Chat log from the meeting on 2017-05-16

In case anyone wonders... chit chat unrelated to OpenSim gets stripped from the chat log for better readability ^^ [11:17] Andrew.Hellershanks @hg.osgrid.org: I was just looking over the changes to the code from this past week. [11:17] Andrew.Hellershanks @hg.osgrid.org: Some HG related changes, some changes related to land, and a few changes related to pgSQL. [11:17] Ubit Umarov: OoopsGrid.. hmm that sounds like my test grid [11:17] Gavin.Hird @grid.xmir.org:8002: I have a question for Ubit on the last weeks changes [11:17] Mike.Lorrey @grid.kitely.com:8002: without all the accepted LSL functions working, so not a valid load test [11:17] Arielle.Popstar @hg.osgrid.org: its has  been a work in progress [11:18] Gavin.Hird @grid.xmir.org:8002: you pushed a change or two to make sure OpenSim exited properly on mono higher than 4.2.2 [11:18] Ubit Umarov: i did ?? [11:18] Gavin.Hird @grid.xmir.org:8002: yes, that worked fine [11:18] Gavin.Hird @grid.xmir.org:8002: removing the Posix.mono.dll file [11:18] Ubit Umarov: ahh that was another issue [11:18] Ubit Umarov: was fail on run [11:18] Gavin.Hird @grid.xmir.org:8002: if you upgrade to say 4.8 Robust now hangs like OpenSim.exe did [11:19] Andrew.Hellershanks @hg.osgrid.org: Removing mono.posix.dll fixed an exception that was being spit out during region startup [11:19] Ubit Umarov: bc that fille did collide with the one on the machine [11:19] Ubit Umarov: that hanging is another issue :( [11:19] Gavin.Hird @grid.xmir.org:8002: ok, but the change also made OpenSim.exe exit properly on mono higher than 4.2.2 [11:19] Andrew.Hellershanks @hg.osgrid.org: The mono versions are coming thick and fast these days. [11:19] Gavin.Hird @grid.xmir.org:8002: unintended benefit.... [11:19] Ubit Umarov: nice [11:20] Andrew.Hellershanks @hg.osgrid.org: Seems not too long ago that 4.2 came out. [11:20] Gavin.Hird @grid.xmir.org:8002: I tried to compile on 5.0 [11:21] Gavin.Hird @grid.xmir.org:8002: since Microsoft decided to place that on my system when I installed Visual Studio 2017 for macOS [11:21] Ubit Umarov: they now own it :) [11:21] Gavin.Hird @grid.xmir.org:8002: it was quickly uninstalled on that system [11:21] Gavin.Hird @grid.xmir.org:8002: :-)) [11:21] Andrew.Hellershanks @hg.osgrid.org: Gavin, you mean .NET 5.0? [11:21] Gavin.Hird @grid.xmir.org:8002: no [11:21] Gavin.Hird @grid.xmir.org:8002: mono 5.0 [11:22] Sheera Khan: Mac OS-X is an UNIX after all ... [11:22] Andrew.Hellershanks @hg.osgrid.org: MS makes VS for something other than Windows? interesting [11:22] Gavin.Hird @grid.xmir.org:8002: Sorry I don't do windows [11:22] Gavin.Hird @grid.xmir.org:8002: yes, they released for macOS this week [11:23] Gavin.Hird @grid.xmir.org:8002: they hope that people will use it write iOS and Adroid applications [11:26] Andrew.Hellershanks @hg.osgrid.org: Any other new (or old) OS issues since last week? [11:27] Ubit Umarov: uff been lost on lsl json mess.. that forgot all the chages done before :) [11:28] Ubit Umarov: hmm let me see the commits :) [11:28] Andrew.Hellershanks @hg.osgrid.org: Ubit, I'm still trying to figure out how you can have hundreds of json tests when there aren't that many LSL ones that are json based. [11:28] Mike.Lorrey @grid.kitely.com:8002: Possibly because nobody has documented JSON functionality properly in the wiki? [11:28] Ubit Umarov: improvements to PGSQL thx to gavin here [11:28] Andrew.Hellershanks @hg.osgrid.org: Its all in the SL wikis of LSL