Bound Branch is old terminology. Most uses of Bound Branches can be described as heavyweight checkouts.
A bound branch is a branch that does not commit a new revision until after the revision has been applied to another branch and first.
A bound branch thusly requires lockstep mechanisms to ensure that if a revision can not be sent to both branches, Unverified: that it is sent to no branches at all
Assuming that branch A is bound to branch B
When a commit is attempted in branch A:
- commit is attempted in branch B commit is attempted then in branch A if either commit fails:
- removing the commit from both branches is attempted
Is this right? Can't we let it stay in branch B, and do update to fix A?
- removing the commit from both branches is attempted
DavidAllouche: I think it's rather like 1. check A is up to date with B 2. commit on branch B 2. pull B into A. If the check or commit failed, nothing appears changed. If the pull fail, A is now out of date and must pull or merge B before being able commit again.
