Here is the simplest workflow I could get to reproduce the problem.

A .leo file and a source code file (python) in version control
(fossil).

The source code file is linked into the outline as @shadow file.

I add a class with a method (named method1()) to the .py file. Save
and commit. Let it be r1.

I add another method (named method2()) to the class. Save and commit
(but commit to a new branch, to trigger a fork). Let it be r2.

I update the workspace back to r1. Actually, I realize that method2()
will call yet another method (method3()), so I revert back to r1 to
implement method3() first, commit it to the trunk, then go back to the
partially implemented method2(), complete it, commit then merge with
trunk (which already has method3()). This is only the reason for
switching to r1 that is not related to the @shadow issue.

After updating the workspace to r1, I restart leo. The outline has a
method2() node with empty body. Why? It shouldn't be there! After r1
there was nothing related to method2() in the workspace.

If I close leo, delete the .leo_shadow directory and restart leo,
everything is well. There is no method2() node in the outline anymore.
So the @shadow cache somehow has higher priority than the actual
files, which I think is not correct.

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/leo-editor?hl=en.

Reply via email to