Time Nick Message 00:28 iio7 Hi. I was wondering which part of the code that "inserts" the item "unknown block" once a mod has been removed and a block goes missing? I can locate the "register" part, but as far as I can tell it justs registers the node. 00:59 _3SidedSquare Hello everyone, I'm trying to set up my IDE with the minetest code, is there an ide anyone can recommend? (I'm useing codeblocks atm) 01:33 ShadowNinja _3SidedSquare: Use whatever you prefer. I used vim, some use geany, and others use Code::Blocks. 01:33 ShadowNinja use* 01:35 hmmmm i use code::blocks 01:35 hmmmm it's nice, but it really could be nicer 01:35 hmmmm sapier uses eclipse 01:52 _3SidedSquare thanks, I'm trying to get codeblocks set up now, is there away to disable compileing the android version? I'm getting 01:53 _3SidedSquare g++ -Wall -g -c /home/akamanah/minetest/minetest/build/android/src/org/minetest/minetest/MinetestAssetCopy.java -o obj/Debug/build/android/src/org/minetest/minetest/MinetestAssetCopy.o 01:53 _3SidedSquare g++: error trying to exec 'ecj1': execvp: No such file or directory 02:06 hmmmm it shouldn't try to target android... 02:06 hmmmm are you using gentoo or something? 02:08 _3SidedSquare nope, ubuntu 02:17 _3SidedSquare I think it's a problem with misconfigured code::blocks, I manged to build it useing cmake/make 02:29 kahrl_ doesn't code::blocks use cmake to build? 02:29 kahrl_ oh, or actually, I think cmake has an option to create a code::blocks project 02:29 kahrl_ you might want to use that 02:46 _3SidedSquare thanks, kahrl, got cmake to build a nice codeblocks project 12:11 kahrl https://github.com/minetest/minetest/pull/1631 just a little mmdb fix 13:30 kahrl out of curiosity: is there an up-to-date version of the VBO patch around? 13:30 kahrl I tried rebasing the newest one I'm aware of 13:31 kahrl it seems to work fine at first but after a while of playing, you'll see mirages 13:31 kahrl i.e. terrain from elsewhere 13:31 kahrl (I suspect that's because of nore's base coordinate changing system) 13:49 Krock how about adding irrklang to support more sound formats? 13:49 sfan5 what is irrklang? 13:49 sfan5 irrlichts sound library? 13:49 Krock seems so. it klings. 13:50 Krock http://www.ambiera.com/irrklang/ 13:50 sfan5 it's not open-source is it? 13:51 Krock ack. a "buy now" site. R.I.P idea. 13:51 sfan5 there is a free version 13:51 Krock does not mean, it's FOSS 13:51 sfan5 but have fun convincing anyone of adding a proprietary library to minetest 13:52 T4im seems freeware by the looks of the license 13:52 T4im "The contents of the irrKlang distribution archive may not be redistributed, reproduced, modified, transmitted, broadcast, published or adapted in any way, shape or form, without the prior written consent of the owner, Nikolaus Gebhardt." 13:52 Krock mhm I don't like to run against walls 13:52 sfan5 may not be redistributed? 13:52 sfan5 welp 13:52 sfan5 how do you include the dll? 13:52 jin_xi i'd look into sending OSC messages via lua or so 13:52 T4im nope.. used yes, not redistributed 13:53 T4im oh well, you can apparently redistribute the compiled dll's 13:53 T4im still 13:53 Krock with the permission of the owner, it wouldn't be a problem 13:53 T4im incompatible with (l)gpl anyway 13:54 kahrl what's wrong with .ogg? 13:54 sfan5 it's not like we need .mp3 or .aac 13:54 Krock kahrl, nothing but it would allow to play .MD and such files 13:55 sfan5 those would make it potentially non-free anyway 13:55 sfan5 md? 13:55 T4im what is md? 13:55 T4im O_o 13:55 sfan5 md is markdown 13:55 sfan5 did you mean modplug? 13:55 Krock *mod 13:55 sfan5 why not libmodplug? 13:55 Krock possibly called modplug, yes 13:56 Krock nvm. it was only a brainstorm 13:57 sfan5 whats wrong with saving them in a .ogg? 13:57 Krock file size. 13:58 sfan5 does every device have enough processing power to play .mod? 13:59 iqualfragile is filesize really an issue? 