On Jun 7, 2010, at 3:29 PM, Henrik Johansen wrote: > > On Jun 7, 2010, at 3:03 33PM, Fernando olivero wrote: > >> Hi Adrian, i've added this test in the SLICE. >> It wasn't there before. >> >> The problem is that in PharoCore-1.1-11383-Beta the new editors aren't >> initialized, and the test fails, >> >> TextEditor initialize. >> SmalltalkEditor initialize. >> SimpleEditor initialize. >> >> I could either add this statements as a test preconditions, or we could >> execute them and have already initialized in the image itself. So we have a >> valid state of the editors in the image. >> >> What is your opinion? >> >> Fernando > Should do some refactoring too, right now actions are stored in a shared > variable for SimpleEditor, and a class inst var for TextEditor. (capitalized > differently) >
YES! > And most of them are initialized, it's just SmalltalkEditor which is not. > Since it doesn't implement initialize directly, it is never executed when > loaded from Monticello (or filed in, for that matter). > With the initialize factored into different methods like that, the initialize > method itself is seldomly updated, it's hard to ensure you've initialized > correctly when using Monticello. > Up to now i've only ported the Editors, and made minor modifcations due to Pharo being diferent than CUIS and Squeak. What you propose and also what Lukas proposed in the past is the next step. > In this case, a postload script for System-Text doing Editor allSubclassesDo: > [:each | each initialize] might have been the simplest solution, but those > don't exist anymore in Pharo. > > Cheers, > Henry > > > > _______________________________________________ > Pharo-project mailing list > [email protected] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
