| 00:01.44 | VIad | also I noticed blizz using strcmputf8i to compare names, but isnt lua natively unicode? |
| 00:02.09 | VIad | if not, why not? most compilers make unicode executables these days to avoid having to use specific api that are utf8 friendly |
| 00:02.35 | Neffi | Lua has no concept of unicode as of 5.1 |
| 00:03.51 | VIad | what is special about strcmputf8i? |
| 00:04.26 | VIad | strcmputf8i(a,b)>0 contra a>b or similar |
| 00:09.20 | VIad | neat, DoesTemplateExist :) |
| 00:14.01 | fewyn | i'm really surprised this name wasn't taken http://us.battle.net/wow/en/character/malganis/Stardew/advanced |
| 00:14.17 | VIad | why is stardew special? |
| 00:14.24 | VIad | sorry for not being american and getting the reference :p |
| 00:15.07 | fewyn | stardew is a name of a game |
| 00:15.13 | fewyn | stardew valley |
| 00:15.14 | VIad | aha, cool |
| 00:15.19 | VIad | my hunter is called overwatch |
| 00:15.20 | fewyn | which i was playing when trying to think of a char name |
| 00:27.57 | foxlit | <PROTECTED> |
| 00:28.21 | VIad | what does it mean, foxlit? |
| 00:29.53 | foxlit | it means I can probably identify api functions only available to glue. |
| 00:30.13 | Simca | hopefully it means he is doing api diffs so i don't have to |
| 00:30.13 | VIad | ah. I get it. |
| 00:30.19 | VIad | :D |
| 00:32.02 | Simca | sorry if i missed discussion of this already, but |
| 00:32.04 | Simca | https://www.facebook.com/notes/total-rp-3/blizzard-killed-storyline/1175979449090348 |
| 00:32.31 | VIad | why are they killing that addon? |
| 00:32.34 | VIad | RPs really love it |
| 00:33.02 | Simca | blizzard killed one of the model APIs, probably by accident, when they messed with things for legion |
| 00:33.49 | Simca | i know there are probably alternative manual ways to get model height (you could render the model in a local modelviewer, record the height of the model at scale 1, store the value, put it into an array in a lua file, and use that) |
| 00:34.25 | Simca | but is their any truth to the problems they're having with animation duration? I don't know anything about animations |
| 00:34.32 | Simca | is there* |
| 00:34.43 | Simca | mainly @foxlit and @Semlar |
| 00:35.58 | Semlar | you can get the animation duration out of the game if you wanted to but models don't have a height |
| 00:36.50 | foxlit | as Semlar already posted somewhere :GetModelFileID() is the new :GetModel() |
| 00:37.29 | Simca | ah, so they just need to adjust their tables to use filedataID properly and their issues are solved |
| 00:37.32 | Simca | figures |
| 00:37.45 | Semlar | there's a tauren the size of a building on the PTR |
| 00:37.58 | Simca | sure, that's scale. |
| 00:38.01 | Semlar | so their model doesn't inherently have anything to do with their size |
| 00:38.21 | Simca | models do have a height by design, all creatures have a scale too |
| 00:39.02 | Simca | creature scale values are available in creaturedisplayinfo |
| 00:39.12 | Simca | or maybe creaturemodelinfo, i forget |
| 00:45.31 | *** join/#wowuidev brykrys (~chatzilla@WoWUIDev/Norganna/Dev/Brykrys) |
| 00:47.53 | Semlar | it looks like the scale can also be set by the server |
| 00:48.14 | VIad | how didnt the change break blizzards model related UI elements? |
| 00:53.41 | foxlit | nothing used the affected method. |
| 00:55.07 | znf | changes Semlar into a mushroom |
| 00:55.25 | VIad | Semlar is the shroom king. |
| 00:56.01 | foxlit | it's almost as if they specifically designed the tozx opening animation to kill video encoders. |
| 00:58.05 | znf | video encoders...? |
| 00:58.13 | znf | in game? |
| 00:58.32 | Semlar | he's moved on to anime |
| 00:59.00 | Semlar | i assume |
| 00:59.07 | foxlit | indeed. I considered moving on to mushrooms, but this seemed better. |
| 00:59.25 | Semlar | we could all use more fungus in our lives |
| 00:59.32 | VIad | fungi |
| 01:01.26 | VIad | karl_w_w: https://www.youtube.com/watch?v=MWNSopUnTC4&t=8m50s |
| 01:06.39 | Simca | VIad / foxlit: it's kind of dumb they changed the name of the function at all though - you literally can't get filenames now from the API so nobody should expect that. they aren't going to rename every single API to have 'FileID' on the end |
| 01:07.13 | foxlit | I rather like this. |
| 01:07.14 | Semlar | to be fair, they did add a GetSpellTextureFileName function and changed GetSpellTexture to return file ids |
| 01:07.18 | VIad | Simca: I expected the behavior to be like SetTexture, where you can feed it a number and it just uses the correct path internally |
| 01:07.21 | Semlar | so this is perfectly consistent with blizzard |
| 01:07.45 | Simca | roflmao semlar |
| 01:08.02 | Simca | classic blizzard |
| 01:08.12 | foxlit | killing the color overloard on SetTexture is meh, though. |
| 01:08.26 | VIad | yeah |
| 01:08.29 | Semlar | do they not have file names internally though? how do they know what they're working on |
| 01:08.51 | Simca | they have file names on their build servers sure |
| 01:08.52 | VIad | maybe they have a list/browser they just click what they want and it automatically puts in the right data. |
| 01:08.58 | VIad | have blizz mentioned what IDE they use? |
| 01:09.02 | Simca | the deploy process strips them |
| 01:09.03 | VIad | or are they on notepad++ |
| 01:09.15 | Semlar | well that's not very nice of them |
| 01:10.35 | Semlar | is CreatureDisplayInfo a cache file or fully populated |
| 01:11.20 | Simca | fully populated |
| 01:11.34 | Simca | Creature.db2 is cache only |
| 01:11.43 | Simca | seems to be 'important npcs' only] |
| 01:11.55 | Simca | i haven't figured out exactly what makes them important |
| 01:13.25 | Semlar | jeff |
| 01:13.56 | foxlit | as an example, garrison followers are important. |
| 01:14.04 | Semlar | they have a guy who just goes through and decides which npcs are important |
| 01:14.20 | Simca | yeah, and battle pets seem to make it in there |
| 01:19.27 | Semlar | any idea what the ID column of creaturedisplayinfo is, is it the npc id? |
| 01:21.58 | Semlar | wowhead has information on ids that aren't in it, so i'm not sure |
| 01:22.53 | Simca | no, it's CreatureDisplayInfoID |
| 01:23.26 | Simca | CreatureCache.wdb, a cache file, maps NPCID:CreatureDisplayInfoID[4] (up to 4 variants supported for any given NPC ID) |
| 01:24.49 | Semlar | wait so you can't actually tie these to npc IDs without having them cached? |
| 01:25.30 | Simca | Creature.db2 has a field for them |
| 01:25.42 | Simca | so you can map the important ones that way |
| 01:26.47 | Semlar | creature doesn't appear to have an entry on the wiki |
| 01:27.11 | Semlar | not that it matters, it just usually has a page |
| 01:32.18 | Simca | well creature is a weird db2; it has less information than the WDB cache file for NPCs |
| 01:32.52 | Simca | so sites with a lot of sources for cache files, like DB sites, won't bother to use creature.db2; they'll just collect creaturecache.wdb from 10,000 people and call it a day |
| 01:33.38 | Simca | also, now the game returns creature tooltips for any creature, even if it isn't nearby (to break NPCScan), you can populate a future CreatureCache yourself |
| 01:33.46 | Simca | a full CreatureCache* |
| 01:34.41 | Semlar | ah |
| 01:34.45 | Semlar | right |
| 01:34.57 | Semlar | a side effect of breaking npcscan |
| 01:35.48 | Simca | yeah, it's quite funny since one of the devs was like "good thing you can't datamine servers" around the same time |
| 01:35.49 | Semlar | does 8916 sound right for the number of npcs |
| 01:36.01 | Simca | and I was thinking "uh, yeah, you just keep thinking that" |
| 01:37.37 | Simca | no, semlar |
| 01:37.58 | Semlar | that's how many entries are in creaturedifficulty and creature.db2 has like 400 less than that |
| 01:38.11 | Semlar | there's 62153 in CreatureDisplayInfo |
| 01:38.50 | Simca | creaturedifficulty probably only applies for creatures that scale with DifficultyID of instance |
| 01:39.08 | Simca | ah, i found an old cache file |
| 01:39.23 | Simca | let me see if i can't read it for you and generate a csv |
| 01:39.36 | Torhal | breaks Semlar's cache. |
| 01:40.39 | Semlar | i can probably read the wdb file, i might have to add a structure for it |
| 01:41.01 | Simca | Semlar: http://www.mediafire.com/download/7w1kubp051n3fn8/creaturecache_20914.zip |
