On 5 Dec 2013, at 07:34, Andre Hora <andrehoraa(a)gmail.com> wrote: Hello, Yes, as Anne said, the rules were extracted from the code history of some systems. The idea is that we learn with past changes at invocation level. For example, if a system is often changing at:1 to first, = nil to isNil, or foo() to bar(), we detect such pattern (applying data-mining techniques) and we set a rule for that. If you want to check them, have a look in the packages. Gofer new url: 'http://www.smalltalkhub.com/mc/andrehora/APIEvolutionMiner/main'; package: 'Rules-Pharo30'; package: 'Rules-Pharo20'; package: 'Rules-Seaside30'; package: 'Rules-Moose'; package: 'Rules-Glamour'; package: 'Rules-Roassal'; load. Some of the rules may be not relevant/important/valid. So, very soon we are going ask you (experts) to validate them and help us. regards, -- Andre Hora _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev

Hello, Yes, as Anne said, the rules were extracted from the code history of some systems. The idea is that we learn with past changes at invocation level. For example, if a system is often changing at:1 to first, = nil to isNil, or foo() to bar(), we detect such pattern (applying data-mining techniques) and we set a rule for that. If you want to check them, have a look in the packages. Gofer new url: 'http://www.smalltalkhub.com/mc/andrehora/APIEvolutionMiner/main'; package: 'Rules-Pharo30'; package: 'Rules-Pharo20'; package: 'Rules-Seaside30'; package: 'Rules-Moose'; package: 'Rules-Glamour'; package: 'Rules-Roassal'; load. Some of the rules may be not relevant/important/valid. So, very soon we are going ask you (experts) to validate them and help us. regards, On Thu, Dec 5, 2013 at 8:29 AM, Anne Etien <anne.etien(a)univ-lille1.fr> wrote: Hi, Andre is working on the discovery of rules while analyzing the history of a software. He thus studied the evolution of Pharo 2.0 to Pharo 3.0, defined some rules and we applied them on Moose. Every month there is a Pharo sprint in the team and we try to always do in parallel a Moose sprint. When we get new rules, we try to apply them. Last time, it was really hard to apply the rules, because critics browser seems broken in Pharo 3.0 and because of GRPackage issue (we opened bug reports). The rules for Pharo30 are here: Gofer new url: 'http://www.smalltalkhub.com/mc/andrehora/APIEvolutionMiner/main'; package: 'Rules-Pharo30'; load. There are also rules for Moose (extracted from all Moose history): Gofer new url: 'http://www.smalltalkhub.com/mc/andrehora/APIEvolutionMiner/main'; package: 'Rules-Moose'; load. All the merit returns to Andre, I am just clicking on button (when it works) to apply the rules. I am quite sure that Andre can give more information. Cheers, Anne Le 5 déc. 2013 à 07:10, Tudor Girba a écrit : _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev -- Andre Hora _______________________________________________ Moose-dev mailing list Moose-dev(a)iam.unibe.ch https://www.iam.unibe.ch/mailman/listinfo/moose-dev