User:Enus Linden/Office Hours/2009 July 11

• [2009/07/10 13:02] Morgaine Dinova: Hiya Enus!
• [2009/07/10 13:02] Latif Khalifa: heya Enus :)
• [2009/07/10 13:02] Enus Linden: hiya!
• [2009/07/10 13:02] Aimee Trescothick: the next stage of world domination
• [2009/07/10 13:02] Enus Linden: how are y'a;; this lovely day?
• [2009/07/10 13:03] Ardy Lay: Warm
• [2009/07/10 13:03] Morgaine Dinova: Fine thanks Enus, how's things, got some sleep at last?
• [2009/07/10 13:03] Fury Rosewood: hot... x.x
• [2009/07/10 13:03] Latif Khalifa: sleep is for the weak ;)
• [2009/07/10 13:03] Morgaine Dinova: Hehe
• [2009/07/10 13:03] Enus Linden: um, got some sleep yeah
• [2009/07/10 13:03] Enus Linden: i dont like sleeping tho
• [2009/07/10 13:04] Enus Linden: but sometimes i cant help it really
• [2009/07/10 13:04] Latif Khalifa: waste of time ;)
• [2009/07/10 13:04] Latif Khalifa: hehe
• [2009/07/10 13:04] Morgaine Dinova chuckles
• [2009/07/10 13:04] Morgaine Dinova: Damn protein bodies, the sooner we can be shod of them, the better
• [2009/07/10 13:04] Enus Linden: so thanks to saijanai, i have something of an agenda today! or at least, some pre-thought out things to bring up
• [2009/07/10 13:04] Latif Khalifa: lol
• [2009/07/10 13:05] Latif Khalifa: heya Eddy, Yohan :)
• [2009/07/10 13:05] Eddy Stryker: hey
• [2009/07/10 13:05] Enus Linden: hi all!
• [2009/07/10 13:05] Yohan Pintens: 'lo
• [2009/07/10 13:05] Enus Linden: welcome welcome
• [2009/07/10 13:05] Latif Khalifa: libomv group in force :P
• [2009/07/10 13:05] Eddy Stryker: maybe enus knows about notecard creation
• [2009/07/10 13:05] Morgaine Dinova: Wow, it's the ninja team :P
• [2009/07/10 13:05] Enus Linden: i know little about notecard creation atm
• [2009/07/10 13:05] Enus Linden: but, i can find out :D
• [2009/07/10 13:05] Yohan Pintens: Ed: did you see my note about encoding in IRC?
• [2009/07/10 13:06] Eddy Stryker: yohan: yep, checked that
• [2009/07/10 13:06] Enus Linden: so, things i was going to mention re: pyogp are...
• [2009/07/10 13:06] Morgaine Dinova listens
• [2009/07/10 13:06] Enus Linden: 1. running functional tests at build time using pyogp
• [2009/07/10 13:06] Enus Linden: 2. maybe moving pyogp to hg and killing buildout
• [2009/07/10 13:06] Enus Linden: 3. poppy testing a dataserver via the pyogp messaging system
• [2009/07/10 13:07] Enus Linden: and another i thought of 4. appearance and other recent cahnges to the trunk
• [2009/07/10 13:07] Enus Linden: and of course 5. enus still sucks at typing
• [2009/07/10 13:07] Morgaine Dinova: \o/
• [2009/07/10 13:08] Latif Khalifa: but being a python programmer, never misses the correct whitespace ;)
• [2009/07/10 13:08] Techwolf Lupindo: TrO{gO DsA[M,bN HyAqR4tDc TgrOo TgYPmE WeIjTyH P;AzWqS,. *
• [2009/07/10 13:08] Enus Linden: grrrrrr ;)
• [2009/07/10 13:08] Morgaine Dinova: Tech: APL? ;-)
• [2009/07/10 13:08] Enus Linden: so 1. i've started tinkering with an internal branch, where pyogp.lib.base comes in as an external and is used in the test.py framework
• [2009/07/10 13:09] Morgaine Dinova: Woohoo!
• [2009/07/10 13:09] Enus Linden: i don;t think this bit is publically available atm, i don't see it in http://svn.secondlife.com/trac/linden/browser/trunk/indra/test
• [2009/07/10 13:10] Enus Linden: but, there is a python test framework there which can spawn a pseudo grid
• [2009/07/10 13:10] Enus Linden: and i've written a couple of proof of concept tests that login to this psuedo grid and so far check that caht and IM works
• [2009/07/10 13:10] Enus Linden: and, so far it's working well.
