Minetest logo

IRC log for #minetest-dev, 2019-03-05

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

All times shown according to UTC.

Time Nick Message
00:00 paramat then when the app is updated it will be good to inform about the big improvements as much as possible
00:00 sfan5 what is left now:  Notify known package maintainers
00:00 sfan5 maybe someone else can do that
00:01 paramat the playstore page should make the improvements very clear, as the app has been rather broken recently
00:07 * sofar is hereby notified
00:08 GreenDimond darn I was gone when it was released
00:09 GreenDimond cool Minetest is updated how about that site screenshot
00:11 sfan5 joined #minetest-dev
00:13 GreenDimond Now I get to make my server work all over again
00:13 GreenDimond I am sooooo excited
00:13 rom1504 joined #minetest-dev
00:13 luk3yx The stable-5 branch of minetest_game appears to still have minetest_game 0.4.17.1
00:14 GreenDimond oof yeah
00:14 GreenDimond paramat, MTG is still stuck in the stone ages
00:14 paramat sfan5 ^
00:14 luk3yx Can someone ping me when it's fixed so I can rebuild my snap?
00:14 paramat or rubenwardy
00:14 sfan5 fixed
00:15 paramat thanks
00:15 GreenDimond wat
00:15 sfan5 luk3yx: you should probably be using tags, not the stable-5 branch
00:15 sfan5 since you'll need to specify the version number in your build script anyway
00:15 sfan5 I guess..
00:16 GreenDimond I guess MTG doesnt need a version bump
00:16 luk3yx The version is auto-detected based on tags, it's weird
00:17 GreenDimond Is there a list of breaking mod changes I can reference
00:17 GreenDimond I need to know what methods mods are using that are gonna break
00:17 sfan5 changelog
00:19 GreenDimond fun fun
00:21 Fixer joined #minetest-dev
00:37 GreenDimond Whens the ubuntu repo getting updated
00:39 rubenwardy wait
00:39 rubenwardy was weblate merged in before the release?
00:40 sfan5 <paramat> translation string freeze started on 23rd feb and there have been no string changes since
00:40 sfan5 i assume yes
00:40 paramat https://github.com/minetest/minetest/commit/17451f15dcd0c524225305e2bc255e8281a99974
00:40 rubenwardy I think he meant source string changes
00:40 rubenwardy oh ok
00:41 paramat see commits on 23rd/24th feb
00:46 paramat however, i don't understand translation stuff
01:26 kaeza joined #minetest-dev
01:43 ordoflammae joined #minetest-dev
01:52 turtleman joined #minetest-dev
01:59 Cornelia joined #minetest-dev
02:04 Miner_48er joined #minetest-dev
02:39 paramat joined #minetest-dev
02:49 fluxflux joined #minetest-dev
02:56 GreenDimond Hm
02:56 GreenDimond just to confirm
02:56 GreenDimond data set by set_attribute is still stored on the player across version change right?
02:56 GreenDimond the meta methods will still be able to access already stored data?
02:57 paramat ^ sofar
03:03 Cornelia joined #minetest-dev
03:05 Cornelia joined #minetest-dev
03:11 Cornelia joined #minetest-dev
03:34 paramat i think yes and yes, see 'Breaking changes and deprecations ' here https://dev.minetest.net/Changelog#0.4.16_.E2.86.92_5.0.0 there's not much mod breakage
03:58 est joined #minetest-dev
04:00 est oh indeed there's been a release
04:00 GreenDimond yuppers
04:00 est so I guess that means updating the ppa then
04:00 sofar currently bumping/testing for clearlinux
04:00 sofar and flatpak/flathub PR
04:00 GreenDimond Yeah a ppa update would be nice
04:00 GreenDimond > le me on xubuntu
04:01 GreenDimond > le me with 2 versions of 0.4.17.1 because my system copy is stuck on 0.4.17.1
04:01 Cornelia joined #minetest-dev
04:02 est https://code.launchpad.net/~minetestdevs/+recipe/minetest-stable
04:02 est it's building
04:02 est also, congrats guys to the 5.0.0 release
04:03 GreenDimond \o/
04:03 est it's a big milestone
04:04 est !rainbow PARTY
04:04 sofar haha, bot fail
04:04 est the bot's gone T.T
04:05 sofar the ninja is hard to find
04:05 GreenDimond yayupdated
04:05 * sofar waits for highly optimized clearlinux build
04:06 * GreenDimond sudo apt updates
04:06 est GreenDimond, not yet
04:06 est it's still compiling
04:06 GreenDimond u sure
04:06 est pretty sure
04:07 GreenDimond It said it was uploading
04:07 GreenDimond and now its on the release page
04:07 GreenDimond are it did lie to me?
04:07 est if sudo apt update works, it's been uploaded
04:08 est it should be done in ~20 minutes
04:08 GreenDimond so it *did* lie to me
04:09 est https://launchpadlibrarian.net/413728711/buildlog_ubuntu-trusty-amd64.minetest_5.0.0-ppa0~ubuntu14.04.1_BUILDING.txt.gz
04:09 est build fail :/
04:09 est > Insufficient gcc version, found 4.8.4.  Version 4.9 or higher is required.
04:09 sofar I'll give you some clearlinux rpms?
04:09 sofar I just finished building
04:09 est No ubuntu 14.04 then
04:09 GreenDimond Who uses 14.04 any more
04:10 est idk it's still supported by canonical
04:10 est but only a matter of months until support ends I think
04:11 est not worth the hassle
04:18 sofar MT build looks good, pushing into our release system
04:25 GreenDimond yoyitsupdated
04:26 est ????
04:28 GreenDimond ouch
04:28 GreenDimond Minetest quit
04:29 GreenDimond yay now it works
04:30 ssieb joined #minetest-dev
04:34 est very nice, then my job is done :)
04:40 paramat thanks
04:40 est thank you for working so hard for the release!
04:41 est also btw, the topic needs an update
04:41 est take care and goodbye
05:03 kaeza joined #minetest-dev
05:09 sofar alright, clearlinux will have minetest-5.0.0 when the next release passes QA :)
05:42 sofar I'm seeing if I can update the flathub flatpak
06:19 sofar running minetest-5.0.0 flatpak right now :)
06:19 sofar will PR it to flathub after a few tyre kicks
06:30 sofar https://github.com/flathub/net.minetest.Minetest/pull/9
06:36 nerzhul nice job for the tag :)
06:36 nerzhul now re-review pr and merge :)
06:36 nerzhul i will produce android build asap
06:36 sofar two of mine are ready to go :D
06:36 sofar I need a tiny patch on top of a third
06:37 sofar all good fun, we should do a merge fest
06:37 sofar let's merge lots of open things XD
06:37 nerzhul link those ready PRs ?
06:39 kaeza joined #minetest-dev
06:40 nerzhul i'm building android release
06:44 sofar #8304 #8135 and #8140
07:11 nerzhul merging #8304
07:11 nerzhul merging #8135
07:12 nerzhul latest i ait
07:12 nerzhul wait*
07:12 nerzhul merging #8255
07:13 nerzhul and #8194
07:14 nerzhul and #7173
07:14 sofar yeah I'll throw 1 commit on #8410 as requested
07:17 nerzhul merging #7719
07:34 nerzhul apk is published in beta on play store and on the gh release page
07:35 nerzhul i push a commit to master to include the android keystore (without pass) to the repository
08:07 kaeza joined #minetest-dev
08:56 ensonic joined #minetest-dev
09:11 nerzhul merging #8298
09:33 proller joined #minetest-dev
09:34 kaeza joined #minetest-dev
09:45 Megaf joined #minetest-dev
09:54 proller joined #minetest-dev
10:48 rubenwardy does contact @ minetest.net actually go to someone?
10:49 Thomas-S @nerzhul: I downloaded Minetest 5.0 from the playstore. It seems like the player model is one node below ground in 3rd person view.
10:50 rubenwardy sounds like it's still shipping 0.4.17 of MTG
10:52 proller joined #minetest-dev
10:53 Thomas-S @rubenwardy: I don't think so because the new fence rails are available. As far as I know they didn't exist in 0.4.17
10:55 * rubenwardy wonders how to make a download manager for Minetest
10:55 rubenwardy argh
10:57 sfan5 merging game#2324
10:57 sfan5 rubenwardy: a progress bar when downloading from CDB would be nice, btw
10:58 rubenwardy yeah
10:58 rubenwardy my plan is to have engine support for queuing multiple downloads, and showing a progress bar fixed to the bottom of the screen
10:58 rubenwardy maybe
10:58 rubenwardy idk
10:59 rubenwardy having a download manager would be useful for both downloading loads of mods (ie: for dependencies), or for downloading images
11:00 rubenwardy I'll have to do some research
11:21 Megaf_ joined #minetest-dev
11:21 nerzhul i updated my mtg to the tag
11:21 nerzhul and i did the build in a cleanall env
11:21 nerzhul hmmm
11:26 proller joined #minetest-dev
11:32 Fixer joined #minetest-dev
11:39 p_gimeno can anyone please update the title?
11:40 Topic for #minetest-dev is now Minetest core development and maintenance. Minetest 5.0.0 released! Chit-chat goes to #minetest. http://irc.minetest.net/minetest-dev/ http://dev.minetest.net/
11:41 p_gimeno thanks :)
11:57 rubenwardy play store is weird     https://i.rubenwardy.com/oM0F4.png
12:31 calcul0n joined #minetest-dev
12:32 ordoflammae joined #minetest-dev
13:35 sfan5 turns out I accidentally built the 64-bit windows build with gc64 luajit
13:35 sfan5 guess we'll find out if it causes any problems (unlikely)
13:41 rubenwardy nice
13:41 Fixer smile.png, downloading it just this second
14:00 Persi joined #minetest-dev
14:36 kaeza joined #minetest-dev
14:40 proller joined #minetest-dev
14:50 benrob0329 joined #minetest-dev
14:55 Darcidride joined #minetest-dev
14:58 YuGiOhJCJ joined #minetest-dev
15:02 rubenwardy is Android beta only currently?
15:04 Megaf joined #minetest-dev
15:05 p_gimeno could anyone with access to the gitlab repo remove the obsolete pull/x/y branches?
15:05 p_gimeno git br -a | sed -nr '/^  remotes\/ogitlab\/pull\//s/  remotes\/ogitlab\///p' | xargs git push -d ogitlab
15:06 p_gimeno sorry, but with gitlab instead of ogitlab, ogitlab is my 'origin gitlab' remote
15:10 Megaf joined #minetest-dev
16:19 argyle77 joined #minetest-dev
17:21 nerzhul joined #minetest-dev
17:28 nerzhul rubenwardy: strange why it has the beta information on the release ?
17:28 nerzhul android is only beta
17:28 rubenwardy did you push to the right channel?
17:29 nerzhul yep
17:29 rubenwardy I'm in the beta testing thing
17:29 rubenwardy oh, so you haven't released 5.0.0 completely on Android yet?
17:30 nerzhul https://i.ibb.co/Y7NTms2/Capture-d-cran-de-2019-03-05-18-29-40.png
17:30 nerzhul ruben: remember you have read only access to the play store console for mt :)
17:31 kaeptmblaubaer joined #minetest-dev
17:34 nerzhul other note: i pushed the keystore to the repository (without the password)
17:38 kaeptmblaubaer joined #minetest-dev
17:58 kaeptmblaubaer joined #minetest-dev
18:12 Calinou making the keystore public doesn't sound ideal, even if the password isn't there
18:12 Calinou (as it makes offline attacks possible)
18:13 rubenwardy it was a very bad idea
18:18 nerzhul rewrite history then ?
18:18 nerzhul i don't think it's a problem as it's for signing releases and it's locked behind dev accounts, not only signature
18:22 sofar flathub took my 5.0.0 update, get your flatpak :)
18:29 benrob0329 joined #minetest-dev
18:39 rubenwardy loads of people are getting the %1s$ or whatever confirmation dialog bug
18:41 sfan5 on Windows?
18:45 ensonic joined #minetest-dev
19:17 kaeza joined #minetest-dev
19:25 ssieb joined #minetest-dev
19:37 fwhcat joined #minetest-dev
19:38 proller joined #minetest-dev
19:49 fwhcat joined #minetest-dev
19:52 fwhcat joined #minetest-dev
19:54 Cornelia joined #minetest-dev
19:57 kaeza joined #minetest-dev
20:02 fwhcat joined #minetest-dev
20:24 Lia joined #minetest-dev
20:51 Cornelia joined #minetest-dev
21:24 kaeza joined #minetest-dev
22:32 paramat joined #minetest-dev
22:35 paramat i'll merge #8303 in 15 mins
22:35 paramat "Change 'num_emerge_threads' default to 1"
22:51 GreenDimond joined #minetest-dev
22:55 sfan5 paramat: in my experience sqlite3 crashes pretty fast when used in a thread-unsafe manner
22:55 sfan5 I can't believe we have been using sqlite3 incorrectly all the time
22:56 paramat ok, i'm delaying dealing with that, won't update docs to mention that
22:56 sfan5 unfortunately I haven't had time to look into how the emerge threads actually push data to the database
22:56 paramat the default is still best 1 for other reasons
22:57 sfan5 since sofar mentioned the thread unsafety, maybe knows how the data actually gets from EmergeThread into the database
22:57 paramat merging 8303
22:58 rubenwardy #8303
22:58 paramat heh
23:06 sfan5 sorry not sofar, it was nerzhul who mentioned the thread safety of sqlite
23:07 sofar I'm with sfan5 here, this is tricky because it's possible someone has a "bad" OS sqlite3
23:07 sofar but, highly unlikely
23:09 sfan5 i'm not sure how #8300 even came to revolve around the backend
23:10 sfan5 to me this looks like a issue in the noise code
23:10 sfan5 sqlite is not even involved
23:11 sfan5 "I got postgresql working, and still experienced the crash." oh yeah of course, who would have expected
23:13 sfan5 paramat: what else speaks for "num_emerge_threads = 1 is optimum for most users"?
23:13 paramat yes, 8300 is about vein ore's use of noise
23:13 sfan5 the only thing I can see is the warning in conf.example about singleplayer and Lua code
23:13 sfan5 though I wonder why this would interfere with singleplayer?
23:14 paramat see https://github.com/minetest/minetest/pull/8066
23:14 paramat see the links, which recommend '1'
23:15 paramat 1 is apparently optimum for many users, and safest
23:16 paramat for example whenever stuff runs in 'on generated'
23:16 nerzhul 1 is safest yes
23:16 nerzhul i'm not very confident about our emerge thread code
23:16 nerzhul i think some parts are no so safe
23:17 sfan5 it sounds like hecks was running a *heavy* lua mapgen
23:18 sfan5 the situation described is very much possible since all emerge threads + the server thread compete for the env lock
23:18 sfan5 for Lua execution that is, the C++ mapgen is no affected
23:23 sfan5 hm, yeah a default of 1 sounds useful
23:23 sfan5 perhaps the automatic detection code can even be removed
23:24 sfan5 it's not very good either, my ryzen counts as 12 "cores" and will cause 10 emerge threads to be spawned
23:24 paramat yeah maybe manual is best
23:26 sfan5 by the way, there seems to a quite bad bug with many items in a single inventory list
23:26 sfan5 https://forum.minetest.net/viewtopic.php?f=6&amp;t=22285
23:39 sfan5 ok looks like someone broke packets larger than 65535
23:39 GreenDimond saywhatnow
23:40 GreenDimond are you saying a breaking change is broken
23:40 nerzhul sfan5: wtf
23:41 paramat sofar if you are able could you review https://github.com/minetest/minetest_game/pull/2125 ?
23:41 GreenDimond > minetest release broke
23:42 GreenDimond > calmly reviews beds
23:43 paramat sofar and https://github.com/minetest/minetest_game/pull/2286
23:46 paramat heh i was half expecting 5.0.0 to be broken
23:47 sfan5 nerzhul: https://github.com/minetest/minetest/issues/8325
23:48 GreenDimond sfan5: Is it just 5.1.0? Or 5.0.0?
23:48 sfan5 5.0 obviously
23:48 GreenDimond Issue says 5.1.0 but k
23:53 paramat both
23:54 GreenDimond so if the bug is in a breaking change
23:54 GreenDimond is the fix going to be a breaking change
23:54 sfan5 what are you even on about
23:57 Lia joined #minetest-dev
23:58 GreenDimond 5.0.0 is a breaking release
23:58 GreenDimond some versions of the dev client werent compat with others
23:59 GreenDimond so if the packet bug is part of a breaking change, would the fix also be a breaking chnage

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