From: Derrick Stolee Date: Tue, 10 Oct 2017 13:11:15 GMT Subject: Re: [PATCH v4 4/4] sha1_name: minimize OID comparisons during disambiguation Message-ID: In-Reply-To: On 10/10/2017 8:56 AM, Junio C Hamano wrote: > Jeff King writes: > >> OK, I think that makes more sense. But note the p->num_objects thing I >> mentioned. If I do: >> >> git pack-objects .git/objects/pack/pack > >> then I have a pack with zero objects, which I think we'd similarly want >> to return early from. I.e., I think we need: >> >> if (p->num_objects) >> return; >> >> Technically that also covers open_pack_index() failure, too, but that's >> a subtlety I don't think we should rely on. > True. I notice that the early part of the two functions look almost > identical. Do we need error condition handling for the other one, > too? I prefer to fix the problem in all code clones when they cause review friction, so I'll send a fifth commit showing just the diff for these packfile issues in sha1_name.c. See patch below. Should open_pack_index() return a non-zero status if the packfile is empty? Or, is there a meaningful reason to have empty packfiles? Thanks, -Stolee diff --git a/sha1_name.c b/sha1_name.c index 49ba67955..9f8a33e82 100644 --- a/sha1_name.c +++ b/sha1_name.c @@ -153,7 +153,9 @@ static void unique_in_pack(struct packed_git *p,         uint32_t num, last, i, first = 0;         const struct object_id *current = NULL; -       open_pack_index(p); +       if (open_pack_index(p) || !p->num_objects) +               return; +         num = p->num_objects;         last = num;         while (first < last) { @@ -513,7 +515,9 @@ static void find_abbrev_len_for_pack(struct packed_git *p,         uint32_t num, last, first = 0;         struct object_id oid; -       open_pack_index(p); +       if (open_pack_index(p) || !p->num_objects) +               return; +         num = p->num_objects;         last = num;         while (first < last) {