On Tue, Feb 10, 2009 at 11:39 AM, Laurent Gautier <lgaut...@gmail.com> wrote: > Dismantling R_runHandlers is indeed opening interesting perspectives.
Quite so. Obviously someone should rewrite R's event loop. Someone else. > Callbacks are the Achilles tendon, I suspect. > I had an independent locking mechanism for fear of something bad happening > (concurrent access to R is "bad", in a Egon-to-Venkman "don't cross the > streams" sense). Oh, I see, you were worried about reentrancy, not just concurrency? Good point, my lock patch prevents concurrent access, but removes the reentrancy protection... maybe that's bad. I don't know. Obviously the garbage collector running twice at once will blow up, but R re-enters itself all the time in some ways (e.g., one function calls another)... If it is need then it could be added back easily enough -- just set a flag in the callback-to-Python code that is checked for by the ENTER_R stuff. -- Nathaniel ------------------------------------------------------------------------------ Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM) software. With Adobe AIR, Ajax developers can use existing skills and code to build responsive, highly engaging applications that combine the power of local resources and data with the reach of the web. Download the Adobe AIR SDK and Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com _______________________________________________ rpy-list mailing list rpy-list@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rpy-list
