uPortal IRC Logs-2008-01-24

Owned by Apereo Infrastructure
Last updated: Jan 24, 2008Version comment

[02:18:00 EST(-0500)] * esm (n=esm@207-53-192-2.dynamic-dsl.qis.net) has joined ##uportal
[08:01:44 EST(-0500)] * jayshao (n=jayshao@pool-68-239-246-163.nwrk.east.verizon.net) has joined ##uportal
[08:59:42 EST(-0500)] * jayshao (n=jayshao@pool-68-239-246-163.nwrk.east.verizon.net) has joined ##uportal
[09:30:15 EST(-0500)] * jayshao (n=jayshao@pool-68-239-246-163.nwrk.east.verizon.net) has joined ##uportal
[10:06:10 EST(-0500)] * athena7 (n=athena7@lumina.its.yale.edu) has joined ##uportal
[10:07:43 EST(-0500)] * clown (n=clown@guiseppi.atrc.utoronto.ca) has joined ##uportal
[10:17:51 EST(-0500)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:26:32 EST(-0500)] * michelled (n=team@142.150.154.199) has joined ##uportal
[10:52:33 EST(-0500)] * esm (n=esm@asdf.dkc.jhu.edu) has joined ##uportal
[11:08:23 EST(-0500)] * arybicki (n=arybicki@pool-71-106-224-185.lsanca.dsl-w.verizon.net) has joined ##uportal
[12:21:26 EST(-0500)] * clown_ (n=clown@142.150.154.101) has joined ##uportal
[14:25:29 EST(-0500)] * clown (n=clown@142.150.154.101) has joined ##uportal
[14:50:52 EST(-0500)] * clown_ (n=clown@guiseppi.atrc.utoronto.ca) has joined ##uportal
[15:39:51 EST(-0500)] * esm (n=esm@asdf.dkc.jhu.edu) has joined ##uportal
[15:52:47 EST(-0500)] * EiNZTEiN (n=einztein@205.241.143.4) has joined ##uportal
[15:57:27 EST(-0500)] <EiNZTEiN> EricDalquist: yesterday I posted an apparent bug here, on the QuickStart distribution package
[15:57:38 EST(-0500)] <EiNZTEiN> at least for me
[15:57:39 EST(-0500)] <EiNZTEiN> [16:24:42] <EiNZTEiN> on build.xml
[15:57:39 EST(-0500)]

<EiNZTEiN> [16:25:15] <EiNZTEiN> line 1251 should read " <arg value="$

Unknown macro: {deploy.home}

"/>" and not " <arg value="../${deploy.home


[15:57:49 EST(-0500)] <EiNZTEiN> when I changed it, everything installed fine
[15:57:52 EST(-0500)] <EricDalquist> ok
[15:57:57 EST(-0500)] <EiNZTEiN> it was keeping the portlets from being installed
[15:57:58 EST(-0500)] <EricDalquist> can you create a jira issue?
[15:58:04 EST(-0500)] <EiNZTEiN> sure, where?
[15:58:12 EST(-0500)] <EricDalquist> http://www.ja-sig.org/issues
[15:58:26 EST(-0500)] <EricDalquist> be sure to describe the version of the quick start you're using
[15:58:39 EST(-0500)] <EiNZTEiN> quick note: I used the uPortal folder from the quickstart BUT I'm not installing the quickstart... but that seems to be irrelevant to this issue
[15:59:04 EST(-0500)] <EricDalquist> oh .... um I would think it would be
[15:59:17 EST(-0500)] <EricDalquist> the uPortal folder from the quick start is setup explicitly to work with the quick start
[15:59:33 EST(-0500)] * arybicki (n=arybicki@pool-71-106-224-185.lsanca.dsl-w.verizon.net) has left ##uportal
[15:59:46 EST(-0500)] <EricDalquist> I don't create the quickstarts but I do know there are changes made in that uPortal folder from the plain source download
[15:59:51 EST(-0500)] <EiNZTEiN> true, but that was creating a "round-robin" link
[15:59:54 EST(-0500)] <EiNZTEiN> as you can imagine
[16:00:01 EST(-0500)] <EiNZTEiN> ../folder/.. << weird
[16:02:39 EST(-0500)] <EiNZTEiN> my paste up there was complete
[16:02:51 EST(-0500)]

<EiNZTEiN> "line 1251 should read " <arg value="$

Unknown macro: {deploy.home}

"/>" and not " <arg value="../$

/.."/>"


