Re: [PATCH] mktree: learn about promised objects

On Wed, Jun 15, 2022 at 02:01:33PM -0700, Junio C Hamano wrote:

Richard Oliver [off-list ref] writes:

quoted

Meanwhile, is it worth considering a '--promised-as-missing' option
(or a config option) for invocations such as 'mktree --missing' that
prevents promised objects being faulted-in? Currently, the only
reliable way that I've found to prevent 'mktree --missing' faulting-in
promised objects is to remove the remote. Such an option could either
set the global variable 'fetch_if_missing' to '0' or could ensure
'OBJECT_INFO_SKIP_FETCH_OBJECT' is passed appropriately.

I didn't spend too much time on thinking this one through myself,
but do we really need a separte option?
[...]
We by grave mistake at 31c8221a (mktree: validate entry type in
input, 2009-05-14) started insisting on inspecting objects even when
allow-mising was given.  I do not think it was sensible, given why
we had "--missing" as an option to allow users to say "you do not
have to be too paranoid".

The codebase is so distant but I think we should probably do a moral
revert/reconstruct of that commit so that the extra paranoia of the
said commit applies only when "--missing" is not in effect, or
something like that.

FWIW, I had the same reaction. I think fixing "--missing" should be the
first step, and would unstick Richard's use case, as I understand it.

There is some value to improving the promisor case, since using
"--missing" is a pretty broad stroke (i.e., you'd fail to actual
corruption of missing non-promisor objects). That could either be
checking the promisor object and type without faulting it in, or just
skipping the type-check for objects after confirming that they're
promisors.

But that can come on top, I think. The use case there is also a bit
narrower.  The local repository does not know about all promised
objects. It can only see the boundaries of objects it doesn't have (so
with --filter=tree:0, for example, a partial clone of a repo with path
"a/b/c" would know about "b" but not "c"). So in the most general case
you'd still have to resort to "--missing", but I suspect in practice
you'd always feed things at that boundary to mktree (otherwise, how
would you even know the oid of "c").

-Peff