| 01:41.10 | Simca | you can try, but I don't think they're documented anywhere |
| 01:41.29 | Simca | even people who take extreme interest in db2s don't pay attention to cache files generally |
| 01:41.46 | Simca | that's the raw file, i'll see about a csv |
| 01:41.51 | Semlar | i'm pretty sure i wrote something in python to read the wdb files but i'd have to find it |
| 01:42.18 | Semlar | is it still just a chunked file? |
| 01:43.10 | Simca | ah, found this in my bookmarks |
| 01:43.18 | Simca | this is what the structure was in june 2014: http://pastie.org/private/wkjjgntk7sx98eywahmcva |
| 01:43.32 | Semlar | thanks |
| 01:43.49 | Simca | classic blizzard with the 11 bit string lengths |
| 01:50.37 | *** join/#wowuidev mckenziemc (~Mark@104-51-145-162.lightspeed.frokca.sbcglobal.net) |
| 01:52.22 | *** join/#wowuidev Seerah (~RyanNL@104-180-66-29.lightspeed.mmphtn.sbcglobal.net) |
| 01:52.22 | *** mode/#wowuidev [+o Seerah] by ChanServ |
| 01:57.18 | mckenziemc | (1:50:00 PM) fewyn: i like my burgers to have some moo in it |
| 01:57.24 | mckenziemc | maybe you should kill them yourself |
| 01:57.49 | Torhal | I think mckenziemc is the type of guy who cuts off what he needs and rides the rest home. |
| 01:59.37 | *** join/#wowuidev p3lim (~p3lim@unaffiliated/p3lim) |
| 02:04.23 | Semlar | oops |
| 02:04.36 | Semlar | i managed to make lucasc use 5gb of ram |
| 02:04.52 | Semlar | i don't think i did that right |
| 02:06.45 | Semlar | looks like it's from using a question mark in the sig on this file |
| 02:07.45 | Semlar | alright fine, i guess i can't just tell it to read it as a dbc file |
| 02:08.54 | Semlar | i suppose it's because it thinks there are 1701729619 fields per row |
| 03:12.32 | *** join/#wowuidev Esamynn|Away (~chatzilla@d154-5-17-3.bchsia.telus.net) |
| 03:18.04 | mckenziemc | you can always download more ram when you need it, Semlar |
| 03:19.47 | mckenziemc | "This Minecraft tutorial explains how to craft a block of iron with screenshots and step-by-step instructions." |
| 03:23.08 | nebula | GetModelFileID, eh, was that added recently, or did I just miss it when looking at the model mt? |
| 03:36.17 | *** part/#wowuidev Seerah (~RyanNL@104-180-66-29.lightspeed.mmphtn.sbcglobal.net) |
| 03:44.37 | Semlar | it was added sometime after may 13th |
| 03:45.23 | Semlar | i just noticed it because i was looking through the metatable to check for anything new |
| 03:50.34 | Semlar | Dave's Industrial Light and Magic Bunny (Medium)(Sessile) |
| 03:51.22 | Semlar | http://www.wowhead.com/npc=38821 |
| 04:03.03 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.red.bezeqint.net) |
| 04:13.27 | fewyn | i sometimes wish i bought a 3d printer that wasn't so finnicky |
| 04:19.27 | *** join/#wowuidev Yoshimo (~anything@unaffiliated/yoshimo) |
| 04:36.11 | mckenziemc | design one |
| 04:38.38 | fewyn | i've already spent money on this one tho |
| 04:47.51 | mckenziemc | apparently not enough |
| 05:03.26 | *** join/#wowuidev ShadniX (dagger@p5DDFFC7E.dip0.t-ipconnect.de) |
| 05:08.41 | Fisker | acts like a tree and hugs Cairenn |
| 05:08.57 | Cairenn | hugs Fisker |
| 05:09.39 | karl_w_w | what's that term you use when you damage something to impede its progress |
| 05:09.48 | *** join/#wowuidev NodusCursorius_ (~NodusCurs@unaffiliated/noduscursorius) |
| 05:09.51 | nebula | sabotage? |
| 05:09.52 | karl_w_w | I'm pretty sure it's a reference to a body part |
| 05:10.24 | kd3 | kneecap? |
| 05:10.30 | karl_w_w | close |
| 05:10.52 | karl_w_w | not achillies either |
| 05:11.17 | nebula | hamstring |
| 05:11.21 | karl_w_w | thanks |
| 05:11.25 | karl_w_w | champion |
| 05:13.14 | kd3 | guh. hard to decide if fireworks or gunshots outside |
| 05:13.42 | Cairenn | =/ |
| 05:13.43 | Semlar | that's what you get for living in detroit |
| 05:14.03 | karl_w_w | heh |
| 05:14.36 | Torhal | wipes Semlar's memory and drops him off in Detroit. |
| 05:14.38 | Cairenn | fyi - metal splinters hurt |
| 05:14.45 | Torhal | Yes |
| 05:14.47 | Torhal | VERY |
| 05:17.16 | Cairenn | Also, metal working destroys manicures ;-p |
| 05:17.33 | Fisker | rude |
| 05:33.29 | Semlar | Simca: how much do i need to throttle the server requests for npc info |
| 05:34.05 | *** join/#wowuidev mitch0 (~mitch@84-236-113-208.pool.digikabel.hu) |
| 05:34.17 | Semlar | i remember writing something for this for the WoD beta but it's probably gone |
| 05:37.26 | Simca | let me see |
| 05:38.28 | Simca | i do 40 NPCs per second |
| 05:38.40 | Semlar | how do you know when to stop |
| 05:39.09 | Torhal | explains programming to Semlar like he's five. |
| 05:39.18 | Semlar | after a certain gap? |
| 05:40.36 | Simca | i use two modes of operation - one works until X preset ID, the other works until a gap of Y. in this case X is 115000 and Y is 2000 |
| 05:41.17 | kd3 | eesh. 2 stream crashes for gdq. today |
| 05:41.44 | Semlar | i know they go up to at least 118334 |
| 05:42.16 | nebula | it's annoying, my tv won't keep the stream open if it's offline |
| 05:42.21 | Semlar | well, assuming those are npcIDs in this file |
| 05:42.56 | Simca | well |
| 05:43.08 | Simca | i often forget to update the max value |
| 05:43.26 | Simca | so it wouldn't surprise me if I was wrong about that part |
| 05:43.55 | Simca | I generally use the gap scanning, it's pretty reliable and 2k IDs is a pretty low number of wrong NPCs to have to scan for it to stop |
| 05:45.01 | Simca | you can probably go faster than 40 NPCs per second; 100 is probably safe. |
| 05:45.24 | Simca | but i generally scan spells, quests, and creatures at the same time, and if i put those all on 100 per second |
| 05:45.31 | Simca | the server goes "GTFO!" |
| 05:46.29 | Semlar | i'm going to try all of them at once |
| 05:46.51 | znf | DiscoMode() |
| 05:47.32 | znf | Torhal, explain me programming like I'm 3 |
| 05:48.25 | Simca | that is probably a good way to crash the game |
| 05:48.35 | Simca | but at least we have a forum for that now |
| 05:49.08 | Simca | "Hey Guill when I try to send 120000 packets in under a second, the game crashes, please fix" |
| 05:49.11 | Torhal | No, znf. I wasn't actually going to explain anything to Semlar - his head is like a sieve. |
| 05:50.06 | Semlar | alright it didn't like doing 40,000 at once |
| 05:50.26 | Semlar | the first 2000 worked |
| 05:50.37 | Semlar | but i'm not sure how many of those actually returned data |
| 05:51.01 | Semlar | my game is currently frozen but i have faith it will come back |
| 05:52.01 | Semlar | any day now |
| 05:52.24 | Semlar | alright i don't think it's coming back |
| 05:53.38 | Semlar | <PROTECTED> |
| 05:53.39 | Stanzilla | not looking forward to fixing all the pet battle addons I'm using for legion :( |
| 05:53.45 | Stanzilla | all abandoned, I guess |
| 05:54.39 | *** join/#wowuidev Esamynn|Away (~chatzilla@d154-5-17-3.bchsia.telus.net) |
| 05:54.41 | Semlar | does the wdb file stay populated indefinitely or what |
| 05:56.21 | Simca | it's wiped every new build i think |
| 05:56.24 | Simca | other than that, yes |
| 05:57.34 | Simca | Stanzilla: BattlePetBreedID isn't abandoned at least |
| 05:57.51 | Stanzilla | are you sad when I say I did not use that? :D |
| 05:57.59 | Simca | nah |
| 05:58.11 | Semlar | aaaaand it disconnected me |
| 05:58.14 | nebula | did something with the pet battle api change that'll break all of them or something? |
| 05:58.24 | Stanzilla | yeah, the names of almost every api |
| 05:58.32 | Simca | not really; journal filters changed some |
| 05:58.35 | Stanzilla | nothing a search and replace cant fix though |
| 05:58.49 | Stanzilla | or was the name change a different part |
| 05:58.55 | Stanzilla | move from no C_ to C_? |
| 05:59.16 | nebula | they've been in a collection since they were implemented |
| 05:59.36 | Stanzilla | then I remembered wrong |
| 06:00.17 | *** join/#wowuidev Elmoe_ (~Elmoe@24.18.131.251) |
| 06:02.19 | *** join/#wowuidev NodusCursorius (~NodusCurs@unaffiliated/noduscursorius) |
| 06:07.54 | Stanzilla | Simca: mainly looking into PetJournalEnhanced but I probably just have to update the libs it uses since it doesnt use a repo and I cant trigger a rebuild with up to date shit. It also uses http://wow.curseforge.com/addons/libpetbreedinfo/ which should be similar to what you did? |
