| Time |
Nick |
Message |
| 00:00 |
|
kaeza joined #minetest-dev |
| 00:30 |
* VanessaE |
smacks hmmmm around a bit with a large OOM error :) |
| 00:55 |
|
OldCoder joined #minetest-dev |
| 01:48 |
|
blaise joined #minetest-dev |
| 01:54 |
|
blaise joined #minetest-dev |
| 03:40 |
VanessaE |
hmmmm: Any chance we can get 0.4.13 to be released in the form of a self-extracting archive of some kind? |
| 03:40 |
VanessaE |
zip or 7z or whatever is convenient |
| 03:52 |
|
Player_2 joined #minetest-dev |
| 04:02 |
hmmmm |
what do you mean? aren't releases already made as .zip |
| 04:13 |
|
blaise joined #minetest-dev |
| 04:37 |
|
paramat joined #minetest-dev |
| 04:49 |
paramat |
now pushing geme#614 |
| 04:49 |
paramat |
game#614 |
| 04:50 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/614 -- Default/trees: Combine sapling ABMs into one ABM by paramat |
| 04:52 |
|
Hunterz joined #minetest-dev |
| 04:54 |
paramat |
complete |
| 05:18 |
|
selat joined #minetest-dev |
| 06:12 |
|
kaeza joined #minetest-dev |
| 06:12 |
paramat |
this one needs discussion game#616 |
| 06:12 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/616 -- Default: Rename tree nodes and textures to underscore versions by paramat |
| 07:00 |
|
Krock joined #minetest-dev |
| 07:04 |
|
lag01 joined #minetest-dev |
| 07:07 |
|
Amaz joined #minetest-dev |
| 07:22 |
|
kaeza joined #minetest-dev |
| 08:00 |
paramat |
this is better game#617 |
| 08:00 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/617 -- Default: Rename pine tree nodes, textures and mapgen aliases by paramat |
| 08:22 |
paramat |
^ sfan5 |
| 08:22 |
|
paramat left #minetest-dev |
| 08:33 |
|
cheapie joined #minetest-dev |
| 08:45 |
|
nore joined #minetest-dev |
| 08:48 |
|
blaze joined #minetest-dev |
| 09:06 |
|
blaze joined #minetest-dev |
| 09:11 |
|
WSDguy2014 joined #minetest-dev |
| 09:40 |
|
deltib joined #minetest-dev |
| 09:59 |
|
SopaXorzTaker joined #minetest-dev |
| 10:25 |
|
est31 joined #minetest-dev |
| 10:38 |
|
twoelk joined #minetest-dev |
| 10:58 |
|
TBC_x joined #minetest-dev |
| 11:03 |
est31 |
nore, can you review some french translation changes? |
| 11:04 |
est31 |
first this https://github.com/est31/minetest/commit/5ffb752d3ae4ba14729d7301be544a6e8fbf51ce |
| 11:04 |
est31 |
and https://github.com/est31/minetest/commit/5622fb2cee6d86f0220ea5773ec78f5f2f6714c3 |
| 11:05 |
est31 |
also somebody needed for russian https://github.com/est31/minetest/commit/247dda5005ae35f6047008938aadd5efcd78a494 |
| 11:05 |
est31 |
say if they are ok, then I'll push them all |
| 11:08 |
est31 |
Me wonders whether "По Ñтому пути мира нет:" is a good translation for "Provided world path doesn't exist: " |
| 11:09 |
est31 |
google translator gives "In this way there is no peace:" |
| 11:11 |
|
lag01 joined #minetest-dev |
| 11:18 |
est31 |
russian looks good too |
| 11:21 |
est31 |
except that single string |
| 11:22 |
est31 |
any russian speakers here? |
| 11:23 |
est31 |
ah |
| 11:23 |
est31 |
google translator gives me for "мира нет" "there is no peace" |
| 11:23 |
|
stormchaser3000 joined #minetest-dev |
| 11:23 |
est31 |
but for "мира" it gives me world |
| 11:24 |
|
stormchaser3000 left #minetest-dev |
| 11:24 |
est31 |
now нет is "no", know that much russian |
| 11:24 |
est31 |
guess its ok then |
| 11:24 |
est31 |
pushing the translations |
| 11:25 |
|
Calinou joined #minetest-dev |
| 11:51 |
VanessaE |
hmmmm: of course they're zip, but they're not self-extracting i.e. you can't just "run" the file to have it extract. Seems like a lot of people who are otherwise intelligent players can't figure out the concept of unzip -> cd to minetest/bin -> execute minetest.exe, so I figured, why not a self-extracting archive that handles all of these steps? (plus a dialog box on first-run that tells the user where to find the |
| 11:51 |
VanessaE |
game next time around) |
| 12:10 |
Calinou |
we should have a proper NSIS installer... |
| 12:10 |
Calinou |
that's all |
| 12:10 |
Calinou |
and portable .zip builds for those who want it |
| 12:16 |
|
