Ignoring a large number of selections

[#1]

Background:  I have a qlikview document that is simple in concept.  Basically just a list of transactions, and a list of the companies that made those transactions.  However, both the transactions and the companies have a lot of attributes.  There's additional complexity in model as well, for example there's parent companies, and a parent might have more than 1 parent company.

Challenge:  I want to do some calculations where I ignore the selections related to companies.  I've previously done this sort of thing with set analysis, but in this case I'm struggling with whether there's a cleaner way of doing things.  At the moment, with set analysis, if I only want to ignore selections from the company side of my data model, I basically have to go through and, for each field, put in something like [Field 1]=, [Field 2]=, etc.  This is both cumbersome, and hard to maintain.  It's no better if I try to use {1} to ignore the selections, since there's many selections on the transaction side of the model that I actually don't want to ignore. 

So is there a good way of doing this?  Some syntax that ignores a whole table at a time?  Should I be using something other than set analysis?

Thanks for your help

[#2 · Accepted solution]

Check this link is awesome. If you have no experience with set analysis refer to the reference manual for more information.

http://community.qlik.com/docs/DOC-1334

[marcus_sommer]

Perhaps you could move some complexity into the script, for example with a hierachy from the companies and their parents/children.

- Marcus

[#4]

Thanks juan pedro, that looks like exactly what I'm after.  The syntax is still fairly readable, and allows me to include/exclude whole tables at a time!