Minetest logo

IRC log for #minetest-dev, 2015-04-09

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:04 hmmmm for SRP accounts, you could make it fail
00:05 hmmmm then recommend modders to use get_password_hash_ex() for all kinds
00:08 paramat joined #minetest-dev
00:09 est31 whatever, later change
00:09 est31 first I shoot at initial support
00:09 est31 so that you can logon onto a server with srp
00:09 est31 and it automatically upgrades upon first login
00:10 est31 the other things can be done later
00:14 paramat i'm hoping to push mgv6 snow biomes #2598 within a day or two, so i'm requesting that file isn't touched until then so i can avoid conflicts. it's ready for review and suggestions now but no rush
00:14 ShadowBot https://github.com/minetest/minetest/issues/2598 -- Mgv6: Add optional snow biomes by paramat
00:15 paramat left #minetest-dev
00:23 Zeno` joined #minetest-dev
00:23 hmmmm ahh!  too many things to review
00:23 hmmmm Zeno`, have you reviewed shadowninja's network changes yeT?
00:24 Zeno` only very briefly
00:24 hmmmm it's large and i had to put it on hold earlier
00:24 Zeno` Wait, which ones?
00:24 hmmmm multiple listen addrs
00:25 hmmmm and implicit ipv6 support
00:25 Zeno` yeah only briefly
00:25 hmmmm and code cleanup of semi-related things
00:34 Zeno` https://github.com/ShadowNinja/minetest/blob/multi-socket/src/test.cpp#L2029 seems as though it should be >= 20
00:34 Zeno` it's still WIP and that's copied code
00:41 Zeno` actually 20 is right. Now, what's for breakfast?
00:57 Megaf_ joined #minetest-dev
02:04 Hijiri joined #minetest-dev
02:29 hmmmm lol
02:30 hmmmm what do you guys think about replacing all instances of std::map<std::string, std::string> with StringMap
02:30 hmmmm I already made the typedef in mapgen.h because it's just sooooooo damn common
02:30 VanessaE anything that reduces code repetition is probably a good thing
02:31 VanessaE (especially something like that, which would probably be optimized away anyway)
02:43 Zeno` it's just a typedef?
02:44 Zeno` I dunno why not... that's what typedefs are for after all
02:46 est31 perhaps we can introduce a new typedef, unordered_map that maps to std::map for < c++11 and std::unordered_map for >= c++11
02:54 hmmmm maybe yeah
02:54 hmmmm I don't know, pilzadam said to keep C++03 support until 2017 which is when ubuntu 12 LTS ends
02:54 hmmmm but isn't that a bit too far off?  i don't want to keep moving the target
02:54 est31 thats why typedef
02:55 hmmmm as i recall, minetest can't compile as C++11 anyway
02:55 est31 perhaps its offtopic right now yea
02:55 hmmmm paramat:  https://github.com/paramat/minetest/commit/41986695ea58a35f521b5ae005a4077aa71fe47d#diff-4a657be8ef68ecadefec5fe941ff8f42R185  ?? why did you add spaces around the negative sign
02:58 VanessaE I'd have read that as "maximum value of an unsigned 32 bit int, minus 1"
03:01 VanessaE in which case the spaces are *technically* correct, if slightly misleading
03:01 hmmmm but that's not correct because it's a unary operator
03:02 hmmmm unary operators are supposed to have no spaces separating them from the operand
03:02 hmmmm and then a cast is a unary operator as well
03:02 VanessaE I'm guessing paramat didn't read it as unary then
03:03 hmmmm there you're impyling that the cast operator is the first argument of a binary operation
03:03 VanessaE something like that, remember my C++ knowledge is way less than yours
03:05 Zeno` the cast isn't even needed
03:09 Zeno` it would be if you were to do ~((u32)0) which is probably how I would have done it
03:10 Zeno` :D
03:10 hmmmm the reason why I do -1 is because it's less characters and we're pretty much guaranteed that signed numbers are stored using two's complement encoding in the standard
03:11 Zeno` Yeah, I guess when using things as a mask I like it obvious. There is no right or wrong; it's personal preference IMO
03:15 Zeno` -1 (without the cast) is less likely to cause a casual mistake though
03:17 hmmmm ahgh
03:18 hmmmm how do you make code comments in a pull request
03:18 Zeno` ?