twoelk joined #minetest-dev |
| 12:19 |
|
nore joined #minetest-dev |
| 12:19 |
|
nore_ joined #minetest-dev |
| 12:20 |
kaeza |
I wrote this long ago, FWIW: https://forum.minetest.net/viewtopic.php?f=3&t=4245 |
| 12:20 |
|
nore_ left #minetest-dev |
| 12:21 |
VanessaE |
the important part of that thread: https://forum.minetest.net/viewtopic.php?p=67020#p67020 |
| 12:22 |
nore |
est31: the translations look good |
| 12:35 |
kaeza |
related: https://github.com/kaeza/minetest/commit/3285bc69afe7ae750af9040616b51757f94dd312 |
| 12:43 |
|
paramat joined #minetest-dev |
| 12:52 |
paramat |
nore sfan5 any comments/approval for game#617 ? |
| 12:52 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/617 -- Default: Rename pine tree nodes, textures and mapgen aliases by paramat |
| 12:54 |
nore |
paramat: looks good |
| 12:54 |
nore |
but I think jungletrees could be changed too |
| 12:55 |
nore |
aliases aren't that expensive |
| 12:57 |
paramat |
yeah perhaps, sfan5 may prefer that too |
| 12:58 |
paramat |
if so there are some issues to consider in 616 |
| 13:00 |
VanessaE |
paramat: I don't recall seeing an answer: |
| 13:00 |
VanessaE |
if A is aliased to B, and C comes along that's aliased to A, will C = B or will the engine throw an error? |
| 13:01 |
paramat |
i dunno |
| 13:12 |
paramat |
that may be another reason to try avoiding aliases. also the jungle node names may be the wrong form but they are consistent as a group and players are used to them |
| 13:12 |
paramat |
how about i push this as is, then we can consider the jungle names later? |
| 13:14 |
|
Krock joined #minetest-dev |
| 13:17 |
|
nrzkt joined #minetest-dev |
| 13:21 |
nore |
VanessaE: I think it will work |
| 13:21 |
nore |
I am not completely sure though |
| 13:24 |
kahrl |
as I recall, it will work as long as the alias from A to B is defined earlier than the alias from C to A |
| 13:24 |
kahrl |
(so that the definition of A can be looked up at the time C is defined) |
| 13:25 |
twoelk |
no alias database? ;-P |
| 13:35 |
|
paramat left #minetest-dev |
| 13:56 |
|
nore joined #minetest-dev |
| 13:57 |
|
alket joined #minetest-dev |
| 14:30 |
|
blaze joined #minetest-dev |
| 14:53 |
|
SopaXorzTaker joined #minetest-dev |
| 14:57 |
|
kaeza joined #minetest-dev |
| 15:59 |
|
SopaXorzTaker joined #minetest-dev |
| 16:03 |
|
hmmmm joined #minetest-dev |
| 17:34 |
|
crazyR joined #minetest-dev |
| 17:40 |
|
lag01 joined #minetest-dev |
| 17:45 |
|
sloantothebone joined #minetest-dev |
| 17:52 |
|
Yepoleb joined #minetest-dev |
| 18:22 |
|
lag01 joined #minetest-dev |
| 18:30 |
|
est31 joined #minetest-dev |
| 18:30 |
est31 |
VanessaE, no, its not transitive: https://github.com/minetest/minetest/blob/990a96578f20244626b6b9f67f8e79a7e2e614ea/src/itemdef.cpp#L222 |
| 18:30 |
est31 |
and here too https://github.com/minetest/minetest/blob/990a96578f20244626b6b9f67f8e79a7e2e614ea/src/itemdef.cpp#L281 |
| 18:31 |
VanessaE |
ok |
| 18:31 |
est31 |
kaeza, it would be great to have an installer |
| 18:31 |
VanessaE |
in which case game#617 is gonna break things. |
| 18:31 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/617 -- Default: Rename pine tree nodes, textures and mapgen aliases by paramat |
| 18:31 |
est31 |
but I have no windows to do it, so you'll have to convince a minetest dev with windows |
| 18:32 |
VanessaE |
kaeza is not here. |
| 18:32 |
est31 |
i know |
| 18:32 |
est31 |
tab complete didnt work :(( |
| 18:35 |
est31 |
BlockMen uploads the windows version at around 00:00 CEST, so in around 3 hours 25 minutes |
| 18:42 |
est31 |
ok hmmmm kahrl nore nrzkt sfan5 Tesseract, last call, to propose any changes to the core engine |
| 18:43 |
est31 |
and I dunno about minetest_game, will have to ask paramat which things he wants in until 0.4.13 |
| 18:43 |
sfan5 |
est31: you'll also need to tell me to make a build |
| 18:44 |
est31 |
which build do you want to do? |
| 18:44 |
sfan5 |
mingw |
| 18:44 |
sfan5 |
and it's not "want" |
| 18:45 |
VanessaE |
make them self-extracting zip/7z please |
| 18:45 |
est31 |
which builds does blockmen make? |
