Smalltalk › Squeak › Squeak - Dev

Odd difference in MC repository connections between 32 & 64 bit images

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

2 messages Options

timrowledge

Odd difference in MC repository connections between 32 & 64 bit images

After having fun running tests I went to publish the changes for a few _ assigns etc and then ran the MC 'check all packages for changes' to make sure I'd saved everything.

On my Pi 32 bit image - which I'm pretty sure is descended from a published-on-squeak.org Squeak5.3-19431-32bit.image - it stopped several times complaining about no baseImage. I compared the state to a 64 bit image on my iMac and almost randomly checked the repositories connected to the problem package. In each case it was lacking the trunk and adding that and then restarting from #checkSomethingorottheriforgetexactly made everything continue happily.

I'm not sure what might have lead to this difference, and it was only five or six packages and ther are other packages lacking the trunk that did not cause a problem. So I'm just throwing it out there in case it makes ssnese to anyone.

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
Strange OpCodes: CDC: Clear Disks and Crash

marcel.taeumel

Re: Odd difference in MC repository connections between 32 & 64 bit images

Hi Tim.

> In each case it was lacking the trunk

That's expected for a release image. They just point to the repositories "squeak53" and "inbox" and some Metacello repositories

Best,

Marcel

Am 11.05.2020 07:31:01 schrieb tim Rowledge <[hidden email]>:
After having fun running tests I went to publish the changes for a few _ assigns etc and then ran the MC 'check all packages for changes' to make sure I'd saved everything.

On my Pi 32 bit image - which I'm pretty sure is descended from a published-on-squeak.org Squeak5.3-19431-32bit.image - it stopped several times complaining about no baseImage. I compared the state to a 64 bit image on my iMac and almost randomly checked the repositories connected to the problem package. In each case it was lacking the trunk and adding that and then restarting from #checkSomethingorottheriforgetexactly made everything continue happily.

I'm not sure what might have lead to this difference, and it was only five or six packages and ther are other packages lacking the trunk that did not cause a problem. So I'm just throwing it out there in case it makes ssnese to anyone.

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
Strange OpCodes: CDC: Clear Disks and Crash