Hopbot log for 2008-09-28 - Helma IRC channel: #helma on irc.freenode.net

2008-09-28:

[21:21] <bard joined the chat room.
[21:23] <bard> folks, does anyone have experience with existdb (exist.sf.net)? it looks like an option to take helma's xml-based persistence beyond prototyping stage
[21:29] <peter_12 joined the chat room.
[21:36] <peter_12 left the chat room.
[22:36] <decke left the chat room. ("Leaving.")
[23:05] <zumbrunn> bard, I remember looking at it some years ago...
[23:07] <zumbrunn> now, with helma-ng, it would probably be quite easy to integrate
[23:25] <decke joined the chat room.
[00:01] <bard> cool
[00:01] <bard> how far is helma-ng from public consumption?
[00:49] <decke left the chat room. ("Leaving.")
[01:46] <bard> zumbrunn: I see existdb can be deployed as a library too (http://exist.sourceforge.net/deployment.html#N10431), so it should not be too hard to integrate it with helma 1.6 either
[02:14] <philmaker joined the chat room.
[02:16] <bard> persevere looks interesting too
[03:00] <bard left the chat room. ("ERC Version 5.3 (IRC client for Emacs)")
[03:27] <jkridner|work joined the chat room.
[04:42] <philmaker left the chat room.
[08:41] <philmaker joined the chat room.
[08:42] <philmaker left the chat room. (Client Quit)
[08:42] <philmaker joined the chat room.
[09:27] <zumbrunn> hi philmaker, I see bard just asked about persevere a few hours ago
[09:27] <zumbrunn> you have experience with that, right?
[09:28] <philmaker> I haven't seen that yet - I'll take a look
[09:30] <philmaker> Oh it's right above
[09:30] <zumbrunn> well, he didn't actually ask a question
[09:30] <zumbrunn> just pointed it out as interesting
[09:31] <zumbrunn> bard: regarding helma-ng.... it will never be more ready for "public consumption", it's just "production use" that it isn't "ready" for, being alpha
[09:31] <zumbrunn> but it is the best time to look at it, work with it and help shape it
[09:31] <philmaker> Persevere is interesting - but when I first worked with it several months ago it did indeed require perseverance.
[09:33] <philmaker> yeo
[09:33] <philmaker> yep
[09:33] <zumbrunn> with helma-ng, pluging all kinds of data backends behind helma should become very simple
[09:33] <philmaker> excellent
[09:34] <philmaker> I have recently used Jaxer and I find Sling and Jackrabbit to be very interesting.
[09:34] <philmaker> I have recently also dabbled with a custom Rhino server.
[09:35] <zumbrunn> yeah, jackrabbit would be on that list of data backends as well
[09:36] <philmaker> righto - sounds great
[09:36] <philmaker> I have glanced at the Helma NG stuff lately.
[09:38] <zumbrunn> currently, there is the super slim filestore module, plus hibernate
[09:38] <zumbrunn> http://github.com/hns/helma-ng/tree/master/modules/helma/filestore.js
[09:38] <philmaker> Chris, I do want to chat with you more. It's kind of late here and I'm having some of these: http://www.sierranevada.com/beers/paleale.html
[09:40] <zumbrunn> the darker the better
[09:40] <zumbrunn> http://github.com/hns/helma-ng/tree/hibernate/modules/helma/hibernate.js
[09:42] <philmaker> I'll probably email you within the next several days about Helma NG, and what I desire in a Javascript server.
[09:42] <philmaker> I bought this book: http://www.amazon.com/Revolution-Manifesto-Ron-Paul/dp/0446537519/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1222587613&sr=8-1
[09:42] <philmaker> tonight
[09:44] <? zumbrunn is thinking whether he should get that, too
[09:44] <philmaker> [09:45] <zumbrunn> I think I kind of know where I am going to disagree
[09:45] <zumbrunn> but I should probably read it
[09:45] <philmaker> I'll listen.
[09:47] <bard joined the chat room.
[09:47] <zumbrunn> well, in a nutshell, I think ron paul, while cleaning up the federal system, would give to much free reign to the capitalistic free market
[09:47] <zumbrunn> bard, you are in the right channel
[09:47] <philmaker> well yes
[09:48] <zumbrunn> it just looks like you are not
[09:48] <bard> lol
[09:49] <zumbrunn> for a shift towards a libertarian system, there also needs to be a transformation in society
[09:50] <zumbrunn> of course, that might happen as a result of the way ron paul would change the federal government
[09:50] <zumbrunn> but that's not likely to be the case, I think
[09:51] <philmaker> Regarding current events in the US and the wall street rescue and the US debt  - it is becoming more and more evident that this US is losing it's sovereignty. Does Switzerland have any national debt.
