So I tried my own fix:

1) create Kernel-nice.779 = fbs.779+ Class>>declare: fix
2) use this to replace Kernel-fbs.779 starting at updates-fbs.239 (I have
overwritten the updates)
3) merge Kernel-fbs.791+Kernel-nice.779 -> Kernel-nice.792
4) use it in latest updates in place of Kernel-fbs.790 and 791

Stage 3) is a no-op except for ancestry: it avoids having a dead commit
(those marked bold because not in ancestry)

Please report if problems...


2013/7/21 Nicolas Cellier <nicolas.cellier.aka.nice at gmail.com>

> Yes that works,you can edit an existing map and save it under same name.
>
>
> 2013/7/20 David T. Lewis <lewis at mail.msen.com>
>
> On Sat, Jul 20, 2013 at 08:02:18PM +0100, Frank Shearar wrote:
>> > On 20 July 2013 18:48, Chris Muller <asqueaker at gmail.com> wrote:
>> > > Hey all, when the trunk is broken, the community is at a full stop.
>> > > Can we please all jump-in and make a push to just solve this?  I'm
>> > > dead in the water and it sounds like others are too.
>> > >
>> > > The status I'm seeing currently is -- when I simply download the
>> > > latest alpha image, Squeak4.5-12641.image, and then "Update Squeak"
>> > > from the menu, it blows up during Morphic-fbs.663 trying to declare:
>> > > PianoRollNoteMorph because it's environment is nil.
>> > >
>> > > How can we fix this?
>> >
>> > I don't know if this is feasible or possible, but can we _delete_
>> > config maps?
>>
>> I'm no expert here, but I think that an existing configuration map can
>> be rewritten by editing it and saving it under its existing name. If so,
>> it may be possible to take a configuration map that is broken with respect
>> to updating from its predecessor, and fix it.
>>
>> Dave
>>
>>
>> > If so, let's delete everything after NNN (where NNN is
>> > the latest map that doesn't load Morphic-fbs.663), and issue a new
>> > config map that loads in this order:
>> >
>> > Kernel-fbs.791
>> > Morphic-dtl.688
>> > Files-fbs.125
>> > System-fbs.569
>> >
>> > What's stopping me from simply _doing_ this is that I don't really
>> > understand the process around config maps. Hence my frequent calls for
>> > help these past few days!
>> >
>> > I'll go and see if I can figure out the value of NNN in the meantime.
>> >
>> > frank
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20130721/dcf98e76/attachment.htm