| 00:25.11 | seebs | Am I missing something obvious, or does UI.Native.Tooltip:GetBounds produce nothing of value? |
| 00:25.32 | seebs | I seem to be consistently getting l=0, r=256, t=0, b=1226. |
| 00:26.08 | Mere | doesn't look helpful |
| 00:27.10 | seebs | Hmm. The obvious tactic of making a larger frame with a smaller frame inside it does not work well if you want the inner frame to be transparent. |
| 00:29.11 | Mere | trying to put a border around it? |
| 00:29.48 | seebs | Yeah. And I can do four thin frames, but in other cases I've done a behind frame and an in-front frame. But I didn't want transparency then. |
| 00:30.15 | Mere | yeah, I thought that when I did my tooltip thing, a border would be nice, but that means 4 more frames... |
| 00:32.44 | seebs | I am trying to figure out how to tell where to put my tooltip relative to a UI native one. |
| 00:32.56 | seebs | And I can't figure out how, since I can't tell how much space I have around my frame. |
| 00:36.58 | Mere | does the tooltip window need to be open for it to give sane bounds? |
| 00:37.22 | seebs | I dunno. I'm checking it in a handler for Event.Tooltip. |
| 00:42.12 | Mere | that would seem it should be valid then |
| 00:42.21 | Mere | what does UI.Native.TooltipAnchor do? |
| 00:42.35 | Mere | or is that something around moveing the tooltip to another place on screen? |
| 00:42.54 | seebs | I think that's what's used for tooltips which aren't tied to the mouse -- but item tooltips are. |
| 00:46.09 | Mere | ah |
| 01:06.10 | seebs | hmm |
| 01:06.15 | seebs | setpoint is not liking my plan. |
| 01:06.35 | seebs | Which is to set BOTTOMRIGHT on a frame, then periodically setpoint TOP and set width. |
| 01:07.13 | seebs | Ohh. |
| 01:08.00 | seebs | Ah-hah. I want an x offset of nil, not 0. |
| 01:08.09 | Mere | heh |
| 01:49.31 | ZorbaTHut | seebs, I know of a few issues with UI.Native.Tooltip, but haven't had time to track them down and kill them |
| 01:51.31 | Mere | zorba: any chance you'll implement libhealcomm functionality in rift? :) |
| 01:52.03 | ZorbaTHut | mere, don't hold your breath :) |
| 01:52.21 | Mere | I just don't want to do it myself ;) |
| 01:52.33 | Mere | not that we've currently got cross player addon comms (that I know of) |
| 01:53.50 | Mere | and I really don't want to try and monitor 20 players all casting, and working out which casts are heals, and then guestimating the size of the heal |
| 01:54.10 | Mere | (actually aren't there some healing pets as well...) |
| 01:54.37 | ZorbaTHut | in fairness, I don't really want to do it either ;) The big problem is that it involves modifying the server ability usage ordering, and that's something which is waaaaay outside of my expertise |
| 01:54.57 | ZorbaTHut | it may be extremely difficult - Rift abilities are complicated and I doubt we have any existing code for predicting what an ability might do |
| 01:55.28 | ZorbaTHut | Really, all of that is easy except for "guesstimating the size of the heal". |
| 01:55.55 | Mere | it's more I don't want to have to list out every heal in the game, and watch for it to start being casted |
| 01:56.25 | Mere | in theory the server knows the expected normal size (as a range) of the heal, and I assume it can allow for personal SP buffs etc |
| 01:56.41 | ZorbaTHut | You're dramatically underestimating how complicated the Rift ability system is. :) |
| 01:56.58 | Mere | oh probably :) |
| 01:56.59 | ZorbaTHut | this is one of those situations where it may actually be easier to implement an estimation system than to work with the real data. |
| 01:57.09 | Mere | lol |
| 01:57.29 | Mere | you mean until it finally goes through the cast path, and gets all the buffs and talents applied, what it'll do isn't known? |
| 01:57.34 | Mere | how do the tooltips know? |
| 01:58.13 | ZorbaTHut | Pretty much, yep. And I think they may just guess, or be designed by the ability people - we've had a few cases where the tooltips and the real effect ended up out of synch. |
| 01:58.45 | Mere | tbh I'd expect the current tooltip values, they're going to be "good enough" |
| 01:58.58 | ZorbaTHut | Now, I could be totally wrong here, but I *do* know the system is complicated. As much as I'd like to provide this, it's just not the best time-to-benefit ratio right now. |
| 01:59.38 | Mere | yeah, I'm kind of going to backseat it for now, as parsing combat logs etc would feel painful |
| 01:59.53 | Mere | displaying the information is easy, making it up is hard :) |
