Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 19 Next »

[09:17:49 EDT(-0400)] * esm (n=esm@asdf.dkc.jhu.edu) has joined ##uportal
[10:10:25 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:44:31 EDT(-0400)] * pberry (n=pberry@waldorf.CSUChico.EDU) has joined ##uportal
[10:46:38 EDT(-0400)] <EricDalquist> morning all
[10:47:52 EDT(-0400)] <esm> hola
[10:48:02 EDT(-0400)] <EricDalquist> I'm taking a look at the api work
[10:48:05 EDT(-0400)] <EricDalquist> (finally)
[10:48:15 EDT(-0400)] <esm> oh, no problem (smile)
[10:48:31 EDT(-0400)] <esm> Firefox plugin of the day: https://addons.mozilla.org/en-US/firefox/addon/28
[10:48:50 EDT(-0400)] <EricDalquist> nifty
[10:48:55 EDT(-0400)] <esm> simple but handy: will allow you to duplicate tabs, either in a new tab in the same window or a new tab in a new window
[10:49:04 EDT(-0400)] <esm> yes, i've wanted "tab detachment" for some time now
[10:49:08 EDT(-0400)] <EricDalquist> yeah
[10:49:27 EDT(-0400)] <esm> You may need http://extensionroom.mozdev.org/more-info/keyconfig
[10:49:52 EDT(-0400)] <esm> because the key to duplicate a tab in a new window (command-shift-n) is already bound tom seomthing else on my mbp
[10:51:31 EDT(-0400)] <EricDalquist> oops
[10:52:11 EDT(-0400)] <esm> it is bound to some crazy thing that I have no idea what it does
[10:56:14 EDT(-0400)] <EricDalquist> I want a FF extension that replaces the default bookmarks system
[10:56:58 EDT(-0400)] <EricDalquist> specifically back it with google bookmarks
[10:57:04 EDT(-0400)] <EricDalquist> and possibly have label filtering
[10:57:19 EDT(-0400)] <EricDalquist> so I can share bookmarks between computer and have them accessible online
[10:57:21 EDT(-0400)] <EricDalquist> (smile)
[10:57:31 EDT(-0400)] <esm> oh
[10:57:40 EDT(-0400)] <esm> well the new delicous plugin does just that
[10:57:47 EDT(-0400)] <EricDalquist> hrm
[10:57:52 EDT(-0400)] <esm> so its backed by delicous instead of google
[10:57:52 EDT(-0400)] <EricDalquist> I might have to look at that
[10:57:57 EDT(-0400)] <esm> i use it
[10:58:08 EDT(-0400)] <esm> http://del.icio.us/help/firefox/extensionnew
[10:58:27 EDT(-0400)] <esm> so
[10:58:43 EDT(-0400)] <esm> with the older delicious bookmark plugin i would tag pages
[10:58:54 EDT(-0400)] <pberry> bookmarks?
[10:58:57 EDT(-0400)] <pberry> what are those? (wink)
[10:59:05 EDT(-0400)] <esm> for example when I come across javadoc i know i will reference again, i tag it as 'api'
[10:59:34 EDT(-0400)] <esm> i have a firefox live bookmark which consumes the rss version of my api tag
[10:59:40 EDT(-0400)] <esm> http://del.icio.us/rss/emetsger/api
[10:59:52 EDT(-0400)] <EricDalquist> nice
[10:59:56 EDT(-0400)] <esm> so that was my poor man's sync
[11:00:10 EDT(-0400)] <esm> then i upgraded to the latest delicious plugin when i got my new mbp
[11:00:19 EDT(-0400)] <esm> and it is much improved, with the features you want
[11:00:26 EDT(-0400)] <esm> but i still use my poor man's sync
[11:01:32 EDT(-0400)] <EricDalquist> (smile)
[11:01:35 EDT(-0400)] <EricDalquist> so ... I'm a little nervous about this pass at the -api refactoring. I'm not happy with how much other stuff gets pulled in.
[11:01:52 EDT(-0400)] <esm> i agree
[11:01:56 EDT(-0400)] <esm> it is pretty bad
[11:02:01 EDT(-0400)] * pberry does battle with awstats
[11:02:08 EDT(-0400)] <EricDalquist> I'm working on the transfer/domain/dao/registry refactoring
[11:02:32 EDT(-0400)] <EricDalquist> and I think we may be better off making a -api module but slowly move interfaces over
[11:02:49 EDT(-0400)] <EricDalquist> then I can keep things like dependency creep in mind when moving interfaces over
[11:03:16 EDT(-0400)] <EricDalquist> since we can spend more time on each part
[11:03:22 EDT(-0400)] <EricDalquist> what do you think?
