[squeak-dev] MC true ancestors false positive?

Chris Muller asqueaker at gmail.com
Sat Feb 16 00:15:07 UTC 2019

> With the small utility that i just added, I did find the merge of the two Monticello-bf.540
> It's Monticello-topa.599 from september 2014.
> So we effectively have both bf.540 packages as valid ancestors, one in trunk, the other in the inbox, but we cannot have them both in same repository.
> We could perform some chirurgy, renaming inbox/Monticello-bf.540 with a branch name => Monticello.partialCommit-bf.540
> But then, we will have to apply this chirurgy to every offspring down to working copy, and republish every offspring package...
> Since 2014, that's quite some progeny!
> I guess that we should also apply this to the inbox offsprings, and let the responsibility of updating any other replication to their owner...
> We can as well ignore the problem, move inbox/Monticello-bf.540 to TreatedInbox, since MC is robust to missing packages...
> Thoughts?

-1 unless you're willing to address the underlying problem of
Monticello generating duplicate names.  Four alphanumeric digits can
get us until 2022...

     36rZZZZ minutes fromNow    " 2022-04-27T03:47:51.947721-06:00"

... after which we'd have to add a fifth, which would last more than a
century.  Please consider it.

In the meantime, I had some thoughts about rectifying this symptom
too.  I think we should make a hack which establishes a variable
somewhere (possibly MCHttpRepository) that can keep an "exceptions
list" for certain versions which will tell it to look in another
repository for the offending version uuid.

 - Chris

More information about the Squeak-dev mailing list