Resource Menu

RE : Explaination of So6 queue and So6 workspace defined
> Can you explain to me how So6 is intended to be used. As an example, I have a project > which contains independent sub modules with source code.

I will try…

> From reading the site, it appears that a so6 queue is the root of the source code.

The So6 queue is the central point of your source code management. So everyone can have access to it. (Only allowed people)

> Then the initial source file uploading is accomplished by a workspace being created > under that queue.


> Any user wanting to modify the code locally and update/commit would need to, first > download the code anonymously to their computer, then create a workspace specific > to them before being able to upload the changes.


If they want to contribute, they just have to create a workspace in an empty directoy and call update on their workspace. They MUST NOT include the same thing at the begining, first they have to share the same object, so the user has to get the shared objects from the system by updating himself first.

> So with 10 users working on 10 computers, you would see 10 workspaces under 1 > so6queue, correct?


> When anyone updates their workspace a patch is created for the so6 queue, > and all workspaces will get the patches when they update, correct?


> Ok, if the above is correct then I have several questions.

Yes, except for the anonymous download...

> First, once the initial workspace is created and the files are uploaded from the > file server (the Libresource server), does this workspace need to remain, > or can it be deleted; are the files stored within libresource, or is it directly tied to > the workspace location on the file server, or is their any other way of uploading > the files for the first time from the file server to the queue without creating a > workspace, so other developers can get the files?

So6 is a versionning tool, so usually you don't remove the initial workspace. But as the data are kept into the server, you can remove any workspace you want, you will be able to retreive the data either by creating a new empty workspace or with the anonymous download.

And unfortunatly, the workspace creation has been made to improve awareness, but if you don't want awareness, then you can use hidden workspace… If you remove the workspace on the server, they can still be used localy to send and retreive files from the server. You only removed the link to access easily to the update or commit action.

> Second, I have a source project with multiple sub modules. From reading this site, > I read that you can create sub workspaces. How is this done? I would like to do > this, but can't find any documentation on how to accomplish this. Can you give > more detail.

A So6 queue take care of the content of a directory so you make as much as workspace you want. You just have to take care to prevent cross synchronisation between the so6 queue. Then you will have to add .ignore.so6 file in some higher workspace to tell the system to don't synchronized directory that are managed by other So6 queue.

> "A so6 workspace can be embedded in a so6 workspace. It allows you to build > composite workspace. It is possible to commit/update in each leaf workspace > and to commit/update the whole workspace."

This is true only if only one personne do it...

> "On this figure, there are two libresource servers. The central workspace has two > subworkspaces. You can update from joram and bonita queue and commit to the > whole jonas queue."

This case is used in process definition with a hierarchical propagation.

> And lastly, if you must keep the initial workspace that was created, then is there > a way to schedule the workspace to be updated automatically; for backup purposes, > or is this necessary?

You can make cron script for scheduled update but as the data are kept on the server, you don't need such system for you.

> Thanks in advance.

You're welcome…

posted by Sebastien Jourdain at Aug 29, 2006 10:36 AM