obsolete marker terminology

Matt Harbison mharbison72 at gmail.com
Thu Nov 21 02:37:11 UTC 2019


On Wed, 20 Nov 2019 12:38:50 -0500, Pierre-Yves David  
<pierre-yves.david at ens-lyon.org> wrote:

> These terms are already used by the normal graph. So we cannot really  
> reuse them directly.
>
> Are you suggesting we use:
>
> - successors-parents,
> - successors-descendants,
> - predecessors-parents,
> - predecessors-ancestors ?
>
>
> On that topic, we started to talk about `evolution` to refer to the  
> various iteration over a changesets. So maybe:
>
> - successor,
> - evolution-child,
> - evolution-parent,
> - predecessor
>
> ?

direct-predecessor and direct-successor seem more clear and explicit than  
evolution-*.

And going one step further, it might make sense that these are looking for  
the direct successor/predecessor in global state (and could therefore be  
empty if something was edited in another clone), and change the existing  
predecessor/successor to find the closest one that is currently in the  
repo.  Similar to what I was wondering about on IRC the other day.



More information about the Evolve-testers mailing list