Hi!
In my quest to use the toolbox for managing the version of moose, I bumped into a serious problem I think. I defined version 4.3.1 in ConfigurationOfMoose using the toolbox. The validator does not complain: (MetacelloToolBox validateConfiguration: ConfigurationOfMoose) isEmpty => true However, executing: (ConfigurationOfMoose project version: '4.3.1') load raises en error: 'No version found for ''1.1'' of ConfigurationOfHealthReportProducer' Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
On 02/16/2011 02:57 PM, Alexandre Bergel wrote:
> Hi! > > In my quest to use the toolbox for managing the version of moose, I bumped into a serious problem I think. > I defined version 4.3.1 in ConfigurationOfMoose using the toolbox. The validator does not complain: > (MetacelloToolBox validateConfiguration: ConfigurationOfMoose) isEmpty > => true > > However, executing: > (ConfigurationOfMoose project version: '4.3.1') load > > raises en error: > 'No version found for ''1.1'' of ConfigurationOfHealthReportProducer' > > Cheers, > Alexandre I'll take a look ... I have a guess, but I'll look first:) I'll also check your other issue... first thing though, i need to check out the hudson metacello test failures. Dale |
In reply to this post by Alexandre Bergel-5
On 02/16/2011 02:57 PM, Alexandre Bergel wrote:
> Hi! > > In my quest to use the toolbox for managing the version of moose, I bumped into a serious problem I think. > I defined version 4.3.1 in ConfigurationOfMoose using the toolbox. The validator does not complain: > (MetacelloToolBox validateConfiguration: ConfigurationOfMoose) isEmpty > => true > > However, executing: > (ConfigurationOfMoose project version: '4.3.1') load > > raises en error: > 'No version found for ''1.1'' of ConfigurationOfHealthReportProducer' > > Cheers, > Alexandre Alexandre, My guess was wrong:) However, I am not able to reproduce the problem. I started with ConfigurationOfMoose-AlexandreBergel.179 and I have gotten through the fetch phase (where the missing version would have been reported) and I'm in the load phase which is still going strong as far as I can tell ... So there must be something specific to your particular image that is causing the problems ... Note that I could reproduce Issue 107 either ... Are there any "dirty" configuration packages in your image? If so they are the possible cause of the problem, especially if the dirty versions are older than the one(s) needed by another configuration... BTW, I got a dependency error during compilation where MAMatrix and others are needed to load Dude, but I assume that these problems aren't related ... Dale |
Can it be an old version of that package in cache?
On Wed, Feb 16, 2011 at 8:42 PM, Dale Henrichs <[hidden email]> wrote:
|
Actually no, but reseting the cache solves another problem with Monticello. So thanks for this :-)
Alexandre On 16 Feb 2011, at 23:44, Guillermo Polito wrote: > Can it be an old version of that package in cache? > > On Wed, Feb 16, 2011 at 8:42 PM, Dale Henrichs <[hidden email]> wrote: > On 02/16/2011 02:57 PM, Alexandre Bergel wrote: > Hi! > > In my quest to use the toolbox for managing the version of moose, I bumped into a serious problem I think. > I defined version 4.3.1 in ConfigurationOfMoose using the toolbox. The validator does not complain: > (MetacelloToolBox validateConfiguration: ConfigurationOfMoose) isEmpty > => true > > However, executing: > (ConfigurationOfMoose project version: '4.3.1') load > > raises en error: > 'No version found for ''1.1'' of ConfigurationOfHealthReportProducer' > > Cheers, > Alexandre > > Alexandre, > > My guess was wrong:) However, I am not able to reproduce the problem. I started with ConfigurationOfMoose-AlexandreBergel.179 and I have gotten through the fetch phase (where the missing version would have been reported) and I'm in the load phase which is still going strong as far as I can tell ... > > So there must be something specific to your particular image that is causing the problems ... Note that I could reproduce Issue 107 either ... > > Are there any "dirty" configuration packages in your image? If so they are the possible cause of the problem, especially if the dirty versions are older than the one(s) needed by another configuration... > > BTW, I got a dependency error during compilation where MAMatrix and others are needed to load Dude, but I assume that these problems aren't related ... > > Dale > -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
In reply to this post by Dale Henrichs
Humm... It could be due to my image. I wish I could write test about this.
I see that in your test you make an intensive use of Gofer. Is this really mandatory? I cannot write tests about Metacello without saving and loading? Cheers, Alexandre >> However, executing: >> (ConfigurationOfMoose project version: '4.3.1') load >> >> raises en error: >> 'No version found for ''1.1'' of ConfigurationOfHealthReportProducer' >> >> Cheers, >> Alexandre > > Alexandre, > > My guess was wrong:) However, I am not able to reproduce the problem. I started with ConfigurationOfMoose-AlexandreBergel.179 and I have gotten through the fetch phase (where the missing version would have been reported) and I'm in the load phase which is still going strong as far as I can tell ... > > So there must be something specific to your particular image that is causing the problems ... Note that I could reproduce Issue 107 either ... > > Are there any "dirty" configuration packages in your image? If so they are the possible cause of the problem, especially if the dirty versions are older than the one(s) needed by another configuration... > > BTW, I got a dependency error during compilation where MAMatrix and others are needed to load Dude, but I assume that these problems aren't related ... > > Dale > -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
On 02/17/2011 07:38 AM, Alexandre Bergel wrote:
> Humm... It could be due to my image. I wish I could write test about this. > I see that in your test you make an intensive use of Gofer. Is this really mandatory? I cannot write tests about Metacello without saving and loading? At the end of the day, if the correct mcz files don't make it into the image, there something wrong with Metacello, so for a large number of the tests, I need to use Gofer-based tests. In this particular case, lets assume that the state of your image _is_ affecting the outcome of your load .... to write a test case, we have to create an image state that allows us to duplicate the problem ... The bulk of the issues that I have dealt with in the last 6 months or so have all been related to image state having an impact on the load and getting those algorithms right, so I have been writing a lot of Gofer-based tests lately.... Dale |
>
> At the end of the day, if the correct mcz files don't make it into the image, there something wrong with Metacello, so for a large number of the tests, I need to use Gofer-based tests. > > In this particular case, lets assume that the state of your image _is_ affecting the outcome of your load .... to write a test case, we have to create an image state that allows us to duplicate the problem ... > > The bulk of the issues that I have dealt with in the last 6 months or so have all been related to image state having an impact on the load and getting those algorithms right, so I have been writing a lot of Gofer-based tests lately.... Ok. Thanks. I am trying to fully understand Metacello. I really want to have a sexy GUI for Metacello :-) Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. |
Free forum by Nabble | Edit this page |