| 02:00.01 | ZorbaTHut | if I were doing this, I'd get a bunch of friends to give me their spell power and tooltip values - I'm gonna guess that *most* of the time, healing is a linear relationship constant + ratio * spellpower |
| 02:00.23 | Mere | and we're on patch 1.7 and ppl have survived without it... |
| 02:00.41 | ZorbaTHut | get them to give you numbers with their gear equipped and unequipped, then you can calculate the constant and ratio. then just watch people actually casting spells and you can get a good estimate of the person's spellpower. from there, you should be able to predict approximate healing values |
| 02:00.42 | Mere | ah, you're assuming I know people in game |
| 02:00.53 | ZorbaTHut | you do! it's this channel. :D |
| 02:00.59 | Mere | heh |
| 02:01.13 | Mere | daft as it sounds, I'm enjoyign writing this addon, and not playing as much |
| 02:01.26 | Mere | my healer ain't even left 50, and I was annoyed with healing in rift ;) |
| 02:01.45 | Mere | I guess coming from WoW, vuhdo is damn amazing |
| 02:02.12 | ZorbaTHut | Seebs can tell you a lot about playing the game so much that you forget to level ;) |
| 02:02.39 | Mere | actually something I was gonig to ask, will we have the ability to render anything in frames? eg, come 1.8 we can work out someone is out of range, vuhdo has the feature to put up a direction arrow to someone out of range/dead |
| 02:02.47 | ZorbaTHut | yeah, WoW's got a whole lot of time put into raid frames, and it'll take a bit for Rift to catch up |
| 02:03.03 | Mere | from what I can tell it's some kind of 3d rendered .blp file I think |
| 02:03.17 | Mere | indeed, wow has had years to get frames |
| 02:03.24 | Mere | and for them to mature |
| 02:03.24 | ZorbaTHut | if you mean a 3d object in a frame, the answer is probably no. But what a lot of the WoW addons did was keep a big database of various arrows, then just show the most appropriate one |
| 02:03.42 | ZorbaTHut | 3d rendered stuff *might* show up but that's another huge "don't hold your breath" thing. |
| 02:04.04 | Mere | that's fair enough, I know it can be handy, but also easy to abuse (eg avr) |
| 02:04.26 | seebs | The ability to display models made for awesome cursor addons. |
| 02:04.33 | Mere | I guess I'm looking at what the limits are, so I know where to go, this is as far as we can go |
| 02:04.38 | seebs | Ohh, hey. I should maybe make a cursor-identifying addon, since I want one. |
| 02:04.58 | seebs | BTW, Zorba, check out the new Baggotry. Finally does something people in general will care about: Tooltips showing inventory across alts. |
| 02:05.15 | ZorbaTHut | yeah, it's not really "abuse" - there's many ways I can avoid that - it's just that the 2d rendering layer we're using makes it quite difficult to do that |
| 02:05.20 | ZorbaTHut | seebs, oh nice, I'll check it out :) |
| 02:05.36 | seebs | It also has autostacking, which is really handy. |
| 02:05.48 | Mere | actually on the minor request list, someone else asked about having a /cast [nomod] abiity, is that something to harass you about, or is it on the list? |
| 02:06.00 | seebs | In general, /bag -S 0 <something that I can figure out what to do with> will try to stack items to max stack size, -S N will try to stack to a fixed size. |
| 02:06.11 | seebs | Oh, I would love /cast [nomod] |
| 02:07.03 | Mere | I suspect that it's macro engine stuff? same as no-one answered my /cast @u64bitUnitId :) |
| 02:07.12 | ZorbaTHut | honestly it's not me but I should probably go ping someone about it |
| 02:07.16 | seebs | And FWIW, even in the case with no comparisons, the bounds reported for ui.native.tooltip are meaningless, which makes it hard to decide where to place a tooltip. |
| 02:07.34 | ZorbaTHut | seebs, I think it worked once but for some reason it's now glitchy. It's on the TODO list |
| 02:08.19 | seebs | Cool. |
| 02:08.24 | Mere | zorba: thanks, be nice to not have odd things going on with macros, especially as raid frames are likely to push some of these odd macro things :) |
| 02:08.32 | seebs | Today had AWESOME forum rage. |
| 02:09.04 | seebs | Trion's worst outage ever, by far, is a little under two thirds the length of typical patch days in other MMOs. Forums are full of people who think Trion reps should be killed for this. |
| 02:09.29 | Mere | there was an outage? |
| 02:09.42 | seebs | Yeah, servers were down for 4-5 hours in effect. |
| 02:09.58 | seebs | Lots of horrible lag and weirdness, connection problems, I think some stuff actually got rolled back, and so on. |
