[RFC] A better way to help users understand and resolve confclits

Guilhem Bichot guilhem at sun.com
Thu Aug 13 13:51:57 BST 2009


Hello Vincent,

thank you so much for creating this proposal.
Here are small comments.

Vincent Ladeuil a écrit, Le 13.08.2009 10:46:
> Hi all,
> 
> Here is a rough draft on some ideas about how we can help users
> better understand the origins of conflicts and resolve them more
> easily.
> 
> It's quite light on the design because I want to write some tests
> first to make sure I've got all the main scenarios covered.
> 
> That means tests that do:
> 1) create conflicts,
> 2) resolve them in various ways,
> 3) commit (i.e. ensure we can commit a *clean* tree without any
>   .OTHER, .moved, .diverted garbage)
> 
> From there I'd like to be able to propose both text and graphical UI[s].
> 
> Feedback welcome,
> 
>          Vincent

I think what can enlighten users most is explanations like you wrote:

bzr resolve --interactive <item>

<base> is <revid>/<revno>
<other> wants to create <item> bzr created <item>.moved
^^^^^^^^ here I'd be verbose and add "but <item> already exists so"

<this> keeps <item>
1) Right, get rid of .moved
2) Wrong, .moved is the good one, get rid of <item>
3) Abort, let me handle that myself.
4) Do nothing, just mark as resolve
^^^^^^^^ good, except that I wonder when (4) is acceptable ^^^

Slightly related is this bug report which also falls in the category of 
"better explanations":
"show base for weave merge"
https://bugs.edge.launchpad.net/bzr/+bug/40412
It should not be discussed here, it's just something to keep in mind.

-- 
Mr. Guilhem Bichot <guilhem at sun.com>
Sun Microsystems / MySQL, Lead Software Engineer
Bordeaux, France
www.sun.com / www.mysql.com



More information about the bazaar mailing list