Source Control/Version Control

[sunny_talwar]

Hey guys -

I hope someone can help me out here. I have been struggling to implement the Source Control for the QlikView dashboards but not entirely sure what would be the best way to do it. Let me first lay out a scenario for you and then it might make sense to explain the issue

Day 1: I created an app with 1 text box object (TX01)

Commit changes -> Revision 1 -> PRJ Includes TX01

Day 2: I created another object, let's say straight table (CH01)

Commit changes -> Revision 2 -> PRJ includes TX01 and CH01

Day3: Removed TX01 and replaced it with a new text box object TX02

Commit changes -> Revision 3 -> PRJ includes TX02 and CH01

Day4: Made changes to CH01

Commit changes -> Revision 4 -> PRJ includes TX02 and CH01 (updated)

Day5: Found out that the changes I made were not right and I need to roll back to Revision 4. Now the concern here is that when I roll back to Revision 3 will I see 2 (TX02, CH01) or 3 (TX01, TX02, CH01) objects?

The issue I see is that although the current revision doesn't have TX01, but the object TX01 was never deleted from the repository. Is Source Control smart enough to know that I am rolling back to Rivision 3 which did not have TX01 in it?

This might not be a question for Qlik Community, but I am not sure where else to go with this.

Thanks,

Sunny

[swuehl · Accepted solution]

Hi Sunny,

in short: Yes, source control should be able to handle this correctly.

That's mainly why Source Control or configuration management was created, right?

We should be able to create a snapshot of a configuration and recall it whenever we like.

(You just need to ensure that all necessary information is part of the configuration items under control, and here is where  the QV fun begins...)

[sunny_talwar]

I guess there is no better way to know this then to test it out. Will give it a shot and let you/everyone know what I found.

[b_garside]

Does anyone have experience using TFS and SVN?  which one would they recommend. Im trying to determine what would work best for my small team of 3. I understand you need a server/service installed for both.  Anything I should consider when evaluating both? I find both confusing in terms of use case and setup.