[16:03:10 EST(-0500)] <EricDalquist> ah
[16:03:16 EST(-0500)] <EricDalquist> ok, yeah go file the bug (smile)
[16:04:05 EST(-0500)] <EiNZTEiN> ok ok
[16:04:22 EST(-0500)] <EiNZTEiN> anyway another question, are there any resources on "best practices" on theme/UI design for uPortal?
[16:04:41 EST(-0500)] <EricDalquist> I'm not sure
[16:04:49 EST(-0500)] <EricDalquist> check with uportal-user@lists.ja-sig.org
[16:05:03 EST(-0500)] <EricDalquist> I stay away from theme/UI work
[16:13:24 EST(-0500)] * EiNZTEiN (n=einztein@205.241.143.4) has joined ##uportal
[16:24:37 EST(-0500)] <EricDalquist> esm: assemblies make my head hurt
[16:32:25 EST(-0500)] <esm> i agree i have a headache now (smile)
[16:32:29 EST(-0500)] <esm> they're the worst
[16:32:38 EST(-0500)] <esm> i just don't understand how they work really.
[16:32:43 EST(-0500)] <EricDalquist> yeah
[16:32:56 EST(-0500)] <esm> single module, no prob.
[16:32:57 EST(-0500)] <esm> oh
[16:33:03 EST(-0500)] <EricDalquist> I think the frustraiting bit is I feel like they can do what I want, but the docs ...
[16:33:07 EST(-0500)] <EricDalquist> yeah
[16:33:13 EST(-0500)] <esm> yeah teh docs kill
[16:33:21 EST(-0500)] <esm> and all the different goals
[16:33:25 EST(-0500)] <esm> single attached
[16:33:25 EST(-0500)] <EricDalquist> like I want a source distribution (figured that out) and I want to include the javadocs (no idea how to do that)
[16:33:26 EST(-0500)] <esm> blah
[16:33:26 EST(-0500)] <esm> http://mail-archives.apache.org/mod_mbox/maven-users/200709.mbox/%3C9AB4FF82-4097-4391-B1DC-9C7A7217D9E0@commonjava.org%3E
[16:33:36 EST(-0500)] <esm> ay
[16:33:50 EST(-0500)] <esm> just delicious that url
[16:33:54 EST(-0500)] <esm> good reference later
[16:34:01 EST(-0500)] <esm> kind of explains why asselmby is fubar
[16:34:11 EST(-0500)] <EricDalquist> (smile)
[16:34:22 EST(-0500)] <esm> doc in an assembly
[16:34:23 EST(-0500)] <esm> hmm
[16:35:23 EST(-0500)] <esm> need a <fileSet> of the javadoc target directory?
[16:35:26 EST(-0500)] <EricDalquist> well it isn't too hard if I run 'mvn javadoc:javadoc assembly:assembly'
[16:35:31 EST(-0500)] <esm> yeah
[16:35:32 EST(-0500)] <esm> ok
[16:35:41 EST(-0500)] <EricDalquist> but I'd like to just have the javadoc run when I execute the assembly
[16:35:45 EST(-0500)] <EricDalquist> so profiles here I come
[16:35:50 EST(-0500)] <esm> and bind javadoc:javadoc t...
[16:35:51 EST(-0500)] <esm> yeah
[16:36:03 EST(-0500)] <esm> ugh.
[16:36:25 EST(-0500)] <esm> i'm sure there's a least one person who understands it.
[16:36:31 EST(-0500)] <EricDalquist> lol
[16:36:36 EST(-0500)] <esm> if they bite it...
[16:36:57 EST(-0500)] <esm> at some point i'll try to clean up some mods to the irc bot
[16:37:07 EST(-0500)] <esm> i have it logging in xml
[16:37:12 EST(-0500)] <EricDalquist> cool
[16:37:24 EST(-0500)] <esm> but ... well
[16:37:31 EST(-0500)] <esm> i forget exactly why i did it
[16:37:40 EST(-0500)] <esm> but i added lifecycle hooks for the message handlers
[16:37:51 EST(-0500)] <esm> oh
[16:37:59 EST(-0500)] <esm> i think it was because....
[16:38:22 EST(-0500)] <esm> right so teh message handlers are getting constructed using reflection or somesuch
[16:38:33 EST(-0500)] <esm> hmm
[16:38:35 EST(-0500)] <esm> i forget
[16:38:44 EST(-0500)] <EricDalquist> (smile)
[16:39:11 EST(-0500)] <esm> oh
[16:39:13 EST(-0500)] <esm> ok
[16:39:20 EST(-0500)] <esm> mh constructed with reflection
[16:39:27 EST(-0500)] <esm> then setter injection
[16:39:38 EST(-0500)] <esm> then init()
