Logs for the RingoJS IRC channel, as logged by ringostarr.

2010-11-10

[08:50] <hannesw> oops... I just discovered that we're _still_ extending native objects

[08:50] <hannesw> http://ringojs.org/api/master/ringo/utils/regexp/index.html

[08:51] <hannesw> https://github.com/ringo/ringojs/blob/master/modules/ringo/utils/regexp.js

[08:51] <hannesw> i think i might have left this in for narwhal compatibility

[08:51] <hannesw> but i think we should fix it

[08:53] <hannesw> maybe even put it in ringo/utils/strings as escapeRegExp() or something?

[09:32] <mcepl> hannesw: it's actually bash >=4.0 :) http://twitter.com/#!/hannesw/status/2276200357564416

[09:34] <hannesw> mcepl: right :)

[09:34] <mcepl> I am getting to addicted to TAB-completion is isn't funny

[09:34] <mcepl> *so addicted

[09:34] <hannesw> or actually bash-completion :)

[09:35] <hannesw> yes, would be really hard to live without

[09:36] <mcepl> git autocompletion, ssh (complets defined and bonjour-available hosts), even fedpkg (packaging tool for Fedora), etc. etc.

[10:01] <hannesw> question: what's the proper name for dash and underscore notation?

[10:01] <hannesw> or better: what would be the proper name for a function converting to it?

[10:02] <hannesw> toDash(), toUnderscore()?

[11:11] <hannesw> went with 'toDashes' and 'toUnderscores'

[13:11] <hannesw> ah, proper html <title> tags on ringojs.org

[13:11] <hannesw> ...mostly

[14:59] <hannesw> now using ringo-modulr to share scripts on server and client

[14:59] <hannesw> https://github.com/hns/ringobot/tree/master/shared/

[15:15] <hannesw> ........test

[17:02] <hannesw_> fyi, i just pushed an updated rhino snapshot to github

[17:03] <tschaub> nice!

[17:03] <hannesw_> sorry it took so long

[17:03] <hannesw_> i had some weird test results for one particular spidermonkey test

[17:06] <hannesw_> haven't solved it, but it's nothing to worry about

[17:22] <earl> so why does ringostarr no longer echo commit messages?

[17:24] <hannesw> earl - good question

[17:24] <hannesw> i have no idea

[17:24] <hannesw> i switched the github hook to the ringojs.org bot

[17:24] <hannesw> didn't work before, doesn't work now...

[17:24] <earl> yeah, just re-added the hook for the ci bot

[17:25] <hannesw> what's ci bot?

[17:25] <earl> my bot does automated builds & tests

[17:25] <hannesw> i didn't remove any hooks, there was just one

[17:27] <earl> maybe the https switched changed something in their webhook calls

[17:27] <earl> probably https urls

[17:27] <earl> yep, that's it

[17:27] <hannesw> ok

[17:29] <hannesw> the github switch to https messed up a lot of things

[17:29] <earl> for the bot, just adapt the url in the config

[17:30] <hannesw> ok

[17:30] <earl> https://github.com/earl/ringobot/commit/82a99d1

[17:37] <hannesw> sorry earl, did the same

[17:41] <hannesw> ok, went back and used your commit :)

[17:46] <earl> np :)

[18:06] <earl> hannesw: are you in between deploying the config changes to ringojs.org?

[18:18] <hannesw> earl: should already be deployed

[18:18] <hannesw> why?

[18:27] <earl> because the url in the ringobot repo is an example url and needs adaptation on deployment

[18:29] <earl> ok, fixed it locally on the ringojs.org machine. feel free to make a commit out of it

[18:36] <hannesw> oh :) okd

[18:39] <hannesw> done

[22:01] <oravecz> I have a module that is shared (the default behavior). I am seeing it get instantiated multiple times every once and a while. Is this expected behavior?

[22:26] <hannesw> oravecz this might be due to garbage collection

[22:26] <hannesw> modules are cached via soft references

[22:27] <oravecz> what might I do to invoke a method in my module prior to GC? I need to destroy a service.

[22:28] <hannesw> we don't have a hook for that currently

[22:29] <oravecz> Would that originate in your code or Rhino's?

[22:30] <hannesw> that would be Ringo

[22:33] <oravecz> So, ReloadableScript I presume?

[22:36] <hannesw> yes

[22:37] <hannesw> at the bottom of that class there's the ScriptCache

[22:37] <oravecz> Cool, I just want to see if I can add some logging to tell me if that is the cause

[22:38] <hannesw> the map.remove() in line 479 should be the last time module is handled before garbage collection:

[22:38] <hannesw> https://github.com/ringo/ringojs/blob/master/src/org/ringojs/engine/ReloadableScript.java#L479

[22:38] <hannesw> you might add some logging there

[22:38] <oravecz> great thanks

[22:47] <oravecz> if it is the problem, what would a good approach be to solve it? I think invoking a 'destroy' function may be cumbersome. Perhaps a module property that makes sure the script isn't put into the cache as a SoftReference?

[23:01] <oravecz> Well, it never executes line 479. And the calls are far apart so it isn't a case where two instances are created at almost the same time.