Re: ds/commit-graph-incremental (was Re: What's cooking in git.git (Jun 2019, #04; Fri, 14))

[Derrick Stolee <stolee@gmail.com>]

On 6/19/2019 10:32 AM, Junio C Hamano wrote:
> Derrick Stolee <stolee@gmail.com> writes:
> 
>> On 6/14/2019 4:50 PM, Junio C Hamano wrote:
>>> * ds/commit-graph-incremental (2019-06-12) 16 commits
>>>  - commit-graph: test --split across alternate without --split
>>
>> Please hold on this one. I've found multiple issues while integrating
>> this with VFS for Git and there are enough to merit a full re-roll.
> 
> Thanks.
> 
>> Please ignore the two patches I sent yesterday as I will incorporate them
>> into the next version of this series.
> 
> I think I picked up "normalize c-g filenames" and queued it directly
> on top of this topic, but I do not recall the other one, which
> probably means I did ignore it so hopefully no harm done ;-)

Do you mean you applied the v6 patches? [1]

The range-diff included some changes that are necessary due to textual or
semantic conflicts with ds/close-object-store, but one change was hidden
in PATCH 10 that is important:
.
10:  2093bab5b1 ! 10:  65b1cc6ae9 commit-graph: allow cross-alternate chains
     @@ -97,6 +97,8 @@
      -	for (i = 0; i < count && valid; i++) {
      -		char *graph_name;
      -		struct commit_graph *g;
     ++	prepare_alt_odb(r);
     ++
      +	for (i = 0; i < count; i++) {
      +		struct object_directory *odb;

If you replayed the entire v6 series on top of ds/close-object-store, then
everything is as I had hoped. Otherwise, I can take a look at what you have
applied after the next cooking email and send new patches.

Thanks,
-Stolee       

[1] https://public-inbox.org/git/pull.184.v6.git.gitgitgadget@gmail.com/T/#u