| 04:38.26 | *** join/#riftuidev seebs (~seebs@67.51.147.18) |
| 04:38.26 | *** mode/#riftuidev [+o seebs] by ChanServ |
| 08:47.21 | *** join/#riftuidev Wildtide (~quassel@host86-135-37-139.range86-135.btcentralplus.com) |
| 08:47.40 | Wildtide | good morning folks |
| 08:54.15 | *** join/#riftuidev Noshei (~quassel@c-67-161-146-244.hsd1.co.comcast.net) |
| 08:54.21 | Noshei | morning wild |
| 09:42.39 | Thorarin | Morning |
| 09:46.08 | Thorarin | I think my brain hasn't quite woken up yet... Any idea why Inspect.Buff.List would return an empty list? |
| 09:46.18 | Thorarin | I'm calling it from an Event.Unit.Add event handler |
| 09:46.39 | Thorarin | And yes, there are buffs on the unit ;) |
| 09:46.42 | Noshei | is the unit fully available? |
| 09:47.06 | Thorarin | I imagine, I just cast a debuff on it? |
| 09:47.35 | Noshei | yes, but the addon system has to see the unit as fully available |
| 09:49.18 | Noshei | I would do a quick check on the units availability before trying to get the buffs |
| 09:50.13 | Thorarin | Guess I'll try that yeah... not sure why I never needed to do that before |
| 09:52.03 | Thorarin | Probably because I was doing the listing from Event.System.Update.Begin *shrug* |
| 10:25.10 | Thorarin | Do you need to track unit availability purely by the events? Or is there some cheap way to check availability? |
| 10:25.22 | Noshei | there is an inxpect now |
| 10:25.46 | Thorarin | I couldn't find one? Know what it's called? |
| 10:25.47 | Noshei | Inspect.Unit.Detail().availability |
| 10:26.02 | Thorarin | Okay, but that involves getting all the details |
| 10:26.14 | Noshei | nope |
| 10:26.26 | Noshei | you can query for only a specific detail |
| 10:26.53 | Noshei | Inspect.Unit.Detail("player.focus.target.pet").availability |
| 10:27.08 | Thorarin | That still gets all the details |
| 10:27.14 | Noshei | nope |
| 10:27.23 | Thorarin | Yes, you just throw the rest away |
| 10:27.24 | Noshei | it will only return availability |
| 10:28.01 | Thorarin | It may be acceptable though, will do some testing |
| 10:28.08 | Noshei | but we don't have to handle tossing out the rest of the details, I'm pretty sure thats build into the api |
| 10:28.29 | Thorarin | No, that's just Lua |
| 10:28.41 | Thorarin | You get a table, then access a key in that table |
| 10:29.00 | Thorarin | Still the entire table is first retrieved |
| 10:45.22 | Thorarin | Now that I use the availability check, the unit is always available... |
| 10:45.26 | Thorarin | Weird :S |
| 11:06.31 | Thorarin | Oh well, seems it's unavailable on /reloadui, so it's good for something :) |
| 11:12.16 | Noshei | you are using availability right, not available and unavailable? |
| 11:12.43 | Thorarin | Yeah |
| 11:12.56 | Thorarin | I can run completely in strict mode ;) |
| 11:15.14 | Thorarin | Man, these performance optimizations are working nicely... But I think I've increased the lines of code by 40% :< |
| 11:15.59 | ZorbaTHut | always one of those inevitable tradeoffs. fast code takes more lines. code that behaves predictably also takes predictably longer. |
| 11:16.13 | ZorbaTHut | better speed = more memory usage, etc. |
| 11:16.22 | Thorarin | Usually yes |
| 11:18.01 | Thorarin | It's quite fun if you run into a piece of code you can refactor to take fewer lines while being more efficient and/or robust |
| 11:18.53 | ZorbaTHut | oh, definitely, but that's usually a sign that past-you was sort of an idiot :V |
| 11:18.58 | ZorbaTHut | (can't tell you how many times that's happened to me) |
| 11:19.30 | Thorarin | Reminds me of a particular conversation :) |
| 11:19.30 | Thorarin | Junior programmer: "Wow, I've written 800 lines of code today!" |