| 18:45 |
sfan5 |
i don't think anyone else has a build env working |
| 18:45 |
sfan5 |
msvc |
| 18:45 |
est31 |
ok |
| 18:45 |
VanessaE |
let's finally make it easy for users to install. |
| 18:45 |
sfan5 |
VanessaE: if you tell me how to do that with p7zip i can do that |
| 18:45 |
est31 |
ok |
| 18:45 |
VanessaE |
sfan5: http://www.wikihow.com/Use-7Zip-to-Create-Self-Extracting-excutables |
| 18:45 |
sfan5 |
i know how to use the gui |
| 18:46 |
VanessaE |
looks pretty simple, but no idea how you'd script it |
| 18:46 |
sfan5 |
there is no 7zip gui on linux |
| 18:46 |
VanessaE |
zeno had something the that though |
| 18:46 |
VanessaE |
damned if I can remember |
| 18:46 |
est31 |
also, we need sb to create a changelog |
| 18:46 |
kahrl |
what about the blockers? |
| 18:46 |
Warr1024 |
making 7z sfx isn't hard... |
| 18:46 |
kahrl |
#3011 #2653 #2145 |
| 18:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/3011 -- ERROR[main]: Lua: error in error handling |
| 18:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/2653 -- (Another) Possible memory leak |
| 18:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/2145 -- Multi-second lag in default singleplayer game |
| 18:47 |
VanessaE |
release can't be made until those ^^^^^ are fixed |
| 18:47 |
Warr1024 |
it's just the 7z sfx module with the archive concatenated after it. |
| 18:47 |
VanessaE |
and any blockers that were for some previous version that has been pushed back |
| 18:47 |
est31 |
the issue has been created january |
| 18:47 |
est31 |
in |
| 18:47 |
est31 |
the last one |
| 18:47 |
VanessaE |
sfan: https://netbeansscribbles.wordpress.com/2015/01/30/creating-a-self-extracting-7zip-archive-multi-platform/ |
| 18:48 |
est31 |
thats before 0.4.12 |
| 18:48 |
Warr1024 |
the *.sfx file is just an exe blob. cat 7z.sfx archive.7z >install.exe |
| 18:48 |
Warr1024 |
there can be an optional config file (plain-text) that can be catted between. |
| 18:48 |
kahrl |
the blocker label was added only 19 days ago though |
| 18:48 |
sfan5 |
VanessaE: thanks ill look at that |
| 18:50 |
hmmmm |
est31: Hold on a minute |
| 18:50 |
hmmmm |
why are you releasing in a known bad state |
| 18:50 |
est31 |
what are your issues? |
| 18:50 |
hmmmm |
how about all the graphical regressions on android |
| 18:51 |
hmmmm |
and then what about the ATI/intel issues |
| 18:51 |
hmmmm |
and then there are these unknown problems with delete_area |
| 18:51 |
est31 |
they existed before too |
| 18:51 |
hmmmm |
and what about the race conditions? |
| 18:51 |
est31 |
graphical regressions on android? |
| 18:53 |
est31 |
is it ATI/intel, and not just old GL API version? |
| 18:53 |
est31 |
also do you mean #2866 with it |
| 18:53 |
ShadowBot |
https://github.com/minetest/minetest/issues/2866 -- Graphical regression. |
| 18:53 |
|
sloantothebone joined #minetest-dev |
| 18:54 |
|
Amaz joined #minetest-dev |
| 18:58 |
est31 |
hmmmm, is it that issue, or some other github issue? |
| 19:01 |
est31 |
The "possible memory leak" one requires imo a deep rewrite of when we delete mapblocks from the scene |
| 19:01 |
est31 |
about 3011, it seems to be of recent origin? |
| 19:01 |
VanessaE |
est31: also #2350 |
| 19:01 |
ShadowBot |
https://github.com/minetest/minetest/issues/2350 -- Wielded is glichted on certain OpenGL versions |
| 19:01 |
VanessaE |
this seems to be affecting more people than was originally though |
| 19:01 |
VanessaE |
thought* |
| 19:01 |
est31 |
I mean, is it our fault, or is it just modders assigning their errors to the engine |
| 19:02 |
VanessaE |
plus there are fps regressions |
| 19:03 |
VanessaE |
the more map that's loaded, the lower your FPS, to the point of the autotuner really having to scale back your view range for the same scene as would have produced higher fps right after login |
| 19:03 |
VanessaE |
and then there's the autotuner itself not scaling back the range properly from time to time as well |
| 19:04 |
est31 |
and about 2350, which options need to be enabled? |
| 19:04 |
VanessaE |
(wanted fps: 30. actual? say 20 with a view range of 100+) |
