Re: [PATCH v5 0/4] Importing and exporting stashes to refs

>
> On 2025-05-09 at 16:53:17, D. Ben Knoble wrote:
> > - Can't we "git push <remote> stash[@{n}]:<branch>" to share a stash somewhere?
> > - And then, doesn't "git stash apply [--index] <arbitrary sha>" work?
> > (At which point you could presumably create a new stash, though I'll
> > admit that's cumbersome relative to dedicated export/import.)
>
> I haven't tried, but it does certainly seem plausible that you can
> import and export them in that way.
>
> > I can see how that doesn't help you quickly export a whole _chain_ of
> > stashes, so I'm not saying "this series seems like the wrong tack"
> > (far be it from me, who doesn't understand your use case, to say
> > that!). Rather, I'm confused about the inability to move a (single)
> > stash across machines.
>
> Let me explain the intended use case here.  At work, many people use
> GitHub Codespaces, which are throwaway development environments.  Since
> one's whole set of stashes cannot be imported or exported, it's hard to
> use stashes effectively in such a case, since they'd be deleted when you
> destroyed the environment.  I like stashes a lot (my personal Git
> development repository has 153), so I want to synchronize all of them
> across.
>
> The other case is people who routinely work on multiple machines.
> (Remote or throwaway development environments, like GitHub Codespaces or
> Devcontainers, are just a special case of this.)  Many of these users
> want to keep their working tree and other state across machines and lots
> of them rely on cloud syncing services, such as Dropbox, to do this,
> which often ends up corrupting the repository (as outlined in the FAQ).
> Providing a way to quickly and easily synchronize the working tree
> across systems, including any stashes, is really important to encourage
> best practices that don't result in data loss or have unpleasant
> security issues (such as untrusted local config).
>
> > Unrelated question: Can we import arbitrary refs into stashes? That
> > is, what happens if the commit structure doesn't look right? (Maybe I
> > should go read the tests and see.)
>
> That doesn't work because the commit used here has to have a fixed
> number of parents, since we need to keep track of the index and the
> working tree.  Stash commits, even the regular ones used in the reflog,
> always have to have a certain structure.
>
> If you try to do that anyway, you get this message:
>
>     % git stash import HEAD
>     error: 3bf235c35ef51d01663f2ab9665026b05b8af1dd is not a valid exported stash commit
>
> I did try to avoid people accidentally destroying data.
> --
> brian m. carlson (they/them)
> Toronto, Ontario, CA