| 11:19.30 | Thorarin | Me: "I've deleted 8000 lines, and it does more than it did this morning :D" |
| 11:19.57 | ZorbaTHut | when I was at Google, I managed to fix a bug in the calculator by deleting 45 lines of code. No code added or changed, just 45 lines straight-up deleted. |
| 11:20.18 | Thorarin | lol |
| 11:20.25 | Thorarin | That is kinda rare |
| 11:20.57 | ZorbaTHut | yeah, I ended up emailing the original writer to ask if I was missing something. His response was something along the lines of "nope, that was just a dumb idea from the beginning" |
| 11:21.12 | Thorarin | Sometimes you can remove a 2-5 line exception case, but 45 lines... |
| 11:22.17 | Thorarin | Hmm, should probably refactor some of this other code to use LibUnitChange |
| 11:22.23 | Thorarin | Best do one thing at a time tho |
| 11:24.04 | Thorarin | I'm not sure I understand the Event.Buff.Change event btw, when do you need that? |
| 11:24.35 | ZorbaTHut | it's pretty rare, but there's a few cases where a buff can change its status without being added or removed. In some cases this will be something like a stacking buff where individual stacks can fall off or be removed. |
| 11:24.36 | Thorarin | Can't think of a situation where existing buffs change... |
| 11:25.26 | Thorarin | Ah, so... Michael Bringhurst should display that behavior? |
| 11:25.59 | ZorbaTHut | buffs being *added* will generally be shown as a remove/add pair. Not always, but generally. |
| 11:26.16 | ZorbaTHut | do the stacks fall off gradually? I actually don't know, I've never paid much attention to that fight :P |
| 11:26.21 | Thorarin | But if you drag him out of the moonlight, stacks gradually decrease |
| 11:26.26 | ZorbaTHut | ah, quite likely then |
| 11:26.42 | ZorbaTHut | no guarantees, sometimes things are implemented weirdly, but that's a place I'd definitely check. |
| 11:26.48 | Thorarin | Half the time I cba and just stand there :D |
| 11:27.16 | Thorarin | It's not like bosses hurt in T2s nowadays |
| 11:45.28 | Thorarin | Hmmmz... |
| 11:45.35 | Thorarin | Buff removal handler is some epic code :< |
| 11:45.48 | Thorarin | More than adding one, interestingly |
| 12:09.17 | ZorbaTHut | obviously I don't know how your system works, but I've been surprised how often setup is clean and simple, but teardown is a colossal pain |
| 12:09.33 | ZorbaTHut | you'd definitely expect it to be the opposite, but it really isn't |
| 12:32.37 | Thorarin | Well, originally KaruulAlert had very basic event handlers, signalling only that something with buffs had changed. Then doing lots of Inspects to figure out what actually changed at certain intervals |
| 12:32.46 | Thorarin | I'm changing it to be much more event based |
| 12:33.14 | Thorarin | Looks like the buff alerts are 5 times more efficient now :> |
| 12:33.58 | Thorarin | It's a bit of a pain with inherited code, at first you don't really want to change too much... |
| 12:34.20 | Thorarin | And for me everything was fine, because I only use a handful of alerts |
| 12:34.35 | Thorarin | But some silly sods actually used 40 of them |
| 12:34.59 | Thorarin | On machines that are at least twice as slow as mine :) |
| 12:35.41 | Thorarin | The coroutines fixed it for the short term, but once that was mostly under control I figured I'd do a proper rewrite |
| 12:36.28 | Thorarin | Only thing that takes much time now is the actual frame updates |
| 12:39.33 | Thorarin | Well, relatively speaking that is |
| 12:41.52 | Wildtide | from some benchmarking i did a while ago, hidden frames are really really cheap |
| 12:42.23 | Wildtide | i don't know how KA works, but having every possible icon pre-prepared, and just changing it's visibility, should be fairly efficient |
