On Tue, Apr 27, 2010 at 7:15 PM, Torsten Bergmann
<astares@gmx.de> wrote:
Hi Mariano,
now he complains about "OmniBrowser-DamineCassou.462"
�(which comes from 1.0 of ConfigurationOfAutomaticMethodCategorizer
� which loads 1.0 of ConfigurationOfOmniBrowser
� which loads OmniBrowser-DamienCassou.462)
Looks like this conflicts with "OmniBrowser-lr.469"
which is noted in version 1.1.3 of ConfigurationOfOmniBrowser
Looks like a conflict:
You now use 1.1.3 of ConfigurationOfOmniBrowser, but the
ConfigurationOfAutomaticMethodCategorizer that you use in
ConfigurationOfPharo still points to 1.0 of ConfigurationOfOmniBrowser.
Yes, but it doesn't matter. In such situation (conflict), Metacello will load the latest version, in this case, 1.1.3.
I cannot mantain all configurations. If you think ConfigurationOfAutomaticMethodCategorizer should have 1.1.3 instead of 1.0, just do it.
�
Lesson learned: take care when you edit configs since you may break other
peoples code.
I do take care, but I cannot test all configurations. Even more, when they are not stable releases.
In addition, I didn't break anything:
OmniBrowser-DamienCassou.462
�
existed at the point I did the configuration. Now, if I browse the repository
http://source.wiresong.ca/obI don't see it. So, maybe Colin removed that version and put Lukas ones and that's why it is failing.
Colin, do you know if this was what happened ?
So I would not call this a reproducable Pharo 1.0 Release ;)
I don't understand. Pharo 1.0 is completly reproducable. I take a PharoCore 1.0 I avaluate:
Gofer new
��� squeaksource: 'MetacelloRepository';
��� package: 'ConfigurationOfPharo';
��� load.
((Smalltalk at: #ConfigurationOfPharo) project version: '1.0-10517') perform: #silently: with: true; perform: #load.
And I have the PharoDev 1.0 stable release in perfect state and without any problem.
Cheers
Mariano
