Re: [Monotone-devel] Re: monotone-viz oddities again?

[Richard Levitte - VMS Whacker]

In message <address@hidden> on Wed, 26 Oct 2005 10:53:17 +0200, Olivier Andrieu 
<address@hidden> said:

oliv__a>  Richard Levitte - VMS Whacker [Wednesday 26 October 2005] :
oliv__a>  >
oliv__a>  > richard> I just noticed that I'm currently getting a
oliv__a>  > richard> disconnected graph in monotone-viz.  Am I alone
oliv__a>  > richard> in this?  Olivier, can you remind me how to
oliv__a>  > richard> generate some debugging output for you?
oliv__a> 
oliv__a> Yes I've noticed this. Actually it's because the graph is no
oliv__a> more disconnected. Someone did a commit&merge on a rather old
oliv__a> revision (from august) which had effect on reconnecting what
oliv__a> was previously two components. I'm not sure what I could do
oliv__a> to "fix" the layout.

OK, so if understand correctly, checking for other possible paths is
only done when there's a disconnection in the graph, correct?  I
understand why you don't do that for all nodes, as I can imagine the
time spent would not make it worth it.  However, here's an idea:
regard apparent leaves as a possible disconnection, and check if
monotone really regards them as leaves.  If not, they should be
treated as a graph disconnection.  Of course, there will be cases when
that doesn't lead any further, but I assume you know how to take care
of that.

oliv__a>  > And oh, I work with the bleeding edge, both of monotone-viz
oliv__a>  > (4e8b670a7090b6a49ea60440be4182b1bb6ca86c) and monotone
oliv__a>  > (e54b9741e204c966f94a71bbccafd6202cea0eb1).
oliv__a> 
oliv__a> Ah, I forgot to push my changes to the venge.net server so,
oliv__a> in fact, you're not so bleeding edge :). I just pushed them
oliv__a> now. With the latest monotone-viz you can view any set of
oliv__a> branches (and not just a branch / a branch + all
oliv__a> sub-branches). If you select n.v.monotone and the branch that
oliv__a> was responsible for the disconnection in the graph (was that
oliv__a> n.v.rewrites.path_handling ?), I think you'll get a nicer
oliv__a> layout.

OK, I'll play with that.

I also noticed that the current real leaf doesn't stay at the bottom
for the moment (because of the disconnected behavior).  While I like
the more compact display, in the current situation, it's really
confusing to have the real leaf somewhere in the middle of the
graph...

Cheers,
Richard

-----
Please consider sponsoring my work on free software.
See http://www.free.lp.se/sponsoring.html for details.

-- 
Richard Levitte                         address@hidden
                                        http://richard.levitte.org/

"When I became a man I put away childish things, including
 the fear of childishness and the desire to be very grown up."
                                                -- C.S. Lewis