• [2009/07/10 13:11] Enus Linden: that's swell. once this hits trunk, every branch that gets built in our build system will be able to run such tests
• [2009/07/10 13:11] Enus Linden: huzzah!
• [2009/07/10 13:11] Saijanai Kuhn: so you could have an agent domain working against that pseudo grid as well, eventuallY/
• [2009/07/10 13:11] Enus Linden: once that hits, i reckon pyogp functionality will grow a bit as people want to test their new thingie on the server side
• [2009/07/10 13:12] Saijanai Kuhn: python AD ,pyhon grid, python client, python EVERYTHING
• [2009/07/10 13:12] Enus Linden: sai, sure, once OGP work kicks back up we'll be sure to start in with this as well
• [2009/07/10 13:12] Techwolf Lupindo: and watches it craws...
• [2009/07/10 13:12] Enus Linden: well, python spawns the actual services necessary
• [2009/07/10 13:12] Enus Linden: it builds them, then instantiates them
• [2009/07/10 13:13] Enus Linden: kinda neat. it's not a full architecture, but enough tostart testing things on, and will expand in time
• [2009/07/10 13:13] Enus Linden: i've been looking forward to this for a while now
• [2009/07/10 13:14] Enus Linden: the great thing is, with one exception, and sort of unittest tests that one writes to test a live grid can be used in this context
• [2009/07/10 13:14] Enus Linden: the exception is the test class needs to inherit from an internal base calss to work in this context
• [2009/07/10 13:14] Enus Linden: so, nice progress here.
• [2009/07/10 13:15] Enus Linden: i will like;y extend an invitation to contribute to tests sometime in the next couple of months or so
• [2009/07/10 13:15] Morgaine Dinova: I'm a bit confused about the "pseudo-grid", and how that's related to PyOGP. Or is there no relation, except in that PyOGP points to it as its target?
• [2009/07/10 13:15] Enus Linden: the latter
• [2009/07/10 13:15] Morgaine Dinova: Aha
• [2009/07/10 13:15] Enus Linden: it is it's own thing, that other test run against as well
• [2009/07/10 13:16] Saijanai Kuhn: how is that different than Poppy's work on a simulator?
• [2009/07/10 13:16] Enus Linden: Poppy's work is not on the simulator, is on the dataserver
• [2009/07/10 13:16] Saijanai Kuhn: ah, ok
• [2009/07/10 13:16] Enus Linden: for the dataserver, no sim or client presence is required
• [2009/07/10 13:16] Morgaine Dinova: Well PyOGP mustn't have a single grid or world as "special" target ... that's the same problem we have with the main viewer, it's not interop friendly.
• [2009/07/10 13:16] Enus Linden: what's he's doing there is using UDPDispatcher to send messages directly to a dataserver instance under test
• [2009/07/10 13:17] Enus Linden: and validating responses that are sent back
• [2009/07/10 13:17] Enus Linden: in test, he also has the convenience of hooking into the dataserver to examine handling as well
• [2009/07/10 13:18] Enus Linden: that's just getting going though, and there may be a need to refactor some things to make it easier, but he's sending packets fine and been able to receive and check them as well, so perhaps not much needs to be done
• [2009/07/10 13:18] Morgaine Dinova: Enus: you might as well start proposing ways for PyOGP to talk to a data service from client side, because that's where Opensim is going and that's where OGP is going.
• [2009/07/10 13:19] Enus Linden: i do plan of refactoring a bunch of the messaging stuff soon, perhaps starting next week, to integrate message.xml conformance (and override of said conformance as needed for testing)
• [2009/07/10 13:19] Morgaine Dinova: SL's data service won't be available externally of course, but one could treat it as if it were. You'd at least have uniform access.
• [2009/07/10 13:19] Enus Linden: Morgaine, there's not really anything to it. one could mimic the messaging implementation in Region() for a new Dataserver() class and use it were it available
• [2009/07/10 13:20] Saijanai Kuhn: have you seen any need to replace the regex dictionary parser with the one that Zero wrote in indra?
• [2009/07/10 13:20] Enus Linden saijanai: kotler is looking at that right now
• [2009/07/10 13:20] Enus Linden: where is zero's btw?
