Since v1, I've took a look at 3 other cases that use has_object_file():
2 don't need lazy-fetching (patches 3 and 4 in this set) and 1 does -
"cat-file -e" - although I think this is a special case. So I still
think that not lazy-fetching when checking object existence is more
likely, and should be privileged with the shorter function name
(has_object() instead of has_object_locally()).
Changes from v1:
- Patch split into 2 (patch 1 and patch 2)
- 2 additional patches that fix bugs by making use of the new function
Jonathan Tan (4):
sha1-file: introduce no-lazy-fetch has_object()
apply: do not lazy fetch when applying binary
pack-objects: no fetch when allow-{any,promisor}
fsck: do not lazy fetch known non-promisor object
Documentation/git-pack-objects.txt | 11 +++++++----
apply.c | 2 +-
builtin/fsck.c | 2 +-
builtin/pack-objects.c | 4 ++--
object-store.h | 25 +++++++++++++++++++++++--
sha1-file.c | 12 ++++++++++++
t/t4150-am.sh | 16 ++++++++++++++++
7 files changed, 62 insertions(+), 10 deletions(-)
--
2.28.0.236.gb10cc79966-goog
