20:58 <boc_tothefuture[> Hi there.. I am trying to get JRuby up and running using JSR233 and hitting some issues. This is running inside of another system that supports multiple JSR233 scripting engines that I am trying to add JRuby support to. I have hit a couple of odd issues getting it up and running.. The current issue I am hitting is that when I do a require 'logger' it errors out saying that "undefined method `basename' for class

20:58 <boc_tothefuture[> java.io.File:Java::JavaLang::Class". I would think that would happen if I had imported java.io.File, but I haven't.. Also, I thought the docs said that ruby classes would prefer ruby objects over Java ones unless you were explicit.

20:59 <headius[m]> hmmm yeah that is a good theory and I'd still suspect something is importing java.io.File into the top level namespace

21:01 <headius[m]> Ruby code will always use Ruby classes, but there's only one entry for "File" in the namespace, and the Java one could overwrite the Ruby one

21:01 <headius[m]> definitely would break a lot of stuff

21:04 <boc_tothefuture[> hmm.. can "unset" that before the require?

21:06 <boc_tothefuture[> For reference.. here is the code: https://gist.github.com/boc-tothefuture/ff57340dd6c4e972592ac977e5cbb91e

21:08 <boc_tothefuture[> Here is the output: https://gist.github.com/boc-tothefuture/81435085834acdecba6affdd819ce211

21:17 <boc_tothefuture[> What i pasted was the only ruby file being loaded.. Not sure what else could import that into the namespace.

21:17 <boc_tothefuture[> I did have to manually set the loadpaths for it to even find the logger which I thought was also interesting.

22:27 <headius[m]> You can't really unset it because it has already been overwritten. I am not sure what would be causing that, since it is clearly something we would not do within jruby itself

22:28 <headius[m]> The code looks fine, could you print out what File is before logger is required?

22:30 <headius[m]> yeah you really shouldn't need to add those class path entries. Logger is a standard library and should just be loadable out of the complete jar

22:30 <headius[m]> If you can make this into a little reproducible repository I can try to test it out