chunk loop start with flow variable

[sanderlenselink]

Hi all,

I would like to split a string into different chunks. I calculated the #chunks for each row.

It looks like topic (see also Use of variable in Chunk loop)

For me the Chunk Loop with Flow Variables behave strange. I expected that the loop reacts on every company (AA, BB, CC, DD and EE). However, it sitcks to 5 (#rows per chunk).
What I’m doing wrong?

Hopefully someone is able to help me reach the final situation (screenshot)

SL_chunk_iter1.knwf (17.4 KB)

Chunk_input.txt (511 Bytes)

[denisfi]

Hi @sanderlenselink,

I saw the file and your flow and I have some doubts.

First, you can use the group loop start to separate company values. Each process can use a rule engine to set the split number to row or chunk options.

After that, I saw that you use a group node to concatenated all rows as one value column. Then you used a split cell to break the information into columns, that’s it you need?

See my example… I can believe that can help you…

SL_chunk_iter1.knwf (61.6 KB)

Tks,

Denis

[sanderlenselink]

Hi @denisfi

thnx for you fast feedback. Yes, your input gives me a great new insight how to solve my problem.

The reason I developed the flow as I did was that I need per Company 3 subgroups (in the real world there will be 50+ subgroups). See screenshot beneath.

Via String Manipulation you can split the first subgroup but with 50+ subgroups that’s not handy

FYI . . . the structure within each row is fixed (e.g. row1 5 times a value for each subgroup). But rows differ so row2 consists of 3 times a value for each subgroup.

Therefore I thought earlier about Chunk in combination with Flow Variable.

So, do you or anyone else have an additional suggestion for my problem.

Many thnx in advance

. . . I’m sorry but the initional screenshot in my first post contained a little omissiong. So here a new one

[denisfi]

Hi @sanderlenselink ,

Well, you can use the chunk loop to split it in groups, not for rows… example:

1- Company “AA” have 21 values, if split as chunk group by 3, you 'll have 7 values each one, right? Company “BB” have 10 values, 1 group will have 4 values and the others will be with 3 values… so go on…

My suggest here is to set a value using the interaction value from the loop to set these silos (1,2,3).

2- Other tip is remove duplicated rows… If you don’t need it, just unique values, before all of it, use a reme duplicate rows ou when you are grouping information, see if its necessary too.

The result of the chunk loop was it:

The result of the company loopy was it:

After that, you can manipulate the values as you wish…

chunk loop flow.knar (116.7 KB)

Hope that its help you.

Thanks,

Denis

[sanderlenselink]

hi @denisfi ,

many thnx for your help . . . it works !!!
Earlier I thought like your solution about a nested loop with unfortunately lacked the overview.
I learned a lot from you . . . muchas gracias

But now I’m struggling with the classical problem that the dataset is not so clean as promised

I will overcome