SEND296/FB1701: iterate over all variables that have a non-extensible codelist, then check the values in the dataset against that codelist

[JozefAerts]

I am still "fighting" with this rule ...

First of all, I need to iterate over all the variables that have an associated codelist according to the CDISC Library for the IG and version,
check whether that codelist is extensible for the given CT-version, and only keep the variable in the iteration when the codelist is not extensible.

I haven't been able to find out how to do this.

Then, for each of these variables, iterate over all the rows in the dataset, and raise an issue when the value is not in the associated non-extensible codelist.

A define.xml should not be involved.

Anyone tried something like this?

[eljanssens]

Hi @JozefAerts FB1701 has already been written and published as it is also applicable for SDTM. It is available here: https://github.com/verisianHQ/core-contributor/tree/main/rules/CORE-002123
So at some point, the SEND296 authorities can be added to this rule.

[JozefAerts]

Great! I was a bit confused, as the status in the old "rule-editor" was still "Draft".
So, what I think to do is to "steal" the Github code, and then use that for a "SEND-only" SEND296 rule, still using the old rule-editor, and then test it on my test SEND files.
BTW, should we expect false positives with the FB1701 rule for values like "MULTIPLE", "OTHER", ... ?

Also, as I have been not much involved with the transfer process, for new rules (for SEND, there are still a lot open, for example, the ones that are "SENDIG-3.0-only", which we gave low priority), what rule editor should I use?

Many thanks for the lightning-fast answer!
Jozef

[JozefAerts]

Ok, doesn't work with the old "rule-editor" and (probably thus the old engine) - getting an error:
"error":"Column not found in data"
"message":"library_variable_ccode_values"
So, yes, I need to switch to the new editor ...