Re: [RFC PATCH 00/19] object-store refactoring 3 (replace objects, main ref store)
- From
Stefan Beller <sbeller@google.com>
- Date
- Apr 9, 2018, 17:39 UTC
- Message-ID
- <CAGZ79kaLDTY7zeG6f2pGE6ZJvScM-FunjLkeAqu_YsUQEoepqg@mail.gmail.com>
- In-Reply-To
- <CACsJy8BKKgfxUGm+SgOAENNsSaMGehU5NVbn6YuebCgUL4S0jQ@mail.gmail.com>
On Sat, Apr 7, 2018 at 2:50 AM, Duy Nguyen <pclouds@gmail.com> wrote:
Show 17 quoted lines
> On Sat, Apr 7, 2018 at 1:21 AM, Stefan Beller <sbeller@google.com> wrote: *
>> diff --git a/repository.h b/repository.h
>> index 09df94a472..2922d3a28b 100644
>> --- a/repository.h
>> +++ b/repository.h
>> @@ -26,6 +26,11 @@ struct repository {
>> */
>> struct raw_object_store *objects;
>>
>> + /*
>> + * The store in which the refs are hold.
>> + */
>> + struct ref_store *main_ref_store;
>> +
>
> We probably should drop the main_ prefix here because this could also
> be a submodule ref store (when the repository is about a submodule).it's still the main ref store of the submodule, then. :) But yes, I agree we should rename it. Now or eventually later?
> worktree ref store is a different story and I don't know how to best > present it here yet (I'm still thinking a separate struct repository).
I imagine that we'd rather want to have arrays of config/worktree path/refs when needed and the worktree is just an index into all of these things?
> But we can worry about that when struct repository supports multiple > worktree.
ok. Thanks for bringing this to my attention.
Thanks, Stefan