| 00:18.48 | *** join/#wowuidev Andols (~yaaic@m77-218-252-170.cust.tele2.se) |
| 00:21.05 | *** join/#wowuidev ShadniX (dagger@p5B3D2410.dip0.t-ipconnect.de) |
| 00:22.11 | foxlit | AGILITY_TOOLTIP = "Increases all charactersâ chance to Critical Hit. The|namount of the bonus is higher per point on Rogues than|nother classes. Agility affects dodge. Rogues get more|nDodge per point of Agility than other classes. Agility|nadds directly to Armor."; |
| 00:22.17 | foxlit | ROGUE_AGILITY_TOOLTIP = "Increases attack power with both melee and ranged weapons, and improves the chance to score a critical hit with all weapons.|nIncreases armor and chance to dodge attacks."; |
| 00:25.34 | *** join/#wowuidev ShadniX (dagger@p5B3D2410.dip0.t-ipconnect.de) |
| 00:26.38 | foxlit | (they're probably unused; the inversion in the amount of attention to Rogues is just weird) |
| 00:31.34 | *** join/#wowuidev Andols2 (~yaaic@m77-218-240-68.cust.tele2.se) |
| 00:44.17 | *** join/#wowuidev Andols (~yaaic@m83-185-94-229.cust.tele2.se) |
| 01:08.55 | *** join/#wowuidev Andols (~yaaic@m83-185-247-8.cust.tele2.se) |
| 01:18.41 | *** join/#wowuidev purl (ibot@rikers.org) |
| 01:18.41 | *** topic/#wowuidev is Happy Holidays! || 7.3.x ToC: 70300 || http://pastebin.com || Battletags: http://goo.gl/2bm7n || #WoWUIDev-commits: CIA & JIRA bots reporting || This channel is logged, via purl |
| 01:22.39 | *** join/#wowuidev Andols (~yaaic@m83-185-95-181.cust.tele2.se) |
| 01:28.44 | Semlar | does anyone know how to check if you're in an area where your pvp talents are active |
| 01:32.08 | *** join/#wowuidev Seerah (~umsin@2601:3c2:101:320:64f9:56d6:4219:6d7f) |
| 01:32.08 | *** mode/#wowuidev [+o Seerah] by ChanServ |
| 01:34.14 | *** join/#wowuidev A-KO (sa@2600:8802:3600:3640:e0d2:eda0:1992:62dd) |
| 01:38.37 | nebula | Semlar: GetZonePVPInfo() == arena/combat maybe? |
| 01:39.26 | Semlar | i'm not actually sure what conditions set your pvp talents to active, because i know entering a pvp world quest area will do it |
| 01:39.40 | Semlar | i assume dueling someone |
| 01:39.52 | nebula | i don't think dueling does |
| 01:48.07 | *** join/#wowuidev Andols (~yaaic@m77-218-240-162.cust.tele2.se) |
| 01:49.41 | Semlar | https://wow.gamepedia.com/Honor_talent this implies that they're active in an arena, battleground, or world pvp |
| 01:50.21 | Semlar | so if they spontaneously activate on getting attacked in the world, how would you detect that |
| 01:51.09 | Semlar | aside from actually watching if you're gotten attacked by another player |
| 01:51.59 | *** join/#wowuidev Funkeh` (~Funkeh`@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh) |
| 01:52.55 | Mr_Rabies | i have a feeling they'll change the activation in bfa since world pvp will only be possible when flagged |
| 01:53.14 | Semlar | i guess i could watch the spell book for one of your pvp talents |
| 01:53.34 | *** join/#wowuidev ShadniX (dagger@p5B3D2410.dip0.t-ipconnect.de) |
| 01:53.37 | *** join/#wowuidev Iriel (~Iriel@173-8-135-209-SFBA.hfc.comcastbusiness.net) |
| 01:53.37 | *** mode/#wowuidev [+o Iriel] by ChanServ |
| 01:53.51 | Semlar | i would think there'd be a function or event for this somewhere |
| 01:56.22 | nebula | i dunno, i remember blizzard talking about not want stats to jump around activating pvp templates for random world pvp/dueling |
| 01:56.28 | nebula | wanting* |
| 01:56.42 | nebula | i assume the talents are tied to the stat templates |
| 01:58.22 | Semlar | i'm almost positive they activate when you're in a pvp world quest area, without scaling your stats |
| 01:59.43 | Semlar | i don't play on a pvp server, so outside of that i have no idea |
| 02:00.44 | *** join/#wowuidev Andols2 (~yaaic@m83-185-84-23.cust.tele2.se) |
| 02:02.43 | Semlar | do you know how to check when you can change talents outside of a rested area? |
| 02:03.02 | Semlar | or do you have to make a list of every buff that lets you change talents |
| 02:03.24 | nebula | probably the latter :p |
| 02:03.35 | Semlar | i know there's the tome and the tree thing |
| 02:03.43 | Semlar | i don't know what else there is |
| 02:14.21 | Mr_Rabies | you could maybe scan talent tooltips or is that a nono in combat |