[09:51] <philmaker> ?
[09:51] <zumbrunn> (no matter whether "libertarian" is flavored "left" or "right")
[09:51] <philmaker> One of the reasons we are preparing the bail out package is to assure foreign investors.
[09:51] <zumbrunn> about 100 billion
[09:52] <philmaker> We have about 70x times that (one the books)
[09:52] <philmaker> one the books
[09:52] <zumbrunn> yeah, but you also have 40x the amount of people
[09:52] <philmaker> on (d'oh)
[09:52] <philmaker> well ok
[09:53] <philmaker> maybe it's ok for the US to have a large debt and a large military
[09:56] <zumbrunn> well, debt really is bad, if it is to big
[09:57] <philmaker> but it's never too big if you have a large military
[09:57] <zumbrunn> the concept of debt in the financial systems at that level are a funny thing, anyway
[09:58] <zumbrunn> if it is to big, you won't be able to afford the large military anymore
[09:58] <philmaker> [09:59] <zumbrunn> no, kidding aside, debt is bad...
[09:59] <philmaker> yeah
[10:00] <zumbrunn> if it gets out of hand, the US could slip towards the status of a "developing country", almost
[10:00] <philmaker> Have you seen the things that are happening with out banking?
[10:00] <philmaker> our banking?
[10:00] <zumbrunn> like what happend to the sowjet union countries
[10:01] <zumbrunn> yes, of course
[10:01] <philmaker> Some of our major banks are failing...
[10:02] <philmaker> ? sowjet union countries ?
[10:02] <zumbrunn> you can send a thank you letter to phil gramm
[10:02] <philmaker> [10:02] <philmaker> My Texas senator
[10:03] <zumbrunn> yeah, like russia, you know, that sowjet union
[10:03] <philmaker> Hmm.
[10:04] <philmaker> I don't know anything about that - and everything I seen is indeed in Russian.
[10:05] <zumbrunn> one could easily argue that the reason for the collapse of the sowjet union was their debt
[10:06] <philmaker> sowjet union = soviet union?
[10:06] <philmaker> We call it soviet union
[10:06] <philmaker> that's interesting
[10:11] <philmaker> Chris, I'm glad to see you here. I'm interested in following up with you here or via email about Helma NG and some customer Rhino stuff I'm playing with - or that I'd like to see in Helma NG.
[10:25] <zumbrunn> yeah, sowjet union == soviet union
[10:33] <philmaker> The new MacBook is supposed to come mod-October here. I am trying to wait patiently.
[10:34] <philmaker> The new MacBook is supposed to come mid-October here. I am trying to wait patiently.
[10:35] <philmaker> It is supposed to have LED backlighting. I hope the display is better than the current MacBook.
[10:36] <philmaker> I have owned several Macs with quality LCD screens and the display/backlighting in the MacBook (consumer) is pretty cheap.
[10:36] <zumbrunn> the problem is that stupid glossy screen, I think
[10:36] <zumbrunn> no wonder they can't light the thing
[10:37] <philmaker> That is perhaps part of it - but I can adjust to that.
[10:37] <zumbrunn> I probably could adjust to it, too - but I don't want to
[10:38] <philmaker> I did use a MacBook for a short time and the display was uneven and very annoying when wokring with lines of code.
[11:11] <philmaker left the chat room.
[11:16] <philmaker joined the chat room.
[11:47] <decke joined the chat room.
[12:49] <philmaker left the chat room.
[13:49] <philmaker joined the chat room.
[14:55] <decke left the chat room. (Read error: 104 (Connection reset by peer))
[14:56] <decke joined the chat room.
[15:32] <bard> chris: following up from yesterday, I managed to get hold of existdb from helma using the java api (from js, though). I'm happily running xpath queries in the app's onStart() now
[15:51] <philmaker left the chat room.
[15:54] <zumbrunn> bard, cool
[15:54] <zumbrunn> did you take a look at the filestore.js code?