| 12:43.06 | Wildtide | which is probably exactly what you already do :) |
| 12:45.59 | Thorarin | Pretty much, except for timer updates |
| 12:46.09 | Thorarin | Which are text frames :P |
| 12:46.29 | Thorarin | Actually, in my case 6 text frames since I outline and shadow |
| 12:46.39 | Thorarin | (optionally) |
| 12:47.59 | Thorarin | Managed to improve the performance there a bit as well, since the original code sometimes set the timer to empty text, then to an actual value again later |
| 12:48.15 | Thorarin | Only having the text change once every second makes it a lot better :) |
| 12:53.22 | Thorarin | Oh, ZorbaTHut, if you're still around... I have a Russian user complaining he cannot enter Russian letters in a text field. He can only copy/paste them in there |
| 12:53.45 | Thorarin | Doesn't make a lot of sense to me however, and I do have a Russian client now, but no keyboard... |
| 12:53.55 | ZorbaTHut | huh, odd |
| 12:54.05 | ZorbaTHut | if you have a high-end windows version you can install the russian language pack |
| 12:54.08 | Thorarin | Very :/ |
| 12:54.12 | ZorbaTHut | I think that might be Ultimate only, though |
| 12:54.45 | Thorarin | Not sure if you can use the on-screen keyboard though? |
| 12:55.12 | Thorarin | I can install Russian language pack, but what do I need it for? :) |
| 12:55.35 | ZorbaTHut | well, you could test it :V but I can test it also |
| 13:00.16 | Thorarin | Well, on the English client the fonts don't show normally, so would have to be the russian one |
| 13:04.34 | Thorarin | I don't see the relation with the language pack though.. I don't need a Russian Windows for RIFT |
| 13:06.15 | ZorbaTHut | I'm assuming there's just some glitch with the addon textfield on Russian clients |
| 13:10.15 | Thorarin | водка! |
| 13:13.10 | *** join/#riftuidev g0urra (~g0urra@unaffiliated/gourra) |
| 13:17.56 | Thorarin | Hmmm |
| 13:18.18 | Thorarin | RIFT seems to ignore the Windows keyboard settings. It types latin characters instead |
| 13:19.26 | Thorarin | I think that is the problem Zorba |
| 13:19.41 | Thorarin | It works in the chat/command line, but not in text fields |
| 13:23.29 | ZorbaTHut | curious |
| 13:25.01 | Thorarin | I don't know who build the russian client, but perhaps you could have someone check it out? |
| 13:25.22 | Thorarin | I will do some more testing, but I doubt I'm causing this somehow :) |
| 13:27.19 | Thorarin | If needed I can make a minimal repro case |
| 13:35.20 | Thorarin | Other addons are showing the same behavior however |
| 13:39.41 | Noshei | oh hey ZorbaTHut I got an error from a user that could only have happened if Event.Mail fired before Event.Interaction with mail = true. Even more interesting is that the Event.Mail had to of returned a valid table or the user would have gotten a different error |
| 13:41.03 | ZorbaTHut | hmm, interesting |
| 13:41.30 | ZorbaTHut | I know there's a fix planned to make it so Event.Mail won't fire in a few cases like that, but I'm not sure if it turned into a 1.9 or a 1.10 fix |
| 13:41.41 | ZorbaTHut | I'll check that out |
| 13:44.08 | Noshei | oh and to add to that my addon was able to Inspect.Mail.Detail() as well before Event.interaction had fired |
| 13:45.26 | Noshei | oh course I'm making the assumption that the execution that I have occur after Event.Interaction wasn't somehow interupted, but I would think that would have thrown an error as well |
| 13:56.54 | Noshei | hey Wildtide, just a thought, but it would be great if you had mouseforward/back events for the scroll bar in the Create Gadget dialog |
| 14:05.38 | Thorarin | Yes! |