| 02:18.32 | Mr_Rabies | Talents have red text that says "Talents can only be changed in a Rested XP area." |
| 02:18.34 | Mr_Rabies | if you can't |
| 02:27.32 | *** join/#wowuidev Andols (~yaaic@m77-218-249-101.cust.tele2.se) |
| 02:41.45 | *** join/#wowuidev Andols (~yaaic@m77-218-244-66.cust.tele2.se) |
| 02:50.59 | *** join/#wowuidev Andols2 (~yaaic@m77-218-252-156.cust.tele2.se) |
| 03:01.27 | *** join/#wowuidev Elmoe_ (~Elmoe@24.18.143.238) |
| 03:03.08 | *** part/#wowuidev Seerah (~umsin@2601:3c2:101:320:64f9:56d6:4219:6d7f) |
| 03:05.58 | *** join/#wowuidev Elmoe (~Elmoe@24.18.143.238) |
| 03:12.17 | *** join/#wowuidev Elmoe_ (~Elmoe@24.18.143.238) |
| 03:16.49 | *** join/#wowuidev Andols (~yaaic@m77-218-242-187.cust.tele2.se) |
| 03:17.38 | *** join/#wowuidev Elmoe (~Elmoe@24.18.143.238) |
| 03:42.37 | *** join/#wowuidev Andols (~yaaic@m77-218-246-25.cust.tele2.se) |
| 04:05.54 | *** join/#wowuidev Andols (~yaaic@m77-218-246-13.cust.tele2.se) |
| 04:31.11 | *** join/#wowuidev Andols (~yaaic@m77-218-246-127.cust.tele2.se) |
| 04:57.20 | *** join/#wowuidev Ermad (~Ermad@pool-98-113-144-112.nycmny.fios.verizon.net) |
| 04:57.34 | *** join/#wowuidev Andols (~yaaic@m77-218-249-221.cust.tele2.se) |
| 05:22.58 | *** join/#wowuidev Andols (~yaaic@m77-218-243-75.cust.tele2.se) |
| 05:26.31 | *** join/#wowuidev pompy (~Mike@c-73-194-183-217.hsd1.nj.comcast.net) |
| 05:34.13 | *** join/#wowuidev pompy (~Mike@c-73-194-183-217.hsd1.nj.comcast.net) |
| 05:34.59 | *** join/#wowuidev ShadniX (dagger@p5B3D24B5.dip0.t-ipconnect.de) |
| 05:45.35 | *** join/#wowuidev ShadniX (dagger@p5B3D24B5.dip0.t-ipconnect.de) |
| 05:48.02 | *** join/#wowuidev Andols (~yaaic@m77-218-246-46.cust.tele2.se) |
| 06:11.46 | *** join/#wowuidev Andols (~yaaic@m77-218-243-178.cust.tele2.se) |
| 06:26.04 | *** join/#wowuidev Ssateneth (ident@unaffiliated/ssateneth) |
| 07:15.34 | *** join/#wowuidev mitch0 (~mitch@85-238-71-175.pool.digikabel.hu) |
| 07:57.39 | *** join/#wowuidev Stanzilla (~Stanzilla@WoWUIDev/WowAce/Stanzilla) |
| 08:03.22 | *** join/#wowuidev VIad (~Vlad@cm-84.213.146.58.getinternet.no) |
| 08:23.51 | VIad | Funkeh`: imho when you enter the raid or the sub-zone of the next active boss, should definetly like pre-load stuff in memory, especially when you got tons of it! :P |
| 08:32.53 | VIad | foxlit: training dummy related, afflication warlocks use them to dot up with permadots and become self heal raidbosses :P |
| 08:33.20 | VIad | they apparently got some permadot, put it on the 70 dummy and walk around the city having fun |
| 09:24.30 | *** join/#wowuidev Varyak (~Varyak@host-188-174-235-114.customer.m-online.net) |
| 10:19.31 | *** join/#wowuidev elvador (~elvador@053-140-067-156.ip-addr.inexio.net) |
| 10:19.53 | *** join/#wowuidev elvador (~elvador@WoWUIDev/WowAce/Elvador) |
| 10:32.23 | *** join/#wowuidev stolenlegacy (~stolenleg@unaffiliated/stolenlegacy) |
| 11:18.59 | nevcairiel | its not exactly raid breaking because what can you do in the city |
| 11:19.04 | nevcairiel | once you get too far away it just stops |
| 11:32.38 | VIad | I was told my warlocks using this the debuff ticks even if out of range and heals |
| 11:33.09 | VIad | I got a warlock, should try it out myself I guess :o |
| 11:33.18 | nevcairiel | out of range is up for definition, sure you can walk around the city, but you cant like travel to the other side of the world |
| 11:33.45 | nevcairiel | once you leave the visibility range it should stop |
| 11:34.59 | VIad | ah ofc :D |
| 11:35.16 | VIad | dunno, there is weirdness in legion man |
| 11:35.32 | VIad | DH darkness cloud on the floor interrupts eyebeam when it procs and absorbs damage :D |
| 11:35.51 | VIad | it's like as if link totem would break your casts because it rebalances your health up or down ;D |
| 11:35.55 | nevcairiel | friends dont let friends play DHs |
| 11:35.58 | VIad | heh |
| 11:36.39 | VIad | when I kill stuff and mount up I keep getting spam "Can't do that while mounted." because one of my healing orbs tries to heal me but I can't because it's actually me casting my own orb heal spell or something... |
| 11:37.58 | nevcairiel | just be happy it doesnt dismount you to heal |
