Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi all,
I am setting up a QlikView deployment pipeline with Jenkins and I have read that as a QVW is a binary file, we are unable to track changes to the file and hence we use the -prj method to break down a QVW into multiple XML files.
From what I understand, deploying the -prj folder will have the below advantages:
- Code changes can be tracked, since we can now track changes of the XML files.
- Saves space since breaking down the -prj folder stores no actual data as compared to pushing the entire QVW through the pipeline.
However I am having trouble with building the pipeline to deploy the -prj folder and recreate the QVW in higher environments. Is it advisable to just stick to deploying the entire QVW, but drop all tables before deployment?
Would appreciate any related advice, thanks.
If you have issues recreating the QVW, make sure to do as follows:
If this is not the issue, please elaborate.
To create the -prj folder it does not matter whether or not the QVW has data. It will obviously save faster if there is no data, but it will not break.
To add to Miguels answer.
If you update the prj files, refresh from your version control system. You need to open and save the qvw file to apply the changes. Just doing a reload with Publisher will not work since the qvb.exe (Publisher) doesn't have a gui. The handling of the prj files is done in the gui part of developer.
Basically you have to build something ( like a C# program or powershell maybe) that opens and saves the qvw file before doing a reload in Publisher.
I haven't worked with Jenkins but I have done this in a large production environment and the process are the same regardless tool.
Thanks Miguel,
We have no issues recreating the QVW manually, its more of an issue with executing these steps via shellscript etc.
This steps will eventually be the final steps in the automated deployment pipeline for our continuous delivery, but I guess you are right. We just have to find a way to automate it.
Hi Jerry,
Thanks for the information, you are right we are actually stuck at the part to use powershell to automate the steps to re-create the QVW.
Would you happen to have a sample powershell script used to execute these steps?
There are partners who have been developing solutions around the SCI/SVN for QlikView and Qlik Sense. You will need to check but In4BI and PlatformManager come to my mind as two of them.
No we used bat files, not recommended..
I think you will have all information here:
Re: recreating Qlikview from prj folder
Re: Information about -prj folder
One thing though, the account that runs the powershell script needs a license.