Re: [PATCH] pathspec: remove check_path_for_gitlink

[Junio C Hamano <gitster@pobox.com>]

Stefan Beller <sbeller@google.com> writes:

> It was a bug, but now people in the outside world consider it a feature.
> Search for "Git fake submodules" and you'll find a few users who use this
> technique successfully.
>
> I do not think fixing this bug would do good. So maybe we just let it slip?

I am OK with leaving it unfixed, iow, we just say this:

    If deep/in is a different repository, whether it is a submodule,
    "git add deep/in/the/tree/is-a-leaf.txt" will give an undefined
    result.

But that is totally different from accepting it as a feature.  If we
were to accept it as a feature (and we will not), then

    I did "git add deep/in/the/tree/is-a-leaf.txt" and have kept the
    path tracked.  Today I did "git add deep/*" and then the path
    disappeared from my project--I now only have deep/in as a
    submodule, which is not what I want.

would become a valid bug report.  I do not want to see that happen.

I.e. I am *NOT* OK with polluting the codebase with a hack to
respond to such a bug report, e.g. by adding a rule that says "if a
file deep/in/the/tree/is-a-leaf.txt is tracked and deep/in is a
repository, 'git add deep/in' must fail".

The stance "It is a bug, but we do not fix it right now.  The
behaviour is undefined" also leaves the door open for a future
enhancement that allows 'git add deep/in/the/tree/is-a-leaf.txt' to
be an equivalent to 'git -C deep/in/the/tree add is-a-leaf.txt'.