On Tue, Mar 3, 2009 at 2:16 AM, Scott Blum <sco...@google.com> wrote: > SourceGen: all this really does is make it so that when you click on a > JsniMethodBody in the debugger, you can see the source.
Is SourceGenerationVisitor specified to be able to recreate the original Java input? Also, why keep the comment braces around the toString version? LGTM > JsModVisitor: two peephole "optimizations", but the real goal is to enable > immutable lists to be traversed, provided you don't actually try to mutate > them. Don't you still need to reset ctx.replaced to false? LGTM > JsSourceInfo: this is just deleting an old file that's no longer needed; > JsSourceInfo was superceded by SourceInfoJs. LGTM > JsParser: The API on JsParser is kind of abysmal. There's no reason for > this to be stateful between calls, so I just turned this into a couple of > static methods and updated the call sites. I'm not 100% done with this > patch-- I actually want to verify/tweak on this to make sure we can get > character-level accurate SourceInfo in the JS AST. But I figured it would > be good to review the basic idea first. LGTM, thought what's the call to Context.exit() for? > Thanks, > Scott > -- Bob Vawter Google Web Toolkit Team --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/Google-Web-Toolkit-Contributors -~----------~----~----~----~------~----~------~--~---