| 11:38.41 | *** join/#wowuidev brykrys (~Bryan@WoWUIDev/Norganna/Dev/Brykrys) |
| 11:41.29 | *** join/#wowuidev zenzelezz (~zenzelezz@cm-84.210.76.194.getinternet.no) |
| 11:43.56 | VIad | :D |
| 11:54.30 | Varyak | "You are in a shapeshifted form" then bloddy shift me out and let me do that thing |
| 11:55.15 | Varyak | every spec besides resto requires you to shapeshift? Though luck |
| 11:56.59 | nevcairiel | most things actually cancel the spec automatically |
| 11:57.09 | nevcairiel | or maybe there was an option do allow that? |
| 11:58.25 | ThiefMaster | resto doesn't shift to a tree anymore? |
| 11:58.45 | nevcairiel | not for years and years |
| 11:59.00 | nevcairiel | i mean, they can, but its purely cosmetic |
| 11:59.01 | VIad | incarnation does, it's like a powerful CD |
| 12:00.16 | *** join/#wowuidev StephenLynx (~Stephen@186.244.245.87) |
| 12:02.21 | nevcairiel | the permanent form lasted through BC and WotLK only, as of cataclysm we have the CD version |
| 12:03.11 | StephenLynx | i used to love to be a walking broccoli |
| 12:03.42 | nevcairiel | and as of MoP we have a glyph to restore the permanent tree form, except without any combat bonus |
| 12:04.31 | VIad | as nev said, you can wazzle dazzle as a tree but no benefits. can't be sheeped I guess. :D |
| 12:04.53 | nevcairiel | not sure if its still poly immune |
| 12:05.01 | nevcairiel | i havent pved as resto druid for years |
| 12:05.02 | VIad | unless its like a fake shapeshift, does appear on the shapeshift bars :d |
| 12:05.21 | VIad | incarnation is the new one with combat benefits |
| 12:05.28 | VIad | and it looks like a ancient of wisdom |
| 12:05.33 | VIad | tbh needs a model update |
| 12:05.37 | VIad | but still not too shaby |
| 12:06.45 | VIad | +15% healing and buffs rejuv, wild growth, regrowth and entangling roots |
| 12:07.01 | VIad | +120% armor for the tanky situations |
| 12:07.34 | VIad | wild growth +2 targets, regrowth is instant cast, rejuv +50% healing and -30% mana cost and instant entangling roots :d |
| 12:58.09 | Gnarfoz | anyone know if Blizzard fixed the raid FPS issues? |
| 12:59.51 | nevcairiel | they did |
| 13:01.57 | Stanzilla | only partially but it is better |
| 13:02.16 | nevcairiel | i have 100+ fps in lfr, so shrug |
| 13:03.22 | Gnarfoz | guess I'll have to wait and see until tomorrow ^^ |
| 13:15.09 | [TOM_RUS] | fps still crap compared to pre 7.3.5 |
| 13:15.18 | *** join/#wowuidev Borkr (~Borkr@mail.seaonics.com) |
| 13:19.36 | AcidWeb | sound issues are not fixed yet too |
| 13:22.41 | Gnarfoz | that's more along the lines of my expectations ^^ |
| 13:51.33 | Funkeh` | TheDanW: any word on fixing the sound issues? Would rather not have to force on the dialog channel and use that. |
| 13:59.22 | Simca | both friends i have with serious fps issues since patch still have it, so i don't think it's fixed |
| 14:00.46 | StephenLynx | is this a thing or blizzard just screwed up 7.3.5 specifically? |
| 14:01.03 | StephenLynx | before I stopped playing things were pretty stable. |
| 14:01.20 | Simca | it's a 7.3.5 issue |
| 14:01.23 | *** join/#wowuidev Funkeh` (~Funkeh`@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh) |
| 14:01.53 | StephenLynx | and in general? is it common for them to release botched patches? |
| 14:04.42 | Constie | Well, if you ask the right people they'll tell you every patch made the game worse. |
| 14:05.32 | Gnarfoz | this is a rather big outlier |
| 14:06.00 | Simca | i'm not going to throw too much shade at them but i think this has been a bit of a trend recently. |
| 14:06.13 | Funkeh` | never like this |
| 14:06.18 | Semlar | it's mostly because they tried to change how many slots were in the backpack |
| 14:06.33 | nevcairiel | yeah they warned us about that |
| 14:06.53 | Funkeh` | yeah, that's totally the issue lol |
| 14:06.55 | Simca | in 7.2, they broke 4 different raid bosses in the current tier (Nighthold) that should have been caught by their playtesting |
| 14:06.57 | Chipzz | how does adding 4 extra slots to the backpack break FPS? :P |
| 14:07.16 | Chipzz | or sound? :P |
| 14:07.16 | StephenLynx | Chipzz, if you knew how bad is general game development |
| 14:07.18 | Semlar | look, the backpack is complicated |
