Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi Guys,
I have a few synthetic keys, from what i read on this community site, it has been advised to get rid of these due to circular referencing and looping that can occur. I am not sure how I can go about doing this, without impact the level of flexibility I wanted my EU to have. Below is a screendump of my relationship table, any help would be greatly appreciated
To answer your questioms
1) The difference between agent and skillgroup is the dimension or granularity. The fields are pretty much the same, except one is with agent information and the other is for the entire operation for the day by half hour. My limited knowledge on QV and its functions oftens means I find very nasty ways around my problems
2) Remove the histogram chart or remove the histogramdata load???
3) Will add the calendar, havent really tried using it before but will see what improvement it makes
4) Am busy merging data as we speak, hopefully I do this right
Thanks so much to all so far for the time you have dedicated to my problem. Much appreciated
Regards,
Byron
Ok, i think I've done everything you asked for. The tables look a lot better but I still need to check if everything is going to work properly. Havent fixed the charts or tables or anything after changes were made. let me know if this is looking right
There goes the synthetic keys!
I still would leave the histogram data out of the datat model, and solve it by calculating the result in a chart object.
Regarding the table merge I was thinking more in line of keeping the original names for the fields, so that you used the table more efficiently. There is no need to differentiate the content on a field name basis as you can separate it based on the AgentID value, since all rows with Null will be rows from the non-agent group.
Based on what you want to present in a chart object, you might need to add simple set expressions or use Aggr() in order to get your datat grouped properly. Try out making your charts and then if you run in to trouble attach the doc and we'll see how to guide you further.
Thanks Toni
Will leave out the histogram data and see if I can somehow get it to count properly without the need for an additional view from SQL.
Take care all, and thanks once again
Byron