| 19:04 |
est31 |
can you add that perhaps as github comment |
| 19:04 |
est31 |
"I can reproduce this with no option enabled" |
| 19:04 |
est31 |
or "only if fsaa is set" |
| 19:04 |
|
Calinou joined #minetest-dev |
| 19:04 |
est31 |
and so on |
| 19:04 |
VanessaE |
no idea |
| 19:04 |
VanessaE |
WSDguy is the latest victim of it |
| 19:10 |
hmmmm |
what's 3011? |
| 19:10 |
hmmmm |
#3011 |
| 19:10 |
ShadowBot |
https://github.com/minetest/minetest/issues/3011 -- ERROR[main]: Lua: error in error handling |
| 19:10 |
hmmmm |
oh |
| 19:10 |
hmmmm |
yeah if we release a minetest where people get error in error handling problems every 3-4 hours that's not too good |
| 19:11 |
hmmmm |
i don't think there needs to be a release right now, personally |
| 19:12 |
hmmmm |
also what ever happened to the Android stuttering issue |
| 19:12 |
nrzkt |
hmmmm: agree |
| 19:12 |
VanessaE |
I agree, hmmmm |
| 19:12 |
VanessaE |
(sorry est :) ) |
| 19:12 |
est31 |
stuttering? |
| 19:13 |
hmmmm |
wasn't there a stuttering using the softfp abi? |
| 19:13 |
hmmmm |
stuttering issue |
| 19:13 |
est31 |
ah that one |
| 19:13 |
est31 |
the "looking around is jittery" one? |
| 19:13 |
hmmmm |
yeah |
| 19:13 |
est31 |
that one wayward says is old, but he only opened the report recently |
| 19:14 |
est31 |
#2998 |
| 19:14 |
ShadowBot |
https://github.com/minetest/minetest/issues/2998 -- [Android] Looking around is jittery |
| 19:14 |
hmmmm |
plus TBC_x is in the process of fixing some race conditions iirc |
| 19:14 |
hmmmm |
what's going on with #2999? |
| 19:14 |
ShadowBot |
https://github.com/minetest/minetest/issues/2999 -- Fixes a freeze on the phone or NativeActivity LockScreen by MoNTE48 |
| 19:14 |
est31 |
thats only an issue if we target a newer api version |
| 19:14 |
VanessaE |
note also the crash-on-exit-to-menu that I get sometimes |
| 19:15 |
hmmmm |
crash on exit to menu usually means the server isn't getting cleaned up correctly |
| 19:15 |
hmmmm |
like a double free or use after free |
| 19:15 |
est31 |
or the client |
| 19:15 |
est31 |
can happen with it too |
| 19:15 |
hmmmm |
it's almost always the server :). |
| 19:15 |
est31 |
I had it multiple times on the client |
| 19:15 |
est31 |
while connecting to a server :) |
| 19:15 |
|
rubenwardy joined #minetest-dev |
| 19:16 |
hmmmm |
:( |
| 19:16 |
VanessaE |
I'm talking strictly about client exit-to-menu while on a server. |
| 19:16 |
est31 |
but dont have much singleplayer experience |
| 19:16 |
est31 |
perhaps its more often there |
| 19:16 |
hmmmm |
well if we can reproduce that |
| 19:16 |
VanessaE |
(though I'm sure it happens on sp also) |
| 19:19 |
est31 |
its hard to fix these bugs |
| 19:19 |
est31 |
the memory leak is one of the easiest to reproduce |
| 19:20 |
est31 |
but the other ones, are all hard to repro |
| 19:20 |
est31 |
and about android, I have no idea how the bug is created |
| 19:20 |
hmmmm |
i don't want to release software with such blatant problems |
| 19:20 |
hmmmm |
why the rush for a release?? |
| 19:21 |
est31 |
I dont even have an idea about #2973 |
| 19:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/2973 -- [Android] No text on main menu |
| 19:21 |
est31 |
hmmmm, do you think we should wait? |
| 19:21 |
hmmmm |
absolutely... |
| 19:21 |
est31 |
how long? |
| 19:21 |
est31 |
6 months? |
| 19:21 |
est31 |
8? |
| 19:21 |
hmmmm |
not how long |
| 19:21 |
hmmmm |
until things get solid |
| 19:22 |
est31 |
we should make a clear list of issues |
| 19:22 |
hmmmm |
i'm not counting months here, i'm counting severity and how common the known bugs are |
| 19:22 |
est31 |
in github |
| 19:22 |
est31 |
so that we can track them |
| 19:22 |
hmmmm |
they have the blocker tag on them don't theY? |
| 19:22 |
est31 |
not just "i remember someone said it" |
| 19:23 |
est31 |
"nobody knows which config to reproduce, but well, lets stop release for it" |
| 19:23 |
hmmmm |
like i said |
| 19:23 |
hmmmm |
if it's a /common/ error, that has more weight |
| 19:23 |
hmmmm |
