Re: [RFC PATCH 2/4] change submodule push test to use proper repository setup

On Wed, Oct 11, 2017 at 08:31:37AM +0900, Junio C Hamano wrote:

Stefan Beller [off-list ref] writes:

quoted

So you propose to make git-add behave like "git submodule add"
(i.e. also add the .gitmodules entry for name/path/URL), which I
like from a submodule perspective.

However other users of gitlinks might be confused[1], which is why
I refrained from "making every gitlink into a submodule". Specifically
the more powerful a submodule operation is (the more fluff adds),
the harder it should be for people to mis-use it.

A few questions that come to mind are:

 - Does "git add sub/" have enough information to populate
   .gitmodules?  If we have reasonable "default" values for
   .gitmodules entries (e.g. missing URL means we won't fetch when
   asked to go recursively fetch), perhaps we can leave everything
   other than "submodule.$name.path" undefined.

My suggestion would be: If we do not have them we do not populate them.
We could even go further and say: If we do not have the set "git
submodule add" would populate then we do not add anything to .gitmodules
and warn the user.

 - Can't we help those who have gitlinks without .gitmodules entries
   exactly the same way as above, i.e. when we see a gitlink and try
   to treat it as a submodule, we'd first try to look it up from
   .gitmodules (by going from path to name and then to
   submodule.$name.$var); the above "'git add sub/' would add an
   entry for .gitmodules" wish is based on the assumption that there
   are reasonable "default" values for each of these $var--so by
   basing on the same assumption, we can "pretend" as if these
   submodule.$name.$var were in .gitmodules file when we see
   gitlinks without .gitmodules entries.  IOW, if "git add sub/" can
   add .gitmodules to help people without having to type "git
   submodule add sub/", then we can give exactly the same degree of
   help without even modifying .gitmodules when "git add sub/" is
   run.

This "default" value thing got me thinking in a different direction. We
could use a scheme like that to get names (and values) for submodules
that are missing from the .gitmodules file. If we decide that we need to
handle them.

Cheers Heiko