c Ceci
on

 

In the new validator v2.0, the message "Variable length is too long for actual data" is becoming an error. I usually keep the length of each variable = the maximum length of all observations within that variable. What about the empty variables? The empty variables have a length of 1. But I keep getting the error messages “Variable length is too long for actual data”. For example, I sometimes leave the QEVAL blank in the SUPP-. If I remove QEVAL from the dataset, I will have a warning message in the validator report “FDA expected variable not found”. If I keep QEVAL in the dataset, I will get the error message. Any suggestions on how to avoid this error message? Thank you!

Forums: SDTM

j Jozef
on February 26, 2015

This might help: http://www.opencdisc.org/forum/rule-sd1082-actually-required

Comment by doetzman and follow up.

s Sergiy
on February 26, 2015

Hi Ceci, 

It's a bug. A fixed Auto-Update version will be available in few days.

Sorry for inconvenience.

Kind Regards, 

Sergiy

i Isabelle
on October 20, 2020

Dear Sergiy, 

we're in 2020 and I'm experiencing the exact same issue as this user in 2005.

Is it to be fixed again?

Or shall I simply provide comments along that issue to the FDA?

 

Thanks a lot in advance for your feedback.

Best regards,

 

Isabelle

Sergiy
on October 20, 2020

Hi Isabelle, 

What version of validation engine do you use?

Thank you,
Sergiy

i Isabelle
on October 21, 2020

The version we use is the following:

FDA (1907.02)

(the latest available)

We already had the issue with the previous one (Validation engine = 1903.01), as we issued preliminary reports in August. But that problem was limited to one variable.

Excess,Variable   1,VSLNKGRP     SD1082     FDAB018        Variable length is too long for actual data

 

Now it seems more widespread accross multiple domains:

Excess,Variable   1, AECONTRT

Excess,Variable   1, AESCONG

Excess,Variable   1, AESDISAB

Excess,Variable   1, AESDTH

Excess,Variable   1, AESHOSP

Excess,Variable   1, AESLIFE

Excess,Variable   1, AESMIE

Excess,Variable   1, CEOCCUR

Excess,Variable   1, ISBLFL

Excess,Variable   1, QSDRVFL

Excess,Variable   1, SUDOSU

Excess,Variable   1, VSBLFL

Excess,Variable   1, VSLNKGRP

 

Hope this will help you figure this out.

Thanks a lot in advance.

Best regards,

Isabelle

 

Sergiy
on October 21, 2020

Hi Isabelle, 

This bug was introduced in OpenCDISC Community v2.0 and fixed in the next release few months later in early 2015.

So, my assumption is that you are still using extremely outdated version of OpenCDISC Community 2.0 with new configs files.

Please be aware that such approach does not work and produces incorrect validation results. Configuration files (e.g., engine 1907.2) must be executed only by version of validator they refer to (1907.2).    

Kind Reagrds,
Sergiy

i Isabelle
on October 26, 2020

Dear Sergiy,

We made some verifications on the versions we use.
Pinnacle21 version: 3.1.0
Validation Engine: FDA (1907.02) (along with the configs files provided at download)
Configuration: SDTM-IG 3.2 (FDA)

To clarify the issue, here is an example of what we experiment with the variable AESDTH.
We set all the value of that variable to "N", then no error message pops out.
Afterwards, we set one value to empty, and the error message SD1082 pops out.


We didn't experiment the issue on that domain with the previous version of Pinnacle21 (3.0.2).


Unlike what I previously stated in one of my message, it is following the upgrade of Pinnacle21 Community 3.0.2 to 3.1.0 that made the problem worse for us (no impact came from the upgrade of engine).
Initially only one domain was impacted with that issue (VS Domain).
After that Pinnacle21 update, all domains were impacted once a variable had a length = 1 and for which at least one value was null.

Thanks a lot in advance for any advise on this matter.

Best regards,

Isabelle

 

Sergiy
on October 28, 2020

Hi Isabelle, 

Unfortunately, we cannot reproduce your reported issue. Could you provide sample dataset?

Thank you,
Sergiy

sergiy at pinnacle21 dot com

Want a demo?

Let’s Talk.

We're eager to share and ready to listen.

Cookie Policy

Pinnacle 21 uses cookies to make our site easier for you to use. By continuing to use this website, you agree to our use of cookies. For more info visit our Privacy Policy.