Clarity

Hi,

Is there any way wherein i can set Time Scaled Virtual Column in a portlet. I need to built a report something like this,

Regards,

Samad Abdul Aleem

Get the data from the timeslices. If it is a custom TSV define the slicerequest for the first.

Your example does not necessarily look like a TSV to me. It does just look like a multi-dimension* portlet though, which we can build in NSQL. 

(* - Dimension 1 is your "Value Dimension", Dimension 2 is the months)

Actually i have a tsv on the object list. I have to create a report based on that. The dates are dynamic and will have to be displayed either monthly or year. Of course these can be read from the time slice tables but the report will have too many columns.  I have six value dimensions and each has got its own custom time slice defined.

Now if i have to read the results in portlet, i will have to create six columns for the month of January alone. If i show the results for the complete year month wise, then i will end up 72 columns.

 I am actually looking for a TSV concept in portlets are well.

Thanks

The way I see it such functionality does not exist in CA PPM.

If I take an object with timescaled values as the data source of a portlet like assignments I do not see the blob fields in the list of fields available for the layout.

If I query the prassignment for the blob fields I get blobs and there is no option to turn them into clear data

I if use a system provider specifically prepared for that like Team Allocations I do get the normally readable data fields but again I do not see the blob fields. However, I do have the option to add a virtual timescaled field

which does not exist for a query based portlet

but exists for an object based portlet like on base on Assigments. However, if I try to add such a field

it is not possible as no value attributes are available like in a specially prepared system data provider.

When looking at the portlets that can display timescaled data I experience that the data is limited to the time range is system timeslices. My conclusion from that is that those system data providers do not have the blob cracker built it, but behind the scenes they do get the data for display with queries from the system slices. While using action trace when storing data it can be seen that they use java code to write directly to the blobs.

Thanks Urmas for explaining it will much details. Really appreciate it.