Jonathan wrote:
Show 5 quoted lines
>What might make git nice as an interoperability tool is that it tracks >the _relevant_ information for the history of a software project. >Example of what is not relevant information and why that matters: > > http://thread.gmane.org/gmane.comp.version-control.git/53494
Hell, that message seems to even suggest a potential solution: use .gitattributes for tracking empty dirs. That way a post-change "git checkout-index" could find the info (from .gitattribute in the index) without confusing any pre-change tool.
For ease of use, some porcelain could possibly be taught to add empty directories into $GIT_DIR/.gitattribute (as well as checking whether an empty-dir entry has turned useless because it now has content, and any other administrative stuff we could want).
But that idea is linked to another issue, which could be the subject of a 1.8 proposal of its own: the handling of .gitattribute, for which only the checked-out version is taken into account:
http://marc.info/?l=git&m=126458888515166
-- Yann Dirson - Bertin Technologies