13:59 sfan5 yes 13:59 T4im ogg has not much to do with filesize.. it's the underlaying codec that has.. vorbis, opus, flac, pcm.. depends on what you use 13:59 sfan5 if the media download takes ages people are more likely not to join a server 14:00 sfan5 right now .ogg = vorbis because we only support that 14:00 T4im ah, ok 14:00 sfan5 opus support would be nice tho' 14:00 Krock life is life. /off 14:01 T4im "Unlike Ogg Vorbis, Opus does not require the definition of large codebooks for each individual file, making it preferable to Vorbis for short clips of audio." 14:02 T4im with short files it might actually make a size difference then 14:03 sfan5 and opus can use less bitrate and be better 14:05 T4im hmm regarding resource downloads.. does minetest already support out of band downloads? 14:06 sfan5 out of band? 14:06 iqualfragile T4im: yes, kinda 14:06 iqualfragile you can use a http server 14:06 T4im ah, ok.. 14:11 ElectronLibre Hello everyone. I would like to have your opinion about a little modification that I sent in a pull-request (Here : https://github.com/minetest/minetest/pull/1591 ). Somebody told me that it was the best place to know why nobody have seen it. Thank you. 14:12 sfan5 1) why did you remove the newline? 2) should be "You can't teleport any other players. (missing privilege: bring)" 14:14 nore I'm ok with the pull if that's fixed 14:14 sfan5 me too 14:15 ElectronLibre Hmmm... The newline.. Oops.. I never let blank lines in my programs, so I removed it here sorry. I change this and I modify the message thank you sfan5 :). 14:15 ElectronLibre *in my files 14:19 kahrl Well, it's been a week I think but I didn't hear back from hosted weblate 14:19 ElectronLibre Can I use "" instead of ''? I'm afraid that the ' of can't could be interpreted as the end of the string... 14:19 kahrl (Unless my crappy webmail provider deleted the email) 14:20 nore ElectronLibre: yes 14:20 kahrl OTOH, if I were to try to install it myself, I couldn't because https://docs.weblate.org/en/latest/admin/install.html sends a wrong certificate :/ 14:21 ElectronLibre Ok. I've modified the file. 14:22 Krock ElectronLibre, a soulation for ' ' would be (example) 'You can\'t teleport' 14:23 ElectronLibre Yes, but I prefer using "". If you want I can modify it again.. 14:35 ElectronLibre The Travic CI build is in progress... Passed. 14:36 kahrl I don't think Travis actually checks the lua syntax :P 14:37 ElectronLibre No. But at least I didn't created problems ^^. I tested it again and again in solo mod, it normally works perfectly. 14:37 kahrl what does normally mean? 14:38 ElectronLibre Sorry, wrong translation. It could works. 14:39 kahrl still not sure I understand ;) can you say it in French? 14:40 ElectronLibre Ça devrait normalement parfaitement fonctionner. 14:42 kahrl ah okay. So are there edge cases where it wouldn't work? 14:42 kahrl Because from the code it looks like it would work all the time 14:45 kahrl OK, not counting the cases where the network would go down in the middle of the operation, etc. 14:46 ElectronLibre Hmm.. What have I to add so? Because it only change that a different message is sent if the player hasn't got bring privilege. I can add a if player ~= nil? 14:51 kahrl well that check would never be satisfied since there is no variable called player in this function 14:52 ElectronLibre I think it's minetest.get_player_by_name(name)~=nil . 14:52 kahrl looks like the code above already checks for that 14:53 ElectronLibre I have to verify if the player exists, haven't I? 14:53 kahrl although yeah, you have a point 14:53 sfan5 kahrl: https://weblate.readthedocs.org/en/latest/admin/install.html 14:53 sfan5 that link works 14:54 kahrl if the code reaches your error message, it might be because both get_player_by_name returns nil and missing bring privilege 14:54 kahrl but I can't think of any case where that would happen and complaining about the missing privilege would be completely wrong 14:55 kahrl so I say your code is fine 14:55 ElectronLibre Thank you :). 14:55 ElectronLibre It's not a very huge code, si there is not a lot of cases. 