• [2009/07/10 13:20] Saijanai Kuhn: its int he indra dierectory...
• [2009/07/10 13:20] Enus Linden: ty
• [2009/07/10 13:21] Enus Linden: which one specifically?
• [2009/07/10 13:23] Enus Linden: indra.ipc.llmessage?
• [2009/07/10 13:23] Saijanai Kuhn: right
• [2009/07/10 13:23] Enus Linden: sweet, i asked him to look for this for at least reference use
• [2009/07/10 13:23] Enus Linden: i'll point it out to him
• [2009/07/10 13:24] Enus Linden: Morgaine: when is dataserver work happening? is it relevant now?
• [2009/07/10 13:24] Morgaine Dinova: Hah, I wish! At current rate of progress on OGP, it'll be 2015
• [2009/07/10 13:25] Enus Linden: awesome, so I have time
• [2009/07/10 13:25] Enus Linden: XD
• [2009/07/10 13:25] Latif Khalifa: Morg, optimist!
• [2009/07/10 13:25] Enus Linden: so that's 1 & 3
• [2009/07/10 13:25] Yohan Pintens: you think it'll be that quick morgaine?
• [2009/07/10 13:25] Morgaine Dinova: But Opensim needs it, and OGP is now chasing Opensim's tail lights, so you might as well :P
• [2009/07/10 13:26] Morgaine Dinova: Yohan: they don't call me Miss Optimist for nothing ;P
• [2009/07/10 13:26] Latif Khalifa: hey Kotler, the appearance guru :D
• [2009/07/10 13:26] Enus Linden: are there docs describing it?
• [2009/07/10 13:26] Enus Linden: or is that not yet defined really?
• [2009/07/10 13:26] Morgaine Dinova: Hiya Kotler
• [2009/07/10 13:26] Enus Linden: anyway, pyogp could be useful for testing whatever given a little attention
• [2009/07/10 13:27] Enus Linden: the second agenda item (we've covered once and three): svn -> hg?
• [2009/07/10 13:28] Enus Linden: poppy would like to move pyogp to hg for various reasons, and we may go that route if in doing so it is 1. still publically accessible 2. able to be used in the testing we've started easily and 3. is painless
• [2009/07/10 13:28] Enus Linden: if we move it, i'd probably take the time to do away with buildout, which would require updated docs about dependencies and such, and consideration toward app/ storage and such
• [2009/07/10 13:28] Latif Khalifa: i don't understand what's with the linden svn hate ;)
• [2009/07/10 13:29] Enus Linden: so, it's just a head's up that we may talk about this more later
• [2009/07/10 13:29] Morgaine Dinova: Enus: re docs, no, because Opensim works as a swarm rather than as a directed entity. But this is the kind of thing that's happening on asset service decoupling -- http://code.google.com/p/cablebeach/wiki/CableBeachCore1_0
• [2009/07/10 13:29] Enus Linden: i don't get it either latif
• [2009/07/10 13:29] Enus Linden: but, eh, it's happening
• [2009/07/10 13:29] Yohan Pintens: its trendy
• [2009/07/10 13:29] Enus Linden: and not worth my time fighting
• [2009/07/10 13:30] Enus Linden: so the 4th and final agenda item is recent checkins i guess
• [2009/07/10 13:30] Saijanai Kuhn: Isn't Merov part of the Mecurial team?
• [2009/07/10 13:30] Latif Khalifa: well then just do it, leave the fights for more important stuff
• [2009/07/10 13:30] Enus Linden: maybe sai?
• [2009/07/10 13:30] Enus Linden: yep latif
• [2009/07/10 13:30] Morgaine Dinova: SVN hate is well deserved. That was last decade, now it's the decade of distributed CMS's. Hg or Git make sense in LL, especially given the freer commit system now on Snowglobe.
• [2009/07/10 13:31] Enus Linden: i'm in the midst of a small scrum right now, where i've asked that *some* amount of unit tests be required on all builds before QA touches a branch. I've have kicked up a fun discussion ;)
• [2009/07/10 13:31] Latif Khalifa: hehe
• [2009/07/10 13:31] Enus Linden: speaking of which, we need more unit tests in pyogp hahaha
• [2009/07/10 13:31] Enus Linden: but that's not production code atm, so we aren't too picky
• [2009/07/10 13:32] Enus Linden: yet
• [2009/07/10 13:32] Yohan Pintens: like I said, trendy
• [2009/07/10 13:32] Wyn Galbraith: Need to automate regression testing Enus, saves a lot of time.