functions [11:29] Arielle.Popstar @hg.osgrid.org claps [11:29] Arielle.Popstar @hg.osgrid.org: thx Gavin [11:29] Mike.Lorrey @grid.kitely.com:8002: yet the OS wiki says those functions are not implemented in OS [11:29] Gavin.Hird @grid.xmir.org:8002: yw [11:29] Gavin.Hird @grid.xmir.org:8002: I am not sure how many use it [11:29] Ubit Umarov: possible bc no even SL did correctly documented it [11:29] Andrew.Hellershanks @hg.osgrid.org: Mike, well that is more of a problem with the OS wiki that may not be up to date. [11:29] Ubit Umarov: yeackk [11:29] Mike.Lorrey @grid.kitely.com:8002: not many use it here because the wiki tells them it hasn't been implemented [11:29] Gavin.Hird @grid.xmir.org:8002: but terrain loading was significantly faster after the upgrade [11:29] Ubit Umarov: i should just delete than and assume we don't have it [11:30] Arielle.Popstar @hg.osgrid.org: lol [11:30] Andrew.Hellershanks @hg.osgrid.org: There are two main wikis that document LSL calls. Sometimes you need to compare the information in both of them. [11:30] Kayaker Magic: Anyone who complains about the OpenSim Wiki is required to help work on it to improve it. TAG YOU'RE IT MIKE! [11:30] Andrew.Hellershanks @hg.osgrid.org: :) [11:30] Ubit Umarov: and see how they where diferent :p [11:30] Arielle.Popstar @hg.osgrid.org: exactly Kay [11:30] Mike.Lorrey @grid.kitely.com:8002: I cannot improve it if I can't find any source material upon which to base user documentation. [11:30] Ubit Umarov: but that second wiki is now gone [11:31] Arielle.Popstar @hg.osgrid.org: you seem to know a lot  about what it should do [11:31] Ubit Umarov: only a few backups here and there [11:31] Kayaker Magic: There are a few copies of it around for the referencing. [11:31] Andrew.Hellershanks @hg.osgrid.org: I will update the OS wiki if I notice something out of date. I have added some pages about newer OSSL functions. I haven't checked the page that lists the status of LSL/OSSL support. [11:31] Leighton.Marjoram @grid.lighthousepoint.co.uk:8002: I would help with the wiki if some one who knew what to write sat with me. [11:31] Andrew.Hellershanks @hg.osgrid.org: Not recently that is. [11:31] Mike.Lorrey @grid.kitely.com:8002: I've written a half dozen software manuals in my career, so I know a bit about technical writing, and if OS devs were a company I'd be docking every programmers pay for not documenting their work [11:32] Andrew.Hellershanks @hg.osgrid.org: I was having a chat with a couple of people about mesh upload. I was saying it needs to be documented for OS and the two different physics engines. [11:32] Arielle.Popstar @hg.osgrid.org: the manual is the code i believe i have heard  it said :) [11:32] Kayaker Magic: In cases where the OS wiki says "not implimented yet" when it has been, you can use the LL Wiki as a source. [11:32] Gavin.Hird @grid.xmir.org:8002: Melanie quote [11:32] Arielle.Popstar @hg.osgrid.org: mmhmm [11:32] Andrew.Hellershanks @hg.osgrid.org: Arielle, yes that has been said on many occasions. It isn't easy to read. [11:33] Arielle.Popstar @hg.osgrid.org: unfortunately when Justin left the wiki's and  opensim site in general stopped [11:33] Ubit Umarov can't even read his code a weak later... so stays quiet [11:33] Arielle.Popstar @hg.osgrid.org: ;) [11:33] Gavin.Hird @grid.xmir.org:8002: is it even possible to add users to it any more? [11:33] Arielle.Popstar @hg.osgrid.org: should be [11:34] Arielle.Popstar @hg.osgrid.org: I believe Dan  has the "keys" [11:34] Ubit