Minetest logo

IRC log for #minetest-dev, 2016-05-12

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

All times shown according to UTC.

Time Nick Message
00:10 est31 no
00:10 est31 its just so slow
00:10 Fixer still building?
00:12 est31 I'm giving it one more try now
00:12 est31 just letting it wait for curl to download
00:15 est31 the curl download is projected for 20 minutes
00:16 est31 paramat, idk when do you want to go to bed
00:17 paramat i'll be up for a few more hours
00:17 est31 can you have a look at this patch https://github.com/est31/minetest/commit/​35f2b30df99c04e4a3ffb31d7f2deae084f63b40
00:17 paramat i'm often semi-nocturnal
00:17 paramat yeah
00:17 est31 I'm testing it right now
00:18 est31 and if it works, i want to push it as well, is that okay?
00:19 paramat looks trivial, +1 if it works
00:21 turtleman joined #minetest-dev
00:48 paramat is 'register on mapgen init' deprecated?
00:51 yang2003 joined #minetest-dev
00:56 yang2003 joined #minetest-dev
01:01 yang2003 joined #minetest-dev
01:07 ElectronLibre joined #minetest-dev
01:16 yang2003 joined #minetest-dev
01:38 est31 okay the build finished now i have to test it
01:44 est31 aha bug!
01:46 est31 well singleplayer is broken
01:46 est31 now I try connecting to servers
01:46 asl97 est31: if you mean on singleplayer, it always been broken
01:47 asl97 i needed to create the world folder before it work
01:47 est31 just the folder?
01:48 asl97 more or less and maybe copy over a copy of minetest_game (which might need to be rename to minetest)
01:48 asl97 it's might just be that my build is weirdly broken
01:55 est31 ah I think I know what the issue is
01:55 est31 I hope :)
01:57 est31 hmm it seems to be a minor bug
01:58 est31 not required to call off the release because of it
02:00 est31 well seems paramat left
02:00 est31 and i dont want to release alone
02:00 est31 ive tried that, it failed miserably
02:10 est31 lol
02:10 est31 just discovering I've failed the same bug already in august last year
02:10 asl97 as long as it doesn't affect most player, i suppose it's fine
02:10 est31 #2971
02:10 ShadowBot https://github.com/minetest/minetest/issues/2971 -- Android: No creation of a new world on first run
02:11 asl97 that seem like the bug
02:31 ShadowNinja <est31> ... 0.5 means we break network compat -- actually, no.  According to semver it just means "new major release, but without major compat breaks"
02:32 est31 ShadowNinja, according to semver we have no stability at all
02:32 est31 http://semver.org/#spec-item-4
02:33 KaadmY well, the public API isn't stable :D
02:33 est31 ShadowNinja, can you tell me whether you have the same bug as me
02:33 KaadmY the lua api at least
02:33 est31 not 100% yeah
02:33 est31 but mostly we try to keep it stable
02:33 asl97 ShadowNinja: i thought the leading 0 is meaningless,  i thought minetest use 0.MAJOR.MINOR.PATCH
02:33 ShadowNinja est31: Yep.  I'd like to break a bunch of stuff when we go 1.0. :-)
02:34 est31 And i dont think we need that :)
02:34 ShadowNinja est31: What bug?
02:35 est31 Try to remove or rename the Minetest folder from the sdcard
02:35 est31 on android
02:35 est31 then get a fresh android build
02:35 est31 dunno where the first version was with this bug
02:35 est31 then start it, and immediately press the start singleplayer button
02:35 ShadowNinja asl97: Theoretically we use semver, but we don't really do it correctly yet, so we have the "0.4.version" joke.
02:35 est31 (dont change tabs for example)
02:36 asl97 semver with a leading 0
02:37 est31 ShadowNinja, the bug is now that there will be an error
02:37 est31 and singleplayer wont start
02:37 est31 can you reproduce?
02:39 ShadowNinja est31: Yep, 3b847c19.
02:39 ShadowNinja (My rewrite with data in the right paths (
02:40 est31 ok
02:40 est31 then its not wrong usage by me
02:41 ShadowNinja est31: Fix: rm builtin/mainmenu/simple_main.lua and use the proper desktop menu.
02:41 est31 because i'd closed #2971 because of that
02:41 ShadowBot https://github.com/minetest/minetest/issues/2971 -- Android: No creation of a new world on first run
02:41 est31 I'll file a new bug, this seems to be a regression
02:41 est31 even though the symptoms are similar
02:42 est31 its nothing we should halt the release for though
02:42 asl97 3b847c19?  github can't find it on minetest/minetest
02:43 ShadowNinja asl97: Might not be published.
02:44 est31 even if, its not inside minetest/minetest
02:44 est31 asl97, look at ShadowNinja/minetest
02:44 ShadowNinja est31: Definitely reopen that though.
02:44 est31 I'll open a new one its cleaner I think
02:44 est31 I can't reproduce with 0.4.13
02:44 est31 so it definitely is a new bug.
02:45 asl97 est31: nope: https://github.com/ShadowNin​ja/minetest/commit/3b847c19
02:45 ShadowNinja est31: GitHub stores the hashes globally, I think torvalds/linux/commit/mtcommithash would work.
