[PATCH 2 of 2] graphlog: draw graphs in give order

vsh vsh426 at gmail.com
Mon Apr 5 05:55:52 CDT 2010


On Mon, Apr 5, 2010 at 1:44 PM, Benoit Boissinot <
benoit.boissinot at ens-lyon.org> wrote:

> On Mon, Apr 05, 2010 at 06:33:46AM +0500, vsh wrote:
> > On Mon, Apr 5, 2010 at 4:43 AM, Benoit Boissinot <
> > benoit.boissinot at ens-lyon.org> wrote:
> >
> > > On Sun, Apr 04, 2010 at 11:39:53PM +0530, Vishakh wrote:
> > > > # HG changeset patch
> > > > # User Vishakh <vsh426 at gmail.com>
> > > > # Date 1270404568 -19800
> > > > # Node ID 4a7f1e4cb05f497e60d957e0fb73a3f4c351b70c
> > > > # Parent  8f5646c4358358a4249273285793544f6e466443
> > > > graphlog: draw graphs in give order
> > > >
> > >
> > > Looking at the test output, it's not exactly nice.
> > >
> >
> > i am not sure exactly what is not nice. i should have mentioned
> > that the meaning of o---+ is also reversed. originally + implied
> > follow to see the parent while in the reverse graph its follow to see
> > children.
>
> So it would be nicer if we don't reverse the meaning of + ?
>

it might be, it makes sense to me though(after the first couple of times,
but that happened with the original glog as well). but it would definitely
take more code. currently i have changed the main algorithm as less as
possible, just in case :)

> >
> > >
> > > > diff -r 8f5646c43583 -r 4a7f1e4cb05f tests/test-glog.out
> > > > --- a/tests/test-glog.out     Sun Apr 04 23:30:24 2010 +0530
> > > > +++ b/tests/test-glog.out     Sun Apr 04 23:39:28 2010 +0530
> > > > @@ -307,6 +307,245 @@
> > > >     date:        Thu Jan 01 00:00:00 1970 +0000
> > > >     summary:     (0) root
> > > >
> > > > +% reverse glog -r :
> > > > +o                changeset:   0:e6eb3150255d
> > > > +|\               user:        test
> > > > +| |\             date:        Thu Jan 01 00:00:00 1970 +0000
> > > > +| | |\           summary:     (0) root
> > > > +| | | |\
> > > > +| | | | |\
> > > > +| | | | | |\
> > > > +| | | | | | |\
> > > > +o---+-+-+-+ | |            changeset:   1:6db2ef61d156
> > > > +|\ \ \ \ \ \ \ \           user:        test
> > > > +| |\ \ \ \ \ \ \ \         date:        Thu Jan 01 00:00:01 1970
> +0000
> > > > +| | |\ \ \ \ \ \ \ \       summary:     (1) collapse
> > > > +| | | |\ \ \ \ \ \ \ \
> > > > +| | | | |\ \ \ \ \ \ \ \
> > > > +o---+ | | | | | | | | | |      changeset:   2:3d9a33b8d1e1
> > > > +|\ \ \ \ \ \ \ \ \ \ \ \ \     user:        test
> > > > +| |\ \ \ \ \ \ \ \ \ \ \ \ \   date:        Thu Jan 01 00:00:02 1970
> > > +0000
> > > > +| | | | | | | | | | | | | | |  summary:     (2) collapse
> > >
> > > The edges above look really weird. Additionally, should the graph look
> > > the same as the previous graph, but reversed? The graph above in the
> > > test output looks much simpler.
> > >
> > > edges look ok in my terminal. which part looks weird?
> > the graph may or may not look similar to original graph in reverse. that
> > depends on the amount of branching. the example in test was contrived to
> > contain many complicated relationships. in other cases/repos it looked
> > reasonable.  glog -r0:tip on the  mercurial repo is much saner and
> > looks similar to reverse of tip:0.
>
> In my opinion, there is still an alignement problem with the '\' edges
> in the example above.
>

i agree about the alignment being off, but graphlog does the same thing in
similar situations AFAIK

>
> Benoit
>
> --
> :wq
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20100405/9e4c2ce2/attachment.htm>


More information about the Mercurial-devel mailing list