03:18 Zeno` click on it
03:18 hmmmm look at what I did https://github.com/minetest/minetest/pull/2598
03:19 Zeno` lol
03:19 hmmmm oh that's what I did, isn't it...
03:19 hmmmm you need to comment from within the Files Changed tab
03:19 hmmmm pfth
03:19 Zeno` yes
03:19 hmmmm i need to get better at this
03:19 Zeno` but it's not tooooo bad
03:20 Zeno` context can still be found by looking at https://github.com/paramat/minetest/commit/41986695ea58a35f521b5ae005a4077aa71fe47d
03:20 Zeno` well, it's kinda bad
03:20 Zeno` heh
03:20 hmmmm okay.  never again.
03:21 hmmmm well, okay, I'll redo it
03:21 prozacgod joined #minetest-dev
03:42 Hunterz joined #minetest-dev
04:31 DFeniks joined #minetest-dev
04:32 paramat joined #minetest-dev
04:32 paramat thanks hmmmm very helpful =)
04:32 paramat the unary operator was a silly mistake
04:40 paramat detecting grass in 'find stone level' also not good but was added to enable this part of addMud https://github.com/minetest/minetest/blob/master/src/mapgen_v6.cpp#L661
04:40 paramat i will remove the grass detection
04:42 paramat that part of addMud would not have been triggered anyway before now
04:43 paramat so perhaps it can be removed..
04:45 paramat yeah it's redundant anyway because of what comes immediately after, will remove
04:53 paramat then perhaps in mtgame i'll add some dry shrub decos in the tundra
05:00 Zeno` is there invalid item id?
05:00 hmmmm hmm
05:00 hmmmm zeno`, no clue
05:01 hmmmm paramat, if it works without the stone detection then great
05:01 paramat okay will test
05:01 hmmmm erm
05:01 hmmmm i mean mud detection
05:01 paramat er yeah, that
05:01 hmmmm dirt_with_grass
05:01 hmmmm whatever you call it =]
05:02 paramat indeed, that, even
05:02 hmmmm i approve of this commit because it doesn't do anything horrible since biomes are already hardcoded in
05:02 hmmmm i'm just concerned about how confusing this would be to users and also how it may possibly hamper efforts to make it biome independent in the future
05:03 paramat yes i feel hardcoded biomes suit mgv6
05:03 paramat i saw in your todo the possibility of adding the biome API to mgv6
05:03 hmmmm you need to compare sections of map before and after your commits just to make sure nothing changed
05:04 hmmmm right now it needs to be done visually because it's pretty difficult to make a unit test which does this
05:04 paramat yes. i feel a bit paranoid about that
05:04 paramat need to carefully check
05:04 hmmmm what I like to do is take a screenshot from a previous suspected version, the current version, and like a 0.4.5 version
05:04 hmmmm the only things different should be the trees
05:05 hmmmm except the trees shouldn't actually be different, they are though because somebody wasn't very careful along the way...
05:06 paramat i plan to start a world in 0.4.12, transfer it to my branch, then check for border discountinuities
05:07 paramat working on mgv6 makes me a little nervous, don't want to screw it up =/
05:08 paramat i'm not sure the biome API would suit mgv6, that's why the pines are hardcoded and not schematics
05:08 paramat stuff like mudflow wouldn't transfer
05:09 paramat but a new mgv6 version made for the biome API, perhaps
05:10 paramat anyway thats for the future..