| 02:10.12 | Mere | ah |
| 02:10.19 | Mere | didn't logon till about 3-4 hours ago |
| 02:10.43 | seebs | That's about when it got fixed. It was up and down and acting weird for a few hours before that. |
| 02:10.48 | ZorbaTHut | fun fact, I was the first one to report the servers getting slow. I happened to be logged in during that time :V |
| 02:10.55 | Mere | I guess it's the "unscheduled" thing that annoys ppl |
| 02:11.41 | seebs | Mostly it's the thing where they've never run a production service in their lives and assume that the problem is lazy people who aren't bothering to push the button labeled "work now". |
| 02:11.53 | Mere | indeed |
| 02:11.54 | seebs | We had one person saying they should just go live with the quirks and wait until Monday to "fix" it. |
| 02:12.12 | seebs | Because no one would mind random 4-second or longer lag during play, or random character deletions. |
| 02:12.37 | seebs | Someone on forums reported a couple of people claiming they got unexpected new characters on their accounts. |
| 02:12.39 | Mere | tbh in any complex system a tweak somewhere could break so many things |
| 02:13.05 | Mere | I remember an ISP that accidently deleted their mail server SAN storage... |
| 02:13.24 | seebs | ohh, nevermind, that's unrelated. |
| 02:13.38 | Mere | they were installing a new one, and were initializing, except they had looked at the old one to see the settings, but didn't switch back to the new one :) |
| 02:13.47 | seebs | Apparently it happened to people who played early on, left, got banned, and came back. So probably RMT people. |
| 02:13.49 | ZorbaTHut | ouch :) |
| 02:13.55 | ZorbaTHut | hope they had good backups |
| 02:14.07 | Mere | they didn't have compelete backups |
| 02:14.15 | Mere | they sent the storage for recovery and got some of it back |
| 02:14.18 | Mere | but not all of it |
| 02:14.22 | seebs | I got to add a paragraph to a user guide to clarify that the procedure for adding a new disk was for adding a factory-new disk, not for moving a disk with existing data on it to a new system. |
| 02:14.43 | Mere | heh |
| 02:14.53 | Mere | but yeah, computers are damn complex |
| 02:15.33 | Mere | it's human to err, it takes a computer to really **** it up :) |
| 02:19.19 | seebs | BTW, Zorba, although I ended up concluding that SetPoint was right, it was confusing that SetPoint("TOP", ..., 0, -3) got a usage error. (I want nil, -3). |
| 02:20.08 | ZorbaTHut | yeah, there's a bunch of places where I want to change "usage error" into something more human-friendly |
| 02:20.30 | Mere | seebs: should you be interested, there's a 0.1.1 beta on riftui for the healing frames, you can actually configure spells and colours (colours need a reloadui to apply right now) |
| 02:21.41 | seebs | Neat! |
| 02:21.58 | seebs | In general, though, this is staying pretty pleasant to work with. |
| 02:25.07 | seebs | Huh. |
| 02:25.10 | seebs | I have just had a crazy thought. |
| 02:25.24 | seebs | Which is: What if I didn't display those tooltips for items that don't have a .stackMax. |
| 02:25.35 | seebs | Because everything I can think of that doesn't have a .stackMax, no one cares about. |
| 02:27.00 | seebs | Things like the pre-order runes, equipment... You basically never care whether alts have them. It's just materials and the like where that matters. |
| 02:27.23 | seebs | Hmm, wait, no, planar essences you might care about. |
| 02:30.08 | Mere | hmm, is there a way to copy and paste a set of commands into rift? or do I have to put them in a macro and run the macro? |
| 02:30.28 | seebs | (going AFK to sit near spouse) |
| 02:53.29 | Mere | going to bed, gn all |
| 08:16.42 | seebs | Huh, interesting. |
| 08:17.07 | seebs | If you have self-cast set to alt, then alt-3 self-casts the spell bound to 3, but alt-shift-3 does NOT try to cast, at all, the spell bound to shift-3. |
| 08:17.54 | seebs | But it does work for alt-ctrl-3. |
| 08:24.54 | *** join/#riftuidev Kreiri (~Miranda@109.227.122.220) |
| 09:05.12 | *** join/#riftuidev Sunspots (~Sunspots@c213-89-195-33.bredband.comhem.se) |
| 09:31.01 | Kreiri | Cairenn, is there a way to receive e-mail notifications for bugs/features on author's portal? |
| 09:31.27 | Cairenn | not at present. RSS feed for it though, iirc |
| 14:30.38 | *** join/#riftuidev Mere (~Mere@buzzbee.plus.com) |
| 15:43.01 | *** join/#riftuidev vect0r (~vect0r@74.117.56.153) |
| 15:43.16 | *** join/#riftuidev vect0r (~vect0r@unaffiliated/vect0r) |
