Hi Levente --

> Yes, but that's not the case with Eliot's changes.

Loading the new method (override) MethodDictionary >> #atNewIndex:put: first will break the image.

Yet, the actual issue was that #at:put: was loaded before #atNewIndex:putValue:. So, I suspect both a bug in the Monticello load order and a more generic problem that might not be so easily fixed.

The issue with the load order might be easily fixed in MCPackageLoader >> #basicLoad:

"..."
"Pass 2: We compile new / changed methods"
methodAdditions do: "..."

Best,
Marcel
Am 21.09.2021 14:53:00 schrieb Levente Uzonyi <leves at caesar.elte.hu>:
On Tue, 21 Sep 2021, Marcel Taeumel wrote:

> Hi Levente --
> > What you described sounds like a bug in the update mechanism.
>
> Maybe. But loading new methods first can also break the system if those new methods override existing ones in the super class.

Yes, but that's not the case with Eliot's changes.


Levente
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20210921/8adacae5/attachment.html>