Using Bazaar – First Steps

These are the things I want to do with Bazaar. I have a network of Win Vista, Win 7 and Ubuntu systems. Click on any picture for a larger view.

1 – Create a project with collocated repository.
Must do this manually due to the bug in creating collocated repositories but it seems to work by using the All command dialog and colo-init.

Creating repository

2 – Check in files the first time.
Use the Add dialog. Click on the Plus Sign next to the red UnVersioned file list or the lower panel. Choose the files / folders to add and click OK.

3 – Mark files as Ignored.
Intermediate files from the compile don’t need to be controlled.
I can’t find a way to set files to be ignored in spite of the indicator that it is supported.

Adding Files

4 – Check out a previous revision.
Looks like I can do that with PULL. But do I need to manually type in the revision number. So I need to look at the Log window and then pick a number and then type that number into the Pull dialog. Odd. I expected to get a list of revisions that I could select from and then just click PULL.

Pull Dialog

5 – Remove a file and commit.
I want to understand what happens when I remove a file and then commit. What I want is for that file to disappear from future revisions. But if I checkout a previous revision, I get the file back, since it’s required in that revision. The Remove function – RCLICK in working tree – seems to delete the folder or file. I presume that committing the changes will cause the file to be removed in that revision, but not in previous revisions.

Working Tree Menu

6 – Retrieve a removed file and add it to the the current revision
I want that file restored so it is carried forward. I’d expect to do this without a lot of fooling around.
This means that I need to be able to choose a single file, or folder in an early revision, that was later removed, and just pull that file / folder into the current working tree. I don’t see a way to do that easily. Looks like I need to pull the entire revision to a new place and then manually copy the file. Then add it again. And then commit. But will the commit cause that file to be confused if I pull a previous revision? I will of course be changing the file in the later revision. That seems like a lot of fooling around to me. Rather than being able to look at a tree of a previous revision and just PULLing that single file / folder from that previous revision into the working tree, I need to manually get the revision into another place and manually copy it into the new working tree. But is that going to work, or is ADDING the file again going to mess up the repository? Are these the right steps to do this?

7 – Does this look right?
I’m not sure if I’ve messed up in creating my repository. Does this look right?

Collocated Repository

8 – For Extra Credit. Shared Repositories?
Can I create a repository on a windows shared disk on another machine and then check out and commit files to that repository? Do I have to run a server on that machine or just clients on the various machines that I’m working from? Please walk me through the steps to accomplish this.