Narrowcast

[138006]

Hi,
In a report I have 2 prompts. Now say each can have 2 values. Then the different sets of report combinations is 2*2. Now, is it possible in MSTR to generate the 4 reports based on these values and schedule them? These numbers can increase as well, so it difficult to create all these reports and then schedule it. Better option will be to create one base report and then appropriately use N'cast to generate different sets of it.

Is this possible in MSTR??
Thanks

 

[138006]

Hi
Anyone has any idea about the question I had regarding MSTR N'Cast:

In a report I have 2 prompts. Now say each can have 2 values. Then the different sets of report combinations is 2*2. Now, is it possible in MSTR to generate the 4 reports based on these values and schedule them? These numbers can increase as well, so it difficult to create all these reports and then schedule it. Better option will be to create one base report and then appropriately use N'cast to generate different sets of it.

Please let me know if this possible ?

 

[Z3]

What is the goal you are trying to reach? To solve a performance problem? There are ways to do that, but I 'd like to now what you are upto.

 

[138006]

Not performance. I have to create a report which has around 8 - 10 prompts. Now the requirement is the create as many number of reports as the combination of the prompts and schedule them. In BO, you can write a VB macro and then schedule the macro than the report.

What to do in MSTR? No out-of-shelf solution in N'Cast.
What else to do then?

Thanks

 

[Z3]

You can do that too with SDK. Consider that doable and done. Now the question is, what do you do with these reports ran by NC? Who wants them?

My basic point is, I still don't know the reason for doing this. You told me it's not performance related.

 

[138006]

Can u please help with the SDK code.

As u understand the number of generated reports is dynamic--depending on the number of prompts and the different values of the prompt.
Instead of copying and saving the different combinations of the same report using hardcoded filters I want the tool to generate that combination of reports. Also rename the reports accordingly for example
Report_1(US)
Report_1(UK) etc.
where UK and US are values of a prompt cqalled Region

Thanks

 

[Z3]

Could you help me understand the reason for doing this? You have 8-10 prompts. If eash prompt has 4 possible answers you will hit 65,536 to 1,048,576 files or emails.

 

[138006]

May not be that much
but it can be a 500 report copies as well. Duplicating that also is an effort. And the requirement is storing in a place--may be the subscription portal etc. All these wont be mailed for sure.
can you please lead me to the solution in SDK??

Thanks

 

[Z3]

Okay, if you have 2 possible answers in each prompt, you will have AT LEAST 512-2048 files, just for this report. The report would be kinda useless when someone wants both answers in some combination. Enough said.

Just a thought - you probably don't need SDK at all. You could use something called "mid-tier slicing" or "dynamic content". The idea is to have a bunch of users with different security filters (on the prompted attributes), then you can run report without prompt and the NC will slice the data accordingly to satisfy each users. Please check the manuals, because my NC 'memory' are long expired.

You will need someone else's help to get the SDK done - I have not touch it in years, but I know there's endless possibility there.

 

[entaroadun]

If you can do BO VB, then you can certainly do MSTR N'cast SDK VB. The N'cast SDK reference explains exactly how to use it. Take a look.

I agree with Z3; your requirement is unclear. What exactly do you need, in terms of report content and delivery? Do you just need to pre-cache a different version of the report for every possible prompt combo? Or do you need to ship the result set somewhere? How do you want the result set stored and organized?

 

[138006]

N'Cast supports VB code?
Even then, BO VB uses lots of VB classes and objects and methods. What will corresponding be in MSTR?

How will I schedule the macros in MSTR N'C?

 

[chael]

Yes, NC supports VB. If you read the manual, or even looked at MicroStrategy's web site, it would be self evident that the API can be accessed in VB.

MicroStrategy's SDK is far more comprehensive than BO...

But to get back to the original basis of your problem, NC may be overkill here if you have OLAP Services...it seems like you just want to generate every possible itereation of a report...why not create an OLAP Services cube (1 report)? Prove to me that you have already thoroughly thought through this problem...

Chael

 

[138006]

I don't have Olap Services. Is it possible in Olap services? As I understand it will fetch data and populate the cube with all values of all prompts. Now depending on the different combinations it can generate reports.
Please let me know if this is correct.

If so, how is OLAP services interfaced with N'Cast? Since I have to schedule these reports in NC I need to know that as well.
thanks

 

[chael]

Well, you never said NC was an absolute requirement...so let's refer back to z3's suggestion that you implement mid-tier slicing into your project. I'm sure there are tech notes available on this topic.

If there is only one report being generated per recipient you can leverage security filters as well...

Here's my advice:
Get a firm grip on the requirements, distill from the requirements the complexity and frequency of your reports, take a stab at implement a small subset of the problem (also called a proof of concept), then take lessons learned and prepare of full scale project plan.

You have not proved to us that you fully understand MicroStrategy's product set or architecture, or your own business needs, and how the two will intersect.

Chael

 

[entaroadun]

If you need to schedule these reports in N'Cast, use Narrowcast. Your Narrowcastor CDs should have the SDK as PDFs. They are very easy to use; if you know enough VB to use BO API, then you can use it.

Crack open the manual and come back after you give it a few tries.