02:46 est31 ShadowNinja, no
02:46 est31 they might share them between all forks of one repo
02:47 est31 but they are not really global
02:47 est31 i doubt that
02:47 est31 in fact sharing commit namespace between all forks of a repo makes total sense from an implementor standpoint
02:48 asl97 this bug affect new installation doesn't it?  a blocker i guess
02:48 Void7 joined #minetest-dev
02:52 hmmmm SIGH
02:53 hmmmm alright i guess 1.5 weeks wasn't long enough
02:53 hmmmm the 18th shall be the day of the release
02:53 hmmmm any objections?
02:53 asl97 i spotted this bug weeks ago, i just thought it was cause of my weird build not including the minetest_game.
02:54 est31 asl97, ShadowNinja, #4111
02:54 ShadowBot https://github.com/minetest/minetest/issues/4111 -- Android: "Provided world path doesn't exist" if chosing singleplayer on first run
02:54 est31 hmmmm, what about sth more close
02:54 est31 like saturday
02:55 hmmmm well, okay
02:55 est31 14th
02:55 hmmmm 14th it is
02:55 est31 its just one bug
02:55 est31 enough time to hunt it down and crush it
02:55 hmmmm i was under the impression that more kept popping up
02:57 est31 nah i dont think so
02:57 est31 its pretty stable and bug free
02:57 hmmmm this is such a fun game
02:58 hmmmm (not minetest, whack-a-mole)
02:58 paramat joined #minetest-dev
02:58 asl97 paramat: hello paramat :D
02:58 paramat no objection, hopefully we can release before then, maybe at the weekend
02:59 paramat hi
02:59 hmmmm i don't get it, do people keep pressing F5 on the logs page?
02:59 est31 yes sometimes i do this
02:59 hmmmm why don't you just keep your irc client in this channel..
03:00 est31 but then people know i am here and i have to answer lots of questions
03:00 est31 and i really only monitor this channel
03:00 est31 with the f5 thing
03:01 est31 but if i open the irc client, i join many other channels
03:05 est31 but I can say that I have a sort of working f-droid build now
03:05 est31 well enough minetest for today
03:11 paramat this gives me time to correct this https://github.com/minetest/minetest/b​lob/master/minetest.conf.example#L1124
03:11 paramat although of course it will differ from translations
03:54 paramat #4112 toodle pip!
03:54 ShadowBot https://github.com/minetest/minetest/issues/4112 -- Documentation: Remove incorrect and excessive mapgen flags text by paramat
03:54 paramat left #minetest-dev
05:13 Hunterz joined #minetest-dev
05:21 jin_xi joined #minetest-dev
05:40 Zeno` joined #minetest-dev
06:44 Lunatrius` joined #minetest-dev
07:03 johnnyjoy joined #minetest-dev
07:15 misprint joined #minetest-dev
07:48 nrzkt joined #minetest-dev
07:59 twoelk joined #minetest-dev
08:24 Amaz joined #minetest-dev
09:05 nore Zeno`: are you there?
09:22 Zeno` barely
09:25 nore I you are enough there then, could you please test #4113? It would be nice to be able to release tonight :)
09:25 ShadowBot https://github.com/minetest/minetest/issues/4113 -- Fix #4111 by Rui914
09:26 Zeno` It's just reverted some of the previous commit, so yes +1
09:27 Zeno` (the previous commit == the one referred to that changed to use indexof
09:30 Zeno` nore: change the commit comment though ;)
09:31 nore Zeno`: it works?
09:31 Zeno` the previous commit worked
09:31 Zeno` testing in 5
09:31 nore Ok, thanks :)
09:40 Zeno` seems fine
09:41 Zeno` I wish they'd make android stuff easier :/ I mean it's not hard, just tedious
09:41 Zeno` deleting everything again to make sure
09:53 Zeno` yep, fixes
09:53 nore Yay!
09:54 nore Now we can merge that and release ^^
09:58 Zeno` who's doing the builds?
10:01 nore I think est31 did the Android one
10:01 Zeno` and Windows? sfan5?
10:02 lisac joined #minetest-dev
10:02 nore Hm, I don't know
10:02 nore Though, I seem to remember that 0.4.13 was released before the builds were done
10:02 nore And then, builds were added
10:03 * nore checks
10:03 nore Yeah, I think it was that
11:19 damiel joined #minetest-dev
11:32 sfan5 Zeno`: yeah i can do for windows (not on windows though)
11:55 naltamur joined #minetest-dev
11:55 naltamur Hello!
12:03 naltamur left #minetest-dev
12:11 STHGOM joined #minetest-dev
12:15 Fixer joined #minetest-dev
12:24 proller joined #minetest-dev
13:02 crazyR_ joined #minetest-dev
13:02 Fixer tweaked my translation a bit, so if you want, you can update it
13:04 crazyR__ joined #minetest-dev
13:07 turtleman joined #minetest-dev
13:25 Fixer !tell millersman I still see gaps after flying far and back to huge lava column after your changes to reflow
13:25 ShadowBot Fixer: O.K.
13:50 Player_2 joined #minetest-dev
13:59 PilzAdam joined #minetest-dev
14:14 srifqi joined #minetest-dev
14:25 misprint joined #minetest-dev
14:38 KaadmY joined #minetest-dev
14:40 Void7 joined #minetest-dev
15:20 Zeno` #4086 might need revisiting
15:20 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
15:25 Zeno` pushing #4113 (after fixing typo) in 10 minutes
15:25 ShadowBot https://github.com/minetest/minetest/issues/4113 -- Fix #4111 by Rui914
15:26 Zeno` difference: http://dpaste.com/121SF99
15:28 Wayward_One Zeno`, did you see my comment on 4113?