| 16:01.38 | *** join/#riftuidev Mere (~Mere@buzzbee.plus.com) |
| 17:30.28 | seebs | I have realized that I could write an alt-tracking addon. Also I have realized that I could call it Alternia and all the Homestuck fans would follow me around typing incomprehensibly. |
| 19:29.47 | seebs | Burrowing... ham? |
| 19:29.52 | seebs | These devs, they are crazy. |
| 19:30.21 | starlon | tilts. |
| 19:52.53 | *** join/#riftuidev Sunspots (~Sunspots@c213-89-195-33.bredband.comhem.se) |
| 20:40.03 | seebs | Talk to the visiting scholar near the gates of Quicksilver College. |
| 20:40.16 | seebs | He says that Pipwhistle believes that the common barkmouse is a cousin of the burrowing ham. |
| 21:58.58 | *** join/#riftuidev Imhothar (5d68a204@gateway/web/freenode/ip.93.104.162.4) |
| 22:02.36 | *** join/#riftuidev Imhothar_ (5d68a204@gateway/web/freenode/ip.93.104.162.4) |
| 22:08.32 | seebs | ZorbaTHut: I am told that MS has hooks to let apps disable the "windows key opens a menu" behavior. Why has Rift not done this wondrous thing? |
| 23:05.38 | *** join/#riftuidev Hyperi_ (hyperi@hyperi.fi) |
| 23:38.55 | Kreiri | I have a problem with quest IDs. XML files have them as long strings of dec numbers, game wants long strings of hex numbers. |
| 23:39.48 | Kreiri | Well, these hex strings are too long for tonumber(n, 16) |
| 23:42.54 | seebs | hmm |
| 23:43.18 | seebs | That is sorta tricky, yes. Hmm. |
| 23:43.40 | seebs | bc is your friend. |
| 23:43.57 | seebs | You could do a one-time conversion in another language, I suppose. |
| 23:44.33 | seebs | Really, though, you just need to get Zorba and whoever did the XML export in a room together with their left wrists tied together and give them sporks. They'll sort it out. |
| 23:46.27 | Kreiri | I just tried to call quest.detal for qXXXXXXXX, where XXXXXXXX - an id from xml file, converted to hex string. Got nothing. quest.detail also chokes on incomplete quest ids, returned by quest.complete. |
| 23:47.17 | Imhothar_ | the ingame qust ids are way larger than the ids in the cml |
| 23:47.40 | Kreiri | they are twice as wide |
| 23:47.55 | seebs | I think Zorba commented on that or something like it already -- there was something getting extended in bit width, and dropping the FFFFs helped. Or maybe that was a different one. |
| 23:49.09 | Imhothar_ | Think that was item types |
| 23:49.14 | Kreiri | When I get ann id like q44BB7EB4xxxxxxxx from inspect.quest.complete, inspect.quest.detail retuns nothing for this id, and nothing for q44BB7EB4 |
| 23:49.33 | ZorbaTHut | here, I'll just explain it, you'll get it yourself eventually |
| 23:49.56 | ZorbaTHut | the first chunk is the quest's ID. the second chunk is a server-provided quest checksum which is required for retrieve information on the quest |
| 23:50.31 | ZorbaTHut | the first chunk is what's going to match with the XML files. The second chunk you can't acquire anywhere unless the addon system deigns to give it to you. So, yes, there's no way to get quest details given what inspect.quest.complete() provides. |
| 23:51.59 | Kreiri | I wrote this script which loads xml file, parses it and writes resulting lua table to file. Right now it's an array of quests; I can make it a dictionary with quest ids as keys, no problem. Only it's 22 MB in size... |
| 23:52.46 | Kreiri | can we load files on demand? |
| 23:53.19 | Kreiri | if yes, I could split it into several files by, say, level range and faction. |
| 23:54.45 | Imhothar_ | the question is, do you really need it |
| 23:57.13 | Imhothar_ | Inspect.Quest.Detail() works for quests you currently have in your log. Saving that you can use the data after receiving Event.Quest.Complete to get the info for the quest qhich was completed |
| 23:58.10 | seebs | Hmm. |
| 23:58.14 | Kreiri | Which does not solve problem of loading info of already completed quests. |
| 23:58.24 | Imhothar_ | true |
| 23:58.39 | Imhothar_ | but loading a 22 MB table doesn't either |
| 23:58.51 | seebs | So, what is the rationale for the magic cookies? Obviously, we could in principle stash the Quest.Detail() results for everything, share them, and end up with All The Stuff. |
| 23:59.14 | seebs | So there is presumably some significant benefit to not allowing Inspect.Quest.Detail() to work on the partial numbers. |
| 23:59.27 | ZorbaTHut | It makes it much harder to do database mining. :) |
| 23:59.55 | ZorbaTHut | no way to dynamically load files yet, although it's something I absolutely want to add |