15:00 * twoelk wonders why ElectronLibre does not simply say: can not 15:00 * ElectronLibre wonders too. 15:02 kahrl sfan5: ah, nice 15:02 kahrl what distro is your server running btw, so I can try installing it on that? 15:03 sfan5 servers.minetest.net uses ubuntu 15:03 kahrl 14.04? 15:04 sfan5 yes 15:05 kahrl alright, I'll try to install it on my test machine then 15:25 ElectronLibre Thank you for having take a look on my pull-request. Bye. :) (A netsplit?) 19:41 sfan5 http://meow.minetest.net/tmp/serverlist_stats_2014-09-2014.txt 19:41 sfan5 statistics for the last 10 days of the serverlist 19:42 sfan5 ^ hmmmm, kaeza, ShadowNinja, celeron55, VanessaE_ 19:43 VanessaE_ 0.4.7 or older -> 281937 52.79% 19:43 VanessaE_ O_o 19:44 kaeza >Mobile -> 175286 71.97% 19:44 Amaz iOS -> 116447 66.43% 19:44 VanessaE_ why am I suddenly reminded of a certain Redmond OS? 19:44 sfan5 looks like we have to focus on mobile 19:44 sfan5 :P 19:45 VanessaE_ particularly the usage of that one we all wanna forget from about 13 years ago.. 19:46 Miner_48er where are these statistics from? 19:46 sfan5 server list 19:46 sfan5 read the first 3 lines... 19:46 Miner_48er which server list? 19:47 sfan5 http://servers.minetest.net 19:52 celeron55 the mobile usage is ridiculous 19:52 sfan5 indeed 19:53 sfan5 and the version adoption is just sad 19:53 VanessaE_ still think co-opting the serverlist for issuing "update your busted-ass install" is a bad idea? :P 19:54 VanessaE_ (I guess only sapier really spoke out against that) 19:54 sfan5 PilzAdam did too 19:54 ShadowNinja celeron55: All ARM usage is counted, so that includes, eg, Rasbery Pi, but not x86 phones. 19:54 VanessaE_ oh yeah 19:55 sfan5 ShadowNinja: I don't think there are 1) many people playing on an ARM devce thats not android or 2) many x86 phones 19:55 ShadowNinja I don't think supporting anything more than the last two stable versions is needed. 19:55 sfan5 "iOS -> 116447 66.43%" 19:55 sfan5 iOS is detectable without any false-positives 19:56 sfan5 Desktop -> 68212 28.00% 19:56 sfan5 still much more than desktop 19:56 VanessaE_ sfan5: I wonder, how many of those were kindles? 19:56 sfan5 how do I detect a kindle? 19:56 sfan5 there's not much to see 19:56 VanessaE_ I'm not sure - it's an ARM device running Android, I think, but that's about all I know of it 19:57 sfan5 [redacted] - - [14/Sep/2014:19:56:10 +0000] "GET /list HTTP/1.1" 200 61514 "-" "Minetest/0.4.10 (Linux/3.1.10-1197491 armv7l)" "-" 19:57 VanessaE_ Mozilla/5.0 (Linux; U; Android 2.3.4; en-us; Kindle Fire Build/GINGERBREAD) AppleWebKit/533.1 (KHTML, like Gecko) Version/4.0 Mobile Safari/533.1 19:57 sfan5 thats how the average android request looks 19:57 VanessaE_ ^^^^^ kindle fire 19:57 sfan5 thats browser not minetest 19:57 VanessaE_ true. 19:58 VanessaE_ haven't the foggiest idea how a request from such a device would look, but I ask because a lot of the players in the Inchra servers use them 19:59 sfan5 the look like any other requests 19:59 VanessaE_ and they always seem to have problems with this or that or the other thing. 19:59 sfan5 but with different linux kernel version number 20:00 ShadowNinja > Mobile Safari/533.1 20:00 ShadowNinja Browsers pretending to be other browsers. 20:00 sfan5 every browser is Mozilla/5.0 20:01 ShadowNinja sfan5: Why? 20:01 sfan5 ShadowNinja: http://stackoverflow.com/questions/5125438/why-do-chrome-and-ie-put-mozilla-5-0-in-the-user-agent-they-send-to-the-server 20:02 sfan5 or directly http://webaim.org/blog/user-agent-string-history/ 20:03 ShadowNinja That shouldn't be needed anymore. 20:04 sfan5 "And thus Chrome used WebKit, and pretended to be Safari, and WebKit pretended to be KHTML, and KHTML pretended to be Gecko, and all browsers pretended to be Mozilla, and Chrome called itself [...], and the user agent string was a complete mess, and near useless, and everyone pretended to be everyone else, and confusion abounded." 