| 06:08.29 | Simca | i would wait on PJE as long as possible |
| 06:08.50 | Simca | the author for that is lazy like me and might do it himself, just like... the day of legion patch |
| 06:10.32 | Stanzilla | maybe it's already enough if I just update the libs anyway |
| 06:10.41 | Mr_Rabies | that was insane |
| 06:10.55 | Mr_Rabies | friends and i ended up amassing about $450 in fireworks |
| 06:10.56 | Stanzilla | also nice hint here, microsoft https://pbs.twimg.com/media/Cmf99EHWgAAVL5T.jpg |
| 06:11.01 | Mr_Rabies | and uh |
| 06:11.03 | Mr_Rabies | it was bonkers |
| 06:16.46 | Stanzilla | yeah looks like updating LibPetJournal was enough, good guy End |
| 06:19.07 | Stanzilla | nebula: I think I had the tradeskill stuff in mind with the C_ move |
| 06:20.59 | Stanzilla | still trying to zoom out my camera every time I log into beta :D |
| 06:21.32 | Semlar | sets stanzilla to first-person mode |
| 06:22.02 | Simca | i am fine the new maximum |
| 06:22.04 | Stanzilla | game already did that for you |
| 06:22.14 | Simca | with* |
| 06:22.37 | Simca | but i think it is dumb that they are trying to force change on people who have done something the same way for 11 years |
| 06:22.38 | Semlar | well it hasn't disconnected me so far doing 500 every 2 seconds |
| 06:24.13 | znf | It's supposed to be raining today |
| 06:24.17 | Semlar | that'll take about 8 minutes to complete |
| 06:25.18 | znf | It's hot as hell |
| 06:25.18 | znf | And sunny as... Being on the sun |
| 06:25.37 | znf | I haven't been up at this early hour in... Years |
| 06:29.47 | Semlar | <PROTECTED> |
| 06:34.45 | Simca | thanks semlar; looks complete |
| 06:34.51 | Simca | let me export that to CSV |
| 06:48.59 | *** join/#wowuidev elvador (~elvador@n.6zj.de) |
| 06:49.20 | *** join/#wowuidev elvador (~elvador@WoWUIDev/WowAce/Elvador) |
| 06:51.19 | *** join/#wowuidev Industrial (~Industria@81.21.139.13) |
| 07:03.33 | *** join/#wowuidev VIad (~Vlad@cm-84.213.146.58.getinternet.no) |
| 07:30.57 | *** join/#wowuidev Miiru (~Miiru@p200300732E01D1009CBFBB1F4152D80B.dip0.t-ipconnect.de) |
| 07:36.59 | *** join/#wowuidev dssd4 (44605bed@gateway/web/freenode/ip.68.96.91.237) |
| 07:49.12 | Semlar | i'm not sure which part of this is the flags simca |
| 07:49.39 | Simca | one sec i'm almost done with my export thing |
| 07:50.45 | Semlar | <PROTECTED> |
| 07:51.27 | Stanzilla | hey my name is benny questgiver |
| 07:52.03 | Simca | i also have the names of two fields added since that pastie |
| 07:52.30 | Simca | one was added in WoD beta: TrackingQuestID - it's a QuestID for the quest (if any) that tracks the mob |
| 07:52.41 | Simca | used for rares and stuff |
| 07:52.42 | Semlar | pastie was too confusing |
| 07:53.03 | Semlar | i had to go through it with a hex editor |
| 07:53.48 | Simca | second was added a couple builds ago: VignetteID |
| 07:54.41 | Semlar | the part where it appears to have optional fields is making this difficult |
| 07:56.19 | Simca | QuestItems bit? |
| 07:57.18 | Simca | http://www.curseforge.com/paste/b153d729 here is my C# parser for it |
| 07:57.57 | Semlar | no it has extra data on the end of it and what it is appears to be based on flags or something |
| 07:58.13 | Semlar | for one of these, it's a subtitle for their name |
| 07:58.20 | Semlar | for other ones i have no idea what it is |
| 07:59.10 | Simca | note what i had to do for CursorName |
| 07:59.37 | Simca | for some reason when its length is 1, it takes up 0 bytes |
| 07:59.45 | Semlar | like "kanrethad" has "master of death", "forest spider" has nothing, and "two-bit thug" has some bytes |
| 08:00.31 | Simca | sounds like title |
| 08:00.58 | Semlar | if it's the title why do some of these have random gibberish |
| 08:01.09 | Semlar | it has to be more than one thing |
| 08:01.44 | Semlar | i'll check your parser |
| 08:01.50 | Semlar | i'm sure i'm doing something wrong |
| 08:03.41 | Simca | sounds like Title, TitleAlt, CursorName, and QuestItems - the last four fields |
| 08:03.46 | Simca | forest spider has nothing |
| 08:03.50 | Simca | kanrethad has a title |
| 08:03.55 | Simca | two-bit thug has a questItem |
| 08:08.12 | Semlar | i don't know what this 11 bit int stuff is at the top because i'm not seeing anything like that with the hex editor |
| 08:09.23 | Semlar | can you even read an 11 bit int with lua |
| 08:10.05 | Simca | same way you can with anything, unfortunately; byte by byte and storing the remainder |
| 08:10.17 | Simca | using bit shifting to combine the remainder with the needed bits from the next chunk |
| 08:10.20 | Simca | it's a huge pain |
| 08:10.52 | Simca | http://www.mediafire.com/download/80kmfjrtbnl348j/CreatureCache_22124_Full.csv |
| 08:10.57 | Simca | finally got my program to work |
| 08:11.16 | Semlar | you mean i have to do math? |
| 08:11.34 | Semlar | that doesn't sound right |
| 08:14.24 | Simca | this is what it uses to read the bits properly as ints |
| 08:14.50 | Simca | http://www.curseforge.com/paste/1b83ef3b\ |
| 08:14.53 | Simca | http://www.curseforge.com/paste/1b83ef3b |
| 08:15.34 | Simca | stores the bit position and current byte, so it knows how to combine and bitshift the next byte |
| 08:16.23 | Semlar | :( |
| 08:17.32 | Simca | working with non-standard units is not fun or performant; that's why most sane people don't do it |
| 08:18.00 | Simca | but blizzard had to save those 5 bits |
| 08:18.44 | nevcairiel | I work in multimedia, everything is bitpacked to save every single bit, so stop whining :P |
| 08:19.07 | nevcairiel | just be happy its constant width fields and not variable length codes |
| 08:19.32 | Simca | what you're talking about sounds like compression |
| 08:19.40 | Simca | which is a little different |
| 08:19.46 | nevcairiel | not necessarily |
| 08:20.00 | nevcairiel | although its of course a goal not to waste space |
| 08:21.57 | Simca | the problem with what blizzard is doing is that manually bitpacking the data with fixed width fields is usually still wildly inferior to actual compression |
| 08:22.18 | Simca | so it's like wanting to save space, and then going halfway to do it |
| 08:25.55 | Simca | though, creaturecache.wdb is 11 years old at this point; one of the only cache files still left in the old format |
| 08:26.18 | Simca | world of computing was probably different in 2003 |
| 08:36.47 | Simca | also Semlar that new VignetteID field finally gives the NPCID:VignetteID relationship you've always wanted |
| 08:38.22 | Semlar | o.o |
| 08:42.42 | Torhal | 2003 + 11 years is 2014. Simca confirmed for time traveler. |
| 08:42.59 | Simca | math wizard |
| 08:43.06 | Torhal | Heh |
| 08:50.55 | VIad | hmm... did you know of this? https://livechat.battle.net |
| 08:51.31 | VIad | interesting |
| 08:51.35 | VIad | I didn't :d |
| 08:52.33 | nevcairiel | support live chat? sure |
| 08:54.12 | VIad | seems like you can also do that from within the game |
| 08:54.16 | VIad | its whitelisted domain |
| 08:54.35 | nevcairiel | you can use most of the support page from ingame |
| 08:56.45 | *** join/#wowuidev Lynxium[] (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 09:00.03 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.red.bezeqint.net) |
| 09:40.01 | karl_w_w | https://www.reddit.com/r/oculus/comments/4r6ek0/masturbation_tips/ |
| 09:40.04 | karl_w_w | that top comment |
| 09:57.22 | mitch0 | karl_w_w: so, are you ordering the kit? satisfied with the answer? :) |
| 10:14.11 | *** join/#wowuidev stolenlegacy (~stolenleg@unaffiliated/stolenlegacy) |
| 10:14.50 | VIad | taximap for legion http://ss.bool.no/upload_0f6ffc16.svg |
| 10:16.01 | VIad | partitioned into the mapID and its fun to see how some loops are closed off |
| 10:18.25 | nevcairiel | the positions seem wrong on that |
| 10:18.35 | VIad | the layout doesnt use any ingame data |
| 10:18.38 | nevcairiel | why is silvermoon city below eastern plaguelands |
| 10:18.52 | nevcairiel | would be far more interesting if it used actual map locations to map out |
| 10:19.25 | VIad | atm I was mostly making sense of the data. luckily only some few changes. like the id field on some moved to the end, kind of odd |
| 10:21.21 | VIad | going to yet again try to calculate taxi times but make it better |