15:28 Zeno` yes
15:28 Wayward_One ah, ok
15:29 Zeno` Wayward_One, try https://www.dropbox.com/s/bcr3hhf​hcmrqc4x/Minetest-debug.apk?dl=0
15:29 Zeno` that has the typo fixed
15:29 hmmmm joined #minetest-dev
15:31 Wayward_One Zeno`, that works
15:33 Zeno` ok, merging really soon
15:33 Zeno` thanks for the confirmation
15:33 Void7 joined #minetest-dev
15:40 Wayward_One yw
15:41 Zeno` pushed
15:42 srifqi waiting for any bot at IRC to notify
15:42 srifqi it just came
15:43 Zeno` silly bots are too slow
15:43 Zeno` :)
15:43 srifqi i'll update my fork and compile :D
15:44 PilzAdam Zeno`, that's not a good commit message
15:44 PilzAdam references to issues should not be in the first line
15:44 Zeno` true
15:44 Zeno` I apologies
15:45 Zeno` want me to change it before the limit is reached?
15:45 srifqi how?
15:45 PilzAdam no, not really necessary here, since it has a valid description
15:45 PilzAdam just keep it in mind for the future
15:45 Zeno` point taken
15:45 Zeno` thanks for pointing it out
15:46 srifqi even if Zeno` really changes the message, will the commit hash changed?
15:46 PilzAdam also never rewrite history if you know that people already fetched the new commit
15:46 PilzAdam srifqi, the hash would have changed, yes
15:46 srifqi :O
15:47 srifqi i was about to fetch it, but paused because i read that and now starting fetch
15:48 Zeno` PilzAdam, hey, at least I made the commit message better than the PR :D
15:48 PilzAdam indeed
15:49 PilzAdam I guess everything is o.k. now
15:49 PilzAdam :-p
15:49 sofar for reference: https://github.com/minetest/minete​st/pull/4113#discussion_r63044961
15:50 sofar getting the "first item of a table" is really easy with "next(table, nil)"
15:51 Zeno` you can do that?!
15:51 sofar yes :D
15:52 Zeno` well there you go
15:52 sofar next() is an iterator, sort of
15:52 Zeno` I agree it's much more readable
15:52 sofar if you do `next(table, i)` you get element i+1
15:52 sofar nil is the start :)
15:52 Zeno` nice
15:52 * sofar admits lua documentation is bizarre, but fascinating nonetheless :D
15:53 luizrpgluiz joined #minetest-dev
15:53 Void7 joined #minetest-dev
15:54 luizrpgluiz hello, what's going on with the update minetest? It has not yet left the new version of the game?
15:54 Krock joined #minetest-dev
15:54 twoelk gotestthecandidate
15:55 Krock -the
15:55 Krock duh. It hasn't been released yet
15:55 Krock *compiles*
15:55 twoelk go-test-release-candidate?
15:56 Krock yeah, I read "got the test the candidate"
15:56 * srifqi is now compiling 9d3d43f
15:57 Zeno` compiling for android?
15:57 luizrpgluiz I've done some performance tests here on my computer to compile Linux Xubuntu 16.04
15:59 Zeno` PilzAdam, is it O.K. if I call you Pilz? :P
16:00 Fixer he has "don't call me Pilz in his signature", not sure why
16:00 luizrpgluiz if you are having trouble compiling the game, maybe I can try to help build the game for Ubuntu Linux
16:03 Zeno` people are having trouble compiling it?
16:06 luizrpgluiz Zeno`: that strange, to compile the game github by Xubununtu Linux and other flavors of Ubuntu, I'm not having trouble compiling and testing the game.
16:06 Zeno` I've not heard of any trouble either...
16:08 luizrpgluiz Zeno`: It should probably be time to create the package is missing or people with plenty of time to compile and create packages for systems compatible with the game
16:21 DonBatman joined #minetest-dev
16:23 paramat joined #minetest-dev
16:23 DonBatman I tested the 4086 pr. This makes it so the hardware keys work. Would be great to have this in 0.4.14
16:23 paramat please can we reconsider #4086 ? in testing it works and seems to not cause new issues
16:23 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
16:24 Zeno` I asked that earlier as well
16:24 paramat i'm just not sure i am qualified to +1 it, although i did before
16:25 DonBatman paramat I have tested it and it works
16:25 srifqi Zeno` I'm compiling for my laptop (using mingw32-make)
16:25 paramat also there's #4112 but it's not essential, up to you
16:25 ShadowBot https://github.com/minetest/minetest/issues/4112 -- Documentation: Remove incorrect and excessive mapgen flags text by paramat
16:25 paramat ok
16:26 Zeno` I don't understand how 4086 actually works, but it does
16:26 paramat DonBatman if you can play for a while with 4086 to check if it causes new bugs that would be useful
16:28 DonBatman Will play more to see if I see any issues
16:28 paramat Wayward_One has played for a while with 4086 with no apparent issues
16:32 Zeno` The only real reason I didn't approve it was because of the way it was initially presented
16:36 hmmmm let's talk about 4086
16:37 hmmmm has anybody with android expertise actually reviewed it and gave their approval?
16:37 hmmmm because the way i see it right now is a bunch of bobbleheads with no real clue what's going on in that change taking somebody's "word for it"
16:38 Zeno` <Zeno`> I don't understand how 4086 actually works
16:38 PilzAdam hmmmm, doesn't Monte have actual "Android expertise"?