05:13 hmmmm yeah future stuff
05:13 hmmmm it can all be modularized in a way that works for everything
05:13 hmmmm nud flowing is so slow because it trashes the cache by doing that radius search
05:14 hmmmm s/nud/mud/
05:14 hmmmm that's sort of why I got rid of it with v7; it didn't actually help make things look much better either
05:15 hmmmm but the speed argument has been blown out of the water now that we have liberal usage of 3d noise which is actually still somewhat slow
05:15 hmmmm i need to find better ways of optimizing it
05:19 paramat i'm happy mud flow isn't in v5/v7, it causes huge problems at y = 47 which i have not yet solved in mgv6
05:23 Zeno` stwange things happening
05:26 * paramat is adbucted by aliensh
05:26 paramat left #minetest-dev
05:28 Zeno` can't be fixed
05:28 Zeno` will just have to stay as it is
05:29 Zeno` wield item being updated when you dig dirt. I guess it's always been like that :3
05:31 hmmmm huh?
05:32 Zeno` because the local inventory changes
05:33 Zeno` (i.e. dirt is added) which causes camera->wield(item); to be called
05:34 hmmmm well i can't check right now because my code is in a state of flux
05:34 hmmmm but okay =]
05:35 Zeno` lol
05:35 Zeno` I see what's wrong
05:35 Zeno` it's not a @$&*@&$@ item id
05:35 Zeno` why call the thing player item id :/
05:37 Zeno` it's a stupid inventory index
05:38 Zeno` oh well I'll fix it later
05:39 Zeno` actually it (the wield item) is being updated ~ every 2 seconds
05:39 Zeno` very odd
05:39 Zeno` just standing there doing nothing
05:43 hmmmm you know it's kind of messy how we have lua things only in the script files
05:43 hmmmm I wonder if it's time to break it
05:44 Zeno` what do you mean?
05:44 hmmmm would be very clear and neat having Schematic::deserializeFromLua() and Schematic::serializeToLua()
05:44 hmmmm in addition to Schematic::serializeToMts() and vice versa
05:44 Zeno` would that necessarily break things?
05:44 hmmmm why's that
05:44 Zeno` I'm asking you
05:45 Zeno` You said it would break "it"
05:45 hmmmm I don't know, I think there's some kind of restriction on including header files
05:45 hmmmm that's sapier's.. thing
05:45 Zeno` Not sure to be honest, but I can't think of a reason why it wouldn't be able to be done
05:46 hmmmm it would certainly break the "standard" of having all lua things ONLY in script/l_*.{cpp, h}
05:46 Zeno` oh yeah, true
05:46 Hunterz joined #minetest-dev
05:46 Zeno` I was thinking of break in a different sense
05:46 hmmmm that too
05:47 Zeno` It's quite nice them being all in one place
05:47 hmmmm there's a comment in big scary capital letters telling you to not include that header in any non-scriptapi files or else things will break
05:47 hmmmm so far i have obeyed it
05:47 Zeno` yeah I was reading that yesterday for some reason :-o
05:47 hmmmm but
05:47 hmmmm what's cleaner
05:48 Zeno` what's the header again?