| 14:11.35 | Noshei | Sweet, I'm the only addon I have that prints my loaded message to the Combat console |
| 14:12.18 | Noshei | well dang it |
| 14:16.17 | Noshei | and scratch that |
| 14:17.02 | Noshei | Inspect.Console.List() returns nil by the point that Event.Addons.Startup.End fires |
| 14:20.17 | ZorbaTHut | yeah I'm definitely leaning towards delaying the addon startup until much later in the startup sequence |
| 14:21.09 | Noshei | I think thats a good idea |
| 14:21.26 | Noshei | make it easier for us to pull stuff during startup |
| 14:21.42 | ZorbaTHut | yeah, and it'll solve issues where certain things - like the console list - simply aren't loaded yet |
| 14:22.22 | Noshei | well that would take care of the feature request I just posted asking for an event for when the console is loaded |
| 14:22.52 | ZorbaTHut | hrm, there should totally be console events anyway, though |
| 14:24.07 | Noshei | can't see much need tbh |
| 14:24.33 | Noshei | unless there is some point where the console is completely unavailable to us after start up |
| 14:24.49 | Wildtide | i'd quite like one to say a console has been added removed, specifically one called wtLog :) |
| 14:25.00 | Wildtide | * added OR removed |
| 14:25.01 | ZorbaTHut | what wildtide just said :) |
| 14:25.43 | Wildtide | Noshei: yep, the lack of mouse wheel handling is something that keeps annoying me too |
| 14:26.02 | Wildtide | i will add it, it's just one of about a million things i need to do :( |
| 14:32.32 | Noshei | oh ZorbaTHut, a small issue I've noticed for a while, and I don't think it's specific to the addon system, sometimes when you take all of the items from a message (generally from the AH) it won't automically delete the message |
| 14:33.05 | Noshei | I've seen this happen through the Native UI as well as within my own UI |
| 14:33.09 | ZorbaTHut | I've noticed that with the Take All button but I always assumed it was a bug in that button |
| 14:34.00 | Noshei | yeah I'm more inclined that it is a timing issue somewhere, but I'm not really sure where or how |
| 14:34.56 | Noshei | but since the native button works basically the same was mine does I would think it's a bug in the base code somewhere |
| 14:38.13 | ZorbaTHut | yup, probably true |
| 14:38.20 | ZorbaTHut | this is likely not going to be a high priority to fix ;) |
| 14:39.00 | Noshei | yeah I figured, of course it might be something super simple that only takes a couple minutes to fix ;) |
| 14:49.07 | ZorbaTHut | it might be, but it'd take more minutes than that to find :) |
| 14:49.20 | ZorbaTHut | but thanks for letting me know, that tells me roughly *where* the problem is |
| 14:49.37 | ZorbaTHut | so if someone tells me it doesn't work I can tell them where the real issue is :V |
| 14:55.40 | Noshei | :) |
| 21:12.36 | *** join/#riftuidev Torhal_ (~torhal@c-71-207-224-241.hsd1.al.comcast.net) |
| 21:12.36 | *** mode/#riftuidev [+v Torhal_] by ChanServ |
| 21:23.59 | *** join/#riftuidev Torhal (~torhal@c-71-207-224-241.hsd1.al.comcast.net) |
| 21:24.05 | *** mode/#riftuidev [+v Torhal] by ChanServ |
| 22:37.53 | *** join/#riftuidev Torhal|work (~torhal_wo@67.58.77.206) |
| 22:37.53 | *** mode/#riftuidev [+v Torhal|work] by ChanServ |
| 23:13.13 | *** join/#riftuidev dolby (~dolby-wow@MMOI/Administrator/pdpc.silver.dolby) |
| 23:13.13 | *** mode/#riftuidev [+o dolby] by ChanServ |
| 23:31.38 | *** join/#riftuidev arton (~arton@tx-184-6-176-223.dhcp.embarqhsd.net) |
| 23:45.07 | Wildtide | lol, thought i'd broken my addon completely there. my cursor had gone weird, and everytime i clicked anywhere, it selected an NPC i couldn't see |
| 23:45.27 | Wildtide | turned out my camera was inside an NPC's body :) |