I would like to see an example of someone showing the workflow for using jj and doing feature branch. I don’t really get that yet. Most examples only show that they commit once and then push. But what if it requires multiple commits.
`jj log` if needed to see what revision ID you need to use
`jj new <starting revision> -m <message>` to create a new change with a given message with <starting revision> as the parent. (you can also use the `--insert-before <revset>` and/or `--insert-after <revset>` to set the parent & child revison(s) for the new revision).
`jj bookmark set <feature 1 name> --revision <revset>` to make a bookmark (a branch name) on the revison you just made.
Make changes on this branch as desired. `jj new` for each new change you want to have its own revision.
If you added more revisions to the branch, `jj bookmark move <feature 1 name> --to '@'` (you can also use the change ID instead of '@', '@' just means "the current change you're editing").
`jj edit <starting revset>` to go back to your starting revset.
`jj new -m <message>` to make a new revset on top of the starting revset (the starting revset now has two children: it has branched)
`jj bookmark set <feature 2 name> --revision <revset>` to name the second branch you just created.
Make changes as desired. `jj new` for each new change you want to have its own revision.
If you added more revisions to the branch, `jj bookmark move <feature 2 name> --to '@'`.
`jj new <feature 1 name> <feature 2 name> -m "merge feature 1 & feature 2"` to make a new revision with the feature 1 & feature 2 bookmark revisions as its parents (this is a merge).
Alternatively, you could use `jj rebase` to move any number of commits to anywhere in the commit graph you want.