05:48 hmmmm i think l_base.h
05:48 hmmmm or l_internal.h
05:49 Zeno` I can't even think why I would have been looking at that
05:49 Zeno` but I was
05:49 Zeno` must have been related to the damage tilt/flash thing
05:49 hmmmm ahh, c_internal.h
05:50 Zeno` strange
05:50 hmmmm and c_converter.h
05:51 Zeno` I was wondering yesterday why the comments didn't explain *why*
05:51 Zeno` heh
05:51 hmmmm i'd need that file at the very least to serialize/deserialize to lua
05:51 Zeno` pretty sure I said to someone it was probably to make them seem more mysterious
05:51 hmmmm lol
05:51 hmmmm maybe relative include paths
05:51 hmmmm i don't know
05:51 hmmmm so what's more organized:
05:51 hmmmm having all lua things in one place
05:52 hmmmm having all <schematic> serialize/deserialize routines in one place
05:53 hmmmm then again there's nothing stopping me from doing both except sanity :-)
05:53 hmmmm add the prototype to (de)serializeTo/FromLua in the Schematic class, define the methods IN scriptapi files
05:54 Zeno` I like all the lua stuff in one place tbh
05:55 Zeno` I've always liked that choice even though it's probably not the one I would personally have made
05:55 hmmmm yeah
05:55 hmmmm same here, i guess we should leave it
05:56 Zeno` maybe
05:56 Zeno` https://github.com/Zeno-/minetest/commit/32a464b7903bfbc34daa44bfd39a4a05806b4905
05:56 Zeno` check out the action stream
05:56 Zeno` I gotta go... have a good one
05:56 hmmmm okay
05:56 hmmmm later
06:06 OldCoder joined #minetest-dev
06:13 cib0 joined #minetest-dev
06:19 hmmmm :D i never cease to amaze myself
06:19 hmmmm I just coded a shitton and it all worked on the first try
06:21 sofar I hate when that happens ;^)
06:40 selat joined #minetest-dev
07:14 hmmmm so anyway I think i came up with a good way of linking different *Managers together
07:15 hmmmm there could be a global instance of ObjDefManager that can "see everything" in a sense, and each item can be accessed by namespaced reference
07:15 hmmmm so you have biomes with a name of say, "morebiomes:crazy_jungle"
07:15 hmmmm the global object name would then be "biomes:morebiomes:crazy_jungle"
07:16 hmmmm and to get the global biome manager it'd just be "biomes"
07:16 hmmmm i realize this sounds a lot like file/device objects in Windows but that's a decent setup that anybody would probably come up with in the end
07:20 est31 as long as it isnt as horribly documented as windows, I've heard horror stories bout that
07:21 est31 sounds good though
07:22 est31 (its also the way linking works in html)
07:32 hmmmm actually you know what
07:32 hmmmm this is a horrible idea.
07:32 hmmmm it's overcomplicated, doesn't solve any real problems, takes more code, and just moves the spaghetti part from compiled code into strings and lookup tables
07:36 est31 second part weighs most I think... Any news from RPC?
07:37 hmmmm no.  i haven't been working on that.
07:38 hmmmm it's funny, i spend half my day writing code for my job and i spread the other half amongst real life things, eating, minetest, etc.  and it's as if i get nothing accomplished anywhere
07:39 kilbith joined #minetest-dev
07:39 est31 sometimes I feel like that too, then I summarize what I actually did, and most times I get larger lists
07:40 est31 but yea my plan for today has been quite ambitious... accomplished only very small parts of it :/
07:41 hmmmm same
07:41 Calinou joined #minetest-dev
08:27 jin_xi joined #minetest-dev
08:28 cib0 joined #minetest-dev
08:38 kilbith joined #minetest-dev
08:55 Krock joined #minetest-dev
08:58 Darcidride joined #minetest-dev
09:01 roniz joined #minetest-dev
09:22 err404 joined #minetest-dev
09:51 * est31 wonders whether we need those additional two messages
09:51 Krock Yours and this message?
09:51 est31 (TOSERVER_PLAYER_NAME and TOCLIENT_AUTH_METHODS)
09:52 est31 in theory, they can be spared
09:52 est31 and perhaps would remove some complexity (and connection time) from the protocol
09:53 est31 one issue remains though: what when we don't want to require users to enter usernames?
09:55 est31 but two additional messages just because of this feature...
09:55 * est31 isn't sure
09:56 Krock Do you think Minetest will ever have a master server for logins?
09:57 est31 I dont know
10:02 est31 I think even if there was we could query it
10:02 est31 and future protocols where we have a master server can have blank usernames
10:03 est31 servers which dont support the masterserver will have to note it somehow
10:03 est31 and then the client has to connect otherwise.
10:04 est31 so the connection fails. Is avoiding that worth the additional lag when we dont implement masterlogin?
