Hopbot log for 2009-05-15 - Helma IRC channel: #helma on irc.freenode.net

2009-05-15:

[7:18] <zumbrunn> hannesw, I noticed that bin/helma isn't executable when untaring the 0.3 builds
[8:21] <simono> oh wow there is a helma-ng release :) *downloading*
[8:25] <simono> impressive.. there's already a lot to see in the demo app itself. hannes in hacking-mode ftw
[9:40] <hannesw> still doing some major cleanup... new helma-ng 0.3 build coming later
[10:32] <hannesw> uploaded freshly built ng 0.3 packages
[10:32] <hannesw> http://github.com/hns/helma-ng/downloads
[10:33] <hannesw> changes from yesterday: unified core/json module, and more reliable log4j initialization.
[10:37] <simono> minor problem: 2 unit tests fail. testEncoding and testReadIndefinite
[10:45] <hannesw> simono: that's not a minor problem! but they pass for me
[10:45] <hannesw> any error messages?
[10:46] <hannesw> (i guess it's because you're running on windows, with something else than UTF-8 as default encoding)
[10:47] <simono> hm, i'm on ubuntu 9.04
[10:48] <hannesw> error messages pls!
[10:48] <simono> http://helma.pastebin.com/m16a2754c
[10:48] <hannesw> tx
[10:49] <simono> my encoding is iso-smth, was forced to change it from ubuntu-default to stay compatible with windows-workgroup
[10:49] <hannesw> wow! looks that the string is broken twice, only in different ways.
[10:50] <simono> :) en_US.ISO-8859-15
[10:50] <hannesw> ok, the second one is because I'm using a java method that is new in Java 1.6, it seems.
[10:50] <hannesw> ok, that explains it.
[10:51] <simono> i'm on java 1.5.0_18
[10:52] <simono> i don't get the encoding-problem. if your files are utf-8 it should work, no matter what my system does
[10:53] <hannesw> no. if my files are utf-8 and your system expects iso-8859-1, breakage is unevitable
[10:54] <simono> ah right
[11:00] <simono> i'll have to read up on this. if i do " $ file test_file.js" my sys still correctly guesses utf8 and so does my editor.
[11:22] <hannesw> simono: both problems should be fixed in git/svn
[11:22] <hannesw> can you please check?
[11:23] <hannesw> simono: helma doesn't do any guessing. it just uses one encoding, which is the platform default encoding.
[11:28] <simono> rhino bug 399347 talks about an "-enc" that can be used, trying that out... checking out svn right after that
[11:29] <simono> hannesw, that i can confirm. it uses platform default encoding. but there should be a way to say "use utf8" .. live javac -encoding
[11:29] <simono> ^^ like javac -encoding=utf-8
[11:30] <hannesw> you can do system.properties["file.encoding"] = "utf-8"
[11:31] <simono> hannesw, all tests pass with HEAD
[11:34] <simono> rhino has -encoding argument that should force JS source files to be read with that encoding
[11:38] <simono> oh ant makes debian package.. thats nice
[11:45] <hannesw> yes! earl contributed that
[11:48] <hannesw> ok, building new packages after lunch...
[12:08] <hannesw> voila: http://github.com/hns/helma-ng/downloads
[15:40] <hannesw> still more testing............
[15:40] <hannesw> http://hensotest.appspot.com/
[20:25] <samwan> Bug 665 says "JSON.js bundled with Helma is old and buggy", but it has a low P5 priority. Does this mean it's unlikely that someone will look into adapting a newer JSON library for Rhino / Helma? I would volunteer to give it a try, but I am not clear on whether the Helma community is shifting effort/attention away from Helma 1.x to Helma NG.
[22:06] <liamMT> I'm having an issue implementing a java interface in js as described at http://www.mozilla.org/rhino/ScriptingJava.html
[22:07] <liamMT> I'm trying var t = new com.some.Interface(myobj)
[22:07] <liamMT> in the shell, t instanceof Interface evaluates to true
[22:07] <liamMT> but when I run it in my app, it evaluates to false
[22:07] <liamMT> any ideas?
[22:37] <liamMT> hannesw: also, module reloading doesn't seem to be working for me at the moment...need to restart to get code changes recognized
[22:38] <liamMT> um - hold that...maybe it is...
[22:38] <hannesw> liamMT: strange (both issues)
[22:38] * liamMT scratches head
[22:38] <hannesw> can you provide some test code i can reproduce this with?
[22:38] <liamMT> yeah - I think the module loading is OK, but I will try to get you something for the interface issue
[22:39] <hannesw> ok
[22:49] <liamMT> gah - I think it was just my problem yet again
[22:49] <liamMT> sorry for the noise :)
[23:01] <hannesw> liamMT: no problem!
[23:03] <liamMT> hannesw: thanks :) while I have you...one more q. the interface I'm implementing (as above) requires a method keySet() like the java.util.Map method except it returns Set<String>
[23:03] <liamMT> instead of Set<K>
[23:04] <liamMT> if I don't define keySet() on my object before passing it into the interface constructor, should I expect the helma jsobject<->Map implementation to work there? or do I need to define it separately due to the slight parameter difference?

 

 

In the channel now:

Logs by date: