How to identify the chain?

[#1]

Hello

I have source with structure: SOMEONE1, SOMEONE2, family_identifier.

SOMEONE in one row can be associated with someone from a different row. Thus, in the family for 3 person. And so on ...

Help me identify the chain effective way?>

[#2]

May you show sample data?

[hic]

This sounds like a hierarchy. If SOMEONE1 has a parent SOMEONE2, then you can resolve this hierarchy using a Hierarchy Load.

See more on http://community.qlik.com/blogs/qlikviewdesignblog/2013/11/11/unbalanced-n-level-hierarchies

HIC

[#4]

in fact I work with numbers:

[#5]

Gertye,

i thought that HierarchyBelongsTo() may be a key, but i understand, that this relation is not hierarchical in fact. I undestand, that someone1 konws someone2 and taht means, that someone2 knows someone1 ....

hmm.

1. How many rows do you have?

2 Are you interested to find "shortest way" (shortest ways) from SMB_X to SMB_Y or maybe all possible ways?

3.How many steps do you expect (i heard somewhere, that from any to any other person in the world you need no more than 7 steps ....)

regards

Darek

[#6]

I think we may need something like Dijkstra's algorithm...

reg

[#7]

Hello Dariusz

1. How many rows do you have?

It's 1 272 055 rows for the start (at the end it can be about 20 000 000 rows).

2 Are you interested to find "shortest way" (shortest ways) from SMB_X to SMB_Y or maybe all possible ways?

I'd like shortest way  (I know how I can make it, but I don't like my way).

3.How many steps do you expect (i heard somewhere, that from any to any other person in the world you need no more than 7 steps ....)

Yes, I don't know where the bottom ;(  Maximum number of people in the family ...mmm.. may be.. 20?

[#8]

thanks, I will try it and let you know then

[#9]

SOURCE			COMBINATIONS			RESULTS
1	2		1	2		1
2	3		1	3		2
2	4		1	4		2
2	5		1	5		2
4	7		1	7		3
			2	3		1
			2	4		1
			2	5		1
			2	7		2
			3	4		2
			3	5		2
			3	7		3
			4	5		2
			4	7		1
			5	7		3

Is it something like this?

[#10]

yes, it was "my way" ))