Umarov: think we did had spam and things like that on the wiki [11:34] Mike.Lorrey @grid.kitely.com:8002: I'm registered on the wiki, but without any source material to work from, I cannot justifiably update or edit anything with any authority. [11:34] Andrew.Hellershanks @hg.osgrid.org: AFAIK, users can be added but random users can not register accounts like in the past. [11:34] Sheera Khan: Mike, you could well try to dock the payments for the devs... wouldn't be too much work though :-( [11:34] Mike.Lorrey @grid.kitely.com:8002: I already do that every week Sheera, Ubits cats hate me [11:35] Andrew.Hellershanks @hg.osgrid.org: I didn't think we have any OS pages that document the details of any LSL functions. [11:36] Mike.Lorrey @grid.kitely.com:8002: the list of LSL functions that have been implemented says that the JSON functions are not implemented, and the two pages on the osJSON functions are nearly blank. [11:37] Ubit Umarov: as now. consider then not implemented [11:37] Mike.Lorrey @grid.kitely.com:8002: all they really say is what their threat level is, as if thats at all important. [11:37] Ubit Umarov: there are functions imported from whitecore back in 2015 by mel [11:37] Ubit Umarov: but.. a disaster [11:38] Ubit Umarov: im working on them now.. but.. uff [11:38] Ubit Umarov: lets see [11:38] Arielle.Popstar @hg.osgrid.org: only 112 errors  you said [11:38] Ubit Umarov: yeap "only" [11:38] Andrew Hellershanks: A quick check of LSL_Api.cs says there are 5 json functions implemented. [11:38] Andrew Hellershanks: Only two in OSSL_Api.cs [11:39] Mike.Lorrey @grid.kitely.com:8002: well how about this: my email is bntholdings@gmail.com. I invite ANY dev to email me complete description and function of things they implement, and I'll turn your coder-speak into understandable documentation. But I don't have ESP. [11:40] Arielle.Popstar @hg.osgrid.org: maybe mention that  in piper mail at opensim-dev? [11:40] Mike.Lorrey @grid.kitely.com:8002: I will need to be able to test any os specific functions on a region somewhere [11:40] James atLLOUD: I'd up my monthly OSgrid donation to support that work. If that was an option. [11:40] Andrew Hellershanks: llJsonGetValue, llJson2List, llList2Json, llJsonSetValue, llJsonValueType, osParseJSONNew, osParseJSON [11:40] Ubit Umarov: the ossl things are a diferent set [11:41] Mike.Lorrey @grid.kitely.com:8002: well I don't need to feed my cats, so send your money to Ubit, as long as he emails me. [11:41] Andrew Hellershanks: Mike, you can check whether the wiki lists those functions I just mentioned as being supported. [11:41] Ubit Umarov: unrelated ( and incompatible most likely, bc of ll idea of json on those functions ) [11:42] Arielle.Popstar @hg.osgrid.org: not sure  if this is helpful [11:42] Arielle.Popstar @hg.osgrid.org: http://lslwiki.digiworldz.com/ [11:42] Ubit Umarov: you really don't read Andrew [11:42] Andrew Hellershanks: Who has time, Ubit? ;) [11:42] Ubit Umarov: well on a meeting one is suposed to read others :p [11:43] Ubit Umarov: hmm wikis like that inwd< [11:44] Ubit Umarov: didn't LL already forced some to close due to copyrights? [11:44] Arielle.Popstar @hg.osgrid.org: no idea [11:44] Ubit Umarov: someone told me that.. [11:44] Arielle.Popstar @hg.osgrid.org: copy it quick then [11:45] Gavin.Hird @grid.xmir.org:8002: Sitesucker [11:45] Mike.Lorrey @grid.kitely.com:8002:  llJson2List     No             Requires Implementation llJsonGetValue     No             Requires Implementation