The paragraph that describes the pull processing seems slightly confusing,
notably the follow-up processing is done by child nodes, not siblings.
I attempted to use a more descriptive language
(No files)
Note:
See TracChangeset
for help on using the changeset viewer.