00:00.31 | ZorbaTHut | btw this isn't in yet, buuuut keep your eyes out for a lua zlib binding to show up in the next PTS |
00:00.37 | seebs | What is the downside of database mining, exactly? I mean, given that we obviously have the ability, across the player base, to develop a very close to complete list. |
00:00.39 | ZorbaTHut | which might help your data requirements a bit |
00:00.39 | seebs | Ooooh. |
00:01.14 | ZorbaTHut | you can develop a complete list of quests that players have seen |
00:01.18 | seebs | And once we have all that done, it's not obvious to me that we're noticably worse off than we would have been if we coulda just used IQD to get it in the first place. |
00:01.28 | ZorbaTHut | this is not the same thing as a complete list of quests in the quest database. :) |
00:01.51 | seebs | Well, everything in IQC is definitionally a quest players have seen. |
00:02.05 | seebs | Do you track shard/world firsts for quests? |
00:02.18 | seebs | If so, "players have seen" is easily determined. Otherwise... hmm. |
00:02.41 | seebs | It would be SUPER convenient if it worked for "anything in IQC", because otherwise people are gonna spam up databases of that and act as though it always worked anyway. |
00:02.43 | Imhothar_ | yes they do |
00:02.51 | Imhothar_ | the XMl files have "firsts" for evry shard |
00:02.54 | ZorbaTHut | there's no reason that IQC doesnt include the checksum, besides "the client doesnt have that data" |
00:03.03 | ZorbaTHut | which it doesn't :) |
00:03.06 | seebs | But I admit, I can't think of any reason for which I *need* to be able to look up a quest that's not in the XML. |
00:03.08 | ZorbaTHut | (hmm, my apostrophe is sticky) |
00:03.36 | seebs | I was thinking it'd make sense to have the server give an answer for any IQD that matches a known-seen quest. Which, given first-tracking, is doable. But slow. |
00:03.57 | seebs | ... but now I have to run out for dinner. Back later! |
00:03.58 | Imhothar_ | I think the issue is people spamming the server with thousands of quest requests |
00:04.11 | seebs | And thanks a TON, Zorba, for working so hard to make the API pleasant and effective. |
00:04.29 | seebs | I continue to have a great deal of fun playing with it. |
00:04.52 | ZorbaTHut | imhotar, a possible issue as well, yeah |
00:06.32 | Imhothar_ | but now i wonder, what would be a practical scenario for zlib in the current state of the API? All I can think of for now is SavedVariables. But needing zlib for savedVariables would mean there was a whole lot of stuff in memory before. And now I might start guesstimating if the zlib binding is just a step in getting something even more awesome soon(TM) |
00:08.06 | Kreiri | does zlib mean that saved vars would not be readable/editable manually anymore? |
00:11.50 | Imhothar | @Kreiri what just come to my mind: You could provide the Quest details for certain areas as separate addon modules for people to download separately |
00:12.47 | Kreiri | when someone wnats to know which wuest they did and which didn't, they want at least names of the quests. |
00:13.19 | Kreiri | at this point all I can do is give them quest ids. |
00:14.52 | Imhothar | well its a start |
00:15.38 | Imhothar | the ids can be typed in the url of http://rift.zam.com/en/quest/#### for example |
00:16.02 | Imhothar | not a perfect solution, but its something to start with |
00:17.25 | Kreiri | typed. manually. decimal representation of 8-digit hex numbers. |
00:17.55 | Imhothar | you could dump the url in a textfield |
00:18.39 | Imhothar | its either that or loading a 22 MB Lua file, you decide |
00:18.46 | Kreiri | Isn't much better. looking at zone quest list and trying to remember what I did not do yet is less bothersome. |
00:18.58 | Imhothar | of course it is |
00:19.58 | Imhothar | how long does loading the file in Rift take? |
00:20.18 | Kreiri | I didn't try loading it in Rift yet. |
00:20.46 | ZorbaTHut | kreiri, it'll just be a pair of functions you can call to compress/decompress data. Obviously compressed data won't be all that readable, but it's up to the addon author to determine what to do. |
00:21.44 | Kreiri | fwiw, file:write part of my script took a couple of seconds. |
00:22.26 | Imhothar | but I wonder, the Quest.xml here has 64MB, how comes a Lua file with only the quest names takes up 1/3 of the original XML? |
00:23.06 | Kreiri | I discarded discoveries data |
00:24.14 | Imhothar | No I mean, how comes it's so big, wouldn't have expected 33% of the XML to consist of quest names |
00:24.44 | Kreiri | names, description, rewards, zone, et cetera |
00:24.52 | Imhothar | ah I see |
00:25.56 | Kreiri | for comparison, file with quest ids only in ["1744483310"] = true format is 121 KB |
00:28.10 | Imhothar | now here we have a use case for sqlite. even if limited to select only |
00:30.21 | Kreiri | an example of quest description in my 22 MB file: http://pastebin.com/EaiXaixX |
00:44.39 | Imhothar | well there's quite some stuff that's probably not relevant for just showing people which quests they haven't completed yet. everything else is on the interwebs |
00:48.36 | Imhothar | gnah, why can't item types have the simple ItemKey as in the XMLs |
00:48.39 | Imhothar | this is driving me nuts |
01:08.37 | Imhothar | is it indended for the the "type" of a rune enchanted item to include the rXXX field? I thought the "type" was meant to represent the raw unmodified item type |
01:52.30 | *** join/#riftuidev Mere (~Mere@buzzbee.plus.com) |
01:53.47 | Mere | hi |
02:07.13 | Imhothar | hi |
04:38.17 | *** join/#riftuidev seebs (~seebs@74.116.16.132) |
04:38.17 | *** mode/#riftuidev [+o seebs] by ChanServ |
04:55.32 | seebs | So I've been thinking about zlib support, and I am not sure what I would use it for. I've thus far simply ignored data size issues for savedvariables because everything I want to save is too small to measure on modern computers anyway. I guess it'd be a lot more noticeable for something like a large auction database. |
08:59.43 | *** join/#riftuidev Kreiri (~Miranda@109.227.122.218) |
12:20.30 | *** join/#riftuidev Mere (~Mere@buzzbee.plus.com) |
12:57.24 | *** join/#riftuidev Mere (~Mere@buzzbee.plus.com) |
13:17.04 | *** join/#riftuidev ZorbaTHut (tocohl@maximilian.pavlovian.net) |
13:17.05 | *** mode/#riftuidev [+o ZorbaTHut] by ChanServ |
14:38.42 | *** join/#riftuidev Imhothar (bcae988f@gateway/web/freenode/ip.188.174.152.143) |
16:14.46 | *** join/#riftuidev Mere (~Mere@buzzbee.plus.com) |
17:56.05 | *** join/#riftuidev Sunspots (~Sunspots@c213-89-195-33.bredband.comhem.se) |
20:51.28 | *** join/#riftuidev Aurorus_ (~Aurorus@ppp121-44-94-47.lns20.syd6.internode.on.net) |
22:25.36 | *** join/#riftuidev Aurorus (~Aurorus@unaffiliated/aurorus) |