This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Either the RubyExecution.RUBY_COMPILER may be black-magically enhanced or we could start to use more than one OutputRecognizer. E.g. public class RubyExecution extends ExecutionService { ... psfs RE_BASE = "^((\\[|\\]|\\-|\\:|[0-9]|\\s|\\,)*)(\\s*from )"; psfs RegexpOutputRecognizer RUBY_COMPILER_NO_EXT = new RegexpOutputRecognizer(RE_BASE + "?(\\S.*)\\:([0-9]+):in.*", // [1] 4, 5, -1); psfs RegexpOutputRecognizer RUBY_COMPILER = new RegexpOutputRecognizer(RE_BASE + "?(\\S.*\\.(rb|rake))\\:([0-9]+).*", 4, 6, -1); ... } // + // NOI18N [1] 'in' to prevent a lot of non-sense matches. Also the RUBY_COMPILER recognizer is used (almost?) everywhere (probably 10+ places). So a method like ExecutionDescriptor.useDefaultRecognizers(boolean) could be considered. Default would be true - in that case RUBY_COMPILER and eventually RUBY_COMPILER_NO_EXT and others would be added by default -> no need to call addOutputRecognizer(RubyExecution.RUBY_COMPILER) on all ED instances.
Agreed
Reassigning this issue to newly created 'ruby' component.
This was fixed this some time ago during large changes in the relevant code.