llJsonSetValue     No             Requires Implementation llJsonValueType     No             Requires Implementation [11:46] Ubit Umarov: list l = llJson2List("[[1,2,3],{\"a\":3,\"b\":[true,\"test\",6]}]");    llOwnerSay((string)l);    verify_list("llJson2List first",l,["[1,2,3]","{\"a\":3,\"b\":[true,\"test\",6]}"]);    list n = llJson2List(llList2String(l,0)); [11:46] Ubit Umarov: 10:57] Object: FAIL test llJson2List l,0: on list entry type comparison on index 0, expected: TYPE_STRING, result: TYPE_INTEGER [10:57] Object: FAIL test llJson2List l,0: on list value comparison on index 0, expected: JSON_TRUE, result: 1 [10:57] Object: FAIL test llJson2List l,0: llList2CSV comparison, expected: ﷖, test, 6, result: 1, test, 6 [11:46] Ubit Umarov: yeackkk [11:47] Ubit Umarov: [11:47] Object: Tests complete. Ran 319 tests with 112 failures. [11:48] Ubit Umarov: was 319 failures and scripts crashs :) [11:48] Ubit Umarov: but still 112 yeack [11:48] dj phil: and the jSonStore work now on 0.9 ? [11:49] Ubit Umarov: guess its like it was on 0.82 [11:49] dj phil: yeah but not sure on 0.9. [11:50] Ubit Umarov: i didn't touch it and don't remember having to during merge [11:50] Ubit Umarov: but my mem.. well [11:51] Ubit Umarov: well last week [11:51] Ubit Umarov: code as added to prevent presence of the same avatar on a grid... [11:51] Andrew Hellershanks: Sounds like a problem with the deserializer in llJson2List [11:51] Ubit Umarov: something i don't understand why we didn't had.. [11:52] Arielle.Popstar @hg.osgrid.org: because  devs  were using  it for testing [11:52] Andrew Hellershanks: Mike, I will update that wiki page to state the functions are implemented but may not be fully working. [11:52] Ubit Umarov: since all code assumes only one presence [11:52] Mike.Lorrey @grid.kitely.com:8002: thank you Andrew. [11:52] Andrew Hellershanks: yw [11:52] Ubit Umarov: well when i give up fixing them [11:53] Andrew Hellershanks: Logger, I know you had an issue you wanted to talk about regarding region crossings. [11:53] Gavin.Hird @grid.xmir.org:8002: Don't let that stand in the way of testing it with more than one Ubit [11:53] Ubit Umarov: ill place a mantis at least with the things i will not do as SL [11:53] Arielle.Popstar @hg.osgrid.org: the more Ubits, the faster  it will be implemented [11:54] Ubit Umarov: another thing added was code to make parcels Dwell ( traffic ) show something [11:54] Gavin.Hird @grid.xmir.org:8002: I want to test that [11:55] Ubit Umarov: osg did started to add a module. for that [11:55] dj phil: hehe me too :) [11:55] Gavin.Hird @grid.xmir.org:8002: did you test it with which viewer? [11:55] logger sewell: yes but not really sure what is causing it we have some water regions and with a physical boat when you cross a region line it will jump straight up in the air at crossing and on one region it will stop you and crash the avatar with no errors or warnings in the log files [11:55] Ubit Umarov: but.. i did add the numbers collect code [11:55] Ubit Umarov: to replace the one on that module [11:55] Ubit Umarov: and of course.. its for all opensim users :) [11:56] Andrew Hellershanks: Too bad we don't have the person behind BulletSim here today. [11:56] Andrew Hellershanks: Anyone else here up to date about BulletSim and vehicle crossings? [11:57] logger sewell: yes Robert may have an idea or we may have to load a few sims with UbOde and try that [11:57] Gavin.Hird @grid.xmir.org:8002: Are there any test scripts that are used for vehicle crossing [11:58] Gavin.Hird @grid.xmir.org:8002: so we have something to verify against? [11:58] Ubit Umarov: seems bullet does vehicle hover diferent for ubOde [11:58] Ubit Umarov: but didn't test [11:58] Mike.Lorrey @grid.kitely.com:8002: I have some physical vehicle scripts to test [11:58] Ubit Umarov: from.. [11:58] Gavin.Hird @grid.xmir.org:8002: it would be great to have a test suite [11:58] Mike.Lorrey @grid.kitely.com:8002: a sailboat and a motorcycle/speeder script [11:59] Gavin.Hird @grid.xmir.org:8002: these are the tests that must pass for any changes / improvements [11:59] Ubit Umarov: ubOde on that as a no standard feature, hover height od zero does not stop it [11:59] Ubit Umarov: only time > 300 [11:59] logger sewell: I have the one we got from here and have tried Kayakers same thing happens on that one sim [11:59] Ubit Umarov: and ubode does negative hover if it is relative to water [12:00] Mike.Lorrey @grid.kitely.com:8002: I know my scripts work on bullet fine, but not on ODE, so testing on ubODE would be useful [12:00] Ubit Umarov: ODE didn't have a lot of vehicle code [12:01] Andrew Hellershanks: Hm... should I mark the LSL json functions as ready for testing or just as partially implemented. [12:01] Kayaker Magic: My boats are non-physical! If it jumps on a crossing it is not a Bullet issue! [12:01] Ubit Umarov: well the dwell time is now possible 2 see on About landmark of fs [12:01] Ubit Umarov: if both regions are on master version [12:02] Ubit Umarov: ( when about lm decides to work :p ) [12:03] Ubit Umarov: and the numbers are not as SL [12:03] Ubit Umarov: they do raise fast.. and decay [12:03] Ubit Umarov: decay faster on high traffic.. then slower in time [12:03] Ubit Umarov: logaritmic +- [12:04] Ubit Umarov: in theory will never return to 0 :p [12:05] Kayaker Magic: My scripts use llSetKeyframedMotion to cross region borders, if Logger is having trouble crossing with that, there is no need to look at physics for the problem. [12:06] Kayaker Magic: I could try crossing with llSetPos and see if the same thing happens. [12:07] Andrew Hellershanks: I did a test a while back using Key Framed Motion (KFM) to move an object across a sim crossing and it worked fine. I wasn't sitting on the object at the time. [12:07] Ubit Umarov: yes we all see that your avatar is good and well :p [12:08] Ubit Umarov: and not lost on the void interregions :) [12:08] Ubit Umarov: yeack that happened to me 3 times at sl [12:08] logger sewell: I'll try to do some more testing and get some more data for next week [12:09] Andrew Hellershanks: You can also try catching Robert on the #opensim-dev IRC channel. [12:10] Arielle Popstar: is  it on .8.2  or .9? [12:10] Andrew Hellershanks: 0.9 [12:10] Arielle Popstar: didnt have the problem on .8.2? [12:10] Ubit Umarov: 82 had crossings? :p [12:11] logger sewell: we didn't try it before Andrew upgraded to 0.90 [12:11] Arielle Popstar: There was  work  done on crossings  during  .8 [12:12] Ubit Umarov: yeap...  scary crossings  :) [12:12] Ubit Umarov: for avatars !! [12:12] Arielle Popstar: thats  why we  have vars [12:13] Ubit Umarov: its good reason yes [12:13] logger sewell: That may be what we'll have to do we have a 100 region var now [12:14] Ubit Umarov: crossings will only work with GRID and Both regions on same backbone  or with very good connections [12:14] Ubit Umarov: here at osg with regions have high ping times.. [12:14] Ubit Umarov: well..  not so nice [12:16] Arielle Popstar: use object teleport  :) [12:16] Ubit