| 14:07.36 | StephenLynx | most of the time developers just hack it enough to work |
| 14:07.37 | Semlar | and it's integral to the whole system |
| 14:07.44 | Simca | semlar stop trolling people |
| 14:07.54 | Funkeh` | It doesn't their obsession with staying under 2GB at the cost of the majority of their playerbase is what breaks fps |
| 14:08.04 | Constie | Many important physics calculations assume you have exactly 16 slots. |
| 14:08.21 | Funkeh` | and 16fps! |
| 14:08.32 | *** join/#wowuidev ikcerog (~ikcerog@12.165.188.174) |
| 14:08.40 | StephenLynx | what do you mean staying under 2gb? you mean RAM? |
| 14:08.51 | Funkeh` | no, hdd usage |
| 14:09.00 | Simca | that's also a troll, he does mean RAM |
| 14:09.05 | Funkeh` | D: |
| 14:09.05 | StephenLynx | kek |
| 14:09.11 | Chipzz | StephenLynx: I have to disagree with that. read a lot of the FrameXML code during 7.x, and it definitely looks like they're cleaning up things and not just hacking |
| 14:09.26 | StephenLynx | cleaning up after 14 fucking years |
| 14:09.37 | StephenLynx | imagine how much technical debt might be |
| 14:09.58 | Funkeh` | game is unplayable form some people but at leas tthe XML is cleaner |
| 14:10.01 | StephenLynx | did you know diablo was meant to be a turn based game? |
| 14:10.14 | StephenLynx | and they changed it to be real time really late in development? |
| 14:10.24 | Chipzz | Funkeh`: well it's the XML with the backpack that's causing all the issues, rite? :) |
| 14:10.26 | StephenLynx | and that they didn't use a depth buffer for it? |
| 14:10.34 | Funkeh` | i'm not sure what you mean after 14 years, every patch features a cleanup of some form |
| 14:10.41 | foxlit | honestly, it's been less than a week. The fact that you expect them to fix everything in like 3 working days is a testament to how well-run the overall operation normally is. |
| 14:10.42 | Funkeh` | it's a non-stop ongoing process |
| 14:11.06 | StephenLynx | my point is that there is cleaning to be done in the first place. |
| 14:11.11 | Simca | foxlit: to be fair, most of this was reported well in advance of it going live. |
| 14:11.49 | StephenLynx | plus, they can't excuse it on a lack of resources. |
| 14:11.49 | Funkeh` | the game is massive, it probably will never stop needing "cleaned up" |
| 14:11.49 | Semlar | are you new to programming stephenlynx |
| 14:12.09 | StephenLynx | no, but I don't tolerate amateurism on it. |
| 14:12.24 | Semlar | but you don't work like, as a professional programmer |
| 14:12.28 | StephenLynx | i do. |
| 14:12.31 | Chipzz | StephenLynx: some of the cleaning isn't bad code per se, but more like "old coding style conventions" |
| 14:12.31 | StephenLynx | did all my life. |
| 14:12.55 | Chipzz | like using name="$parentFoo" instead of parentKey="Foo" |
| 14:12.56 | Semlar | so you just write perfect code that never needs to be fixed |
| 14:13.03 | Semlar | and you do that in a reasonable time frame |
| 14:13.09 | Chipzz | not necessarily a bug, but less global vars so cleaner code |
| 14:13.24 | StephenLynx | no, but I follow strict methodology when I work and nothing I have to fix is a mistery. |
| 14:13.45 | StephenLynx | and I don't get weird shit breaking all over the place over some unrelated change. |
| 14:13.59 | foxlit | but less global vars so cleaner code << meh, there's no benefit. |
| 14:14.01 | nevcairiel | then you havent worked on a big enough and old enough project ever before |
| 14:14.01 | Semlar | you have any projects that are 14 years old? |
| 14:14.10 | Simca | the code base of WoW has been getting significantly better over time. they've done a lot for it. |
| 14:14.21 | nevcairiel | with probably thousands of developers over the better part of two decades, stuff adds up |
| 14:14.35 | Funkeh` | it's not so much the age itself as it is the size imo, the age just compounds the problem |
| 14:14.46 | Chipzz | foxlit: conceptually there is. also I see lot of 'local APISomeFunc = SomeFunc' being used in addons due to massive amount of globals |
| 14:15.21 | Chipzz | although they would need to fix translation strings to not be global in order to have a real effect I guess |
| 14:15.23 | Funkeh` | eh? |