[16:40:17 EST(-0500)] <esm> cause the properties of the mh i defined in the xml config
[16:40:33 EST(-0500)] <esm> and then they are set on the class
[16:41:01 EST(-0500)] <esm> and i needed to know when the setters had all been called so i could initiaize some state
[16:41:05 EST(-0500)] <esm> safely
[16:41:13 EST(-0500)] <esm> anyway
[16:41:19 EST(-0500)] <esm> it needs some cleanup
[16:42:33 EST(-0500)] <esm> so the idea is now i can use cocoon or an xsl to search and transform the logs
[16:43:11 EST(-0500)] <esm> so every time the mh is flush()ed, i have to serialize a complete well formed xml doc
[16:43:23 EST(-0500)] <esm> so there is this dom in memory
[16:43:43 EST(-0500)] <esm> and messages get added to the dom and flushed periodically
[16:43:55 EST(-0500)] <EricDalquist> interesting
[16:44:09 EST(-0500)] <esm> but I have to figure out how to tell the mh when it is time to totally clear the dom
[16:44:36 EST(-0500)] <esm> e.g. it just keeps on getting appended to
[16:44:53 EST(-0500)] <esm> so I could flush the dom and then re-read in the file i just flushed, but that is really nasty
[16:45:09 EST(-0500)] <esm> like when the file rotates
[16:45:25 EST(-0500)] <esm> day1 i log to foo-day1.log
[16:45:36 EST(-0500)] <EricDalquist> why not use sax events to add things to the XML file
[16:45:39 EST(-0500)] <EricDalquist> I've done that in the past
[16:45:52 EST(-0500)] <esm> ah
[16:45:56 EST(-0500)] <EricDalquist> you use a SimpleDateFormat string to describe the file for rotating
[16:45:57 EST(-0500)] <esm> so... the reason i didn't go that way
[16:46:04 EST(-0500)] <esm> i did that for date rotating
[16:46:15 EST(-0500)] <EricDalquist> and on each flush you open "log-'YYYMMDD'.log"
[16:46:22 EST(-0500)] <esm> right
[16:46:27 EST(-0500)] <esm> oh
[16:46:27 EST(-0500)] <EricDalquist> run it through a SAX stream with your own filter
[16:46:31 EST(-0500)] <esm> right.
[16:46:35 EST(-0500)] <esm> so re-read the file each time
[16:46:52 EST(-0500)] <EricDalquist> and when endElement is called for the root element
[16:47:03 EST(-0500)] <EricDalquist> write out all of your new XML log elements
[16:47:04 EST(-0500)] <EricDalquist> yeah
[16:47:18 EST(-0500)] <EricDalquist> but unless you have multi-MB log files that shouldn't be too bad
[16:47:31 EST(-0500)] <esm> well i have to insert my new message elements before the closing root element is emitted
[16:47:34 EST(-0500)] <esm> right
[16:47:37 EST(-0500)] <EricDalquist> yup
[16:47:40 EST(-0500)] <esm> ok
[16:47:48 EST(-0500)] <EricDalquist> so in endElement:
[16:47:53 EST(-0500)] <EricDalquist> if (isLastElement){
[16:48:01 EST(-0500)] <EricDalquist> send new sax events for new log messages
[16:48:07 EST(-0500)] <esm> ok
[16:48:08 EST(-0500)] <EricDalquist> super.endElement(lastElement)
[16:48:09 EST(-0500)] <EricDalquist> }
[16:48:15 EST(-0500)] <esm> maybe i'll switch.
[16:48:17 EST(-0500)] <EricDalquist> else {
[16:48:24 EST(-0500)] <EricDalquist> super.endElement(element)
[16:48:25 EST(-0500)] <EricDalquist> }
[16:48:44 EST(-0500)] <EricDalquist> plus it is so much easier than working with the DOM APIs
[16:48:52 EST(-0500)] <esm> i'm using jdom right now, so its pretty easy
[16:48:55 EST(-0500)] <EricDalquist> which is about as much fun as working with maven assemblies
[16:48:56 EST(-0500)] <EricDalquist> ah
[16:49:14 EST(-0500)] <esm> for each document fragment dom.append( frag )
[16:49:35 EST(-0500)] <esm> then emit via a writer
[16:49:38 EST(-0500)] <esm> so its pretty easy
[16:49:43 EST(-0500)] <EricDalquist> ah
