data set 1 is a big table with 95 entries, where every entry is an Image except the last 91 column, which is a counting column (1, 2, 3, 4, 5, 6, 7, 8, 9).
data set 2 is a table with a lot of entries (mostly numbers) and also a counting column with entries from 1 to 9, where every counting number appears several times (i.e. the table has more than 9 rows).
I would like to do the following: Create a Report with 9 pages, where on page n, we have the pictures corresponding to count n from data set 1 and the table entries of count n from data set 2.
How is this possible?
I already know how to group tables and I am able to group the pictures and the table separately, but I do not know how to do it simultaneously.
Take a look at this workflow on the EXAMPLES server: 05_Reporting/01_BIRT/08_BIRT_Example_Advanced
I think you might be able adapt Grouping Method 2 in the workflow (it’s actually the first set of outputs in the report) to produce what you need.
This workflow is an extension of what we’ve already done with the Austin restaurant data at https://www.knime.com/reporting/birt. I plan to do a full how-to writeup on this advanced workflow at some point, but hopefully just having the nodes and the report layout will help get you where you need to go.
unfortunately I am unable to Access the Examples Server from work, and the zip file is just huge. (Is it possible to download only one example workflow instead of all at once?).
Anyway, I managed to solve my Problem in the following way:
Join the two tables already in the workflow on the counting column
Put the complete table in the Report and afterwards delete the columns with the Images
Group by the counting column and make a page break after every group
Create aggregations of the Images-columns, where for every Group we just take the first Image (This is necessary because the Images were multiplied in the joining procedure)
insert the corresponding images in the Group footer row
It took me a while to figure this out, but now it roughly looks as intended
You can indeed browse the available example workflows and download them separately via either the KNIME Community Hub on the KNIME website, or via nodepit:
@gab1one has a great pointer to our node guide, which is useful when you can’t get to the EXAMPLES server for whatever reason.
In this particular case, the advanced BIRT workflow I mentioned is on the EXAMPLES server, but the node guide hasn’t been updated to include it yet. So I attach it here.
thank you really much for your help I looked in the workflow and saw that you managed to do the simultaneous grouping on the zip-code in the grid at the top. I tried to replicate this and it worked whenever I placed the items in the grid horizontally next to each other. But when I placed them vertically (i.e. one item at the top cell of the grid and one item at the cell below) then it did not work anymore. Then I first get several pages for the first item and then several pages for the second item and I do not get them together on the same page.
The same happens btw. to your workflow as well when I drag and drop theimage-table from the right top cell to the bottom left cell.
Is there a way to solve this behaviour? It’s not that important since I managed a way to do what I want, but the grid solution would give me a little bit more freedom in designing the report, if it worked.
The grouping is taking place at the Table level in both the top left and top right grid cells, and not the Grid level. Also, I am forcing page breaks for both tables. The upshot of this is that if you move displayed tables or images to other grid cells, the grouping will not be maintained. If you play around with the page breaks, it might make the behavior more clear.
I’ve already started writing the how-to page for this workflow… I’ll be sure to include a link when it gets posted to our reporting pages.
EDIT: The advanced BIRT writeup has been published and can be found here.
yes, I think I understood this already. But this still leaves the question open to me if it is possible to place the items on top of each other while maintaining the simultaneous grouping. At least I think it should be possible