a common bug, like the error in error handling one, might hbe hard to reproduce, but it's so common and fatal that it doesn't matter |
| 19:24 |
est31 |
what is that bug precisely about |
| 19:24 |
est31 |
because I see multiple different problems here |
| 19:24 |
hmmmm |
it's PROBABLY a mod problem |
| 19:24 |
est31 |
1. we do "error in error handling" at all |
| 19:25 |
est31 |
and 2. servers occasionally get it |
| 19:25 |
hmmmm |
but it's our problem because there's no way to tell where it came from or to fix it |
| 19:25 |
est31 |
so is 2. our fault= |
| 19:25 |
est31 |
? |
| 19:25 |
est31 |
is that the "blocker"? |
| 19:25 |
est31 |
or is 1. it? |
| 19:25 |
hmmmm |
that's one of the blockers |
| 19:25 |
hmmmm |
it's not an error in error handling, it's a regular runtime error that happens inside of a lua callback that originated from inside of a lua api call |
| 19:26 |
hmmmm |
anyway, https://github.com/minetest/minetest/issues/3023 |
| 19:26 |
hmmmm |
we should check this out too |
| 19:26 |
est31 |
hmmmm, you still havent answered my question |
| 19:26 |
est31 |
is it 1 or 2 |
| 19:26 |
est31 |
or both |
| 19:27 |
hmmmm |
it's both |
| 19:27 |
hmmmm |
we shouldn't ever get an "error in error handling" error |
| 19:27 |
hmmmm |
and 2. it is our fault because there's no way to diagnose it |
| 19:27 |
est31 |
they are different issues, and should be tracked by different github tickets |
| 19:27 |
hmmmm |
what are LUA_ERRERR should be LUA_ERRRUN instead |
| 19:28 |
est31 |
its the fault of the lua runtime, no? |
| 19:28 |
hmmmm |
it might just be a luajit problem for all we know |
| 19:28 |
est31 |
its like blaming your distro for a driver problem |
| 19:28 |
VanessaE |
{" |
| 19:28 |
hmmmm |
if it is a lua problem we should still find a way to work around it |
| 19:29 |
hmmmm |
right now it's unacceptable for people to see "error in error handling" every 4 hours and have their servers shut down |
| 19:29 |
est31 |
that doesnt apply to all downstream problems |
| 19:29 |
est31 |
also its no reason to stop a release |
| 19:29 |
est31 |
we arent perfect |
| 19:29 |
hmmmm |
i don't advocate a release at all. |
| 19:30 |
est31 |
you gonna add blocker to every issue? |
| 19:30 |
hmmmm |
to the issues that are blockers |
| 19:30 |
est31 |
perhaps you are right |
| 19:31 |
est31 |
lets blow 0.4.13 off |
| 19:31 |
hmmmm |
i'm just tired of minetest having a reputation for being a crashy piece of shit |
| 19:31 |
hmmmm |
it's a slippery slope argument |
| 19:31 |
hmmmm |
i am not saying we block on every issue, just the ones that are severe and common |
| 19:31 |
hmmmm |
severe meaning things that crash or are security problems |
| 19:32 |
hmmmm |
commmon meaning happens to like 3-4 people in a regular manner |
| 19:32 |
hmmmm |
on a regular basis of playing |
| 19:32 |
est31 |
you added that precision assert, didnt you? |
| 19:32 |
hmmmm |
yes |
| 19:32 |
hmmmm |
and it caught something :) |
| 19:32 |
est31 |
so it must be reverted then |
| 19:32 |
hmmmm |
no, fix the bug |
| 19:33 |
est31 |
it crashed the game |
| 19:33 |
est31 |
you convert small glitches into crashes, and then complain about minetest being crashy |
| 19:33 |
hmmmm |
that's easy to fix |
| 19:33 |
est31 |
this should give a log msg to errorstream |
| 19:33 |
hmmmm |
crashing behavior is better than subtle problems |
| 19:33 |
est31 |
certainly no |
| 19:34 |
hmmmm |
why?? |
| 19:34 |
hmmmm |
you can fix a crash |
| 19:34 |
est31 |
imagine if minetest had an error, do you want your os to freeze? |
| 19:34 |
hmmmm |
that's obviously not what happens |
| 19:34 |
est31 |
well, but should os developers have the same attitude |
| 19:35 |
hmmmm |
OS authors can control what happens in their environment |
| 19:35 |
hmmmm |
not the individual applications that run |
| 19:35 |
est31 |
we neither |
| 19:35 |
hmmmm |
all we need to do is get a backtrace, find the API it originates from, and then make the API fail |
| 19:35 |
hmmmm |
what's going on here is like |
| 19:36 |
|
MinetestForFun joined #minetest-dev |
| 19:36 |
hmmmm |