[15:55] <zumbrunn> http://github.com/zumbrunn/helma-ng/tree/master/modules/helma/filestore.js
[16:00] <bard> haven't taken a look at helma-ng at all so far... is its status published anywhere? I've had a look at trac but it only shows a handful of tickets in the milestones for 0.3 and 0.4
[16:00] <bard> does filestore.js store json?
[16:02] <zumbrunn> yes, it serializes to json
[16:03] <zumbrunn> but if you look at the code, it would kind of show the way of how one could put other data backends, like existdb, behind helma-ng
[16:03] <zumbrunn> (in a way where js objects become automatically mapped against the data backend)
[16:04] <bard> cool, but see above question on helma-ng status, I'm more than fine with tinkering with the framework but I still want to deliver an app in the meantime
[16:04] <zumbrunn> the easiest way to get helma-ng is to simply go here:
[16:04] <zumbrunn> http://github.com/hns/helma-ng/tree/master/
[16:05] <zumbrunn> and click the download link
[16:05] <zumbrunn> then build it with "ant package"
[16:06] <bard> how complete is it compared to helma 1.6.*?
[16:06] <zumbrunn> it is "different"
[16:06] <zumbrunn> the goal with helma-ng is a much smaller core
[16:06] <zumbrunn> doing less in java
[16:07] <zumbrunn> and instead adding more js modules around it
[16:07] <bard> I like that
[16:07] <bard> I'm already trying to stay away from java as much as possible
[16:07] <bard> I'd just like to avoid being blocked by chunks of basic functionality missing
[16:08] <zumbrunn> in a sense, the core works perfectly well for "production", but that of course depends on you requirements
[16:08] <zumbrunn> and since it is alpha, everything is still subject to change
[16:08] <bard> I'll try it
[16:10] <decke left the chat room. ("Leaving.")
[16:10] <decke joined the chat room.
[16:22] <philmaker joined the chat room.
[16:26] <bard> btw, is helma 1.6 moving to github too?
[16:50] <zumbrunn> I think it would be a good idea
[16:50] <zumbrunn> what do you think?
[16:51] <bard> I'm moving all my projects there, I guess that's a "yes"
[16:51] <zumbrunn> [17:09] <bard> is there any glaring error here http://helma.pastebin.com/d3b3bd6c ? I thought defineLibraryScope would make 'xpath' persistent across requests, but it still shows up as undefined when I call actions...
[17:16] <zumbrunn> what happens when you move defineLibraryScope('xpath'); into the onStart function?
[17:18] <bard> app.log(xpath) at the beginning of an action says " is empty", both if defineLibraryScope is just before "xpath = ...", and if it's after
[17:19] <zumbrunn> hmm, strange
[17:19] <zumbrunn> try with a global prefix:
[17:19] <zumbrunn> global.xpath = collection.getService('XPathQueryService', '1.0');
[17:22] <bard> no luck
[17:23] <zumbrunn> how about that:
[17:23] <zumbrunn> http://helma.pastebin.com/f45f4dd54
[17:25] <decke1 joined the chat room.
[17:27] <bard> I have no idea why it works, but it does
[17:27] <zumbrunn> I'm not sure either
[17:29] <zumbrunn> meaning, I'm not sure why the other ways didn't work
[17:29] <decke left the chat room. (Connection timed out)
[17:33] <bard> ah
[17:33] <bard> found it
[17:34] <bard> uhm, no, false negative.
[17:34] <bard> I'll stick to startExistDb()
[17:59] <philmaker left the chat room.
[18:49] <peter_12 joined the chat room.
[19:06] <bard> *ouch* seems like rhino's parser is more pushy than spidermonkey's: http://helma.pastebin.com/d21be6e53
[19:12] <zumbrunn> ouch indeed
[19:12] <zumbrunn> is there a way to work around that?
[19:13] <peter_12 left the chat room.
