Re: [PATCHv3] gpg-interface: check gpg signature creation status

[Jeff King <peff@peff.net>]

On Wed, Jun 15, 2016 at 09:17:54AM +0200, Michael J Gruber wrote:

> As for the flexibility:
> We do code specifically for gpg, which happens to work for gpg2 also.
> The patch doesn't add any gpg ui requirements that we don't require
> elsewhere already.
> More flexibility requires a completely pluggable approach - gpgsm
> already fails to meet the gpg command line ui.

Does it? I haven't run it in production, but I did some tests with gpgsm
a few months ago and found it to be a drop-in replacement, at least with
respect to git. Though I don't think that matters one way or the other
for the current discussion; it _does_ do --status-fd.

> In any case, "status-fd" is *the* way to interface with gpg reliably
> just like plumbing commands are *the* way to interface with git reliably.

Fair enough. I've generally found its exit code pretty reliable. It's
unclear to me if the problem you saw was because gpg was exiting 0 but
producing no signature, or if your debugging was masking its exit code.

Either way, I do not mind using --status-fd; it seems like it should be
more robust in general. It's the tip commit in the series I'm about to
post.

> As for the read locking:
> I'm sorry I have no idea about that area at all. I thought that I'm
> doing the same that we do for verify, but apparently not. My
> strbuf_read-fu is not that strong either (read: copy&paste). I trust
> your assessment completely, though.

Yeah, you're right that the deadlock thing is also a possibility on the
verification side. I'm not sure whether it's possible to trigger in
practice or not. See the analysis in the series.

> As for the original problem:
> That had a different cause, as we know now (rebase dropping signatures
> without hint). I still think we should check gpg status codes properly.

Yep, I agree.

-Peff