somebody misuses CreateFile() and it causes some subtle data corruption where it's not supposed to |
| 19:36 |
hmmmm |
whereas the correct behavior would be to fail |
| 19:37 |
hmmmm |
from what I understand you just want to get rid of the assertion because you want a release fast |
| 19:37 |
hmmmm |
i don't understand the obsession behind wanting a fast release |
| 19:37 |
est31 |
its not a fast release |
| 19:37 |
est31 |
this is planned for months |
| 19:37 |
hmmmm |
it's not ready |
| 19:38 |
hmmmm |
how about #2145 |
| 19:38 |
ShadowBot |
https://github.com/minetest/minetest/issues/2145 -- Multi-second lag in default singleplayer game |
| 19:39 |
|
VanessaE joined #minetest-dev |
| 19:39 |
est31 |
that seems easy to reproduce, no? |
| 19:39 |
hmmmm |
I personally can't, but multiple people can reliably and it ruins their minetest experience |
| 19:40 |
est31 |
if it actually ruins it, they have motivations to bisect it, perhaps? |
| 19:41 |
est31 |
even if they did few steps, that would be already helping |
| 19:41 |
hmmmm |
a lot of these people are not developers |
| 19:41 |
est31 |
our asserts only fail in debug mode, no? |
| 19:41 |
est31 |
you dont have to be a developer to do a bisect |
| 19:42 |
est31 |
only git and build knowledge |
| 19:42 |
est31 |
and time |
| 19:42 |
est31 |
and you have to trust yourself |
| 19:42 |
hmmmm |
if we can catch any of these people on IRC that would be possible |
| 19:42 |
est31 |
bisects are partly hard because you have to answer the question every build |
| 19:43 |
est31 |
hmmmm, <est31> our asserts only fail in debug mode, no? |
| 19:43 |
hmmmm |
yes |
| 19:43 |
hmmmm |
that doesn't mean ignore it |
| 19:43 |
hmmmm |
this is so easy |
| 19:43 |
hmmmm |
you get the backtrace |
| 19:43 |
est31 |
but a blocker tag isnt justified |
| 19:43 |
hmmmm |
you find the API screwing up |
| 19:43 |
est31 |
its easy yes |
| 19:43 |
hmmmm |
you add if (pos.X > thing || pos.X < -thing || ...) return 0; |
| 19:43 |
hmmmm |
PROBLEM SOLVED |
| 19:44 |
est31 |
but not a super serious which ruins fun for everybody |
| 19:44 |
est31 |
it also depends on people who arent us |
| 19:44 |
est31 |
and it cant be reproduced easily |
| 19:44 |
hmmmm |
alright fine, remove the blocker tag... |
| 19:47 |
hmmmm |
what's going to happen though, is when one of these bugs happen again without getting fixed, the object or whatever is going to magically warp to -2147483 |
| 19:48 |
|
kilbith joined #minetest-dev |
| 19:48 |
est31 |
well it is odd, and a bug, but no reason to put the engine to a grinding halt |
| 19:48 |
hmmmm |
just fix bugs |
| 19:48 |
est31 |
but I agree, its good to have the assert |
| 19:48 |
hmmmm |
this is an excellent opportunity to get minetest more stable |
| 19:48 |
est31 |
it helps people who have debug builds |
| 19:48 |
hmmmm |
the minute this gets released there is going to be nothing but features features features |
| 19:48 |
rubenwardy |
we need more stability in Minetest |
| 19:49 |
est31 |
so you dont want to blow the release off, but you want an extended freeze? |
| 19:49 |
hmmmm |
of coure |
| 19:49 |
hmmmm |
course |
| 19:49 |
hmmmm |
I want the freeze to be extended to the point where vanessae, rubenwardy, and everybody else complaining can say "yes, I would call this software reliable" |
| 19:49 |
VanessaE |
didn't I say a week wasn't enough like, a month ago? |
| 19:50 |
hmmmm |
I want minetest server uptime to not be measured in hours |
| 19:50 |
hmmmm |
this is so pathetic |
| 19:50 |
hmmmm |
this is like RoR |
| 19:51 |
rubenwardy |
I currently have to kill my server every 12 hours, as minetest.delete_blocks causes blocks to not emerge |
| 19:51 |
hmmmm |
rubenwardy: could you help me with this |
| 19:51 |
hmmmm |
i desperately want to get minetest.delete_area to work too |
| 19:52 |
rubenwardy |
What sort of help are you looking for? |
| 19:52 |
hmmmm |
testing things etc. |
| 19:52 |
rubenwardy |
ofc then |
| 19:53 |
hmmmm |
maybe it would be better to go into PM |
| 19:53 |
hmmmm |
mind if I PM you? |
| 19:53 |
rubenwardy |
that's fine |
| 19:57 |
|
FR^2 joined #minetest-dev |
| 19:58 |