• [2009/07/10 13:32] Morgaine Dinova: SVN only works well when you have a close high priest team, like on libomv and Opensim. Community development really benefits from a more distributed approach.
• [2009/07/10 13:32] Morgaine Dinova cackles
• [2009/07/10 13:32] Enus Linden: Wyn: that's what I'm starting off with the pyogp tests: they will do regression tests at a high level of features
• [2009/07/10 13:32] Enus Linden: if fail, send it back to the dev right away
• [2009/07/10 13:32] Yohan Pintens: Morgaine: we're not close at all, infact lkalif hates me.
• [2009/07/10 13:33] Morgaine Dinova chuckles
• [2009/07/10 13:33] Enus Linden: so: recent checkins: kotler committed some of the appearance code to trunk and has continued iterating and extending it
• [2009/07/10 13:33] Enus Linden: josh then was able to validate that his sit/stand/fly stuff was working
• [2009/07/10 13:34] Wyn Galbraith: Great ides Enus.
• [2009/07/10 13:34] Wyn Galbraith: *idea
• [2009/07/10 13:34] Enus Linden: i think i'm going to try and find time to start working with the code again this weekend/next week
• [2009/07/10 13:35] Enus Linden: i've freed up some of my daily schedule that should help make that a bit easier
• [2009/07/10 13:35] Enus Linden: of course, when we deploy and sh*t hits the fan, that takes time away, but all things considered i should be able to help out more again now :D
• [2009/07/10 13:35] Kotler Linden: I also commited so code that I refactored to clean it up a little and standardizes some constructor and method calls
• [2009/07/10 13:35] Morgaine Dinova: It should me much easier to create unit tests once the refactoring's been done, and pyogp can be treated as an installed external library
• [2009/07/10 13:35] Morgaine Dinova: be*
• [2009/07/10 13:36] Enus Linden: morgaine: true
• [2009/07/10 13:37] Enus Linden: and now, whenever i make a change, i'll make sure unit tests are there
• [2009/07/10 13:37] Enus Linden: so, that's it for me...
• [2009/07/10 13:37] Enus Linden: what else?
• [2009/07/10 13:38] Morgaine Dinova: I noticed that eventlet 0.8.11 came out. Is there any change to pyogp for that?
• [2009/07/10 13:38] Enus Linden: i missed that morgaine, and i need to check.
• [2009/07/10 13:38] Latif Khalifa: how far did Kotler get with the appearance stuff?
• [2009/07/10 13:38] Enus Linden: using that should let pyogp work on windows again!
• [2009/07/10 13:38] Enus Linden: is there an egg for it morgaine? not that I like eggs, but we're using them now so....
• [2009/07/10 13:38] Kotler Linden: I have been able to get wearable packets and I will soon begin parsing the infomation out of them
• [2009/07/10 13:39] Latif Khalifa: parsing assets is fun
• [2009/07/10 13:39] Enus Linden: still need to start thinking about baking textures i guess :(
• [2009/07/10 13:39] Kotler Linden: Latif: )
• [2009/07/10 13:39] Latif Khalifa: Enus, you need to start thinking about implementing asset download/upload
• [2009/07/10 13:39] Morgaine Dinova: Enus: Don't know, I just stumbled on it by accident while Googling :P
• [2009/07/10 13:39] Latif Khalifa: and all 5 ways to do it
• [2009/07/10 13:40] Kotler Linden: I started groundworkd for an asset manager
• [2009/07/10 13:40] Latif Khalifa: and you will need all of them for appearance to work
• [2009/07/10 13:40] Latif Khalifa: need to download wearable assets (done in one way)
• [2009/07/10 13:40] Kotler Linden: just grabbing wearables so far though
• [2009/07/10 13:40] Latif Khalifa: texture assets (another way)
• [2009/07/10 13:40] Morgaine Dinova: That's cool Kotler
• [2009/07/10 13:41] Latif Khalifa: and then texture uploads (currently two ways depending on which sim version)
• [2009/07/10 13:41] Kotler Linden: Latif: yeah your right still need to look into how that works
• [2009/07/10 13:41] Latif Khalifa: look at libomv code ;)
• [2009/07/10 13:41] Kotler Linden: i have been :)
• [2009/07/10 13:42] Enus Linden is busy seeing what happens on windows w/ pyogp and eventlet
• [2009/07/10 13:42] Morgaine Dinova Kotler: look at the LL documentation!
