[PATCH 4 of 4] changeset_printer: hide manifest node for workingctx

Matt Mackall mpm at selenic.com
Fri Apr 3 10:55:21 CDT 2015


On Fri, 2015-04-03 at 22:31 +0900, Yuya Nishihara wrote:
> On Thu, 02 Apr 2015 15:39:08 -0500, Matt Mackall wrote:
> > It's worth considering defining __cmp__ on ctxs. Obviously the semantics
> > of ctxa < ctxb are not a perfect match for a mere partial ordering.. but
> > we do have a partial ordering that we regularly use as a shortcut and we
> > should make it convenient.
> 
> Is your suggestion something like this?
> 
>   def __cmp__(self, other):
>       return cmp(intrev(self), intrev(other))

Like this.

> or
> 
>   def __cmp__(self, other):
>       if self == other:
>           return 0
>       elif other in self.ancestors():  # naive implementation
>           return -1
>       else:
>           return 1
> 
> I agree it will be sometimes convenient, but I'm not sure how often we'll
> do "ctxa < ctxb".

We already compare revs in a bunch of places.

-- 
Mathematics is the supreme nostalgia of our time.




More information about the Mercurial-devel mailing list