est31 |
hmmmm, so can you perhaps at least put a list together, of issues which need fixing? |
| 19:58 |
est31 |
then we dont release until they are resolved |
| 19:58 |
hmmmm |
well I think I listed my own concerns already |
| 19:58 |
sfan5 |
est31: some germany translations were not that nicely translated "Fehler in Lua ist aufgetreten (z. B. wegen mod):" (grammatik, z.B. gehört zusammen, Mod ist ein Nomen); can we import the german translations again? |
| 19:59 |
hmmmm |
VanessaE, rubenwardy, sfan5, everybody else, could you compile a list of issues you'd like to see fixed before a release? |
| 19:59 |
VanessaE |
if it were up to me, that list would have every issue that is "medium priority" or higher, including "blocker" |
| 19:59 |
est31 |
VanessaE, so you say you dont want any feature changes for the engine until the bugs are fixed |
| 19:59 |
est31 |
also there are feature changes among them |
| 19:59 |
VanessaE |
est31: correct. |
| 20:00 |
est31 |
VanessaE, you know they are hundreds? |
| 20:00 |
VanessaE |
hmmmm: my three main concerns are the "error in error handling", the OOM issue, and the more-map-loaded-makes-lower-FPS regression |
| 20:00 |
VanessaE |
est31: I know. |
| 20:00 |
VanessaE |
and how did they get into the hundreds? |
| 20:00 |
VanessaE |
by not being fixed. |
| 20:00 |
VanessaE |
frankly I think this is an "I knew this day would come" situation |
| 20:00 |
hmmmm |
yeah those are pretty big concerns |
| 20:00 |
sfan5 |
hmmmm: the two blockers, the possible memory leak is not of such high importance imo |
| 20:00 |
est31 |
the implementors ran away and didnt care |
| 20:00 |
hmmmm |
sfan5: two blockers? |
| 20:01 |
est31 |
and now it should be my job to fix it? |
| 20:01 |
sfan5 |
hmmmm: (#2145 and #3011) |
| 20:01 |
ShadowBot |
https://github.com/minetest/minetest/issues/3011 -- ERROR[main]: Lua: error in error handling |
| 20:01 |
hmmmm |
yeah I figured as much |
| 20:01 |
VanessaE |
est31: not your job, but everyone who has the C++ skill and write access to the repo. |
| 20:01 |
hmmmm |
#3011 is top priority for me right now |
| 20:01 |
ShadowBot |
https://github.com/minetest/minetest/issues/3011 -- ERROR[main]: Lua: error in error handling |
| 20:01 |
est31 |
3011 has to be split up |
| 20:02 |
est31 |
I'll add a new one, and make it a blocker |
| 20:02 |
|
kilbith left #minetest-dev |
| 20:03 |
est31 |
so VanessaE you say I should say "fuck off " to people who demand features in here? |
| 20:03 |
VanessaE |
no... that's a little bit harsh of a response. |
| 20:04 |
est31 |
thats basically what you suggest |
| 20:04 |
VanessaE |
but given the project's current health, something a bit less harsh is in order for a while at least. |
| 20:04 |
VanessaE |
a week's freeze clearly isn't enough |
| 20:04 |
est31 |
btw can you help with other bugs |
| 20:04 |
est31 |
do you have android devices? |
| 20:05 |
VanessaE |
and given how long it takes to go from one release to another, it's about time to say "enough's enough" and make a release where all the really important bugs and most of the less important ones have been fixed |
| 20:05 |
hmmmm |
hey guys |
| 20:05 |
hmmmm |
what if we do a Linux release model |
| 20:05 |
VanessaE |
I have a tablet that's not capable of running MT, and a phone that isn't mine. |
| 20:05 |
hmmmm |
even release numbers are stability releases, odd are feature releases |
| 20:05 |
VanessaE |
hmmmm: that's not a bad idea. |
| 20:06 |
VanessaE |
but it still leaves unanswered the question of how much gets done for this release (which by that def, would be 'stability') |
| 20:06 |
VanessaE |
or wait, no |
| 20:06 |
VanessaE |
it'd be a feature releaase |
| 20:06 |
hmmmm |
well at the very least the untraceable lua errors |
| 20:06 |
VanessaE |
(was thinking 0.4.14) |
| 20:08 |
est31 |
VanessaE, you'd like now to have 1.0.0 dont you :) |
| 20:08 |
VanessaE |
est31: no. |
| 20:08 |
VanessaE |
1.0.0 would imply lots of *features* that are still on the drawing board. |
| 20:08 |
VanessaE |
I'm not talking features, I'm talking about bugs. |
| 20:09 |