• [2009/07/10 13:42] Yohan Pintens: use WinGridProxy, you can see the packets flow in their pure glory
• [2009/07/10 13:42] Morgaine Dinova ducks
• [2009/07/10 13:42] Latif Khalifa: Morg, hahaha
• [2009/07/10 13:42] Kotler Linden: hah hah
• [2009/07/10 13:43] Latif Khalifa: libomv is the best documentetion on the protocol you will get ;)
• [2009/07/10 13:43] Kotler Linden: I do when I can find it
• [2009/07/10 13:43] Saijanai Kuhn: someone had a wireshark module for SL packets written in Lua, but Infinity says she never got it to work..,.
• [2009/07/10 13:43] Morgaine Dinova chuckles --- not the first company in this situation, by several million :P
• [2009/07/10 13:43] Latif Khalifa: Sai, WGP for the win
• [2009/07/10 13:43] Saijanai Kuhn: shedule chatting with jhuliman/Eddy Stryker also, Kotler
• [2009/07/10 13:44] Latif Khalifa: just come to #libomv-dev on efnet
• [2009/07/10 13:44] Yohan Pintens: WGP is specific to the SL Protocol, no wireshark plugin can come close.
• [2009/07/10 13:44] Enus Linden: sai: i've used that module
• [2009/07/10 13:44] Enus Linden: works fine by me, but, wgp is kinda great
• [2009/07/10 13:44] Squirrel Wood: Hand over all your nuts! :p
• [2009/07/10 13:44] Saijanai Kuhn: WGP is the interface you wrote?
• [2009/07/10 13:44] Kotler Linden: ok
• [2009/07/10 13:44] Latif Khalifa: sai, wgp is gui version of libomv's GridProxy
• [2009/07/10 13:45] Yohan Pintens: Sai: yes, and its present tense still :)
• [2009/07/10 13:45] Saijanai Kuhn: ah, ok
• [2009/07/10 13:46] Latif Khalifa: I'll make a binary build for those who are to lazy to build it
• [2009/07/10 13:46] Enus Linden: :)
• [2009/07/10 13:46] Enus Linden: but building is fun!
• [2009/07/10 13:46] Yohan Pintens: lol it takes 20 seconds to build, thats beyond lazy if someone doesn't take that amount of time.
• [2009/07/10 13:46] Alexandrea Fride: and you can drink allot of coffie while waiting on it :p
• [2009/07/10 13:47] Kotler Linden: http://xkcd.com/303/
• [2009/07/10 13:47] Latif Khalifa: works on both .net and mono
• [2009/07/10 13:47] Alexandrea Fride: rofl kotler
• [2009/07/10 13:48] Latif Khalifa: #1 excuse for bugs "but it works fine on my system"
• [2009/07/10 13:48] Morgaine Dinova: WGP is nice and useful, but it's a disaster when you try to get it to follow much traffic. It can't keep up, curls up and dies, because its display callbacks are in its widget handling threads, and there's no load control.
• [2009/07/10 13:48] Latif Khalifa: Morg?
• [2009/07/10 13:48] Morgaine Dinova: GridProxy itself doesn't have that problem.
• [2009/07/10 13:48] Yohan Pintens: Morgaine: upgrade that 8086 machine to a 286 or something
• [2009/07/10 13:49] Morgaine Dinova: 2.6GHz Core2Duo should be enough
• [2009/07/10 13:49] Saijanai Kuhn: better than my GUI for packet display in pygop. Of course that was compsing the GUI on the fly from the packet dit
• [2009/07/10 13:49] Wyn Galbraith: We had that saying at Autodesk Latif " works on my machine " end of bug
• [2009/07/10 13:49] Yohan Pintens: I'm connected through WGP now and have all packets and capabilities turned on without issues
• [2009/07/10 13:49] Yohan Pintens: so as lkalif said, works on my machine.. not a bug :)
• [2009/07/10 13:49] Morgaine Dinova: Yohan: you're probably on Windows
• [2009/07/10 13:49] Wyn Galbraith: HAHA
• [2009/07/10 13:50] Latif Khalifa: yeah mono sucks, especially at gui stuff
• [2009/07/10 13:50] Kay Dreier: hey enus - how are you?
