YES! That would also solve the problems with Kathy's test engine. Indeed, we could just replace test engine with rackunit in *sl. -- Matthias
On May 17, 2011, at 9:06 AM, Matthew Flatt wrote: > At Tue, 17 May 2011 08:57:26 -0400, Matthias Felleisen wrote: >> >> While we're at it: any chance of getting a test-on-entry-point for Rackunit >> as >> I described a while ago is available in Python? > > I think we would want to implement it as a `#lang' mix-in > > #lang testable racket > ... > > where `testable' extends the module's runtime configuration to run > tests. The runtime configuration action is used only when the module is > the main module of a program (e.g., run via `racket -t'). > > I don't think there current runtime-configuration hook is enough, > though. It supports a pre-`require' configuration, but `testable' would > need some sort of port-`require' action. > _________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/users