[11:03:29 EDT(-0400)] <esm> sounds good to me
[11:03:36 EDT(-0400)] <esm> so we are back to modularization (smile)
[11:03:41 EDT(-0400)] <EricDalquist> yeah
[11:04:50 EDT(-0400)] <esm> so do you want to start the -api module in your branch then
[11:05:00 EDT(-0400)] <esm> or should we just put an empty project into trunk
[11:05:00 EDT(-0400)] <EricDalquist> hrm
[11:05:26 EDT(-0400)] <esm> or wait and do nothing until we have something to modularize
[11:05:37 EDT(-0400)] <EricDalquist> I'm not sure ... I'm already in the situation where little changes are causing big ripples throughout the framework (I knew that would happen)
[11:05:57 EDT(-0400)] <EricDalquist> let me think about it a little more ....
[11:06:07 EDT(-0400)] <esm> right obviously thats a bad thing
[11:06:11 EDT(-0400)] <esm> no problem (smile)
[11:07:02 EDT(-0400)] <EricDalquist> do you think you would have any cycles to play with getting an Apache Directory instance embedded in the trunk?
[11:08:36 EDT(-0400)] <esm> yeah
[11:08:38 EDT(-0400)] <esm> def
[11:08:55 EDT(-0400)] <EricDalquist> that would be wonderful
[11:08:58 EDT(-0400)] <esm> i'm pushing out pluto 1.1.3 now but that is about done
[11:08:58 EDT(-0400)] <esm> ok
[11:09:01 EDT(-0400)] <esm> is there a jira for it
[11:09:11 EDT(-0400)] <esm> i'll add one if not
[11:09:11 EDT(-0400)] <EricDalquist> I'm not sure if I made one
[11:09:17 EDT(-0400)] <esm> ok i can look
[11:09:27 EDT(-0400)] <EricDalquist> this dao stuff is going to take me a while, probably until the end of next week
[11:09:30 EDT(-0400)] <esm> btw, can we get that random portlet off of the jira homepage
[11:09:34 EDT(-0400)] <esm> the one withthe 302 redirect
[11:09:46 EDT(-0400)] <EricDalquist> so I'm trying to think of tasks that can be done in parallel without causing too many merge headaches
[11:09:48 EDT(-0400)] <EricDalquist> um
[11:09:50 EDT(-0400)] <EricDalquist> I'll see
[11:10:05 EDT(-0400)] <esm> cool thanks, that will help mylar work with JA-SIG's jira
[11:10:30 EDT(-0400)] <esm> you have to be logged out to see it. I opened an issue in the infrastructure project on it
[11:10:36 EDT(-0400)] <EricDalquist> all gone
[11:10:52 EDT(-0400)] <EricDalquist> yeah I've been playing with mylar a bit
[11:10:55 EDT(-0400)] <EricDalquist> looks pretty cool
[11:11:20 EDT(-0400)] <esm> oh
[11:11:21 EDT(-0400)] <esm> yeah
[11:11:23 EDT(-0400)] <esm> wow thanks
[11:11:28 EDT(-0400)] <esm> i just re-loaded and saw it was gone
[11:11:48 EDT(-0400)] <esm> yeah mylar seems to be cool. The online presenetation is worth the hour
[11:11:56 EDT(-0400)] <esm> i'll go back and see bits and pieces of it again
[11:23:22 EDT(-0400)] <esm> hmmm i may have been using the wrong url for jasig's repo the entire time
[11:23:33 EDT(-0400)] <esm> http://www.ja-sig.org/issues/secure/ worked
[11:23:33 EDT(-0400)] <EricDalquist> ?
[11:23:40 EDT(-0400)] <esm> for the mylar task repository
[11:23:43 EDT(-0400)] <EricDalquist> ah
[11:25:16 EDT(-0400)] <esm> crap
[11:25:25 EDT(-0400)] <EricDalquist> ?