20:05 sfan5 it shouldn't, right 20:05 sfan5 but it is 20:37 VanessaE_ [09-14 16:37] btw, I reverted my minetest_game on mt-deezl to an earlier version, and the lag problem has gone way down 20:38 VanessaE_ where "earlier" means pre-_next 20:38 VanessaE_ so it's not just me. 20:46 hmmmm agh 20:46 hmmmm people are casually picking up minetest for the mobile platform and thinking, "what a shitty game." 20:46 hmmmm how difficult is it to install mods and so on with the mobile platform? 20:46 VanessaE_ it's a pain in the ass. 20:47 VanessaE_ because you have to get them from the store, which doesn't always work, and when you get them, they're often out of datte 20:47 VanessaE_ date* 20:47 VanessaE_ and when installed, they are not enabled by default 20:47 Megaf Segmentation fault 20:47 Megaf thats really annoying 20:47 VanessaE_ Megaf: offtopic. 20:47 Megaf not really, it happens because of bad code 20:47 Megaf in minetestserver 20:48 VanessaE_ Megaf: this topic is about the client, not the server. 20:48 ShadowNinja The reason they're out of date is that you have to manually package and approve them so that you know thay're secure. My mod security patch would fix that. 20:48 VanessaE_ ShadowNinja: yeah but you remember the ...spirited discussion we had about that. 20:49 VanessaE_ it's too G*d damned much work for the modder to constantly have to re-update their mods on the store. 20:49 VanessaE_ no live git pull, no sale. 20:49 ShadowNinja VanessaE_: Yea, what you could do is add a DL URL. Modt mods would point it to github's auto-archive link. 20:50 ShadowNinja Most* 20:50 VanessaE_ besides, you know your patch, which is probably sufficient, won't be enough to sway iqualfragile's opinion that us modders are somehow up to something nefarious. 20:50 VanessaE_ s/us/we/ 20:50 iqualfragile reading 20:50 VanessaE_ hah 20:53 ShadowNinja VanessaE_: Doesn't matter if the modder's up to no good if he can't do anything (other than cause craches and lockups, which are bad but not horrible). 20:54 ShadowNinja On Android it doesn't really matter though, apps are well sandboxed at the OS level. 20:54 ShadowNinja (Assuming you don't request a bunch of privileges) 20:57 Megaf OffTopic, how do I make sure that Im using LuaJIT? cmake says: -- Performing Test VALID_LUAJIT_VERSION - Success and -- LuaJIT version ok 20:58 VanessaE_ hah.. found the phrase I wanted... מלאך טויט קומט (malech muvis kimt, "angel of death comes") 20:58 ShadowNinja Megaf: Check if the "jit" global is available? 20:58 ShadowNinja s/?/./ 21:00 Megaf ShadowNinja: I dont get it 21:00 iqualfragile sorry, the radeonsi drivers don't really seem to have that whole dpm thing figgured out yet… 21:01 iqualfragile VanessaE_: I just assume that there can be nefarious modders 21:01 iqualfragile and assuming anything else would be insane 21:01 iqualfragile ShadowNinja: yes, having these patches would be great 21:01 iqualfragile that would enable me to add an "create version on git commit" option 21:02 ShadowNinja Megaf: `if jit then print("LuaJIT is being used!") end` 21:05 ShadowNinja iqualfragile: That would be better, but it would be best if you simply made it use the archive for a branch. Those that make stable releases would point it to `stable` and those that are rolling release would point it at `master`. You should use a simple archive link for this so that it will work with GitHub/BitBucket/Gitorious/etc. 21:05 Megaf I still dont get it... the system have jit libs and cmake found them 21:05 Megaf theres is no jit command 21:06 iqualfragile can this be merged please: https://github.com/minetest/minetest/pull/1631 21:06 ShadowNinja Megaf: Not jit command, jit table in the Lua environment 21:06 ShadowNinja iqualfragile: Why is the rating called "value"? 21:07 iqualfragile ShadowNinja: because its the value of the mod 21:07 ShadowNinja Shouldn't the MMDB send it as "rating"? That's much more descriptive 21:07 ShadowNinja I'd consider a mod's "value" to be it's code. 21:08 ShadowNinja Unless you mean it in the sense of "percieved value". 