| 10:26.11 | [TOM_RUS] | VIad, hunter flavor taxi nodes missing |
| 10:26.57 | VIad | maybe my dbc were older |
| 10:27.00 | VIad | Simca gave them to me |
| 10:27.52 | [TOM_RUS] | pretty sure they have been in for a while |
| 10:33.04 | [TOM_RUS] | nodes 1838-1850 |
| 10:52.00 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 11:19.48 | VIad | then they are there, [TOM_RUS] |
| 11:19.59 | VIad | bottom right part of the image |
| 11:20.12 | VIad | left of the obvious broken shore nodes, its in its own closed loop :p |
| 11:35.51 | Semlar | Simca: can you paste what you have for all the length vars for kanrethad, npcid 29 |
| 11:36.07 | Semlar | i have "2" for the titleLength and that doesn't make any sense |
| 11:36.08 | Simca | k |
| 11:37.03 | *** join/#wowuidev stolenlegacy_ (~stolenleg@chello080109199104.5.graz.surfer.at) |
| 11:37.03 | *** join/#wowuidev stolenlegacy_ (~stolenleg@unaffiliated/stolenlegacy) |
| 11:40.10 | Semlar | why do they need like 10 different name entries anyway |
| 11:41.35 | Simca | i've seen nameAlt used in case where it like |
| 11:42.14 | Simca | 'Enchanter' and nameAlt is 'Enchantress' |
| 11:42.47 | Semlar | different gender prefixes? |
| 11:44.04 | Simca | yeah |
| 11:47.26 | Simca | Name is 10 chars, Title is 16 chars |
| 11:47.31 | Simca | for that ID |
| 11:48.36 | Simca | 2 << 3 = 16 |
| 11:48.38 | Semlar | titleLength is 16? |
| 11:48.40 | Simca | probably not a coincidence |
| 11:48.52 | Simca | yeah |
| 11:50.47 | Semlar | what does shifting it by 3 do |
| 11:50.53 | Simca | are you resetting the bit position at the end of a record? |
| 11:51.02 | nevcairiel | shifting by 3 is multiplying by 8 :p |
| 11:51.41 | Semlar | should i be doing that |
| 11:52.06 | Simca | if you are implementing it like i was, yes. |
| 11:53.04 | Semlar | well i'm not but clearly i'm not doing it right |
| 11:53.32 | Semlar | http://hastebin.com/bexuyayeve.lua this is what i was using |
| 11:53.41 | Simca | 11 * 10 + 6 + 1 is 117, which is 3 off of the nearest bit, so those 3 bits need to be 'skipped' |
| 11:53.47 | Simca | 3 off the nearest byte* |
| 11:54.07 | Simca | if you didn't skip them, your next read by bits would be off by 3 shifts |
| 11:55.15 | Semlar | i don't like bits |
| 11:55.26 | *** join/#wowuidev stolenlegacy (~stolenleg@chello080109199104.5.graz.surfer.at) |
| 11:55.26 | *** join/#wowuidev stolenlegacy (~stolenleg@unaffiliated/stolenlegacy) |
| 11:56.12 | Semlar | i should just copy what you have |
| 11:56.13 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.red.bezeqint.net) |
| 11:57.55 | *** join/#wowuidev Lynxium[] (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 12:02.50 | Simca | make sure to note that 'leader' is actually 1 bit, even though i call it a boolean |
| 12:03.29 | Semlar | <PROTECTED> |
| 12:06.43 | Simca | try adding a function to the end that is like: function flush() local bitRem = bitPos % 8; bitPos = bitPos + bitRem; end |
| 12:06.55 | Simca | and call it after your last bit-based read from each record |
| 12:09.21 | Semlar | what, just at the end of the readXBits function? |
| 12:12.50 | Simca | nah, after the nameAlt3Length readXbits line |
| 12:13.16 | Semlar | right now i'm not reading the lengths correctly |
| 12:14.43 | Simca | not sure I know enough about how lua numbers work |
| 12:14.52 | Simca | only number type by default is double, yeah? |
| 12:15.03 | Semlar | yes |
| 12:15.29 | Simca | i don't even know what would happen if you bitshift a double |
| 12:15.45 | Alram | i'll bitshift ur double Simca |
| 12:15.48 | Simca | certainly not anything like what you intended |
| 12:15.58 | Semlar | it should handle bitwise operations fine |
| 12:17.05 | Simca | it must only apply the shift to the fraction bit and keep the exponent bit constant |
| 12:17.11 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 12:19.58 | Simca | ok so looking at your function; i think it is only taking 1 bit |
| 12:21.29 | Semlar | i pretty much copied the function from http://stackoverflow.com/questions/8011700/how-do-i-extract-specific-n-bits-of-a-32-bit-unsigned-integer-in-c |
| 12:21.53 | Semlar | mask = ((1 << X) - 1) << startBit; isolatedXbits = value & mask |
| 12:22.36 | Semlar | i don't know if the way i'm incrementing "bitPos" is correct |
| 12:23.05 | foxlit | purely hypothetically, if I told you to read 0 bits, would you expect your position in the stream to change? |
| 12:24.07 | Semlar | alright so i drop + 1 but it still isn't right |
| 12:24.25 | nevcairiel | foxlit: most optimized bitreaders forbig reading 0 bits =p |
| 12:24.29 | nevcairiel | forbid* |
| 12:24.37 | nevcairiel | but no, it should not change |
| 12:25.28 | foxlit | Semlar: if you were 7 bits into a stream, and I told you to read 8 bits, what would your code do? |
| 12:27.49 | Semlar | well, theoretically it should be taking the last bit from the first byte and the first 7 from the next one |
| 12:28.31 | Simca | your code needs to shift result after you apply the mask |
| 12:28.57 | foxlit | setting aside the issue of defining what "last" and "first" are in relation to bytes and bits, the linked function only reads one byte at position 0 and then does things to it. |
| 12:29.59 | Semlar | ah |
| 12:30.55 | Simca | yes, assuming bitPos is 0 when the function first starts |
| 12:31.11 | Simca | I assume he starts it off at the proper value where titleLength for NPCID 29 is though |
| 12:31.28 | Simca | which in the case of his file... is |
| 12:31.31 | foxlit | if you were paying attention, we set bitPos = 7 for the purposes of illustrating the error. |
| 12:31.48 | Semlar | i need to add 1 to the numBytes |
| 12:32.19 | foxlit | sometimes you do; other times you don't. |
| 12:34.27 | Semlar | bitPos % 8 + numBits |
| 12:34.53 | Simca | yeah |
| 12:35.20 | Simca | local numBytes = math.ceil((numBits + (bitPos % 8)) / 8) |
| 12:37.46 | Simca | this bitfield is left to right, which i think is the other concept foxlit was trying to point out |
| 12:38.23 | Simca | that uint_le may not be the proper function as that assumes little endian, which is not what we're after here. |
| 12:40.47 | Semlar | erm |
| 12:41.35 | Simca | titleLength would be 0000 0010 0000 0000 as 2 bytes (11 bits, from the front) |
| 12:41.47 | Simca | if you read the first byte only, and as LE |
| 12:41.53 | Simca | it would be 2 instead of 16 |
| 12:42.21 | nevcairiel | bit-wise reading is typically byte by byte- so LE or BE dont usually apply |
| 12:42.38 | nevcairiel | (or you could consider it BE, if you must) |
| 12:42.42 | Simca | he's pulling 2 bytes at a time |
| 12:43.27 | Simca | (well, depends on the number of bits asked for and the position, but generally, 2 bytes) |
| 12:44.45 | foxlit | to be clear, I don't actually have an opininon on whether LE is appropriate here. It's just something that needs to be checked. |
| 12:48.04 | Simca | only other problem i see without trying to figure out exactly what that C code you took is trying to do is that you need to shift the result after applying the mask |
| 12:48.57 | Simca | the C code samples on StackOverflow are perfectly happy with leaving their results shifted out of place, but since you're using it as a number of characters to read, you need it to be at the right place |
| 12:49.25 | Simca | local result = bit.lshift(bit.band(value, bit.lshift((bit.lshift(1, numBits) - 1), bitPos % 8)), bitPos % 8) |
| 12:50.04 | foxlit | take a drink for every time you write bit. in this code. |
| 12:50.33 | Semlar | my irc client turns 8) into an emoticon |
| 12:50.49 | Simca | oh i'm dumb, that should be an rshift at the start |
| 12:51.03 | Simca | since you're moving it back to zero |
| 12:51.53 | Semlar | what am i moving back to zero |
| 12:52.34 | Simca | the result. your mask takes the section out of the byte(s) that you want, but it doesn't move them. |
| 13:00.07 | Semlar | i think i need some pancakes |
| 13:02.20 | Semlar | pancakes don't care about byte order and what-not |
| 13:27.10 | Semlar | pancakes acquired |
| 13:29.51 | *** join/#wowuidev MaxFrost (~MaxFrost@c-174-53-177-161.hsd1.mn.comcast.net) |
| 13:31.59 | *** join/#wowuidev Lynxium[] (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 13:35.16 | *** join/#wowuidev A-KO (A-KO@2601:143:4201:b940:78db:f8dd:a392:60a8) |