[11:25:55 EDT(-0400)] <esm> http://www.ja-sig.org/issues/secure/ didn't work. But after deleting and re-adding http://www.ja-sig.org/issues/ now works
[11:28:22 EDT(-0400)] <EricDalquist> so with the ldap integration
[11:28:28 EDT(-0400)] <EricDalquist> part of it will be adding logic to the user dao
[11:28:46 EDT(-0400)] <esm> UPT-307 btw
[11:28:47 EDT(-0400)] <esm> ok
[11:28:49 EDT(-0400)] <EricDalquist> so it can create users that are authd to ldap but the portal database doesn't know about
[11:28:50 EDT(-0400)] <EricDalquist> yeah
[11:28:51 EDT(-0400)] <EricDalquist> I saw it
[11:29:05 EDT(-0400)] <EricDalquist> I'm subscribed to all the jasig jira notification lists
[11:30:56 EDT(-0400)] <esm> thats UserDaoImpl in org.jasig.portal.security.dao.springjdbc
[11:31:12 EDT(-0400)] <EricDalquist> yeah
[11:31:30 EDT(-0400)] <EricDalquist> well I haven't really looked at exactly what changes will need to be made
[11:31:37 EDT(-0400)] <EricDalquist> but that dao will need to be able to create new users
[11:31:58 EDT(-0400)] <esm> ok thanks good to know
[11:37:21 EDT(-0400)] <EricDalquist> found a gmarks addon for firefox
[11:37:33 EDT(-0400)] <esm> noice
[11:37:34 EDT(-0400)] <EricDalquist> has a toolbar replacement and lets me filter on labels (big grin)
[11:43:27 EDT(-0400)] * pberry (n=pberry@waldorf.CSUChico.EDU) has joined ##uportal
[14:50:35 EDT(-0400)] <EricDalquist> anyone still awake?
[14:52:04 EDT(-0400)] <esm> kind of
[14:52:12 EDT(-0400)] <esm> i have to be awake for a 4pm meeting
[14:52:19 EDT(-0400)] <EricDalquist> (smile)
[14:52:21 EDT(-0400)] <esm> a FRIDAY, FOUR PM MEETING
[14:52:28 EDT(-0400)] <EricDalquist> (sad)
[14:52:32 EDT(-0400)] <esm> kill -9
[14:52:46 EDT(-0400)] <EricDalquist> sudo ./cancelMeeting?
[14:53:01 EDT(-0400)] <EricDalquist> I'm looking at the portlet preferences interfaces in up3
[14:53:21 EDT(-0400)] <EricDalquist> and wondering if using a LinkedHashMap instead of a List would be good?
[14:53:33 EDT(-0400)] <EricDalquist> except there is no interface to represent it
[14:53:36 EDT(-0400)] <EricDalquist> just a concrete impl
[14:54:47 EDT(-0400)] <esm> hmm
[14:55:05 EDT(-0400)] <EricDalquist> there is some implied ordering from the portlet.xml
[14:55:14 EDT(-0400)] <EricDalquist> but it would be handy to have O(1) access to preferences
[14:55:32 EDT(-0400)] <EricDalquist> or perhaps I'm worrying too much
[14:55:47 EDT(-0400)] <EricDalquist> and the overhead of looking through the list for a specific preference isn't very high
[14:56:01 EDT(-0400)] <esm> yeah sounds a little like premature optimization
[14:56:23 EDT(-0400)] <EricDalquist> yeah (smile)
[14:56:47 EDT(-0400)] <esm> linkedhashmap implements Map
[14:56:54 EDT(-0400)] <EricDalquist> yeah
[14:57:21 EDT(-0400)] <EricDalquist> but I'd like it if there was an OrderedMap extends Map interface
[14:57:56 EDT(-0400)] <esm> ... because then the client of the interface knows it doesn't have to deal with ordering?
[14:58:10 EDT(-0400)] <EricDalquist> hrm
[14:58:13 EDT(-0400)] <esm> why is OrderMap extends Map important
[14:58:20 EDT(-0400)] <EricDalquist> yeah ... I think I'm over thinking this
[14:58:51 EDT(-0400)] <esm> well, its friday afternoon too
[14:58:58 EDT(-0400)] <esm> so thinking goes much slower
[14:58:58 EDT(-0400)] <EricDalquist> very true
[14:59:37 EDT(-0400)] <esm> you could use LinkedList if you want O(1) time, right?
[15:00:02 EDT(-0400)] <esm> or is list not O(1)
[15:00:05 EDT(-0400)] <EricDalquist> well that would be O(N) since you'd have to iterate over possibly N items to find the right one
[15:00:13 EDT(-0400)] <esm> right right
[15:00:24 EDT(-0400)] <EricDalquist> but hopefully people don't have thousands of preferences
[15:00:33 EDT(-0400)] <EricDalquist> so N should be very small
[15:00:35 EDT(-0400)] <EricDalquist> and not really matter
[15:00:36 EDT(-0400)] <esm> heh yeah i don't think we have to worry about that yet
[15:00:52 EDT(-0400)] <EricDalquist> ok ... back to the refactoring (smile)
[16:44:56 EDT(-0400)] <esm> later all have a good weekend

  • No labels