16:38 hmmmm he does not qualify
16:38 paramat est31 seems to have some understanding, looking at the thread
16:39 hmmmm what would you say if nobody on the openssl project had knowledge on eliptic curve cryptography for instance
16:39 hmmmm and somebody presented a change that modified the eliptic curve routines
16:39 hmmmm that's basically what i'm seeing here
16:39 hmmmm there is no meaningful peer review that can be performed
16:39 paramat is sapier contactable?
16:39 Zeno` who is our android expert?
16:39 hmmmm nobody
16:40 PilzAdam <hmmmm> and somebody presented a change that modified the eliptic curve routines <- more like "and a researcher in this area"
16:40 hmmmm i'd rather have no feature than one with known major problems and no support
16:40 PilzAdam Zeno`, Monte is most knowledgeable in this field, at least thats what I am seeing currently
16:40 Zeno` I would agree
16:41 Zeno` with that, PilzAdam
16:41 Zeno` and it does work
16:41 PilzAdam hmmmm, seems to have a tendency to not trust Monte
16:41 Zeno` I just don't know why or how
16:41 hmmmm have you tested?
16:41 hmmmm you mentioned in the thread you haven't tested it and we need to find somebody who can test
16:41 hmmmm you're right i don't trust monte
16:41 PilzAdam hmmmm, why?
16:41 hmmmm it's nothing personal, it's just that when somebody is pushy like that with their code it makes me question them
16:41 Zeno` I have and so have Wayward_One, DonBatman and someone else
16:42 PilzAdam hmmmm, you are "pushy" with your mapgen code
16:42 hmmmm what do you mean
16:42 PilzAdam s/are/were/
16:42 luizrpgluiz left #minetest-dev
16:42 Zeno` if none of us are android experts and the PR fixes the problem, then... *shrug*
16:42 PilzAdam for a rather long time you were the only person who touched the mapgen code and merged without approval of other devs
16:43 PilzAdam this is nothing bad, because you have expertise in this area
16:43 * srifqi just done compiling, now he is testing
16:43 hmmmm this was before we had any semblance of a peer review model
16:43 hmmmm it used to be the wild west
16:43 PilzAdam and Monte has a large userbase who tests his Android port, so I am ok with him working on the Android port without the strict oversight of core devs that we have for regular PRs
16:44 hmmmm well
16:44 PilzAdam we always had the peer review model
16:44 hmmmm i hate to say it but it seems like i'm going to have to read up about android to the point where it's possible to review this
16:44 paramat Zeno` i think sapier was our android expert, but is not around
16:44 PilzAdam the "2 core devs must agree" rule was in place before you even became a core deb
16:44 PilzAdam *dev
16:44 hmmmm you sure you're remembering correctly?
16:45 PilzAdam yes
16:45 hmmmm i'm not saying you're wrong, it's just that i do not remember that sequence of events
16:45 PilzAdam I was one of the 3 initial core devs that c55 nominated and the rule was in place from day 1 on
16:45 paramat nope, the rule was mapgen maintainer could merge on their own approval
16:46 PilzAdam paramat, the maintainer rule was added later and wasn't really used anywhere except in mapgen
16:46 paramat just saying hmmmmm acted correctly during that rule
16:46 PilzAdam (and hmmmm started working on mapgen before the maintainer stuff happened, IIRC)
16:46 hmmmm pretty sure there were no actual rules or regulations in place, things used common sense
16:46 hmmmm yes I did
16:47 hmmmm long before that
16:47 PilzAdam paramat, I'm not trying to say that hmmmm broke the rules
16:47 hmmmm in fact I'm the one who came up with the maintainer rule because it just makes sense
16:47 hmmmm i'd have these commits where nobody would know what was going on and have no way to give meaningful feedback
16:48 PilzAdam hmmmm, this is the same situation that Monte is in, currently
16:48 hmmmm i can't *force* anybody else to become knowledgable in that area because this is an open source project
16:48 PilzAdam he has more knowledge than any core dev when it comes to the Android port
16:48 hmmmm right
16:48 hmmmm but the thing is
16:49 proller joined #minetest-dev
16:49 hmmmm we struck down the "subject area maintainer" rule
16:49 paramat however note monte says he is 'not a programmer', he employs a programmer
16:49 Void7 joined #minetest-dev
16:49 hmmmm i wish people looked at my commits back when i was working on the mapgen full time by myself
16:49 PilzAdam paramat, does it matter, though?
16:50 Zeno` don't two core-devs already agree that it should be committed?
16:50 paramat PA not sure
16:50 hmmmm in any case, how about this
16:50 hmmmm new rule:
16:50 Krock hmmmm is king.
16:51 hmmmm if a core developer wants to review a commit before merge, even if it has a certain amount of approvals, you need to delay for say... uhh.. 24 hours to give the other dev a chance to review it
16:51 hmmmm ??
16:51 sofar that's what I would do
16:51 Krock hmmmm, a bed suggestion for a new rule
16:51 PilzAdam hmmmm, you don't need a rule for that, it's common sense
16:51 Krock *bad
16:51 nore seems fine with me
16:52 PilzAdam hmmmm, if a core dev comments on a PR and says that they need additional time reviewing then people usually wait for that
16:52 hmmmm so krock is against it, nore and sofar like it, pilzadam thinks it should be obvious (but it clearly isn't!)
16:52 hmmmm yeah but the thing is, that hasn't been formalized at all.  it's up to the committer's good will to obey that request
16:52 Krock err no. I like it
16:52 sofar well, tbh, I find it obvious, but that's my personal work ethic, too
16:52 PilzAdam hmmmm, s/commiters/mergers/
16:52 hmmmm mergers yeah that's what i meant
16:53 hmmmm so okay give me some time, i'm going to do due dillegence on 4086
16:53 PilzAdam and mergers are always core devs, who are core devs because they are trusted to have good will
16:53 paramat rule seems fine to me, we would usually wait if someone requested
16:54 hmmmm so let's stop talking about reviewing the commit and start talking about the commit itself
16:55 hmmmm 4086 does three things that i can tell:
16:55 PilzAdam I trust Monte enough to give him my vote on Android related issues
16:55 hmmmm stops the default onBackPressed action from happening
16:55 hmmmm changes a key mapping from back to "cancel"
16:55 hmmmm and it seems like the last one sets some kind of status bit and doesn't pass along the event to lower level.. things
16:56 hmmmm i need to learn enough about the android UI to make a judgement if these things actually do what they seem like they do, and whether or not they're desirable behavior
16:56 hmmmm somebody else should join along with me just in case
16:58 Zeno` why don't we merge and deal with the flak later? :)
16:58 Zeno` since the current behaviour is broken anyway I mean
16:59 hmmmm because if something gets merged as-is and it happens to have problems with it that aren't immediately fatal, nobody goes to follow up on them
16:59 DI3HARD139 joined #minetest-dev
16:59 Zeno` we could drop Android support I suppose
17:00 hmmmm honestly that's what i would prefer lol
17:00 Zeno` yeah, so no loss if the PR breaks anything :P
17:01 RichardTheTurd joined #minetest-dev
17:01 RichardTheTurd I 2nd that
17:02 DI3HARD139 btw. Hi everyone?
17:02 DI3HARD139 -?
17:03 sofar PilzAdam: I'd really like it if Monte would actually test our build, not what he has built himself...
17:06 hmmmm hmmm, does anybody know what "finishes the current activity" means in context of Android?
17:07 Krock hi DI3HARD139
17:07 hmmmm from what i understand about android, and it isn't terribly much, an activity is roughly equivalent to a modal dialog box, not the application itself
17:14 Zeno` I admire your diligence, hmmmm, but I think we should just merge it
17:14 Zeno` I doubt it will kill anybody
17:14 paramat lets see what est31 thinks
17:17 hmmmm this is a bad precedent
17:17 hmmmm i know it's not going to kill anything but this is a total shitshow
17:17 hmmmm no testers
17:17 hmmmm no developers
17:18 Zeno` wait, heaps of people have tested it
17:18 hmmmm you take a patch offered by a random person and pray it works
17:18 hmmmm if something in this process doesn't get fixed we're going to continue running up against this issue
17:19 Zeno` we have no android expert
17:19 hmmmm why release software that has no support
17:19 Zeno` well, there is support in a way. We have a fix and it's been tested
17:20 Zeno` without getting a new core dev committed to android I don't see much else we can do apart from trusting test results
17:20 Hunterz joined #minetest-dev
17:21 stormchaser3000 joined #minetest-dev
17:21 paramat i feel we should consider dropping android support, with apologies to those who enjoy it
17:21 Zeno` paramat, and that's probably 90% of the userbase lol
17:21 Zeno` it's a crazy situation
17:21 sofar minetest isn't a product. The word "support" is entirely overrated for the project.
17:22 hmmmm ugh
17:22 hmmmm well i don't care, it's up to you guys what to do here
17:22 sofar someone doing a android store build should test and fix before publishing
17:22 sofar that's not necessarily up to us
17:23 hmmmm i can guarantee this will be the first of many similar cases, however
17:23 Zeno` similar android cases?
17:23 hmmmm yes
17:23 sofar I don't disagree. hmmmm
17:23 Zeno` possibly
17:24 hmmmm if we were a business, right now we'd be scrambling to find an android developer
17:24 hmmmm but guess what, this project relies on the goodwill of random strangers
17:24 hmmmm oh ffs
17:24 paramat indeed
17:24 sofar we're all strangers, more or less
17:25 PilzAdam if we were a business then where is my salary?
17:26 hmmmm did somebody mention earlier that monte48 was selling multicraft?
17:27 hmmmm maybe there could be a donation box on the play store page so we can scrap something together to hire a part-time android dev
17:27 Zeno` At this point, if this was a business, a manager would step in and either say "merge it" or "delay" and there would be no arguments :P
17:28 hmmmm [01:22 PM] <hmmmm> well i don't care, it's up to you guys what to do here
17:28 hmmmm :)
17:29 hmmmm i changed my mind on caring
17:29 paramat yes monte48 makes money from multicraft
17:30 hmmmm i wonder how plausible it would be to gather up as many developers as we can and implement a dual license scheme to generate revenue off of people using minetest to profit, and put that toward android development
17:32 hmmmm first off, would everybody here agree to that kind of license change?
17:32 proller joined #minetest-dev
17:32 Zeno` not really... it goes against the "freedom" of free software
17:33 KaadmY #minetest-dev: "from here comes the reasons for not merging a 2-line PR"
17:40 paramat i don't feel our android situation is important enough to do this
17:41 paramat it would be less headache to just drop android
17:42 Zeno` this is silly :(
17:42 Zeno` just merge it
17:42 Zeno` if we're going to even consider dropping android support then who cares?
17:43 Wayward_One What about creating some sort of "Android developers wanted" post on the forum?
17:43 Zeno` anyway it already has more than 2 approvals
17:43 paramat i seem to remember a new person who wanted to work on android
17:44 Wayward_One s/wanted/needed
17:44 paramat yeah a forum post would be good
17:45 paramat in the thread est31 is discussing issues, so i think at least let's wait for his opinion
17:50 rubenwardy joined #minetest-dev
17:58 MoNTE48 joined #minetest-dev
17:58 MoNTE48 Hello all :)
17:58 Void7 hi
17:58 MoNTE48 I just read irc log. It was quite funny, ha ha. So many crazy things in one place.
17:59 MoNTE48 If you post a link to donate on pley store description or app, google blocked dev account
18:00 naltamur joined #minetest-dev
18:00 naltamur Hello guys
18:01 MoNTE48 If the problem is that I make money on the LGPL code, then you do not have a clue what the LGPL (this is NOT GPL). I can go away and never come back. I do not get anything to help you guys. I just want to do good.
18:03 paramat you making money is not a problem for us :)
18:03 paramat it is of course allowed
18:04 MoNTE48 @PRMT it's nice to hear. So few people who understand what the license is used. :)
18:07 MoNTE48 Technically, I can do 2 things. Create MultiCraft: Minetest Edition. Version with vanilla minetest_game without ads. Or I can try to continue to create pr and give people who do not understand Android judge my work.
18:08 hmmmm ideally there'd be an in-house android developer
18:09 MoNTE48 You rightly said, "ideally"
18:09 MoNTE48 Therefore, you can either accept my help, or asking me to leave.
18:09 hmmmm we are accepting your help
18:10 hmmmm but it's not reliable enough, no offense
18:11 MoNTE48 The issue is that none of the core devs can not check the correctness of my code. Many do not even have android device
18:11 hmmmm and that's why releasing minetest on android was a mistake
18:11 hmmmm no support whatsoever
18:11 MoNTE48 I understand that I have a bad reputation, and I do not dev. If interested, now i work alone, I do not have C ++ / lua dev.
18:12 hmmmm i could force myself to get interested in android development and work on it myself but i'd rather not
18:12 MoNTE48 :)
18:13 MoNTE48 onBackPressed() need to press the Back button does not return us to the prev activity, which had already finished.
18:13 hmmmm the least i could ask for right now is an independent tester
18:13 hmmmm somebody who has a variety of android devices and could ensure that it's not just going to break in a different way
18:14 Darcidride joined #minetest-dev
18:14 hmmmm i suppose i could install android x86 in a VM and do testing myself, but that's a bit of a project in itself
18:14 Wayward_One right here
18:15 DonBatman hmmmm I test on an lg g3 and a nexus 4
18:15 MoNTE48 android x86 ... you need x86_64 build, not x86 (I think, but am not sure)
18:15 hmmmm cool
18:15 hmmmm have you guys tested #4086 to see if it works?
18:15 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
18:15 Wayward_One i have 3 different devices, 5 counting two androidx86 forks
18:16 MoNTE48 I have a lot of devices, just when there is a problem on some device, I just go to the store and buy it to see :)
18:16 MoNTE48 @Wayward_One
18:16 MoNTE48 @Wayward_One, x86 build work on Android-x86 on 64 bit CPU?
18:16 DonBatman I have tested 4086 and could find no issues
18:19 Wayward_One MoNTE48, they both support the normal minetest builds, if that's what you mean
18:20 MoNTE48 @Wayward_One thank
18:21 Wayward_One np
18:22 MoNTE48 If my little changes can not take Minetest for a few weeks, major changes are not accepted ever :)
18:28 paramat hmmmm, Wayward_One has played for a while with 4086 applied and found no issues
18:29 MoNTE48 I made a number of significant changes in depends libs, but none of the core devs will not be able to approve them - none of the core devs do not have android devices. ** Smile **
18:38 naltamur guys Can anyone send me a android build stable?
18:39 naltamur MoNTE48: ?
18:40 Amaz joined #minetest-dev
18:40 DonBatman this is a link from Zeno with 4086 included
18:40 DonBatman https://www.dropbox.com/s/bcr3hhf​hcmrqc4x/Minetest-debug.apk?dl=0
18:40 paramat bbl
18:40 paramat left #minetest-dev
18:41 naltamur thanks
18:55 Fixer joined #minetest-dev
19:19 ThatGuyWho joined #minetest-dev
19:20 ThatGuyWho joined #minetest-dev
19:21 ThatGuyWho joined #minetest-dev
19:39 Fixer i will also try zenos apk with a fix on my lenovo, dropping the support of android with that much people on it... it is not like this bug is end of the world bug, or it is?
19:47 Fixer ok, i will not test, that apk was for for different bug testing :)
19:56 rubenwardy joined #minetest-dev
20:21 ElectronLibre joined #minetest-dev
20:27 nrzkt joined #minetest-dev
20:30 est31 joined #minetest-dev
20:31 est31 Glad to see #4111 was fixed in the meantime
20:31 ShadowBot https://github.com/minetest/minetest/issues/4111 -- Android: "Provided world path doesn't exist" if chosing singleplayer on first run
20:32 DonBatman est31 I tested 4111 and couldn't find any issues
20:34 est31 nrzkt, can you test whether you can build minetest for android?
20:34 est31 just wondering, not that you discover at release time that it doesnt work
20:34 est31 nrzkt, watch out the curl download is very slow. its a curl hosting problem.
20:34 est31 be patient it takes around 30 mins to download it
20:34 nrzkt est31, i can test it yes
20:35 est31 rubenwardy, was it you who agreed to do the MSVC builds?
20:35 est31 after blockmen left?
20:35 rubenwardy No
20:35 nrzkt no problem for me to download curl est :p
20:38 est31 damn its fast now
20:39 est31 well thats good
20:47 Void7 joined #minetest-dev
20:48 Fritigern joined #minetest-dev
21:00 est31 hrmm because I remember somebody bringing up the MSVC situation once and somebody agreeing to do the msvc builds
21:01 ptv joined #minetest-dev
21:02 est31 would be good if we found out who will do the msvc build for minetest
21:02 rubenwardy Krock?
21:03 est31 I wonder what krocks microsoft visual studio version is
21:03 est31 maybe more recent versions produce faster code
21:03 rubenwardy It's one that can build to XP
21:03 rubenwardy so not most recent
21:04 est31 He is trustable, thats a big plus
21:05 est31 s/code/binaries/
21:06 Fixer Krock only builds 32 bit builds iirc
21:06 Fixer can you build 64 bit from 32bit os?
21:15 sfan5 probably not
21:20 nrzkt est31, build was in progress i'm fixing the end thing
21:20 nrzkt seems i haven't the build tools, strange
21:21 est31 nrzkt, which end thingß
21:21 est31 ?
21:21 nrzkt i'm building my native android app with android sdk v23.0.2 too and it's working well
21:21 nrzkt oh MT wants 23.0.3
21:21 nrzkt i should update
21:22 nrzkt make[2] : on quitte le répertoire « /home/nerzhul/Devel/minetest-stock/build/android »
21:22 nrzkt find: ‘./Minetest/games/minetest_game/.git’: Aucun fichier ou dossier de ce type
21:22 nrzkt what is this ?
21:22 est31 that can be ignored
21:22 est31 its harmless
21:22 est31 not sure from which find it comes from
21:22 nrzkt MT game is copyied from somewhere ?
21:22 est31 yes
21:22 nrzkt but where ? because i didn't specified it
21:23 est31 it is copied from games/minetest_game to build/android/src/main/assets/Minetest/games/
21:23 est31 long path :)
21:23 nrzkt maybe could be good to have a variable
21:24 est31 there is a variable in the makefile
21:24 est31 GAMES_TO_COPY
21:24 nrzkt but not in the local.properties :)
21:24 nrzkt i don't thing it's good to modify makefile
21:24 nrzkt ok build success
21:24 est31 yeah
21:24 nrzkt it doesn't mean it works well but it build
21:25 est31 feel free to test it :)
21:25 nrzkt i'm rebuilding in release mode now
21:25 nrzkt i don't have time atm but maybe later
21:25 nrzkt but where did i signed the release build ? nothing told me
21:26 nrzkt strange
21:27 nrzkt okay it was started and seems to work on my phone
21:27 est31 nrzkt, i think its not signed
21:27 nrzkt except i didn't sign the APK and i'm in release mode
21:28 nrzkt very strange something is missing, it was checked in the previous process
21:28 est31 for me it created Minetest-release-unsigned.apk
21:28 nrzkt miss, in fact i have a script for me
21:28 nrzkt :p
21:28 nrzkt but i'm use it's not a unsigned
21:28 nrzkt oh ... i uploaded minetest release from august sorry
21:29 nrzkt ok i see the unsigned
21:31 nrzkt okay it's good for me
21:31 nrzkt except the version number :p
21:32 nrzkt and the shaders :)
21:47 proller joined #minetest-dev
22:00 stormchaser3000 joined #minetest-dev
22:01 paramat joined #minetest-dev
22:04 troller joined #minetest-dev
22:04 paramat est31 we are considering #4086 what's your opinion? (see logs)
22:04 ShadowBot https://github.com/minetest/minetest/issues/4086 -- Fix locked hardware buttons on Android by MoNTE48
22:09 torgdor joined #minetest-dev
22:09 est31 I'll test it
22:09 est31 and if it works, I'll take a closer look
22:09 est31 but tbh its not the kind of patches i want
22:09 est31 non obvious changes without any explanation
22:09 est31 it has nothing to do with monte48 being unpopular or something
22:09 est31 just a pr without any explanation is bad
22:10 est31 even if the code were perfect, i still didnt knew what it was doing
22:10 est31 and that is really bad
22:10 paramat ok
22:11 kaeza joined #minetest-dev
22:12 paramat the bugs it fixes are not blockers
22:22 kaeza blah. got free day and somebody locks translations, and now release is delayed
22:22 kaeza :(
22:22 est31 sorry about accidentially keeping the translations locked
22:24 est31 man this gradle is such a lame duck
22:24 est31 how many files does it have to compile, five?
22:24 est31 it takes approx. 30 seconds for it
22:24 est31 i hate gradle to be honest but google gives us no choice
22:24 est31 we had to switch one way or another
22:26 kaeza really, why does every effing language have its own build system/whatever?
22:26 kaeza make is all you need
22:27 Fixer est31, what about updating translations once more, just for good luck? %)
22:27 est31 kaeza, thats what I dont understand either
22:28 est31 every language tries to do "the perfect" build system
22:28 est31 and for some languages build system and package manager are even the same tool
22:28 est31 like rust for example
22:29 est31 and this affects interpreted languages too
22:29 est31 just look at python
22:29 est31 pip
22:29 est31 or npm
22:30 kaeza you know, it always reminds me of https://xkcd.com/927/
22:32 Fixer ahaha, yes
22:32 Fixer thinked about that pic beforehand
22:36 nrzkt joined #minetest-dev
22:36 Guest37757 joined #minetest-dev
22:39 est31 nrzkt2, can you publish the apk somewhere as release candidate?
22:39 est31 then Fixer can test it
22:39 Fixer ( ͡° ͜ʖ ͡°)
22:42 nrzkt2 http://www.megafileupload.co​m/jikM/Minetest-release.apk
22:43 Fixer oh lol
22:43 nrzkt2 didn't get time to upload it on my server XD
22:43 Fixer why you hate me that much ._.
22:43 Fixer downloading
22:44 Fixer are you sure that this site will not inject malicious code in apk?
22:44 Fixer nrzkt2, give me its md5 please
22:44 nrzkt2 don't know but i uploaded it fast
22:44 nrzkt2 md5 is shit
22:44 est31 sha256sum
22:44 est31 thats better
22:44 nrzkt2 5fbf580ffda7a1069751adb48a103c0d073593651d0​4576753527beca4c9a8146b848ef8baa97b112d5a2f​710e2ec988b14be7951bf2d2f6bcd761edca421f43
22:45 nrzkt2 here is the sha512
22:45 Fixer ok thanks
22:46 Fixer it is ok
22:46 Fixer let me try it
22:47 Fixer no wait, tablet is in use, i will check it in 20 min or so :)
22:47 Fixer nrzkt2, it includes that patch that fixes hardware buttons?
22:47 nrzkt2 it's the master bare metal
22:47 nrzkt2 9d3d43f3340030db8855ebc394e969f10d00dc30
22:48 nrzkt2 commit ^
22:48 Fixer ok
22:52 Megal joined #minetest-dev
22:54 EvergreenTree joined #minetest-dev
23:01 EvergreenTree Hey everyone!  I was looking into #3565 to see if I could come up with a fix, but I'm stuck
23:01 ShadowBot https://github.com/minetest/minetest/issues/3565 -- Death message repeats until respawn
23:02 EvergreenTree It appears that this bug only happens with nodes that contain a damage_per_second field, but not with drowning
23:02 electrodude512 joined #minetest-dev
23:02 EvergreenTree I found the place in the source code where damage_per_second is handled, but I'm not sure if that is where the bug is:  https://github.com/minetest/minetest/​blob/master/src/environment.cpp#L2553
23:04 crazyR joined #minetest-dev
23:04 EvergreenTree I checked, and it looks like this code hasn't been changed since 2011.  Either the bug has been around that long, or I'm looking in the wrong place
23:04 EvergreenTree Does anyone have any ideas?
23:05 est31 perhaps
23:05 est31 you can try to bisect to find the commit that caused the bug
23:07 EvergreenTree Does that mean I have to recompile for every time to decide whether a commit is good or bad?
23:07 est31 yes
23:08 est31 but the closer you get the less you have to recompile
23:09 est31 EvergreenTree, I assume you know about the -j option for make?
23:09 EvergreenTree yes
23:10 EvergreenTree I have never actually used bisect before, so time to learn
23:10 EvergreenTree (i'm not terribly experienced with git)
23:10 Fixer i besected once
23:10 Fixer bisect*
23:11 Fixer just google it, it is pretty easy, i think it needs just two git commands for bad and good commit hash
23:12 Fixer it is very useful
23:12 Fixer it helped me to find some bug commits very fast
23:37 Fixer hmm, so I've installed it but it is old version :/
23:38 Fixer !tell nrzkt2 it is old version: 6a1047d8 why you posted it?
23:38 ShadowBot Fixer: O.K.
23:39 Fixer 21 aug 2015 build :/
23:39 Fixer i will try zeno's apk instead
23:48 Fixer how can I delete singleplayer map in android? :/ i remember a button but now it is gone?
23:50 Fixer android: can't go back from change keys window :/
23:50 Fixer android: save/cancel not working
23:51 Fixer android: no world delete button
23:53 est31 Fixer, android has no change keys window
23:53 DonBatman Fixer use the file manager
23:53 est31 and what do you mean with save/cancel
23:53 Fixer est31, it has
23:53 DonBatman goto minetest/worlds folder and delete
23:53 est31 Fixer, can you make a screenshot
23:53 Fixer est31, go to settings tab and down right see "Change keys"
23:54 est31 ahhhh
23:54 est31 I see
23:54 est31 confirmed
23:54 est31 yeah its issue worthy
23:54 KaadmY how to use git bisect: https://xkcd.com/1597/
23:54 Fixer DonBatman, but is not there was a button? :/
23:54 est31 KaadmY, lol
23:55 Fixer KaadmY, just be careful
23:55 KaadmY pretty much how I use git ;)
23:55 Fixer once i mistyped git checkout --.
23:56 KaadmY musta been fun!
23:56 Fixer recloned everything again lol
23:56 Fixer android: shaders are not working on lenovo b8000-h kitkat 4.4.2
23:56 est31 Fixer,  that is known
23:57 Fixer android: navigating in "Advanced Settings" is fucking horrible
23:58 est31 Fixer, do you want to file bugs or should I
23:58 Fixer android: server list scrollbar and buttons to scroll should be bigger (tested on 10" screen)
23:58 DonBatman Fixer try having big fingers and doing it on a phone
23:58 Fixer est31, it is 3:00 here, can you file it?
23:58 est31 okay
23:58 DonBatman I use a wireless keyboard/mouse
23:59 est31 its only 2:00 here :)
23:59 Fixer let me connect to some server

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