10:06 est31 ~tell nrzkt what do you think should be done in protocol 25? When the username is sent in TOSERVER_INIT, we can spare two additional messages. Therefore I would move playername sending there, and authentication (e.g. via srp, or public key) happens after TOCLIENT_HELLO. Does that sound good? Can I start coding?
10:06 ShadowBot est31: O.K.
10:32 Hijiri joined #minetest-dev
10:51 rubenwardy joined #minetest-dev
10:55 ninnghazad joined #minetest-dev
10:55 ninnghazad left #minetest-dev
11:06 cib0 joined #minetest-dev
11:19 SudoAptGetPlay joined #minetest-dev
11:45 rubenwardy joined #minetest-dev
12:12 Darcidride joined #minetest-dev
12:25 Player_2 joined #minetest-dev
12:42 roniz joined #minetest-dev
13:00 selat joined #minetest-dev
13:07 Zeno` joined #minetest-dev
13:07 Zeno` I have a problem
13:08 Zeno` About a week ago I stated here that there was what I thought an invalid DMCA takedown request on one project
13:08 Zeno` I did this because the project was compliant with the LGPL
13:09 Zeno` Now I am getting emails from strangers complaining they got takedown notices!
13:09 Zeno` And I have NFI how to respond
13:09 Zeno` stupid logs
13:10 Robert_Zenz joined #minetest-dev
13:11 Zeno` celeron55, wtf should I do?
13:13 Zeno` shit, it's equivalent to irrlicht issuing minetest a DMCA takedown request
13:24 celeron55 i don't know; it would be useful to know who is issuing such notices
13:25 Zeno` nerzhul
13:27 celeron55 eh, what? is he like spamming lawful projects with takedown notices?
13:27 celeron55 that doesn't even make any sense
13:28 Zeno` This is a notification that your application, MultiCraft II — Free Miner!, with package ID com.MoNTE48.MultiCraft2, has been removed from the Google Play Store.
13:28 Zeno` REASON FOR SUSPENSION: Alleged copyright infringement (according to the terms of the Digital Millenium Copyright Act)
13:28 Zeno` subject_lr_dmca: Your Request to Google
13:28 Zeno` full_name: Loïc Blot (nerzhul)
13:28 Zeno` companyname: minetest
13:28 Zeno` represented_copyright_holder: core-developper and maintainer
13:28 Zeno` contact_email_noprefill: loic.blot@unix-experience.fr
13:28 Zeno` country_residence: FR
13:29 Zeno` dmca_affirmations_authorized: agree
13:29 Zeno` dmca_affirmations_penalty: agree1
13:29 Zeno` signature: Loïc Blot (nerzhul)
13:29 Zeno` There is one
13:29 Zeno` want another 3?
13:30 rubenwardy Are we sure the follow LGPL?
13:30 rubenwardy *they?
13:30 Zeno` The did when they sent me that email
13:30 kilbith the original source was not shown in the description, so not
13:30 Zeno` I compiled the silly thing
13:30 Zeno` kilbith, it was
13:30 celeron55 so okay so why are these people sending complaints to *you*?