| 14:15.38 | Semlar | what are you talking about |
| 14:15.41 | Funkeh` | :s |
| 14:16.04 | Semlar | you want them to localize all of their globals? |
| 14:16.29 | Simca | the 2 major issues in 7.3.5 (FPS is awful for some and sound cuts in and out) are the result of 2 distinct changes (as far as i know): their new DB optimizations (WDC1) effectively doubling the amount of calls to access compressed data (for FPS issues) and an unintentional change to sound channel priorities (for sound issues) |
| 14:16.44 | Chipzz | Semlar: no |
| 14:17.01 | Simca | so it's not a ridiculous mystery due to a cryptic code base or anything (again, as far as i know) |
| 14:17.10 | StephenLynx | what if they made thei client FOSS? :thinking_emoji: |
| 14:17.10 | nevcairiel | globals by themself arent really that bad, there isnt exactly another option to pass API to addons |
| 14:17.20 | Funkeh` | Simca: I'm not so sure it is unintentional, that's what worries me, since they've known about it since patch day and it wasn't in any of the client fixes |
| 14:17.35 | StephenLynx | nevcairiel, you can use namespaces. |
| 14:17.43 | nevcairiel | just because its high priority for you Funkeh` doesnt mean its for them |
| 14:18.00 | Chipzz | nevcairiel: how about a lot of the API moving in to C_*.* ? |
| 14:18.03 | nevcairiel | StephenLynx: global access is basically just a table lookup, so are namespaces .. so, performance wise its the same shit |
| 14:18.16 | StephenLynx | of course. |
| 14:18.26 | StephenLynx | but it makes your code more organized and helps avoiding conflicts. |
| 14:18.31 | Chipzz | nevcairiel: except the sheer number of globals makes the table lookup go slower |
| 14:18.37 | Funkeh` | nevcairiel: they spent the time to identify the issue, but not resolve it? wat |
| 14:18.47 | foxlit | it's a bloody hash table. |
| 14:18.49 | Semlar | why would more globals make the lookup slower |
| 14:18.56 | Semlar | that's the whole point of an index |
| 14:19.15 | Simca | i think you guys are missing the point a little on code cleanup/organization. all of your examples on both sides are Lua/XML things, and the UI is not what is causing problems for the game. In fact, if I had to guess, I'd guess the UI code is by far the cleanest part of the game's code. |
| 14:19.20 | StephenLynx | lets not forget 7.3.5 broke it for linux |
| 14:19.42 | Semlar | also global lookups aren't exactly expensive operations |
| 14:19.43 | StephenLynx | my desktop died before I could find a way to run it again |
| 14:19.46 | Simca | you're pointing at issues like sound and then giving Lua examples; Lua is not breaking the sound. |
| 14:20.11 | Semlar | no, the backpack is breaking the sound |
| 14:20.21 | Semlar | we've been over this |
| 14:20.30 | Funkeh` | small indie company |
| 14:20.48 | Semlar | the number of sound channels is directly tied to the number of slots in the default backpack |
| 14:20.57 | StephenLynx | Funkeh`, ikr |
| 14:20.58 | Simca | inversely, of course |
| 14:21.10 | StephenLynx | just a couple dudes hacking away on their garage |
| 14:21.14 | Constie | They're both multiples of 2. It makes sense. |
| 14:21.40 | Semlar | it was the one constant that held the entire game together |
| 14:21.55 | Funkeh` | consistency? in wow? |
| 14:22.14 | Semlar | now mobs can't walk up and down stairs |
| 14:22.26 | StephenLynx | wait |
| 14:22.29 | StephenLynx | is that for real? |
| 14:22.50 | Constie | Semlar always lies. |
| 14:22.51 | Chipzz | not present in the current version but I saw this for example: |
| 14:22.52 | Chipzz | /Applications/World of Warcraft/Interface/AddOns/Ovale/dist/BanditsGuile.lua:local API_GetTime = GetTime |
| 14:23.09 | Simca | they've fixed mobs walking up and down stairs mostly since patch day |
| 14:23.13 | Simca | but yes there were issues |
| 14:23.14 | Semlar | i did a mythic court of stars earlier and the disguised guy just got stuck on the stairs and stood around for a while |
| 14:23.18 | StephenLynx | jesus christ |
| 14:23.44 | Semlar | we couldn't even do the first boss in vault because the adds that he spawns were all evading |
| 14:23.55 | Simca | mobs colliding with things caused them to despawn or evade and then despawn for some reason |
| 14:23.56 | Semlar | but the ones that spawned in the fountain could still attack |
| 14:24.15 | Semlar | and they would just kill the group in seconds |
| 14:24.18 | Simca | Attunmen is still not fixed FYI. Court is, Vault is, but that isn't |
| 14:24.33 | Simca | did lower kara last night; he despawned when he collided with a wall during the horse charge sequence |
| 14:24.48 | Semlar | i need a list of all of the known mythic+ bugs i should be aware of |
| 14:25.07 | Funkeh` | there are no bugs if you don't use the 4 backpack slots |
| 14:25.20 | Semlar | i should take the authenticator off my account |
| 14:25.20 | Simca | the NL bug with Dargul's lava attack going through the crystal spikes is fixed |
| 14:25.40 | Simca | we did NL last night too; no issues |
| 14:25.40 | Funkeh` | also only group with non authenticator folk |
| 14:25.55 | Funkeh` | make an addon for re-create LFR for non bagslot folk |
| 14:26.02 | Chipzz | foxlit: so why do you think the above was done? Ovale is hardly the only addon I've seen this pattern in? |
| 14:26.02 | Funkeh` | LFG* even |
| 14:26.15 | Semlar | i'm going to call it GoQueue |
| 14:26.25 | nevcairiel | Chipzz: its done because many addon devs believe in micro-optimizations |
| 14:26.42 | nevcairiel | if you run that particular function from OnUpdate, there might be some small point to that, but shrug |
| 14:26.44 | Funkeh` | It saves a table lookup to _G |
| 14:26.47 | foxlit | Chipzz: people do silly things, and people like to copy silly things without actually thinking about whether they're necessary or not. |
| 14:27.13 | Semlar | to be fair, most addon authors aren't professional programmers |
| 14:27.18 | Simca | were you or were you not the person who coded data into ASCII to save vertical line space, foxlit |
| 14:27.25 | Chipzz | Funkeh`: I get that. the point is why is saving that table lookup needed in the first place? |
| 14:27.27 | Simca | just saying, in terms of unnecessary optimizations |
| 14:27.28 | Simca | lol |
| 14:27.37 | foxlit | it almost never is |
| 14:27.45 | Funkeh` | Chipzz: depends what you're doing with the call itself |
| 14:27.49 | nevcairiel | Chipzz: the point is that its not "needed", people like to do that anyway |
| 14:27.50 | Funkeh` | if you use it once, then no |
| 14:27.52 | Constie | I could maybe see someone doing that with the "API_" prefix to make it more clear that that's where the function is from, maybe. I usually see it done without that prefix, though. |
| 14:28.11 | Funkeh` | but it can also be used as a poor mans hook protection |
| 14:29.18 | Semlar | i'm going to start localizing api functions in my addon, but name them after different api functions |
| 14:29.34 | Simca | excellent |
| 14:29.43 | Chipzz | Funkeh`: that is of course assuming the function wasn't hooked before you got a reference to it in the first place :) |
| 14:29.53 | Funkeh` | "poor mans" |
| 14:29.55 | Simca | and change the names at different points in the code too |
| 14:30.06 | Semlar | yeah, gotta switch it up |
| 14:30.07 | foxlit | accessing globals and locals are both conceptually constant-time operations, and I have a hard time believing that there's any code where the the impact of this is significant -- Amdahl's law should more or less reduce the benefit to nothing unless your code does nothing but access globals. |
| 14:30.51 | Semlar | i made an addon once that tried to hook every function in _G to output debug information |
| 14:31.05 | Semlar | it didn't go very well |
| 14:31.23 | Simca | i only do it if the function is referenced in a perf-heavy loop, but yeah it's probably rarely ever needed there too |
| 14:31.35 | Funkeh` | something something secure something something taint |
| 14:32.15 | Funkeh` | Simca: I do it to every API to try get the client over 2GB to annoy blizz |
| 14:33.04 | Constie | I think you might be more annoyed than they are. |
| 14:33.05 | Alram | sonofabitch |
| 14:33.22 | Simca | i wonder how much space addons have to use to push the 32-bit client to crash |
| 14:33.23 | Chipzz | $ grep . Localization.lua -c |
| 14:33.23 | Chipzz | 15126 |
| 14:33.32 | Chipzz | and that's not even complete |
| 14:33.42 | Chipzz | 15K global variables containing strings |