Umarov: object teleport is a extension to crossing code :p [12:16] Andrew Hellershanks: Any other thoughts about sim crossings with vehicles? [12:17] Andrew Hellershanks: Any other OS topics for today? [12:17] Gavin.Hird @grid.xmir.org:8002: As I said it would be good to have aet of test scripts [12:18] Ubit Umarov: a reason why it is nasty if regions are outside view range [12:18] Ubit Umarov: on the json im using one from the Others ;) [12:18] Andrew Hellershanks: btw, logger, the person to talk with in IRC is misterblue [12:19] dj phil: We need working pathfinding functions, like SL :p [12:19] Ubit Umarov: no we do not [12:19] Ubit Umarov: :p [12:19] Andrew Hellershanks: yes, that would be nice. [12:20] Ubit Umarov: noo [12:20] Gavin.Hird @grid.xmir.org:8002: I will remove the pathfinding code for the opensim viewer [12:20] dj phil: yes yes, we need :) [12:20] Andrew Hellershanks: Ubit, what is the issue with pathfinding functions? [12:21] Ubit Umarov: what pathfinding functions? [12:21] Gavin.Hird @grid.xmir.org:8002: The viewer nags the server constantly for pathfinding stuff [12:21] Ubit Umarov: we never had those [12:21] dj phil: all of course ! [12:21] Gavin.Hird @grid.xmir.org:8002: also on region crossings [12:22] Gavin.Hird @grid.xmir.org:8002: so the code will go [12:22] Ubit Umarov: we did had stats on pathfinding..  strange.. not having it..  but was a glitch :) [12:22] dj phil: Aurora have llPursue if i remember ... [12:22] Andrew Hellershanks: Do the KFM functions eliminate the need for pathfinding? [12:22] Gavin.Hird @grid.xmir.org:8002: experience keys is something we probably should have [12:22] Ubit Umarov: diff thing andrew [12:23] Andrew Hellershanks: yeah. Just wasn't sure how similar/different they are [12:23] Ubit Umarov: navmesh and pathfind [12:23] Kayaker Magic: KFM moves you, but doesn't help you figure out where to move. [12:23] Arielle Popstar: Dahlia  looked  into pathfinding if  i remember right [12:23] Andrew Hellershanks: pathfinding is the lazy persons KFM? ;) [12:23] dj phil: Carousel don't move you with KFM ... [12:23] Ubit Umarov: dahlia made pathfinding [12:23] Ubit Umarov: script side [12:23] dj phil: ha ! [12:24] Andrew Hellershanks: script side? Hm... [12:24] Kayaker Magic: No, pathfinding tells you where you can KFM without passing through objects in the way. [12:24] Ubit Umarov: its does what name tells [12:24] Ubit Umarov: u tell a npc go there [12:25] Ubit Umarov: and he goes around walls etc [12:25] dj phil: nice for games :) [12:25] Ubit Umarov: and with speeds that depend on parameters on the navmesh etc [12:25] Ubit Umarov: get human players and u have no need for it :P [12:26] Ubit Umarov: humans are good on pathfind [12:26] Ubit Umarov: well not me lol [12:26] Kayaker Magic: You could also use pathfinding to navigate an autonomous vehicle around. [12:26] Kayaker Magic: The vehicle could be run using physics or KFM [12:26] Arielle Popstar: not  sure  if tghis  is of value [12:26] Arielle Popstar: https://github.com/opensimworld/active-npcs/blob/master/README.md [12:26] Gavin.Hird @grid.xmir.org:8002: well...https://www.youtube.com/watch?v=flkgNn50k14 [12:27] Gavin.Hird @grid.xmir.org:8002: check this video [12:28] dj phil: human is bad ^^ [12:28] Kayaker Magic: LMAO! [12:28] Gavin.Hird @grid.xmir.org:8002: In a lagged sim anyone could use pathfinding [12:29] Andrew Hellershanks: About time to wrap up todays meeting. I need to get going soon. [12:30] Andrew Hellershanks: Any other last minute items?