On Sat, May 12, 2018 at 04:13:59PM +0200, Duy Nguyen wrote:
> > Should this one be global in the first place? Can we tie it to say
> > "struct rev_info" or something? I'd somehow anticipate a far future
> > where object flag bits used for traversal book-keeping would be moved
> > out of the obje
On Sat, May 12, 2018 at 3:58 PM, Junio C Hamano wrote:
> Jeff King writes:
>
>> On Sat, May 12, 2018 at 10:00:22AM +0200, Nguyễn Thái Ngọc Duy wrote:
>>
>>> diff --git a/revision.h b/revision.h
>>> index b8c47b98e2..72404e2599 100644
>>> --- a/revision.h
>>> +++ b/revision.h
>>> @@ -6,6 +6,7 @@
>
Jeff King writes:
> On Sat, May 12, 2018 at 10:00:22AM +0200, Nguyễn Thái Ngọc Duy wrote:
>
>> diff --git a/revision.h b/revision.h
>> index b8c47b98e2..72404e2599 100644
>> --- a/revision.h
>> +++ b/revision.h
>> @@ -6,6 +6,7 @@
>> #include "notes.h"
>> #include "pretty.h"
>> #include "diff.h
On Sat, May 12, 2018 at 10:00:22AM +0200, Nguyễn Thái Ngọc Duy wrote:
> diff --git a/revision.h b/revision.h
> index b8c47b98e2..72404e2599 100644
> --- a/revision.h
> +++ b/revision.h
> @@ -6,6 +6,7 @@
> #include "notes.h"
> #include "pretty.h"
> #include "diff.h"
> +#include "commit-slab.h"
>
Instead of relying on commit->util to store the source string, let the
user provide a commit-slab to store the source strings in.
It's done so that commit->util can be removed. See more explanation in
the commit that removes commit->util.
Signed-off-by: Nguyễn Thái Ngọc Duy
---
builtin/fast-exp
5 matches
Mail list logo