On Thu, Jun 26, 2014 at 12:44 AM, stepharo <stepharo(a)free.fr> wrote: Yes...circularity should produce an error ... Do you have an idea of what could be the problems? Not really, however, if you are running with Pharo3.0 I understand that the Metacello tests are not green (they've never been green in my github/travis world ... but at one point in time I understand that they were green in Pharo3.0) ... so I do not trust Metacello in Pharo3.0 ... With that said, the failing tests may have nothing to do with this problem ... however without a baseline of green tests any attempts at fixes would be highly suspect ... > I will check with smaller configurations but this is a daunting task. I understand ... > Snapshotcello is not working anymore and I’m trying to understand why. If you are not using the Metacello Preview api for doing the record (Metacello new .... record.) then you should be ... the old Metacello would try to calculate the current version of packages and that is just not practical ... the Metacello Preview records the version that is loaded by Metacello into a registry and Metacello .... this save tons of execution time and complexity ...

With that said, I don't know it that is the source of your problem ...

_______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

Well, I was under the impression that the Metacello Preview is loaded into Pharo3.0. As I stated earlier I have never seen a green test result for Pharo3.0 in travis-ci (my development environment) the implication of this is that I do not know whether or not the Metacello Preview is functional in Pharo3.0...the tests run green[2] for all other platforms (from Pharo1.1, Pharo2.0, Squeak 4.4+ and GemStone) so I am comfortable that Metacello Preview itself is functional on those platforms ... I would hate to start a wild goose chase with trying to load different things when it might be something fundamentally broken in Pharo3.0. I would like to have green tests for Pharo3.0[1] and I think that getting tests green is the first order of business ... Perhaps Snapshotocello will work once the tests are green in travis-ci? If not at least we will have known base from which to work... Dale [1] https://github.com/dalehenrich/metacello-work/issues/243 [2] https://travis-ci.org/dalehenrich/metacello-work/builds/28156626 On Thu, Jun 26, 2014 at 7:15 AM, Tudor Girba &lt;tudor(a)tudorgirba.com <mailto:tudor@tudorgirba.com>> wrote: Hi, On Thu, Jun 26, 2014 at 4:11 PM, Dale Henrichs &lt;dale.henrichs(a)gemtalksystems.com <mailto:dale.henrichs@gemtalksystems.com>> wrote: On Thu, Jun 26, 2014 at 12:44 AM, stepharo &lt;stepharo(a)free.fr <mailto:stepharo@free.fr>> wrote: Hi dale I'm trying to obtain the record of Configuration of Moose in the latest version of moose ((ConfigurationOfMoose project version: #development) ignoreImage: true; record) and I get a kind of endless loop until my vm simply crash (I opened a Trasncript). Do you handle cycles in configurations? Yes...circularity should produce an error ... Do you have an idea of what could be the problems? Not really, however, if you are running with Pharo3.0 I understand that the Metacello tests are not green (they've never been green in my github/travis world ... but at one point in time I understand that they were green in Pharo3.0) ... so I do not trust Metacello in Pharo3.0 ... With that said, the failing tests may have nothing to do with this problem ... however without a baseline of green tests any attempts at fixes would be highly suspect ... I will check with smaller configurations but this is a daunting task. I understand ... Snapshotcello is not working anymore and I'm trying to understand why. If you are not using the Metacello Preview api for doing the record (Metacello new .... record.) then you should be ... the old Metacello would try to calculate the current version of packages and that is just not practical ... the Metacello Preview records the version that is loaded by Metacello into a registry and Metacello .... this save tons of execution time and complexity ... As I mentioned before, by looking at the trace, it looks like most of the time is being spent on doing just that: figuring the version. So, this might very well be the problem. How do I load the preview API? Cheers, Doru With that said, I don't know it that is the source of your problem ... Stef _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch <mailto:Moose-dev@iam.unibe.ch> https://www.iam.unibe.ch/mailman/listinfo/moose-dev -- www.tudorgirba.com <http://www.tudorgirba.com> "Every thing has its own flow" _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch <mailto:Moose-dev@iam.unibe.ch> https://www.iam.unibe.ch/mailman/listinfo/moose-dev _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev