Ahh, thank you for the recommendation, I will have a look at it. And more information about why I'm looking into this...
I'm basically sick of dealing with the poor error messages from either JVM or ASM when emitted bytecode doesn't match expected. I'm hoping that a higher-level library would handle this better, but still give me the ability to do lower-level bits when I need do. On Sun, Oct 18, 2009 at 2:54 PM, Bob McWhirter <[email protected]> wrote: > I know we used (maybe still do?) Janino in Drools. > > But now we've moved to MVEL for more power and control and such. Might be > worth investigating MVEL. > > http://mvel.codehaus.org/ > > -Bob > > > On Oct 18, 2009, at 3:51 PM, Charles Oliver Nutter wrote: > >> I've been trying to mentally prepare myself for a reworking of the >> current compiler or a JVM backend for the new compiler, and I'm >> following up on a recommendation at JVMLS to look at Janino. >> >> http://www.janino.net/ >> >> Janino is an "Embedded Java Compiler". It essentially provides what >> ASM does but at a higher level; you produce new or manipulate existing >> Java ASTs and then it dumps out the bytecode. >> >> The potential here would be to have a backend with less raw bytecode >> emission, localized compilation optimizations (i.e. if we emit an >> inefficient AST, Janino can in theory improve it), and less disconnect >> between either the Ruby AST or the new Ruby IR and the backend. >> >> I have not started looking into it yet, but I was wondering if anyone >> had any experience with it. It would be worth looking at if anyone >> else is interested in the compiler. >> >> - Charlie >> >> --------------------------------------------------------------------- >> To unsubscribe from this list, please visit: >> >> http://xircles.codehaus.org/manage_email >> >> > > > --------------------------------------------------------------------- > To unsubscribe from this list, please visit: > > http://xircles.codehaus.org/manage_email > > > --------------------------------------------------------------------- To unsubscribe from this list, please visit: http://xircles.codehaus.org/manage_email
