ID of the workflow)
- ......
It's best practise to cover your data from direct outside access by some kind of service,
so this is not a workaround, but the best way to do keep (orthogonal) subsystems separate from each other.
Success, Uli
Author: ulrich-merkel (ulrichmerkel@web.de)
Local Administrator
Thanks,
We already provide reporting and history within our application that uses UnifaceFlow and we provide that through the Compuware API's, but as I said, it is some of our customers who have made investments in reporting tools, such as Business Objects etc, that want to create reports, but the creation of data 'Universes' relies on SQL access.
Paul
Author: chapelhow (paul.chapelhow@b-plan.com)
Local Administrator
Hi Paul,
i recently went through a couple of reporting tools and found some which can use CSV and other flat files as well.
On the other hand: what about providing a "report database" without the uniface obstacles
where you use XSLT etc. to load the data from XML files after you have generated them (see above).
Success, Uli
Author: ulrich-merkel (ulrichmerkel@web.de)