• [2009/07/10 13:51] Yohan Pintens: yeah I tend to use what works best, not whats trendy :)
• [2009/07/10 13:51] Enus Linden: well thanks kay, and you?
• [2009/07/10 13:51] Kay Dreier: fine thanks ;-)
• [2009/07/10 13:51] Enus Linden: so, i've not much more for today myself...
• [2009/07/10 13:51] Morgaine Dinova: Yohan: and I'me connected through GridProxy right now, which is also runnin under Mono, with no issues. Whoever wrote the GUI part of WGP did it poorly.
• [2009/07/10 13:51] Kay Dreier: where is your head? *laughs*
• [2009/07/10 13:52] Enus Linden: a bug ate my head
• [2009/07/10 13:52] Kay Dreier: *giggles* okay
• [2009/07/10 13:52] Saijanai Kuhn looks around for Qarl's ant..
• [2009/07/10 13:52] Wyn Galbraith: a but ate my brain
• [2009/07/10 13:52] Latif Khalifa: Morgaine, no need for insults
• [2009/07/10 13:52] Wyn Galbraith: *bug
• [2009/07/10 13:52] Yohan Pintens: Enus: any reason the PlacesReply packet would be often not including the Simname field since the recent update? (the field is sent, its just now often blank)
• [2009/07/10 13:52] Elbereth Witte shoots the bird.
• [2009/07/10 13:53] Yohan Pintens: Latif: I just consider the source :)
• [2009/07/10 13:53] Enus Linden: Yohan: not sure.
• [2009/07/10 13:53] Enus Linden: which recent update?
• [2009/07/10 13:53] Yohan Pintens: my estate was upgraded in the last couple days
• [2009/07/10 13:53] Kay Dreier: 10 minutes you have to endure *laughs*
• [2009/07/10 13:53] Enus Linden: Tuesday we deployed to centrals, and yesterday we deployed to some sim hosts
• [2009/07/10 13:53] Yohan Pintens: so 1.27.something
• [2009/07/10 13:54] Enus Linden: k
• [2009/07/10 13:54] Latif Khalifa: I had fun with the capability that sends you back group membership info... it sends you different type of packets back depending on when you invoke it lol
• [2009/07/10 13:54] Enus Linden: did you notice it prior to yesterday Yohan?
• [2009/07/10 13:54] Morgaine Dinova: Latif: bad code is bad code. If it had been tested on Linux, it would have emerged immediately that the graphics code can't handle anything but the slowest packet rate.
• [2009/07/10 13:54] Yohan Pintens: Enus: I can guarantee you it did not occur prior
• [2009/07/10 13:54] Enus Linden: so it started yesterday for sure eh?
• [2009/07/10 13:54] Enus Linden: just isolating the package to look at
• [2009/07/10 13:54] Latif Khalifa: Morgaine, and you know this is WGP poor code versus Mono poor implementation of GUI on X?
• [2009/07/10 13:55] Yohan Pintens: Enus: yes
• [2009/07/10 13:55] Enus Linden: kk yohan.
• [2009/07/10 13:55] Enus Linden: i'll see about taking a look?
• [2009/07/10 13:56] Enus Linden: that should be a . not a ?
• [2009/07/10 13:56] Yohan Pintens: I have a bot that checks group owned land via PlacesQuery/Reply, and prior to the simulator upgrade the Simname field always contained the proper simname. This is testing with 15 groups and aboout 5,000+ parcels total between them
• [2009/07/10 13:57] Yohan Pintens: I noticed it when I ended up with duplicates in the database, I was uniquely indexing on Simname, GlobalX, GlobalY and OwnerID. I dropped SimName from the index to correct the issue for me, but it still exists for you :)
• [2009/07/10 13:57] Enus Linden: ty Yohan, will take a look
• [2009/07/10 13:57] Enus Linden: and now, i gotta run to my next meeting!
• [2009/07/10 13:57] Yohan Pintens: good luck with that :)
• [2009/07/10 13:57] Enus Linden: thanks ;)
• [2009/07/10 13:58] Latif Khalifa: Enus, you might want to take a look at AgentGroupDataUpdate message too, sometimes ListInProfile field comes in GroupData map, and sometimes in NewGroupData ;)
• [2009/07/10 13:58] Enus Linden: and thanks all for making it