[16:50:26 EST(-0500)] <EricDalquist> I'm seriously tempted to just write an ant script to do the assembling (tongue)
[16:51:49 EST(-0500)] <esm> http://uportal.pastebin.com/m76634517
[16:51:50 EST(-0500)] <esm> yeah
[16:51:57 EST(-0500)] <esm> i hear you and would not at all blame u
[16:52:06 EST(-0500)] <EricDalquist> I mean I'm going to have to do a bunch in ant for the quickstart
[16:52:24 EST(-0500)] <esm> so yeah my problem i have to solve (either by switching to sax ro whatever) si that doc just gets bigger and bigger and bigger
[16:52:26 EST(-0500)] <EricDalquist> and just putting together a source copy with javadocs is apparently too much to ask
[16:52:29 EST(-0500)] <esm> yeah
[16:52:36 EST(-0500)] <esm> and ant will do it right-quick
[16:52:40 EST(-0500)] <EricDalquist> yup
[16:53:03 EST(-0500)] <EricDalquist> I already have the main build.xml all set which provides nice hooks back into the maven poms if I need it
[16:53:15 EST(-0500)] <EricDalquist> yeah
[16:53:22 EST(-0500)] <esm> how many people are goign to be cutting releases anyway
[16:53:28 EST(-0500)] <EricDalquist> I think read -> sax filter -> write is the best idea
[16:53:34 EST(-0500)] <EricDalquist> much smaller memory footprint
[16:53:46 EST(-0500)] <EricDalquist> since you should be able to use streams the whole way through
[16:54:41 EST(-0500)] <esm> yeah. i wasn't too concerned about it, cause lets be honest, if you have a meg of chat convo a day, thats a lot of text. but i see the advantage. plus i suppose if u're logging many channels
[16:54:53 EST(-0500)] <esm> is that why u threw in the rand on the flush tiimer task
[16:54:57 EST(-0500)] <esm> so they won't all flush at once
[16:55:10 EST(-0500)] <EricDalquist> yeah
[16:55:13 EST(-0500)] <esm> nice
[16:55:20 EST(-0500)] <EricDalquist> the confluence flush timer does the same thing
[16:55:32 EST(-0500)] <esm> yeah i'm using the same one
[17:02:29 EST(-0500)] <esm> its fun interviewing grad cs students
[17:02:37 EST(-0500)] <esm> we interviewd a couple today
[17:03:33 EST(-0500)] <esm> there are some basics missing from their coursework....
[17:04:19 EST(-0500)] <EricDalquist> heh ... yeah I remember running into that in the past when helping with interviewing
[17:04:36 EST(-0500)] <EricDalquist> more cs programs really need a well defined 'Software Engineer' track
[17:05:02 EST(-0500)] <EricDalquist> which is more like a mechanical/civil/etc. engineering degree in types of course
[17:05:15 EST(-0500)] <EricDalquist> like requirements/design/development
[17:05:33 EST(-0500)] <EricDalquist> a decent balance of math/theory and practical coding that includes design patterns
[17:05:49 EST(-0500)] <esm> yeah
[17:06:04 EST(-0500)] <EricDalquist> IDEs & senior design projects would be a + too
[17:06:18 EST(-0500)] <esm> they need to know the difference between static and instance variables.... and what polymorphism is (smile)
[17:06:36 EST(-0500)] <EricDalquist> heh
[17:07:09 EST(-0500)] <esm> one guy worked on some cool projects
[17:07:19 EST(-0500)] <esm> so we're hopeful he has time to work with us
[17:30:17 EST(-0500)] * clown_ (n=clown@guiseppi.atrc.utoronto.ca) has left ##uportal
[17:39:45 EST(-0500)] * michelled (n=team@142.150.154.199) has left ##uportal
[21:03:57 EST(-0500)] * jayshao (n=jayshao@68.239.246.163) has joined ##uportal
[22:33:49 EST(-0500)] * deuce (n=deuce@ip68-3-31-85.ph.ph.cox.net) has joined ##uportal
[23:27:01 EST(-0500)] * esm (n=esm@207-53-192-232.dynamic-dsl.qis.net) has joined ##uportal