| 13:38.18 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 13:38.36 | Neffi | steals Semlar's pancakes |
| 13:38.47 | Semlar | too late |
| 13:38.54 | Neffi | its never too late |
| 13:38.59 | Neffi | gets the scalpel |
| 13:39.04 | Semlar | no don't |
| 13:39.09 | Neffi | !!! |
| 13:52.46 | karl_w_w | annoying bug http://i.imgur.com/JDQiF7D.gifv |
| 13:58.12 | *** join/#wowuidev Lynxium[] (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 13:58.38 | Semlar | twiddles neffi's bits |
| 14:00.14 | Neffi | wat |
| 14:00.32 | Neffi | karl_w_w: rip your monitor |
| 14:02.45 | Semlar | o it accidentally works neffi |
| 14:03.33 | Neffi | like everything you make? |
| 14:03.38 | Semlar | how dare you |
| 14:04.23 | Semlar | i gave up trying to do it wrong and copied simca's code |
| 14:05.46 | Neffi | :| |
| 14:05.47 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.red.bezeqint.net) |
| 14:06.47 | Semlar | <PROTECTED> |
| 14:17.28 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.cablep.bezeqint.net) |
| 14:25.54 | Neffi | I just woke up Semlar |
| 14:26.10 | Semlar | i haven't slept neffi |
| 14:26.14 | Neffi | I dont want to think about bitwise logic until I've had my coffee |
| 14:26.19 | Neffi | omg go slep |
| 14:26.19 | mckenziemc | then how did Neffi wake you up |
| 14:26.24 | Semlar | i'm being haunted by bits |
| 14:26.51 | Neffi | Semlar, I can't find a terminal color scheme to match my GTK/WM theme. This is stressful. |
| 14:27.00 | Semlar | black on black |
| 14:29.08 | mckenziemc | Semlar on Neffi |
| 14:30.26 | *** join/#wowuidev Lynxium (~Lynxium@bzq-84-109-128-19.red.bezeqint.net) |
| 14:31.16 | Neffi | this kinda works http://i.imgur.com/RFnSNLJ.png |
| 14:31.38 | Neffi | but it can be hard to read |
| 14:34.43 | *** join/#wowuidev Noshei (~quassel@c-24-9-171-99.hsd1.co.comcast.net) |
| 14:34.55 | Semlar | pandora is playing the theme song from the beverly hillbillies |
| 14:35.25 | Semlar | come and listen to a story bout a man named jed |
| 14:35.29 | Neffi | plz |
| 14:35.39 | Semlar | poor mountaineer barely kept his family fed |
| 14:36.11 | Semlar | i can't not upthumb this |
| 14:36.27 | Neffi | :\ |
| 14:36.45 | Semlar | this is what happens to all my stations |
| 14:37.07 | Neffi | you can't escape what you are Semlar |
| 14:37.10 | Neffi | you are a hillbilly |
| 14:37.31 | Semlar | well the first thing you know old jed's a millionaire |
| 14:39.51 | *** join/#wowuidev Kultzi (~ironi@85-76-39-18-nat.elisa-mobile.fi) |
| 14:46.47 | Semlar | 93,927 entries |
| 14:51.02 | Semlar | here neffi http://www.wowhead.com/npc=76601/darrens-test-npc-3 |
| 14:52.53 | *** join/#wowuidev Funkeh` (~Funkeh`@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh) |
| 14:54.42 | *** part/#wowuidev mckenziemc (~Mark@104-51-145-162.lightspeed.frokca.sbcglobal.net) |
| 14:55.44 | VIad | Semlar: there are tons of scenescripts.dbc used npcs as well |
| 14:56.35 | *** join/#wowuidev Yoshimo (~anything@unaffiliated/yoshimo) |
| 14:58.29 | *** join/#wowuidev mckenziemc (~Mark@104-51-145-162.lightspeed.frokca.sbcglobal.net) |
| 15:03.27 | karl_w_w | 2 hours of steam sale left |
| 15:05.20 | *** join/#wowuidev Primer_ (~daniel@ceregatti.org) |
| 15:09.18 | VIad | hasn't been that interesting for me this year |
| 15:09.24 | *** join/#wowuidev Tristit1a (~tristitia@static-ip-69-64-50-196.inaddr.ip-pool.com) |
| 15:16.15 | Neffi | thanks for the heads up |
| 15:18.27 | Neffi | Steam controller: yes or no? While it's still on sale |
| 15:18.42 | Neffi | I'm iffy on the weird thumb pads. I play with a PS3 controller right now |
| 15:22.17 | mikma | neffi: yes. |
| 15:22.27 | mikma | steam controller's REALLY good |
| 15:22.41 | *** join/#wowuidev g0urra (~g0urra@unaffiliated/gourra) |
| 15:23.02 | mikma | propably the best gamepad i've ever tried. and i've been playing with xbox xbox360 ps2 ps3 ... |
| 15:25.38 | Semlar | neffi http://lua-users.org/wiki/SortedIteration this is an incredibly slow implementation of "orderedPairs" |
| 15:28.30 | Semlar | it has to iterate over the table to look for the next value on every iteration |
| 15:32.59 | Neffi | <PROTECTED> |
| 15:39.42 | mckenziemc | iterated over the sorted key table, so not as bad? but still bad |
| 15:39.49 | mckenziemc | iterate* |
| 15:40.12 | Semlar | no it's super slow, i had to write my own |
| 15:40.44 | Semlar | it was taking.. minutes to iterate over this table with 90k entries |
| 15:41.22 | mckenziemc | "super slow" doesn't necessarily mean for the reason you think |
| 15:43.25 | Semlar | it's literally looping over the entire table on every iteration |
| 15:43.41 | mckenziemc | clarify what you think "the table" means |
| 15:44.40 | Kalroth | the oak table? |
| 15:45.24 | Neffi | Semlar: I don't get it. Just dump the keys in an array and sort the array, then iterate the array in order |
| 15:46.19 | Semlar | i just made 2 copies of the table, sorted one and had the other one point back to the sorted index |
| 15:46.23 | nevcairiel | thats really what this thing is doing, just hidden behind its back |
| 15:46.39 | nevcairiel | its probably inefficient because its trying to do magic |
| 15:46.43 | mckenziemc | which is what this does, except that it searches from 1 to the last key before returning the next |
| 15:46.52 | mckenziemc | in the sorted keys table |
| 15:47.10 | Semlar | it's iterating over the ENTIRE sorted copy of the table EVERY SINGLE LOOP |
| 15:47.27 | nevcairiel | yes because its trying to do it transparently |
| 15:47.32 | mckenziemc | well yes |
| 15:47.39 | Semlar | it isn't even breaking when it finds the match |
| 15:47.41 | nevcairiel | if you keep the hash and array table separately then you can do it more efficiently |
| 15:47.46 | mckenziemc | now that you've said it the right way, i see what you're saying about that mistake |
| 15:48.13 | Neffi | I saw what he was saying, I'm just sitting here wondering why the fuck they had to make this so complex when the answer is as simple as what I just said |
| 15:48.15 | nevcairiel | if you try to write orderedPairs like this, best you can do is add the break :p |
| 15:48.38 | nevcairiel | otherwise, you can of course do better if you manually index |
| 15:48.46 | nevcairiel | but not as a drop-in replacement for pairs |
| 15:48.52 | Neffi | Semlar: Do what I said, and then for each element return temp[i], original[temp[i]] |
| 15:48.53 | mckenziemc | you can do better than adding break |
| 15:48.59 | Semlar | i did that neffi |
| 15:49.01 | Neffi | that gives you ordered k, v |
| 15:49.38 | Semlar | well, no, i made 2 lookup tables and it does no looping |
| 15:49.46 | nevcairiel | the problem this Lua wiki thing has is that it tries to be a drop-in replacement for pairs thats ordered, which is just problematic |
| 15:49.58 | nevcairiel | or you could add a third table to map key back to index |
| 15:50.07 | nevcairiel | and then index+1 back to next key |
| 15:50.08 | nevcairiel | :D |
| 15:50.59 | mckenziemc | actually even using break would be a mistake, wouldn't it? state is the last value, not the last key? |
| 15:51.02 | nevcairiel | anyway doing it manually without requiring it a drop-in replacement for pairs is clearly going to be more efficient |
| 15:51.38 | nevcairiel | state is the last key |
| 15:52.02 | mckenziemc | oh nvm, got confused on the sorted table |
| 15:52.59 | Neffi | Semlar: Steam wants like $10 for shipping on the controller. Amazon has it with free Prime shipping, but then it told me this: Usually ships within 1 to 2 months. |
| 15:53.04 | Neffi | why is things like this |
| 15:53.42 | Semlar | "usually" |
| 15:54.03 | Neffi | "$34.99 FREE Prime Shipping once available" <- the last part being bolded for emphasis |
| 15:54.11 | Semlar | it's out of stock |
| 15:54.15 | Neffi | yup |
| 15:54.23 | Neffi | and it looks like it usually takes a while to get back in stock |
| 15:54.35 | Semlar | did you check if there were other suppliers |
| 15:54.40 | mckenziemc | with people wanting to save on shipping, most likely |
| 15:54.47 | Neffi | I can buy it straight through Steam store but then I pay shipping |
| 15:55.10 | Semlar | <PROTECTED> |
| 15:55.51 | Neffi | it shows it as in stock on that page |
| 15:55.55 | Neffi | but not on the main product page |
