Skip to main content
Woohoo! Qlik Community has won “Best in Class Community” in the 2024 Khoros Kudos awards!
Announcements
Nov. 20th, Qlik Insider - Lakehouses: Driving the Future of Data & AI - PICK A SESSION
cancel
Showing results for 
Search instead for 
Did you mean: 
Danijel
Partner - Creator III
Partner - Creator III

Number formats are lost

Hello all,

we always have a strange behavior with number formatting when we publish applications.

For example, we define some dimensions as date in the properties of the document. Then, when the application is published and a user accesses it, the date becomes a number again.

Does anyone know this behavior and would happen to know a solution?

 

Greetings

Labels (1)
5 Replies
NitinK7
Specialist
Specialist

Can you please try once using manual conversion to date format like below.

Date(dimension,'DD/MM/YYYY')

Danijel
Partner - Creator III
Partner - Creator III
Author

Hi,

with a manual adjustment in the script is working, however, this means a significant additional effort for us.

On the other hand, it should still not be the case that formats disappear. Therefore, I would still like to find out where the problem comes from.

marcus_sommer

You mean those fields aren't converted/formatted in any way during the load-process - just purely as the come from the sources and the wanted formatting is then set within the document properties? And if you format them within the script explicitly with measures like date(field) it will work with / without the document settings?

- Marcus

Danijel
Partner - Creator III
Partner - Creator III
Author

Hi @marcus_sommer 

right, that's exactly how it behaves. Perhaps a supplement. We use our own tool to bring the data models from a source to QlikView via the Qlik API.

However, so far I have not found any indication that this changes the settings in the document settings.

marcus_sommer

Thinkable are various things, like:

  • different releases between clients in dev and prod and maybe also in regard to the server
  • different data - in the real data there might be blanks, NULL's, strings, other invalid dates, mixing different formats, mixing dates and timestamps and in the end getting mixed values which may not happens during the development (different datasets, section access, ...)
  • data might be differently loaded - without specifying an interpretation/formatting directly in the load-statement Qlik used the standard formatting-variables from the main-script-tab, if they are missing or invalid Qlik takes the appropriate settings from the OS (and this is probably quite different between the machines)
  • prj-folders could be also cause some impact   

I could imagine that in the end defining an explicit interpretation/formatting during the load may the most suitable approach to ensure the wanted results. And with a smart approach like:

https://qlikviewcookbook.com/2016/07/touchless-formatting/ 

the needed efforts are not necessary too expensive and could be also applied within more or less automated loading-routines.

- Marcus