Question

Hi all. We love Skipper and use it on an hourly basis.

Our most recent project contains a Git branch where we're developing a new feature which requires new entities and database tables. That means I need to update my Skipper model...

... But part of the way through development of this new feature I may need to make a change to the ORM model in the Git trunk. So we'll have two different versions of the Skipper model - one to support the trunk and one to support the branch.

Can Skipper cope with this? Is anyone else using Skipper within a SVN/Git environment with branching? Any tips?

Thanks
Dave Small

Answer 1

Hi Dave,

This is definitely possible thanks to ability to split project to different modules. I will describe you our workflow which we're using on our project (Skipper itself, licensing server and some other projects):

As versioning tool we're using SVN. We have our project split to several modules and each module has it's own Skipper model file. Thanks to this it is pretty easy to create new development branch.

When we need to develop new feature we simply create new project where all modules are linked via svn:externals and only the module which will be updated has it's own branch also linked to project.

After that, we simply create new project with root in development directory and let Skipper scan all schema files. Together with schema files Skipper find also skipper.module.xml files and creates new project with all existing modules.

Thanks to this you can work on trunk modules where necessary and also have isolated specific module which needs to be developed separately.

More information about modules and splitting / sharing modules between multiple project is available in our help: http://help.skipper18.com/how-to-use/application-concepts/shared-modules