| 15:57.18 | Semlar | well what does it say the delivery would be if you add it to your cart from there |
| 15:57.45 | Neffi | "Estimated delivery: July 25, 2016 - Aug. 17, 2016" |
| 15:58.01 | Neffi | now that's precision |
| 15:58.49 | mckenziemc | kids these days |
| 15:58.58 | mikma | yeah steam takes a good amount because of shipping but it'll arrive *fast* |
| 15:59.08 | Neffi | haha |
| 15:59.18 | Semlar | compared to that it would be fast |
| 15:59.46 | Neffi | I'm not even sure I want the thing. Only buying ti because it's on sale and it looks like it might be OK |
| 16:00.15 | Semlar | don't do it neffi |
| 16:00.18 | Neffi | I didn't |
| 16:00.18 | mckenziemc | why do you think they but it on sale? |
| 16:00.29 | Neffi | because they put everything on sale? |
| 16:00.32 | Neffi | (almost) |
| 16:00.37 | mckenziemc | for the same reason |
| 16:00.48 | mckenziemc | to get people to buy things they don't know if they really want |
| 16:00.50 | Neffi | I had a few friends say it was amazing, and others say it was awful. It looks like the reception to it is very polarized |
| 16:01.14 | Neffi | I will get my PS3 controller working on Linux instead |
| 16:02.00 | Infus | I don't know where I found it, but this sortedPairs() http://pastebin.com/Q7fwXNzG is efficient and a drop in replacement for pairs |
| 16:02.47 | Neffi | well sure that's a way to do it |
| 16:03.06 | Neffi | but there are people who are going to claim you're being wasteful by throwing out a closure each invocation, even though it's not really that bad |
| 16:03.49 | Neffi | why not have it take an optional third argument, a sorting function to pass to table.sort? it's a trivial addition |
| 16:04.16 | Infus | it already supports a sort function |
| 16:04.42 | Neffi | oh, my mistake |
| 16:04.51 | mckenziemc | Infus: not quite drop-in; you don't have an equivalent to next, which is what the other one was trying to implement and i think what nevcairiel meant by problems with trying to make a drop-in |
| 16:04.54 | Neffi | yes, that is a fine solution right there |
| 16:06.10 | nevcairiel | Infus: your iterator assumes its always called in order, which might work with a for loop, but is not guaranteed :) |
| 16:06.51 | nevcairiel | pairs/next work much more flexible by using the key as a state - which is where it becomes a problem |
| 16:13.22 | mckenziemc | but what do you mean by not guaranteed, nevcairiel? just that there may be other use cases that need the next style, or some case where an iterator function will behave unexpectedly? |
| 16:13.55 | Infus | yeah, it's not perfect, but it works for "for", and is convenient |
| 16:14.29 | Neffi | The Lua definition doesn't guarantee that the iterator function will be called sequentially, is what he's saying |
| 16:15.12 | mckenziemc | i don't understand |
| 16:16.36 | Neffi | Imagine a situation where the itator is invoked once, and the values cached, then invokved again. The 2nd invocation's returns get presented to the first iteration of the for loop, then the cached values. |
| 16:16.49 | Neffi | There is no such situation in any implementation of Lua that I know of, but the spec doesn't disallow this behavior |
| 16:16.50 | mckenziemc | then that's a design error |
| 16:17.01 | mckenziemc | oh |
| 16:17.04 | Neffi | No it's not. Such a design would be compliant against the language definition |
| 16:17.09 | mckenziemc | lua itself isn't guaranteed not to do that? |
| 16:17.15 | Neffi | doesn't appear so |
| 16:17.19 | mckenziemc | interesting |
| 16:17.23 | Neffi | the reference manual @ lua.org is the official language definition |
| 16:17.43 | mckenziemc | i don't know why they'd allow that but that explains the guarantee problem i guess |
| 16:18.07 | Semlar | stop breaking things neffi |
| 16:18.12 | mckenziemc | i know, and i don't see that statement but didn't search for it specifically |
| 16:18.12 | Neffi | It's kind of just pedantic, but it's technically right |
| 16:18.26 | Neffi | you almost certainly will never run into such an issue |
| 16:18.59 | Neffi | Semlar: I am trying to be like you for a day |
| 16:19.00 | Neffi | ;) |
| 16:19.18 | mckenziemc | do you know where the manual says it? |
| 16:19.22 | Semlar | that's a slippery slope neffi |
| 16:19.30 | Neffi | the manual _doesn't_ say it |
| 16:19.33 | Neffi | ;) |
| 16:19.34 | Semlar | you may find you can't stop |
| 16:19.51 | Neffi | here is where the semantics of the 'for' loop are presented: https://www.lua.org/manual/5.1/manual.html#2.4.5 |
| 16:21.04 | mckenziemc | then is the concern legitamate? o.O |
| 16:21.16 | Neffi | depends what your definition of legitimate is |
| 16:21.44 | Neffi | if we're specifically talking about WoW Lua then we might as well drop the manual and go on observed behavior, because it is a specific implementation which will adhere to its own behavior |
| 16:21.46 | mckenziemc | why would you be worried about that? |
| 16:21.53 | *** join/#wowuidev Ressy (Ressy@WoWUIDev/WoWAce/ARL/Ressy) |
| 16:22.39 | mckenziemc | if there's no statement of potential issue, how is it different from saying "an implementation might do this strange convoluted thing just because no mention was made preventing it?" |
| 16:22.53 | mckenziemc | or is the problem just that - an implementation might do it |
| 16:23.23 | Neffi | the potential issue isn't just theoretical. it may pop up in the future |
| 16:23.56 | znf | Local businesses suck. |
| 16:23.58 | Neffi | you avoid undefined behavior as a rule not because it's fun to be pedantic, but because observed behavior may change at any moment, and your only guarantee that something will work as expected is to stick to codified behavior only |
| 16:24.17 | Neffi | maybe the very next minor version bump will change the behavior! |
| 16:24.20 | mckenziemc | i'm not trying to be dismissive, just attempting to clarify nuance |
| 16:25.20 | *** join/#wowuidev Xinhuan (~Xinhuan@HearthSim/Community/Xinhuan) |
| 16:26.15 | Infus | I'm saying the reference manual says the for loop is equivalent to this code, and the code I posted works will always work with the equivalent code; so on that base I feel confident in that code |
| 16:26.19 | znf | I went to the mechanic today, went for a check-up, he gave me a list of parts to buy. He recommended me a local shop for parts. I go there, ask for the parts (didn't even ask for a quote, just wanted to buy/order them directly). He gets my phone number, tells me that he'll call me That was at 10am. I call him at 12, he says he hasn't received an official quote. I wait until 5pm, he tells me he "forgot" because he had a busy day. Tells me he will do |
| 16:26.19 | znf | some overtime and get back at me... It's 7:30pm now, no call. That should teach me from trying to make local business some money. Could have ordered them online and be done with it. |
| 16:26.25 | Neffi | mckenziemc: Honestly this debate kinda goes out the window altogether in WoW |
| 16:26.40 | mckenziemc | i'll have to take your position on this as indication that it doesn't fall into the farfetched enough to not worry about category |
| 16:26.45 | Neffi | There is no guarantee whatsoever because it implements its own dialect of Lua and it has absolutely no official documentation or guarantees |
| 16:26.55 | Neffi | So there is no defined behavior against which to code |
| 16:27.07 | mckenziemc | yeah, but i have interest in language design |
| 16:27.16 | Neffi | So it's perfectly reasonable to just go on observed behavior, IMO, in WoW |
| 16:27.42 | Neffi | But elsewhere, you never know what will change in the specific implementation you use int he very next version bump, let alone who might choose to do what in some other implementation |
| 16:28.14 | mckenziemc | unit tests! |
| 16:28.28 | Neffi | Especially since JIT compilers are all the rage now, you find alternate implementations popping up |
| 16:29.02 | Neffi | If your code is supposed to work in "Lua 5.1" then somebody runs in it an implementation which is perfectly compliant to the definition of Lua 5.1, and it's utterly broken, then your code is by definition wrong |
| 16:29.55 | mckenziemc | sounds a little ham-fisted |
| 16:30.11 | mckenziemc | claiming it's not possible for there to be an oversight in the design or remarks |
| 16:30.31 | mckenziemc | but could be considered an issue of nuance |