[19:13] <zumbrunn> the parser in rhino was recently redone by some google folks, if I remember correctly
[19:14] <bard> I've just found this: http://groups.google.com/group/netscape.public.mozilla.jseng/browse_thread/thread/de00c62284f33e1d/064190d0f3a62d31?q=e4x&rnum=20&pli=1
[19:14] <zumbrunn> but that isn't in rhino cvs yet
[19:14] <zumbrunn> but it might mean that the problem will go away if they have done that bit differently
[19:14] <bard> the message above is from 2004 though (and I'm not really sure what it's suggesting
[19:15] <bard> ah
[19:24] <zumbrunn> maybe the fix/workaround igor is suggesting there is something helma should be doing anyway
[18:38] <bard> :)
[18:39] <bard> for the record, I tried the solution described by igor (by hacking rhino src, I didn't bother with a clean implementation) and it works
[18:39] <bard> and for hopbot (can I call you hoppy?), we're talking about this: http://groups.google.com/group/netscape.public.mozilla.jseng/browse_thread/thread/de00c62284f33e1d/064190d0f3a62d31?q=e4x&rnum=20&pli=1
[18:49] <zumbrunn> the hopbot got some of its memories back, but is all confused about the time zones
[18:49] <zumbrunn> oh well, things should be back on track now
[21:02] <hannesw> The JVM language summit would have been an interesting event to be:
[21:03] <hannesw> http://www.sauria.com/blog/2008/09/27/the-first-annual-jvm-language-summit/
[21:03] <hannesw> http://www.tbray.org/ongoing/When/200x/2008/09/25/JVM-Summit
[21:50] <bard> what is the best way to compile helma (1.6) when you're tinkering with it? ant compile + changing the classpath to ./classes?
[22:01] <hannesw> I'd say ant jar
[22:01] <hannesw> shouldn't be much slower than ant compile
[22:02] <hannesw> actually both could be much faster if we would let ant choose which files need to be recompiled
[22:04] <hannesw> but (at least in helma 1) all compiled classes are deleted before compiling, which may be overkill in most cases...
[22:15] <bard> uhm... after recompiling rhino and moving rhino/build/rhino1_7R1/js.jar to helma/lib/js.jar, I get "Exception in thread "Thread-6" java.lang.NoClassDefFoundError: helma/scripting/rhino/RhinoCore". are there special steps to follow to rebuild rhino for helma?
[22:15] <bard>
[22:16] <zumbrunn> yes, you need the token.patch
[22:16] <zumbrunn> http://dev.helma.org/static/files/1165/token.patch
[22:18] <zumbrunn> the following page should list the patches that haven't gone into rhino proper yet
[22:18] <zumbrunn> http://dev.helma.org/wiki/Rhino+Bugs+%26+Patches/
[22:18] <zumbrunn> but I believe the metaobject.patch is obsolete
[22:19] <zumbrunn> helma 1.6.x contains the JSAdpater patch from the phobos project instead
[22:19] <bard> I think I read that hannes was replacing that with a jomp-based variation
[22:19] <zumbrunn> and helma-ng's rhino has a metamethod patch that isn't listed their yet
[22:19] <bard> ah there it is :)
[22:19] <zumbrunn> :-)
[22:20] <zumbrunn> the one in helma-ng isn't exactly like JOMP
[22:21] <zumbrunn> it doesn't require the use of a __metaobject__ object
[22:21] <zumbrunn> it works directly on any js object
[22:22] <zumbrunn> anyway, the token.patch should be what you need
[22:22] <bard> hm, looks like it's not :( same error
[22:23] <bard> but I see helma/scripting/RhinoCore stuff in helma svn repo
[22:23] <zumbrunn> hmm, then I don't know what is going on
[22:23] <bard> maybe it needs to be compiled and put into rhino.jar
[22:24] <zumbrunn> what are you doing exactly?
[22:24] <zumbrunn> rereading what you said before...
[22:24] <zumbrunn> js.jar would replace rhino.jar
[22:24] <bard> yes, I did that
[22:25] <zumbrunn> ok, but you said "helma/lib/js.jar" which would be incorrect
[22:25] <zumbrunn> you need to rename it to rhino.jar, I think
[22:27] <bard> did it, just reported incorrectly. checking something else...
[22:29] <zumbrunn> Helma at the OpenExpo in Zurich last week:
[22:29] <zumbrunn> http://zumbrunn.com/static/mochazone/openexpohelma.png
[22:30] <bard> my guess was right... there's a bit more in helma's rhino.jar than rhino's: http://helma.pastebin.com/d46812c8c
[22:32] <zumbrunn> it has been a couple of month since I've built a rhino for Helma 1.6.x based on the current rhino cvs version
[22:32] <zumbrunn> but it worked fine for me back then
[22:34] <bard> *sigh*
[22:34] <bard> it's my stumble-on-arcane-issues day
[22:46] <bard> moreover the class it complains about is definitely there, in lib/helma.jar

 

 

In the channel now:

Logs by date: