Time Nick Message 08:46 yunfan do you have any protocol libraries or others that can let me build an headless bot? 08:52 VanessaE yunfan: you mean like irc + irc_commands mods? 08:55 yunfan VanessaE: nope, i mean control real role to exploring server 08:56 yunfan it might be very interesting to made some intelligence ai 08:57 VanessaE not sure I understand the purpose 08:59 VanessaE you can tell the client to just do random movements, but that still requires a "head". you could also just write a mod that would do the same thing; there are a couple of old mods that are basically primitive AIs that so stuff like building things 08:59 yunfan VanessaE: yep, but still i dont want to do real rendering 09:00 yunfan and for eg, i might offer dig worker service to player, which i will host a bot who can dig as the player require and get the tool from arround's chest 12:29 yunfan VanessaE: any protocol lib? 12:36 sfan5 yunfan: no 12:36 yunfan sfan5: why ? 12:37 sfan5 what you'll need to do is rip out or disable most of the UI code in client.cpp and write your bot based on that 12:37 sfan5 "why?" 12:37 sfan5 nobody wrote a protocol lib yet 12:39 yunfan sfan5: then what the client use? 12:40 sfan5 it uses the code we have 12:41 sfan5 but you can't just take network/connection.cpp and use it standalone 12:49 yunfan why cant? as i know , many such game have a separate protocol library, which just manage connection and send/receive events 12:53 sfan5 yunfan: might be the case, but we don't have one 12:55 yunfan ok, then can you show the document for protocol? want to build my own 12:56 rom1504 yunfan: https://github.com/minetest/minetest/blob/master/src/network/networkprotocol.h 12:57 yunfan rom1504: checked, not desired 12:58 rom1504 "not desired" ? 12:59 rom1504 that's the best description of minetest protocol currently 12:59 rom1504 and it's pretty much enough to implement too, imho 12:59 rom1504 *it 12:59 yunfan i need a overview of that, and that .h file only lead me to read more code 12:59 rom1504 ok, write that overview then 13:00 rom1504 it doesn't exist currently afaik 13:00 yunfan looks like finally i need to rtfc 13:01 rom1504 you only need to read that enum + its comments https://github.com/minetest/minetest/blob/master/src/network/networkprotocol.h#L158 13:01 rom1504 I wouldn't say that's really code 13:03 yunfan what about auth? 13:03 yunfan and init action 13:04 rom1504 that's about implementing the client, not the protocol 13:04 rom1504 and for that pretty sure you'll have to read the code indeed 13:05 rom1504 (which is much better than going the same for minecraft, where we have to guess everything, and no code to read) 13:08 nrzkt yunfan just read the code. 13:08 nrzkt you have comments explaining interactions 13:35 yunfan rom1504: nope, there's a wiki page which describe the extractly step of mc's protocol, also dont compare to the worse one 13:35 rom1504 yunfan: which was guessed 13:35 rom1504 it's in no way official 13:36 yunfan rom1504: yes, but they works 13:39 rom1504 yes they work for 1.8, because all the guessing work was done. But for example it's not done for 1.9 and mcpe. 13:39 yunfan http://dev.minetest.net/Network_Protocol rom1504 this is what i need, thanks to someone writting it 13:40 rom1504 okay, I thought you found this already 19:33 RealBadAngel hi folks 19:35 RealBadAngel whats up? 19:35 Fixer good 19:35 Fixer neutral 19:36 RealBadAngel one neutral, one left, any up? 19:36 RealBadAngel ;) 19:37 RealBadAngel two left.... damn shall i change the nick or what? 19:37 sapier https://github.com/minetest/minetest/pull/3461 any objections to merging this? 19:38 est31 nope, it can go it 19:38 est31 in* 19:38 est31 squashed ofc 19:38 sapier ofc 19:38 Fixer just 11 months and bug is fixed :} 19:40 RealBadAngel better late than never 19:40 Fixer yes, some bugs are for decades 19:40 Fixer also, far map looks died ._. 19:40 Fixer dead* 19:41 est31 fortunately minetest is not decades old 19:43 sapier well just stop complaining about it and start fixing bugs ;-P 19:58 RealBadAngel so, who has tested water shaders? 19:58 RealBadAngel hands up please.... 19:59 sapier shaders don't work on android ;-) 20:00 RealBadAngel i know 20:00 RealBadAngel and propably never will 20:00 RealBadAngel until GPUs from boxes go mobile 20:01 sapier come one, 2030 ain't soon enough ;-P 20:01 est31 RealBadAngel, there is opengl es 20:01 est31 just use it 20:01 RealBadAngel im kinda jokin, i was ZX Spectrum coder 20:01 RealBadAngel and then SAM Coupe 20:01 RealBadAngel theres a progress, slight ;) 20:02 RealBadAngel est31, ive bought tablet even 20:03 sapier well I'm quite sure mobils will by some time support opengl7 or whatever it's called by that time .. but that's not gonna happen the next years 20:03 RealBadAngel but still im tryin to code some generic things 20:03 RealBadAngel PC is gaming platform 20:03 sapier depends 20:03 RealBadAngel android can be 20:04 Calinou OpenGL ES 3.x is great, but Irrlicht doesn't support it 20:04 sapier I guess casual games on android have already more usage time then real pc games 20:04 Calinou same for Vulkan 20:04 Calinou sapier, maybe not. 20:04 Calinou average game on Steam sells 30k copies, average mobile game is installed less than 50 times 20:04 RealBadAngel definitely not 20:04 Calinou mobile gaming is a bubble 20:05 Calinou apart of Real Racing and Temple Run, people don't really play 3D games on mobile 20:05 RealBadAngel its good for farm games, diamonds or something like that 20:05 sapier I don't think so nintendos gameboy is almost dead and I don't think ppl do play less then by the time it was sold 20:05 RealBadAngel a few yrs behing pc platform 20:06 sapier for sure there are games that wont be played mobile 20:06 RealBadAngel MT is one of them 20:06 sapier not sure about that 20:07 RealBadAngel u can test and taste it on mobile, right 20:07 sapier minetest game concept is quite simple 20:07 est31 RealBadAngel, sapier minetest's ripoffs have more than one million users, perhaps even more 20:07 sapier there's no need for realtime 20:07 est31 they just took the source code, rebranded minetest, added some stuff of their own, and did some marketing 20:07 RealBadAngel but in the long run, playing it on mobile is just a joke 20:07 sapier the only problem is controls but that ain't something unfixable 20:07 est31 bam, they get huge user count 20:07 RealBadAngel mass effect 20:08 sapier RealBadAngel: remember "socalled mobiles" today have bigger screens then original pc's ;-) 20:08 est31 hah 20:08 est31 yeah :) 20:08 RealBadAngel i am aware 20:08 Calinou Minetest is unplayable on my LG G3 (2560×1440, 5.5") 20:08 RealBadAngel but still 20:08 sapier I can't understand why ppl put dishes to their pants and call it cellphone but ppl do 20:08 Calinou text is tiny, it's hard to look around, you have no idea how to split stacks 20:09 sapier Calinou: don't use uhd screens ;-) 20:09 Calinou it's not Ultra HD, it's QHD 20:09 RealBadAngel about splitting stacks, there should be an input and output inv to do that 20:09 RealBadAngel two fingers trick is dead stupid 20:10 sapier Calinou: you can be glad controls aren't as tiny as text it was heavy fighting to get dpi code into minetest ;-) 20:10 RealBadAngel those screenies are just to small anywaY 20:10 RealBadAngel pixels wont help you 20:10 est31 sapier, it has introduced lots of bugs 20:11 sapier of course ;-) as everything introduced to code that just doesn't honor it 20:11 sapier e.g. localization ;-) 20:12 sapier crazy 20:12 Calinou the dreaded Irrlicht GUI… 20:12 sapier couldn't those irrlicht guys have fixed the texture bug prior me updating the texture hack to their recent code ... and not one day after it 20:13 est31 lol 20:13 RealBadAngel why not having irr bundled? 20:13 est31 one patch less to maintain? 20:13 est31 great 20:13 est31 RealBadAngel, it is bundled, for android and windows 20:14 est31 and now lets stop discussing 20:14 RealBadAngel so lets do it overall 20:14 est31 I want to code tonight 20:14 * est31 codes 20:14 sapier https://github.com/minetest/minetest/pull/3454 any objections? 20:15 est31 the curl removal and the html removal should be in separate commits 20:16 sapier I'd squash those commits ;-) 20:16 est31 other than that, havent reviewed the pr 20:16 RealBadAngel #3391 .... 20:16 est31 sapier, the issue is the changes are unrelated to what the pr does 20:16 sapier I'll call the commit "Android: Update Makefile with recent library versions" ;) 20:17 est31 yeah, removing html is not related to the commit 20:17 est31 so the commit should not do it :) 20:17 sapier well I'll remove it and let html in ther ;-) 20:18 sapier I'm not gonna squash and then resplit things beeing 90% identical if I remove things from that commit i'll not readd them 20:19 est31 its easily doable with git knowledge 20:19 RealBadAngel btw, wheres c55 with his farmap? 20:20 est31 remember the git hashes of the commits you want to have separate 20:20 est31 then do git rebase -i and squash everything, except for the commits you want to have separate, they should be excluded 20:20 sapier doable with git is same as doable ... git can do everything ... except splitting a single commit or rearanging commits in a easy way 20:20 est31 then do git cherry-pick them 20:20 est31 rearranging works great 20:20 est31 git rebase -i is your friend 20:21 sapier rebase can't rearrange commits (as far as I know) 20:21 est31 if there is a merge conflict it will try to resolve it automatically 20:21 est31 lemme try 20:21 sapier LOL automatic resolving of conflicts? in a ideal world maybe 20:21 est31 its called 3 way merge 20:22 est31 if stuff is separated with large enough distance inside a file, git can do it 20:22 sapier yes and that's the type of merge which quite often produces useless results (if it's not a trivial merge) 20:24 sapier I'm gonna try to resort commits via rebase but if it fails, no chance I'm gonna do this manual 20:24 est31 okay then 20:25 est31 but if you decide to squash then point it out in the commit msg 20:25 RealBadAngel sapier, i do always git reabase -i HEAD~x 20:25 RealBadAngel to squash x of commits 20:25 RealBadAngel -i is interactive, follow on screen 20:26 est31 and there seem to be trailing whitespaces http://paste.ubuntu.com/14077929/ 20:26 RealBadAngel git push -f 20:26 est31 they should be fixed 20:26 RealBadAngel pushes it all 20:26 sapier interestingly it's also capable of resorting commits 20:28 est31 otherwise the pr is fine with me 20:29 sapier ok history rewritten better this way? 20:29 est31 you know android very well, I trust you :) 20:29 est31 lemme see 20:29 est31 fine 20:30 sapier trailing whitespaces on patches created with diff? sorry but I'm not gonna fix automatic patches just because some other tool believes them to be wrong 20:31 est31 ill redownload 20:32 sapier guess I should remove the comment about texturehack patch too 20:32 est31 ? 20:33 est31 ah that one 20:33 est31 I see 20:33 est31 yea good idea 20:33 sapier done 20:35 sapier github seems to have issues with displaying diffs for diffs 20:35 est31 can you re-run a clean rebase with --whitespace=fix ? 20:35 sapier what is it supposed to do? 20:36 est31 to remove any whitespace errors you added 20:36 est31 like trailing spaces 20:36 est31 havent found any 20:36 est31 but git complains 20:37 sapier my editor is configured to remove trailing spaces so the only ones can be within patches and I don't even want git to mess around in patches 20:38 est31 makes sense 20:39 sapier as of the paste you posted it complains about all the new patches wich are created using diff -Naur a b >patch 20:39 est31 mhhh 20:39 est31 well then i guess its okay to merge 20:39 est31 i can try to download and build it 20:39 sapier would be goof 20:39 sapier good 20:40 sapier the only open issue seems to be monte's issue but I'm not sure if it's really related to some new thing 20:41 RealBadAngel sofar fences... 20:42 RealBadAngel anybody checked it? 20:43 sapier << merging android fixes ... additional ones can be done later I don't think there's gonna be a release tomorrow 20:43 est31 pls wait 20:43 est31 now I download and test it already 20:43 sapier but waiting for est31 to confirm build prior pushing it ofc 20:43 est31 hrmmmm 20:43 est31 patching file source/Irrlicht/COGLESDriver.h 20:43 est31 Hunk #1 FAILED at 39 (different line endings). 20:44 est31 and I did not do anything to the whitespaces 20:44 est31 I said --ignore-whitespace in my am thingy 20:44 est31 sapier, did you remove the irrlicht directory from deps/ before building? 20:44 sapier yes did a clean_all ... but doing it once again atm 20:47 sapier https://gist.github.com/sapier/563634e7e9513c408a7a 20:47 est31 sapier, which ndk version do you use? 20:47 sapier r10e as mentioned in readme 20:47 sapier but did use r9d before no difference 20:48 est31 it does this for me https://gist.github.com/anonymous/15153117bc1e87619424 20:48 est31 it doesnt fail 20:48 est31 but gives the error 20:48 sapier well no difference prior changing to gcc 4.9 now you need something more new then r9d of course 20:49 sapier Invalid toolchain name: arm-linux-androideabi-4.9 20:49 sapier you don't have a ndk with gcc 4.9 20:49 est31 I have r9d 20:49 sapier which doesn't have r9d 20:49 sapier 4.9 20:50 sapier switch to 4.9 was one of the later commits I already did switch to r10e by that time 20:51 est31 btw this line should be removed from readme "E.g. prerelease Minetest Android builds have been 0.4.9.3, while the first 20:51 est31 official version most likely will be 0.4.10.4" 20:51 sapier there's no reason to stay at r9d r10e and matching sdk still compile apk for old android systems while I'd expect 4.9 build better code 20:51 sapier True but not related ;) 20:52 est31 yea 20:57 est31 seems I'll have to update http://dev.minetest.net/Android then as well 20:57 sapier yes 20:58 est31 okay ndk is extracted 21:06 sapier great xml and png files are still there 21:06 est31 ? 21:07 sapier argh ... built old version 21:08 sapier once again 21:21 est31 okay it builds 21:33 est31 sapier, there is a regression 21:33 est31 fonts get displayed weirdish 21:38 est31 sapier, http://pasteboard.co/6PNG3YR.png 21:38 sapier interesting 21:38 sapier what device do you have? 21:38 est31 galaxy s2 21:39 sapier did you delete your minetest folder? 21:40 est31 no 21:40 est31 trying that 21:40 est31 we really should get rid of minetest copying stuff 21:40 est31 thats totally ugly 21:40 sapier true but for what I know irrlicht still doesn't support loading textures from assets 21:41 est31 okay it works when I remove that dir 21:42 est31 strange 21:42 sapier I wonder what causes this 21:44 est31 well, fine with me then, good for merge 21:45 est31 minetest seems to work more or less 21:45 est31 (all stuff I tested worked) 21:45 est31 ah no 21:45 est31 now I have a suspicion 21:46 est31 wait with the merge 21:46 sapier ok 21:46 est31 yeah 21:46 est31 I can reproduce the bug 21:46 est31 it seems to be curl related 21:47 sapier what happens? 21:47 est31 when I uncheck "public serverlist" then restart minetest 21:47 est31 then check the thing again 21:47 est31 the fonts mess up 21:48 est31 out of some reason they also mess up when I go into the credits tab 21:49 sapier hmm doesn't happen for me ... maybe because I don't get a serverlist anyway 21:49 est31 sometimes they only mess up a bit 21:50 est31 idk though whether its a regression of this pr 21:50 sapier are you sure it's a new issue? ;-) 21:50 est31 yeah, checking that 21:51 sapier E/Minetest(19486): 2015-12-17 22:50:51: ERROR[AsyncWorker-0]: servers.minetest.net/list?proto_version_min=13&proto_version_max=26 not found (Couldn't connect to server) (response code 0) 21:51 sapier ahh wait ... maybe it's a firewall issue 21:53 est31 okay building master now 21:53 est31 ... my master 21:53 est31 which is 6133b2bc4 21:54 sapier ok loading serverlist causes fonts to be broken 21:55 RealBadAngel i do really love comments like BlockMen's ones 21:56 RealBadAngel for fuck sake, try the effect, not just THINK what it does 21:56 sapier guess update ain't ready for pushing Ill just push the removal commits 21:56 RealBadAngel asking for pixels in shaders math is just plain dumb 21:57 RealBadAngel BlockMen, officially, please do try it 21:57 RealBadAngel otherwise, shut up 21:57 est31 great you can reproduce 21:57 est31 gtg now, will be around again in about three hours 21:58 est31 then i can tell whether its a regression 21:58 est31 if you want to be there sapier when I come back its great, if you dont want to be there no prob either 21:58 est31 either way dont merge :) 21:59 sapier I'll merge the non lib commits only 21:59 RealBadAngel and please do note i cant do screenshots for your pleasure, its an animation 21:59 est31 fine 22:08 nrzkt sapier, on which PR are you working ? 22:13 sapier android only 22:13 sapier seems like fonts are broken after serverlist is loaded 22:14 sapier at least in updated version 22:17 nrzkt okay 22:18 sapier https://github.com/minetest/minetest/pull/3454 I'm gonna merge the other commits of this one only for the moment 22:20 sapier https://github.com/minetest/minetest/pull/2132 nrzkt can you check this out and tell me if it's ok? 22:24 sapier ok corruption doesn't happen on current lib version so it's most likely related to some of them 22:24 sapier I'll push re removing fixes now 22:25 nrzkt sorry but i can't test it on my own phone , it doesn support MT, MT crash at start 22:26 sapier what phone do you have? 22:26 nrzkt Ascend Mate 2 22:26 nrzkt Android 4.2 22:28 sapier can you provide a "adb shell logcat | grep Minetest" output? 22:28 nrzkt not atm sorry :)