[PATCH 1 of 2] bookmarks: don't use bookmarks.listbookmarks in local computations

[Matt Mackall]

On Tue, 2013-01-29 at 09:18 -0600, Kevin Bullock wrote:
> On 29 Jan 2013, at 1:55 AM, Matt Mackall wrote:
> 
> > On Mon, 2013-01-28 at 21:24 -0600, Kevin Bullock wrote:
> >> On 27 Jan 2013, at 5:59 PM, Pierre-Yves David wrote:
> >> 
> >>> On 27 janv. 2013, at 22:15, Kevin Bullock wrote:
> >>> 
> >>>> # HG changeset patch
> >>>> # User Kevin Bullock <kbullock at ringworld.org>
> >>>> # Date 1359318277 21600
> >>>> # Branch stable
> >>>> # Node ID 14fe65a26547fe63963f4e6062b26f07a9443860
> >>>> # Parent  b404fa103253960824500066dffa2b7d6cff33ee
> >>>> bookmarks: don't use bookmarks.listbookmarks in local computations
> >>>> 
> >>>> bookmarks.listbookmarks is for wire-protocol use. The normal way to get
> >>>> all the bookmarks on a local repository is repo._bookmarks.
> >>> 
> >>> repo._bookmarks.values() and bookmarks.listbookmarks(repo) have different returns.
> >>> (divergent bookmark are excluded from listbookmarks)
> >>> 
> >>> I would make sense to test that neither filtering nor revset ignore them.
> >>> 
> >>> Patches (1 and 2) looks good to me otherwise.
> >> 
> >> Crewed the series, along with a test that bookmarks prevent changesets
> >> from being hidden (didn't appear to already be one). Thanks for the
> >> review.
> > 
> > Is there a bug that needs to be marked testing for this?
> 
> Pretty sure there's not. I just noticed it while tracking down the
> issue fixed in the following patch. How likely is it that someone
> would have a divergent bookmark on a secret changeset?

Dunno, just making sure we didn't fail to connect some dots.

-- 
Mathematics is the supreme nostalgia of our time.