| 16:30.31 | Neffi | Not to mention, _because_ JIT compilers are all the rage, implementations may choose to behave strangely during cases of undefined behavior, if it lets them generate faster code |
| 16:31.18 | foxlit | moves all of Neffi's files to /dev/null. It's super-fast! |
| 16:31.22 | Neffi | and then of course if you look at AOT languages, compilers have been optimizing forever, so you always had to be careful there |
| 16:31.49 | *** join/#wowuidev Kultzi (~ironi@85-76-39-18-nat.elisa-mobile.fi) |
| 16:31.53 | Neffi | <mckenziemc> claiming it's not possible for there to be an oversight in the design or remarks <- I made no such claim though |
| 16:31.55 | Neffi | What do you mean? |
| 16:32.02 | mckenziemc | not you |
| 16:32.12 | Neffi | You think the spec specifically not making such a guarantee is an oversight? |
| 16:32.24 | mckenziemc | the general idea "then your code is by definition wrong" |
| 16:32.27 | mckenziemc | not necessarily |
| 16:32.37 | Neffi | It very well might be, and maybe it probably is, but the spec is the spec, and you have to conform to it if you want your code to be "correct" |
| 16:33.38 | mckenziemc | where "correct" may be hard to identify until a problem actually comes up |
| 16:34.08 | *** join/#wowuidev disc (disc@2a02:1660:410d::66) |
| 16:34.43 | Neffi | may be. you have to be careful |
| 16:34.51 | Neffi | nevcairiel identified that easily enough ahead of time |
| 16:35.07 | Neffi | I guess that's the difference between knowing the language and not |
| 16:35.35 | Neffi | Not everyone is going to be a master at every language they use, so sometimes you have to listen when somebody tells you "wait that's wrong" |
| 16:35.54 | Neffi | even if there is no observed bug at all! |
| 16:37.31 | nevcairiel | waiting for bugs is a bad way to write really |
| 16:37.38 | nevcairiel | i rather understand what i'm doing and do it right :p |
| 16:38.05 | mckenziemc | i'm not at my best thinking ability right now, but I think what i'm trying to say, to use an analogy, is that it would be silly to add an ad-hoc modification to a scientific hypothesis because of suspicion a future test "might" prove the hypothesis wrong |
| 16:38.47 | nevcairiel | but its easy to come up with a call to the function that breaks it right now :) |
| 16:38.54 | Neffi | but we're not really dealing with hypothesis here |
| 16:39.13 | Neffi | We already _know_ it's wrong. The truth is well established. Now we're just arguing if it matters in practice |
| 16:39.23 | nevcairiel | the point is, iterators are supposed to be state-less, as the state is passed in externally |
| 16:39.36 | nevcairiel | it happens to work in one case, but that doesnt mean its correct |
| 16:39.38 | Neffi | If I wanted to do some odd/clever shit in Lua and virtualize away for loops _right now_ I could break that |
| 16:40.00 | AcidWeb | There is some fast way to check if frame X overlap ANY other frame? |
| 16:40.02 | mckenziemc | but should it be the use of iterator, or the expectation that it will work in a situation that perhaps should not assume an iterator is not used, that is the problem? |
| 16:40.36 | mckenziemc | if i can break something, is it the thing that's defective, or that i shouldn't be breaking something with known limitations? |
| 16:41.04 | Neffi | I see what you mean |
| 16:41.08 | nevcairiel | well that comes then back to the point that you didn't write a proper iterator, but only one that works in specific circumstances |
| 16:41.14 | nevcairiel | so its not a drop-in replacement |
| 16:41.18 | nevcairiel | which was the original argument :p |
| 16:41.29 | Neffi | You are making this particular implementation and saying "this only works under these circumstances", so you're putting forth your definition for the constraints of this particular function |
| 16:41.30 | mckenziemc | that i agree with |
| 16:41.40 | mckenziemc | it's the tangent i'm focusing on |
| 16:41.58 | Neffi | I think it is still defective, because we could write a Lua implementation right now in which for loops work differently, but still adhere to spec, and break your function |
| 16:42.15 | Neffi | So your function is failing to adhere even to the constraints you place on it |
| 16:42.20 | Infus | and I say again: this iterator works with the for loop as it is specified in the reference manual; a lua implementation which doesn't follow that specification is wrong |
| 16:42.57 | nevcairiel | no its not |
| 16:43.04 | nevcairiel | lua specifies how iterators are supposed to work |
| 16:43.08 | mckenziemc | my proposed answer is that there's room for personal interpretation in Occam's razor :p |
| 16:43.17 | Infus | it also specifies how for loops work |
| 16:43.37 | Neffi | My proposed response is that interpretation is irrelevant |
| 16:43.59 | mckenziemc | iterators, or how the for loop works with what you give it, nevcairiel? |
| 16:44.00 | *** join/#wowuidev JTF195 (~JTF195@rapturewerks.net) |
| 16:44.00 | Neffi | If you don't adhere to the spec _as stated literally_ then you are wrong by definition |
| 16:44.05 | nevcairiel | Infus: irrelevant, the argument is that its a faulty iterator, irrespective if it happens to work in this particular for loop |
| 16:44.15 | mckenziemc | wrong, or non-compliant |
| 16:44.17 | Infus | it is a faulty iterator, no doubt about that |
| 16:44.23 | mckenziemc | denotation, connotation |
| 16:44.27 | Infus | but it's perfectly acceptable as use in a for loop |
| 16:44.34 | Infus | since it follows the specification of the for loop |
| 16:48.16 | Neffi | Steam's window is broken :( |
| 16:48.26 | Neffi | my compositor bitches about it and it seems to hurt performance when I have Steam open |
| 16:48.56 | Neffi | Semlar: why can't everything just work perfectly |
| 16:49.18 | Semlar | because that would be boring |
| 16:52.30 | mckenziemc | but is it steam's fault or your OS? ;) |
| 16:53.12 | Neffi | probably everyone's fault |
| 16:53.39 | Neffi | probably least of all the OS. maybe the compositor is broken |
| 16:56.50 | Semlar | probably gabe's fault |
| 16:57.30 | Neffi | yes, gabe personally |
| 17:04.40 | *** join/#wowuidev Seerah (~RyanNL@104-180-66-29.lightspeed.mmphtn.sbcglobal.net) |
| 17:04.40 | *** mode/#wowuidev [+o Seerah] by ChanServ |
| 17:13.44 | Semlar | simca, nothing uses name2 or name3 and only a single unit uses name1 |
| 17:15.12 | Semlar | i bet it's for other languages, maybe they have different variations on the names for different contexts |
| 17:33.25 | nevcairiel | some languages have a buttload of plural forms |
| 17:34.42 | *** join/#wowuidev MaxFrost (~MaxFrost@c-174-53-177-161.hsd1.mn.comcast.net) |
| 17:35.31 | mckenziemc | do they call them buttloads too? |
| 17:35.45 | nevcairiel | ie russian as different forms for 1, (234) and (567890) |
| 17:43.13 | foxlit | gendered nouns might be closer to home. |
| 18:12.35 | Simca | it's quite possible it was designed for data that simply no longer exists in the same form |
| 18:13.30 | Simca | huge swaths of npcs were removed in Cata during the old world redesign |
| 18:44.40 | *** join/#wowuidev Yoshimo (~anything@unaffiliated/yoshimo) |
| 18:49.55 | Cairenn | Happy Independence Day neighbours! |
| 18:55.17 | *** join/#wowuidev stolenlegacy (~stolenleg@chello080109199104.5.graz.surfer.at) |
| 18:55.17 | *** join/#wowuidev stolenlegacy (~stolenleg@unaffiliated/stolenlegacy) |
| 18:55.20 | karl_w_w | merry will smith day! |
| 18:55.38 | Cairenn | lol |
| 18:57.31 | karl_w_w | the day when our lord and saviour rose up and confirmed his big-screen status |
| 19:24.55 | Fisker | hugs Cairenn |
| 19:25.11 | Cairenn | hugs Fisker |
| 19:27.36 | mckenziemc | hugs Cairenn and builds a wall around Denmark. |
| 19:27.47 | Cairenn | hugs mckenziemc |
| 19:27.56 | karl_w_w | hugs Denmark and makes Fisker pay for it |
| 19:28.18 | *** join/#wowuidev Simca (~Simca@c-73-28-210-11.hsd1.fl.comcast.net) |
| 19:28.24 | mckenziemc | :O |
| 19:36.28 | *** join/#wowuidev Grum (~grum@irc.grum.nl) |
| 19:37.14 | znf | Has anyone figured out how world bosses actually work in Legion? |
| 19:38.17 | karl_w_w | you walk up to them and they try to kill you |
| 19:38.27 | znf | Spawn frequency I mean |
| 19:38.33 | znf | There's 11 of them |
