From: Johan Herland <johan@herland.net>
Date: Fri, 22 Oct 2010 13:32:24 GMT
Subject: Re: [PATCHv4 05/21] notes.h/c: Clarify the handling of notes objects that are == null_sha1
Message-ID: <201010221532.24436.johan@herland.net>
In-Reply-To: <20101021175404.GC11328@burratino>

On Thursday 21 October 2010, Jonathan Nieder wrote:
> Johan Herland wrote:
> >   Allow combine_notes functions to request that a note be removed,
> >   by setting the resulting note SHA1 to null_sha1 (0000000...).
> >
> >   For consistency, also teach note_tree_insert() to skip insertion
> >   of an empty note when there is no note to combine it with.
>
> I guess it's probably also worth clarifying that the term "empty
> note" means null_sha1 rather than empty_blob_sha1.
>
> i.e. something like:
>
> 	For consistency, also teach note_tree_insert() to skip insertion
> 	of an empty note (a note with val_sha1 equal to null_sha1)
> 	when there is no note to combine it with.

Agreed. Incorporated.

> The previous behavior iiuc would have been for write_notes_tree() to
> write invalid trees after such an operation.

Indeed, you are correct. I've added a note about this to the commit 
message.


...Johan

-- 
Johan Herland, <johan@herland.net>
www.herland.net