Discussions
Categories
Groups
Community Home
Categories
INTERNAL ENABLEMENT
POPULAR
PUBLIC CLOUD
PRIVATE CLOUD
Quick Links
MY LINKS
HELPFUL TIPS
Back to website
Home
Intelligence (Analytics)
Highlighting rule with dimension not in crosstab
h3llraid3r
Hi,
I got another sticky Problem that I can't solve...
My Data-Set is build up like that:
server-name,server-group,value,time
(For explanation: one server-group contains multiple servers)
This results in a Data Cube like that:
Dimension server-name
Dimension server-group
Dimension time
Summary value
My crosstab is build like that:
|time|server-group1|server-group2|server-group3|...
|yyyy|xxxxxxxxxxxxx|xxxxxxxxxxxxx|xxxxxxxxxxxxx|...
with the time-value as row indicator, server-group als column indicator and the values as fields to be summarized.
That works perfect so far.
The aggregation works fine for all server-groups, but I need a highlight rule that provides following:
- If one (or more) server in a server-group has a value of 90, mark the whole server-group red
- If the avg of a server-group is higher than 50, mark the whole server-group yellow
The second point works fine, because BIRT can process the values that are also shown in the value-field of the crosstab.
But I'm stuck with the first rule, because I can't find a possibility to create a rule that refers to a column/dimension that is not displayed in the cross-tab. I tried to add the server-name column in the crosstab and hide it, but that results in rows/columns that I don't want to have.
Someone got an idea?
Find more posts tagged with
Comments
mwilliams
You could create a list of the servers that have a value over 90 when fetching the data in your onFetch of your dataSet and then in your crosstab, you'd just check the server dimension value and if that server is listed, you'd highlight the row how you'd like. Let me know if that works for you.
h3llraid3r
I got that working yesterday, a little bit confusing, but the output is like I expected. For those who have the same problem, here is my solution:
I added the server-name-dimension to the column indication and grouped the values by time and server-name. That lead (as expected, but not as I wanted) to several columns in the report at the end. I solved that by hiding the value-fields and the server-name-field and setting the width of the value-field to 0 in. Then I created grand totals on row and column, adjusted the fields I want to aggregate on and everything was fine.
Looks a bit crazy, but who cares
mwilliams
Did you ever try just adding server-name as an attribute in your crosstab?
h3llraid3r
I tried that, sure. But it didn't work for me, because the aggregated values were wrong.
mwilliams
If I was understanding correctly before, I think the way I described might be able to avoid the blank areas you have to hide, but I could be wrong. It will involve some script, but not much, but if the way you did it works for you, then there's no reason to change it!