builtin/clone.c: don't ignore transport_fetch_refs() errors
authorTaylor Blau <me@ttaylorr.com>
Thu, 3 Dec 2020 18:55:13 +0000 (13:55 -0500)
committerJunio C Hamano <gitster@pobox.com>
Thu, 3 Dec 2020 20:42:29 +0000 (12:42 -0800)
commitaab179d937379e28c67dcab0a46fdba05c11d919
tree6d7a91e60b0c5f4fba63936f842fbf3fdccb0bb2
parent3a0b884caba2752da0af626fb2de7d597c844e8b
builtin/clone.c: don't ignore transport_fetch_refs() errors

If 'git clone' couldn't execute 'transport_fetch_refs()' (e.g., because
of an error on the remote's side in 'git upload-pack'), then it will
silently ignore it.

Even though this has been the case at least since clone was ported to C
(way back in 8434c2f1af (Build in clone, 2008-04-27)), 'git fetch'
doesn't ignore these and reports any failures it sees.

That suggests that ignoring the return value in 'git clone' is simply an
oversight that should be corrected. That's exactly what this patch does.
(Noticing and fixing this is no coincidence, we'll want it in the next
patch in order to demonstrate a regression in 'git upload-pack' via a
'git clone'.)

There's no additional logging here, but that matches how 'git fetch'
handles the same case. An assumption there is that whichever part of
transport_fetch_refs() fails will complain loudly, so any additional
logging here is redundant.

Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
builtin/clone.c