Are datasets still flat in NAV 2013?

Miklos_Hollender

I thought the biggest change here is with the hierarchical report designer that the dataset too will get properly hierarchical and with that the report designer gets similar to the old report designer, with headers, lines etc. having their own sections, repeaters, whatevers.

I have checked in the SQL Report Builder in R2 and datasets were still flat. Is it the same in Visual Studio? I try to avoid installing such a huge program unless necessary.

If yes that would be a very bad news because this is the very reason I have put off upgrading to the RTC in 2009, I am sure that you cannot properly make real life documents with half a dozen conditional footers and stuff like that in flat datasets. Why is the designer hierarchial then?

If this is true, how do you folks even use it? Even the laughably simple demo database documents are difficult to create with flat data sets, anything real life which tends to be 3 times as complicated in Classic report designer every Lines section repeated 10 times with various conditionals - downright impossible.

lvanvugt

Hi Miklos,

If this is another way of saying "it isn't straight orward to switch from classic report to rdlc report design" I fully agree. However if you mean "it's not possible to create proper reports in RDLC" I do not agree. It's a matter learning and wanting to learn and if you do you will find your way just quite easy (I even dare to say, just as easy).
I recall starting with NAV (or Navision in those days) it took me quite some time to feel at home in creating reports. With RDLC it seemed somewhat the same situation.

davmac1

Unlike Classic reports, we will see constant improvements in the report designer since it is tied to the release of Visual Studio in use at time of development.
Since NAV is the flea on the elephant's back, new VS releases will sometimes break the NAV code. Overall though, only 1 step back for every 3 steps forward.
I am hopeful that a future release will introduce multiple datasets per report which will allow more interesting reports.

Miklos_Hollender

@Ivanvugt learning means learning from somewhere. Is there anywhere a real documentation how to for example handly many kinds of conditional footers with these flat datasets?

@davmac1 I am not talking about more interesting reports but talking about not losing document functionality when you upgrade. And not about multiple datasets but properly hierarchical datasets - as the dataset designer in itself is hierarchcial.

Please both of you consider that it is not about "nice things", because I am not talking about new projects, I am simply talking about upgrading without pissing off users. Given that none of the RTC functionality is really needed in my workplace (OK the web client is nice but we have already built web based sales order entry etc. it is easy), I am already arguing for the upgrade from the viewpoint of "I hate it as much as you do but we must do or lose maintenance and maybe Classic will not run in Windows 8 or 9". So the basic minimum I must do is to make documents exactly the same as before in order not to piss people off. Because the benefits we cannot really use. All the new functionality stuff we could easily develop before.

davmac1

Could you use a single footer and have the NAV code push out the appropriate data to the footer fields?
If the fields occupy different positions, you can make the ones not visible that don't apply.
You may have to stack fields on top of one another, but with document outline, you can still find them and edit them.

lvanvugt

Miklos Hollender wrote:

@lvanvugt learning means learning from somewhere. Is there anywhere a real documentation how to for example handly many kinds of conditional footers with these flat datasets?

Agree that documentation is not superfluous. But that applies also to other matters like "classic" reporting. You have to collect info from various spots, learn by doing and invest by attending some courses on the subject. Steven Renders and Dynamics MVP Claus Lundstrøm are good resources on this specific matter.
Regarding the flat datasets: it's all a matter of filtering out the right "section". And yes, this isn't that obvious when switching from classic to RDLC as you need to learn how to create the right dataset; and to read it right. To me RDLC reporting start with that: understanding and creating the right dataset; and then how to filter out the right section for a specific part of your report.
In case of existing reports that you need to transform to RDLC reports: most often the existing layout does not provide you the right (i.e. complete) dataset to be able to get all needed data displayed in RDLC. So you first need to start there: what data is needed (more), before you even think of the RDLC layout setup.
What is very worthwhile to do is analyse some of the standard reports. I know it looks like a jungle at the start as there are so many controls, some embedded in others or on top of them etc. But unravelling it pays off.

Miklos_Hollender

By the right datasets you basically mean that every dataitem should have a clear unique key sent down so you can group on that?

lvanvugt

Miklos Hollender wrote:

By the right datasets you basically mean that every dataitem should have a clear unique key sent down so you can group on that?

That's one part of the story I think. The right dataset is also getting all the needed data (i.e. field values and captions) in the dataset efficiently to avoid redundancy and overload of data sent to the report viewer.

Miklos_Hollender

Could you be a bit more specific about this?

clausl

I suggest you participate in my "How to write RDLC reports in NAV 2013", Pre-conference class at NAV Tech Days:
http://www.navtechdays.com/2013/sessions.asp#class2 there should be room for 4 more people, before fully booked.

/Claus Lundstrøm