| 19:39.03 | znf | But not all of them seem to be up at the same time, so do they have a fixed number of bosses that spawn? Do they rotate troughout the week. |
| 19:39.12 | karl_w_w | they're always spawned, just in a different location in the 4th dimension |
| 19:39.35 | znf | So you have no clue. Got it. |
| 19:39.52 | karl_w_w | you know too much |
| 19:39.56 | karl_w_w | good day sir |
| 19:40.05 | nebula | aren't they tied to world quests? |
| 19:40.26 | nebula | and the quest is around for 3-7 days |
| 19:40.51 | znf | I don't remember doing the world quests for them this week (maybe just one?) |
| 19:41.26 | znf | The others I just went and killed. Also some people claim that 3 of them are up at any time/week, but I recall doing 4 this week. |
| 19:41.36 | nebula | just going off of info i remember when they were talking about world quests and how you don't need to micromanage them, events like world bosses stuck around for several days |
| 19:42.11 | znf | I'm going go track them next reset |
| 19:42.26 | znf | also Sigils work in a weird way, so far this week I think I got at least 5 |
| 19:42.32 | znf | Wonder if it's a bug or... |
| 19:48.50 | znf | so, blizzard this expansion was all "yeah, no more weapons!" |
| 19:48.51 | znf | yay! |
| 19:48.56 | znf | but you need to get relics! |
| 19:48.58 | znf | oh... |
| 19:48.59 | znf | 3 of them |
| 19:49.00 | znf | oh... |
| 19:51.53 | mckenziemc | that's one way of keeping things "new and exciting", at least for those who don't know too much |
| 19:53.09 | mckenziemc | help Semlar |
| 19:53.15 | mckenziemc | i have a new litter of spiders :( |
| 20:03.58 | fewyn | how's everyone doing |
| 20:04.21 | fewyn | hope everyone is having a good 4th |
| 20:04.26 | fewyn | if you live in the land of freedom |
| 20:04.34 | fewyn | otherwise have a good day! |
| 20:09.13 | mckenziemc | it's still the 4th everywhere else that uses the Gregorian calendar |
| 20:09.19 | mckenziemc | time zones aside |
| 20:27.23 | fewyn | ugh i'm level 73 and i've done all the dungeon quests so far... i don't wanna quest |
| 20:27.32 | fewyn | there's alliance messing with the zones in lk :( |
| 20:28.00 | fewyn | i wonder... are the midsummer fires decent xp? |
| 20:28.22 | *** join/#wowuidev Sajaki (~Sajaki@242.145-66-87.adsl-dyn.isp.belgacom.be) |
| 20:33.59 | fewyn | hrm only 8400 xp |
| 20:36.28 | relaxok | lua question.. say you have something like local blah = myfunction() and that function is tbl = {0, 0}. return tbl; then something does for id, data in pairs(blah).. wouldnt id and data be 0 and 0? |
| 20:36.54 | nevcairiel | no, i would be 1 and 2, and data would be 0 and 0 |
| 20:37.21 | relaxok | can you explain how that works exactly? |
| 20:37.35 | Neffi | pairs(blah()) you mean? |
| 20:37.39 | Neffi | pairs(blah) would give you an error |
| 20:37.42 | nebula | id is the index, indexes start at 1 |
| 20:38.04 | relaxok | the error was that data was nil i think |
| 20:38.06 | nevcairiel | nebula: blah = myfunction(); already calls the function |
| 20:38.09 | Neffi | % for k, v in pairs{0, 0} do print(k, v) end -- will print 1,0; 2,0 |
| 20:38.10 | yalb | Neffi: 1 0; 2 0 |
| 20:38.19 | nevcairiel | er Neffi |
| 20:38.21 | nevcairiel | damn tab complete |
| 20:38.22 | nevcairiel | :D |
| 20:38.28 | Neffi | {0, 0} is really { [1]=0, [2]=0 } |
| 20:38.32 | nebula | was gonna say ;; |
| 20:38.53 | Mr_Rabies | i really hope they don't repeat valor for 7.0 :( |
| 20:38.53 | relaxok | let me make sure the error is what i think it is |
| 20:39.21 | Mr_Rabies | earning a necessary currency exclusively through content i hate is just the worst :( |
| 20:39.22 | nebula | Mr_Rabies: probably not until the end of content patches/the expansion |
| 20:39.27 | Neffi | nebula: Oh, I see. I thought it was a function declaration. My poor eyes |
| 20:39.33 | Neffi | nevcairiel* -- woops |
| 20:39.35 | nebula | >.< |
| 20:39.37 | Neffi | haha |
| 20:39.56 | Neffi | wise choice |
| 20:41.12 | n3bula | valor has been their way to increase player power instead of nerfing bosses like they did in the past |
| 20:41.43 | relaxok | sorry, you guys are right im sure.. data is expected to be a table in this code and its a nuber, which makes sense |
| 20:46.44 | relaxok | number* |
| 20:47.20 | relaxok | C_SharedCharacterServices.GetUpgradeDistributions() ha |
| 20:50.53 | Mr_Rabies | valor's nice, but the means to obtain it sucks, n3bula |
| 20:51.11 | Mr_Rabies | if you could get some via normal raiding it wouldn't be a huge deal |
| 20:51.19 | Mr_Rabies | but i really, really hate lfr |
| 20:51.28 | *** join/#wowuidev MentalPower (~textual@23.252.53.177) |
| 20:51.32 | nevcairiel | so do mythic dungeons, faster and more efficient way t oget valor |
| 20:51.33 | n3bula | if they do do valor again, i'm sure it's sources will be as diverse as AP is |
| 20:51.55 | Mr_Rabies | mythic dungeons also suck |
| 20:52.07 | Mr_Rabies | i really am so tired of wod's dungeons |
| 20:53.50 | Mr_Rabies | they're only fun to roflstomp on heroic mode |
| 20:55.07 | nevcairiel | if you raid hfc, you also roflstomp on mythic |
| 20:55.47 | nevcairiel | back when valor was still needed in higher quantities, we just did all wod instances in like 90 minutes straight through |
| 20:56.30 | nevcairiel | now .. i barely do the weekly event once in a while in case of rare upgrades |
| 20:56.40 | Mr_Rabies | you'd be surprised |
| 20:56.54 | Mr_Rabies | i get absolute paste-eaters anytime i try pugging mythics |
| 20:57.02 | nevcairiel | i have a guild :P |
| 20:57.04 | Mr_Rabies | it takes like 45 minutes to clear a single damned dungeon |
| 20:57.26 | Mr_Rabies | same here, but we can't really align our schedules outside of raid hours |
| 21:49.12 | *** join/#wowuidev Yoshimo3743 (~anything@2001:a61:11b:4e01:5c40:7d2a:d25e:4f55) |
| 21:52.52 | *** join/#wowuidev Yoshimo (~anything@unaffiliated/yoshimo) |
| 22:47.50 | *** part/#wowuidev Sajaki (~Sajaki@242.145-66-87.adsl-dyn.isp.belgacom.be) |
| 23:16.59 | Mr_Rabies | i don't have a lot of faith in blizzard here but i really hope the casc garbage collection comes out soon :( |
| 23:21.05 | Mr_Rabies | also not really sure why the prepatch was so damned huge on the ptr, it was like 20GB....no way the expansion is actually that large at start, yeah? |
| 23:22.15 | bshelden | I dunno. They edited some geometry in Stormwind and Ironforge. That means all of EK entirely, right? |
| 23:26.08 | fewyn | also typically they don't do major compression or anything |
| 23:26.11 | fewyn | iirc |
| 23:26.33 | fewyn | ptr is pretty much same thing as legion beta i think right? |
| 23:26.38 | *** join/#wowuidev Iriel (~Iriel@173-8-135-209-SFBA.hfc.comcastbusiness.net) |
| 23:26.38 | *** mode/#wowuidev [+o Iriel] by ChanServ |
| 23:26.39 | fewyn | except legion is closed off |
| 23:28.13 | Mr_Rabies | yeah, the files are all there, just drm'd out |
| 23:29.15 | fewyn | https://www.youtube.com/watch?v=iMUB1uYLv3M |
| 23:29.22 | fewyn | if anyone wants to see akita puppies |
| 23:29.34 | fewyn | (they were born yesterday) |
| 23:52.19 | mckenziemc | function ATAN2(x,y) return atan2(y,x) end |
| 23:52.29 | mckenziemc | converting spreadsheets to lua... |
| 23:53.03 | Simca | converting spreadsheets to lua arrays is quite easy |
| 23:53.10 | Simca | i maintain all my lua data in spreadsheets |
| 23:53.17 | mckenziemc | i'm converting the logic |
| 23:53.36 | mckenziemc | that's a temporary line |
| 23:53.39 | Simca | what, like rewriting Lua in Visual Basic Advanced? |
| 23:53.48 | mckenziemc | >.> |
| 23:54.08 | Torhal | mckenziemc: http://wow.curseforge.com/addons/libspreadsheet/ |
| 23:54.45 | mckenziemc | http://www.esrl.noaa.gov/gmd/grad/solcalc/calcdetails.html |
| 23:55.05 | mckenziemc | not for an addon, Torhal |
| 23:55.32 | Torhal | ¯\_(ã)_/¯ |
| 23:55.39 | Simca | aren't there far more common languages for math and science programming than lua? |
| 23:56.18 | Neffi | Julia is Lua-like |
| 23:56.31 | Neffi | I don't know that it's "common" though |
| 23:56.40 | Neffi | I hear Fotran is still very common for math/science stuff |
| 23:57.25 | mckenziemc | i don't know them, Simca |
| 23:58.52 | mckenziemc | and i don't think it would really help much |
| 23:59.59 | mckenziemc | these are actually from a book but i don't have it nor do i need it enough |