21:08 iqualfragile at the moment values contents is just positive revievs-negative reviews 21:08 iqualfragile in my mind that can be called rating 21:08 ShadowNinja In which case the MMDB is really talking about how much value the reviewers think it has. 21:09 iqualfragile but i planned on using an a bit more sophisticated algorithm later on 21:09 iqualfragile not sure if that should still be called rating then, im not a native speaker 21:09 iqualfragile ShadowNinja: yes, that was the idea 21:09 ShadowNinja I think it should be stored in "rating", and should probably be a simple mean average if reviews. 21:09 ShadowNinja of* 21:10 iqualfragile reviews are positive or negative 21:10 ShadowNinja Oh, right. You should add a 1-5 star system or similar too. 21:10 iqualfragile no 21:10 iqualfragile youtube does not use that anymore for a reason 21:11 ShadowNinja Being? 21:11 iqualfragile people don't really use it correctly 21:13 ShadowNinja *shrug* whatever. The meaning shouldn't change though. You could make it a 0-1 float or a integer than can be (theoretically) infinitely positive or negative. 21:22 ShadowNinja Hmmm, a Json::Value is converted into a struct ModStoreModDetails, which at some point is converted into a Lua table. Shouldn't the conversion be direct? It would require a lot less conversion code and Lua can check the validity of the data with much less code. 21:27 iqualfragile ShadowNinja: what is porting::path_user? 21:28 ShadowNinja iqualfragile: It's .minetest/ or PATH_RUN_IN_PLACE/ or wherever user-specific data shoulc be stored. It should be writable too. 21:29 iqualfragile so what exactly stops me from putting my malicious code into one of the mods that are declared secure? 21:33 VanessaE_ what stops a person from doing the same on the forums? 21:33 iqualfragile ? 21:33 VanessaE_ you're acting as if the store is the *only* way to install a mod 21:34 iqualfragile wtf? 21:34 iqualfragile this is not about the mod store 21:34 iqualfragile its about sapiers patches 21:34 iqualfragile s/sapier/ShadowNinja/ 21:34 VanessaE_ eh 21:35 * VanessaE_ just shuts up 21:36 ShadowNinja iqualfragile: Nothing. But things won't work. For example os.execute() is nil and require() always thrown a lua_error(). 21:36 ShadowNinja throws* 21:36 iqualfragile ShadowNinja: as far as i understood there is a user configuralbe exemption list, right? 21:37 blaise malicious in the machines security sence, or the server? 21:37 ShadowNinja iqualfragile: Yep. 21:37 iqualfragile on these mods its possible to os.execute()? 21:37 ShadowNinja iqualfragile: Yes, and other things of the like. 21:38 iqualfragile and its possible to edit the init.lua of trusted mods from untrusted mods when they are in the user path? 21:39 ShadowNinja iqualfragile: Hmmm, yes. I think I'll just disallow all path_user access other than to worlds/ and the mod's modpath. 21:40 ShadowNinja But on run_in_place builds path_share == path_user, so that will have to go too. 21:40 ShadowNinja path_share is usually readOnly anyway, so that probably won't be an issue. 21:41 iqualfragile run in place is afaik how most people run their servers 21:41 ShadowNinja Oh, I've got to finish up this minetest.compress() patch forst though... 21:41 iqualfragile oh, removing it won't be a problem 21:41 iqualfragile sorry, missunderstood 21:43 iqualfragile sfan5: do the linux numbers only include desktop linux or also android? 22:38 ShadowNinja iqualfragile: https://github.com/ShadowNinja/minetest/commit/da0578e2116c58e53100671ee69f016f76333ed6 and https://github.com/minetest/minetest/pull/1633 23:10 ShadowNinja My security pull is officialy the most commented Minetest issue of all time with 61 comments, second place goes to sfan's get_player_sysinfo pull with 57, and far behind in third is Selat's circuit simulator pull with 30. 23:13 VanessaE_ heh 23:13 ShadowNinja sfan5: How many index page fetches were there? 23:13 ShadowNinja (Of the server list) 23:14 ShadowNinja I think that would be more interesting becuase the index page doesn't automatically get refreshed.