13:30 kilbith no, i'm sure of not
13:31 Zeno` celeron55, the only reason I can think of that they are sending them to be is because of the .ru logs of this channel
13:31 kilbith the link in the desc pointed to the Multicraft repo, not the Minetest repo
13:31 Zeno` yeah of course they linked to the multicraft repo
13:31 kilbith and there has been not even mention of the name "Minetest" at all
13:31 kilbith so the DCMA is valid
13:32 celeron55 that's the first thing that doesn't make sense
13:32 Taoki joined #minetest-dev
13:32 celeron55 eh what
13:32 Zeno` I really wish I'd kept the clone of the repo that I built to see if it built
13:32 celeron55 LGPL doesn't care about names
13:33 rubenwardy LGPL doesn't want attribution, it wants code sharing
13:33 Zeno` because it really did say that it was based on minetest even though they didn't have to
13:33 celeron55 people come on, don't use legal mechanisms like DMCA takedown notices if you do not understand the law
13:33 celeron55 you end up doing illegal things
13:33 celeron55 and also don't teach other people things that are incorrect
13:34 rubenwardy Nerzhul shouldn't act on behalf of Minetest like this without talking to other people
13:34 rubenwardy He's rogue
13:34 celeron55 he isn't acting on behalf of minetest; he can't be because minetest isn't a legal entity
13:35 celeron55 also it's possible that multicraft2 wasn't even using new enough code to contain any of his contributions which makes it absolutely derpy
13:35 rubenwardy <Zeno`> companyname: minetest
13:36 celeron55 well that's just plain illegal independently of anything else being or not being
13:36 rubenwardy I'm not talking about legal entities, I mean as a community/developing team.
13:36 celeron55 he has been a core developer so he can state that; it's informative
13:37 Zeno` There's a really cool project on the forums
13:37 Zeno` can't remember the name of it, but I think the author is Kenney
13:38 Zeno` is he gonna get a DMCA takedown as well when he charges $0.40 per download?
13:38 Zeno` I'm pretty sure he's expecting to make money from his wokr
13:38 Zeno` work
13:49 SudoAptGetPlay joined #minetest-dev
13:59 Taoki joined #minetest-dev
14:52 VanessaE joined #minetest-dev
14:56 celeron55 anyway i hope nrzkt reads this log or someone reminds him of this and he will learn by himself
14:57 VanessaE zwhat'd I miss?
14:57 VanessaE -z
14:57 celeron55 you're clearly missing this URL 8): http://irc.minetest.ru/minetest-dev/2015-04-09
14:58 VanessaE already on it :P
14:58 VanessaE (I was just being lazy :) )
14:58 VanessaE wait what?
14:58 VanessaE DMCA?
14:58 VanessaE O_o
15:01 VanessaE *facepalm*
15:05 hmmmm joined #minetest-dev
15:21 hmmmm joined #minetest-dev
15:42 MarisD joined #minetest-dev
15:56 kilbith joined #minetest-dev
16:10 err404 joined #minetest-dev
16:22 Hunterz joined #minetest-dev
16:24 disableclouds joined #minetest-dev
17:00 SudoAptGetPlay joined #minetest-dev
17:02 est31 joined #minetest-dev
17:03 MinetestForFun joined #minetest-dev
17:11 cib0 joined #minetest-dev
17:22 Player_2 joined #minetest-dev
18:20 Megaf joined #minetest-dev
19:00 Megaf joined #minetest-dev
19:15 cib_ joined #minetest-dev
19:24 Hijiri joined #minetest-dev
19:27 Gethiox joined #minetest-dev
19:52 err404 joined #minetest-dev
20:00 Amaz joined #minetest-dev
20:07 AnotherBrick joined #minetest-dev
20:24 casimir joined #minetest-dev
21:17 cib0 joined #minetest-dev
21:22 Casimir joined #minetest-dev
21:23 AnotherBrick joined #minetest-dev
21:38 AnotherBrick joined #minetest-dev
21:48 Casimir joined #minetest-dev
22:14 everamzah joined #minetest-dev
22:23 AnotherBrick joined #minetest-dev
22:38 AnotherBrick joined #minetest-dev
22:53 AnotherBrick joined #minetest-dev
23:08 AnotherBrick joined #minetest-dev
23:08 Hijiri joined #minetest-dev
23:13 Megaf_ So, I don't have to mention that my game is based on minetest? How is that so?  Of course I will give all the credits to a piece of software that my software is based on.
23:23 AnotherBrick joined #minetest-dev
23:38 AnotherBrick joined #minetest-dev
23:53 AnotherBrick joined #minetest-dev

| Channels | #minetest-dev index | Today | | Google Search | Plaintext