| 14:33.59 | Chipzz | that might start to degrade the performance of a global variable lookup |
| 14:34.30 | foxlit | why? It's not like it has to go through all the globals to find yours. |
| 14:36.17 | Chipzz | no, but the performance of a hash table is still relative to its size. not O(n), O(log(n)) I think, but still |
| 14:36.29 | foxlit | you're wrong. |
| 14:36.54 | Chipzz | maybe I am. enlighten me |
| 14:36.56 | foxlit | hash tables are useful exactly because their (amortized) performance does not depend on their size. |
| 14:38.19 | Chipzz | assuming perfect distribution |
| 14:38.26 | Chipzz | https://stackoverflow.com/questions/12020984/hash-table-why-is-it-faster-than-arrays |
| 14:38.39 | Simca | hash table lookup is amortized O(1) |
| 14:39.15 | Chipzz | performance is O(1) for perfect case but O(n) with collisions |
| 14:39.39 | Semlar | :thinking: |
| 14:39.52 | Simca | i don't know of any that would even allow collisions |
| 14:40.27 | Simca | i think the std c++ version throws an exception, and the C# version just silently rejects the data |
| 14:41.12 | foxlit | Chipzz: you solve this by forcing a rehash long before you get to an O(n) chain. |
| 14:41.27 | nevcairiel | indeed, a hash table optimized for read performance just re-hashes when needed |
| 14:41.39 | Chipzz | foxlit: right. that is assuming Blizzard (or lua internally) does that |
| 14:41.57 | Semlar | how likely do you think a hash collision is |
| 14:42.13 | nevcairiel | that depends how big the hash is and how many elements you put in there |
| 14:42.16 | nevcairiel | eventually they happen |
| 14:42.30 | Chipzz | given that all translation strings are all uppercase, more likely than your average hash |
| 14:42.49 | nevcairiel | if a proper hash is being used, that shouldnt matter |
| 14:42.56 | Chipzz | *translation variable names |
| 14:42.59 | Simca | CASC is based on MD5 signatures; if they ever have a collision in MD5 there; it may break the entire system |
| 14:43.10 | Simca | it would be amusing to see that try to be mitigated |
| 14:43.29 | nevcairiel | md5 collisions in real data arent that common, its just when you try to attack it that its easy to cause |
| 14:43.36 | Simca | yeah i know |
| 14:43.48 | Simca | it's not very likely but it would be so damaging if it did happen |
| 14:43.49 | Semlar | not that common is an understatement |
| 14:44.48 | Simca | the jenkins name hash they use would have some issues as well with a collision (and it has half the complexity of MD5) |
| 14:44.51 | Semlar | the odds of an unintentional hash collision are astronomically low |
| 14:45.32 | Semlar | that's why we use hashes |
| 14:46.06 | Alram | didnt you manage to get one of those to collide Simca |
| 14:46.08 | Alram | while bruteforcing |
| 14:46.24 | Alram | i dont remember if that was an actual collision or not |
| 14:46.44 | Simca | yeah several times but only one of those times was 'legit' - the first cases i did were not forcibly uppercase, which blizzard does before using jenkins hash |
| 14:47.16 | Alram | its a good thing that was for a file we already knew i think |
| 14:47.21 | Simca | but yeah i accidentally found one name hash collision that could actually occur |
| 14:47.21 | Alram | otherwise that would totally end up in listfiles |
| 14:47.32 | Simca | yeah, i could totally submit it |
| 14:47.33 | Semlar | stop hacking simca |
| 14:47.38 | Simca | i wonder what your server would do |
| 14:47.42 | Alram | it would accept it |
| 14:47.49 | Alram | if it was an unknown file beforehand, that is |
| 14:47.49 | Simca | :kappa: |
| 14:47.55 | Alram | seeing it wasnt it didnt but still :D |
| 14:57.21 | foxlit | Simca: [...] and the C# version just silently rejects the data << this also sounds extremely wrong |
| 14:57.51 | Simca | i'm pretty sure it's right but i can check |
| 14:58.29 | Simca | ah i am wrong |
| 14:58.35 | Simca | it does also throw an exception |
| 14:58.53 | foxlit | the equivalent scenario is two objects that are not equal but have the same hashcode. |
| 15:00.28 | foxlit | and the sane behavior for Dictionary would be to accept both as distinct keys |
| 15:02.18 | foxlit | (maybe it throws if you try to Add the *same* object twice? That'd have more to do with a perceived logic error in your code than any issue with hash collisions, though.) |
| 15:04.37 | Simca | if they have the same hashcode; it'll throw |
| 15:04.41 | Simca | the hashcode is the key |
| 15:05.56 | Simca | for hashtables anyway |
| 15:06.13 | Simca | i don't know about hashsets |
| 15:06.29 | Simca | those appear different but MSDN doesn't say how |
| 15:13.59 | foxlit | Simca: https://dotnetfiddle.net/9tuuwV |
| 15:27.42 | [TOM_RUS] | Hashtable and any other hash based containers in C# checks both GetHashCode() and Equals() when comparing objects |
| 15:40.43 | *** join/#wowuidev Wobin (wobin@www.thewonder.dog) |
| 15:40.48 | *** join/#wowuidev mserrano (~mserrano@euler.derig.org) |
| 15:40.48 | *** join/#wowuidev ThiefMasterrr (thief@indico/developer/thiefmaster) |
| 15:40.48 | *** join/#wowuidev Sharparam (~Sharparam@do.sharparam.com) |
| 15:40.49 | *** join/#wowuidev steino (steino@faxtrola.ixo.no) |
| 15:40.50 | *** join/#wowuidev Primer (~daniel@ceregatti.org) |
| 15:41.35 | *** join/#wowuidev sylvanaar (sid18093@pdpc/supporter/active/sylvanaar) |
| 15:41.38 | *** join/#wowuidev Telshin (sid114244@gateway/web/irccloud.com/x-btzotbllvjemepto) |
| 15:41.43 | *** join/#wowuidev nevcairiel (nev@WoWUIDev/WoWAce/Ace3/nevcairiel) |
| 15:45.28 | *** join/#wowuidev hizuro (~hizuro@WoWUIDev/hizuro) |
| 16:09.18 | *** join/#wowuidev Gnarfoz (~smallbrai@WoWUIDev/WowAce/Gnarfoz) |
| 17:04.57 | *** join/#wowuidev Ssateneth (ident@unaffiliated/ssateneth) |
| 17:48.46 | *** join/#wowuidev Varyak (~Varyak@ppp-82-135-82-69.dynamic.mnet-online.de) |
| 17:51.14 | *** join/#wowuidev Simca (~Simca@c-73-55-2-30.hsd1.fl.comcast.net) |
| 18:25.17 | *** join/#wowuidev schlumpf (~schlumpf@ip5f5ae550.dynamic.kabel-deutschland.de) |
| 18:35.26 | *** join/#wowuidev Primer (~daniel@ceregatti.org) |
| 18:47.47 | *** join/#wowuidev Jitta (~Jitta@ip5b414e46.dynamic.kabel-deutschland.de) |
| 19:21.06 | *** join/#wowuidev Funkeh` (~Funkeh`@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh) |
| 20:21.19 | *** join/#wowuidev stolenlegacy (~stolenleg@unaffiliated/stolenlegacy) |
| 20:34.12 | *** join/#wowuidev stuf (stuf@heatenin.gs) |
| 21:37.36 | foxlit | "The Site is currently in Read-Only mode, we will be back momentarily! Go Back to where I was..." |
| 21:41.02 | foxlit | except this "read-only" mode redirects e.g. https://wow.curseforge.com/projects/pawn -> https://wow.curseforge.com/read-only so it's more like "fuck off, we're busy" mode. |
| 21:43.53 | Alram | yeah mmoc is having maintenance too |
| 21:46.56 | Funkeh` | https://www.curseforge.com/news |
| 21:46.58 | foxlit | momentarily also apparently means an 8-10 hour expected maintenance window pending any unforeseen complications. |
| 21:54.45 | foxlit | (the downtime is fine, but the signaling is somewhat bad. They could have just gone with "Maintentance[link news thread] in progress, some features may be unavailable.") |
| 22:26.42 | *** join/#wowuidev Funkeh` (~Funkeh`@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh) |
| 22:40.07 | *** join/#wowuidev Resike (~Resike@62.77.234.203) |
| 22:52.13 | *** join/#wowuidev Kinesia (~Kinesia@115-64-111-2.tpgi.com.au) |
| 23:02.42 | *** join/#wowuidev brykrys (~Bryan@cpc104562-craw8-2-0-cust239.16-3.cable.virginm.net) |
| 23:02.42 | *** join/#wowuidev brykrys (~Bryan@WoWUIDev/Norganna/Dev/Brykrys) |
| 23:10.48 | *** join/#wowuidev Kinesia23 (~Kinesia@115-64-111-2.tpgi.com.au) |
| 23:27.05 | *** join/#wowuidev Bibi_ (sid144515@gateway/web/irccloud.com/x-lnwdytjqopqoynim) |
| 23:27.15 | *** join/#wowuidev Gnarfoz (smallbrain@WoWUIDev/WowAce/Gnarfoz) |
| 23:27.19 | *** join/#wowuidev sylvanaar (sid18093@pdpc/supporter/active/sylvanaar) |
| 23:28.15 | *** join/#wowuidev Fisker (~ballmer@how.i.met.your.developersdevelopersdevelopers.dk) |
| 23:32.30 | *** join/#wowuidev NickG365 (~NickG365@2607:5300:60:6e29:472:6425:3733:0) |
| 23:36.53 | Kaelten | foxlit yeah it's a little weird in some of there |
| 23:36.55 | Kaelten | we're currently back online with some of the sites and services |
| 23:36.57 | Kaelten | doing a bit more verification before we turn the rest of things back on |