VanessaE |
if it's in the issue list and it's not a feature request, it needs addressed |
| 20:09 |
est31 |
some are a mixture |
| 20:10 |
est31 |
"copy paste doesnt work" |
| 20:10 |
|
Niebieski joined #minetest-dev |
| 20:10 |
est31 |
"mobs are shit" |
| 20:10 |
hmmmm |
I wouldn't consider those bugs personally |
| 20:10 |
VanessaE |
copy&paste is a bug but that's one of those low-prio ones. |
| 20:10 |
VanessaE |
"mobs are shit" depends on adding a new feature |
| 20:10 |
VanessaE |
(client-side modding) |
| 20:11 |
est31 |
so you want to exclude everything that depends on features? |
| 20:11 |
est31 |
also we should wonder about which gfx targets we want to have |
| 20:12 |
est31 |
e.g. should we support the earliest buggiest implementation of opengl that exists on some old computer out there in the world |
| 20:12 |
est31 |
note that im not rba and want to have super old version |
| 20:12 |
VanessaE |
Everything that is at least "medium priority" and which doesn't require implementing some new feature in order to resolve to the satisfaction of the *reporter* |
| 20:12 |
est31 |
err super new |
| 20:13 |
est31 |
so you say we shouldnt release until 3018 is fixed? |
| 20:13 |
hmmmm |
alright guys problem |
| 20:14 |
est31 |
thats e.g. one of these shit bug reports |
| 20:14 |
hmmmm |
HybridDog's method to reproduce a double fault doesn't cause the same error |
| 20:14 |
est31 |
"if you click here click there move here move there, then this happens, and only hypothetically" |
| 20:15 |
hmmmm |
the second thing causing the runtime error is the thing that causes a LUA_ERRERR, which gets rethrown as a LuaError, which gets caught by the first pcall |
| 20:15 |
hmmmm |
which then runs the error handler, succesfully printing the backtrace |
| 20:15 |
est31 |
I think the problem is minetest has far too much features in order to be called stable |
| 20:15 |
est31 |
we should only consider bug reports from people who |
| 20:15 |
est31 |
1. use gdb backtraces |
| 20:16 |
est31 |
2. have NO MODS except minetest_game |
| 20:16 |
VanessaE |
est31: impossible |
| 20:16 |
est31 |
and 3. give us steps to reproduce and their config |
| 20:16 |
VanessaE |
you need mods to reproduce some errors, particularly those which depend on an API call minetest_game doesn't uise |
| 20:16 |
est31 |
otherwise its "impossible" to fix them |
| 20:16 |
est31 |
and even then its hard |
| 20:17 |
est31 |
if they say "it happens on my hardware", they should buy us an example, so that we can test, or help us fix it |
| 20:17 |
est31 |
but if they say "please fix it, minetest is so super buggy shit, i hate you people", why should we do anything for them? |
| 20:18 |
est31 |
I mean its just a git bisect |
| 20:18 |
hmmmm |
because it's our software |
| 20:18 |
VanessaE |
come on, enough with the slipper slope |
| 20:18 |
VanessaE |
+y |
| 20:18 |
hmmmm |
if there's any problems at all I'd be willing to bend over backwards to make it work |
| 20:18 |
hmmmm |
a bug is like... embarassing |
| 20:18 |
hmmmm |
even if it's not caused by me |
| 20:18 |
hmmmm |
whenever I see a new error report I get a feeling of failure |
| 20:20 |
est31 |
hmmmm, how would you fix the "sometimes its a bit laggy" bug? |
| 20:20 |
hmmmm |
the 5 second lag thing? |
| 20:20 |
est31 |
the one that has been mentioned here to be a blocker |
| 20:20 |
est31 |
#2145 |
| 20:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/2145 -- Multi-second lag in default singleplayer game |
| 20:20 |
hmmmm |
look at callgrind outputs, but the links to the ones wayward mentioned are gone |
| 20:21 |
hmmmm |
I didn't get the vibe that it's "sometimes a bit laggy" from that issue report |
| 20:21 |
hmmmm |
sounds to me like it's a consistent problem that happens constantly and makes minetest unplayable |
| 20:22 |
est31 |
first, lets make sure the person feels the same |
| 20:22 |
est31 |
second, how would you fix it |
| 20:22 |
VanessaE |
and if they say it's "laggy" find out if they mean server<->client latency or low FPS |
| 20:22 |
hmmmm |
how it gets fixed depends on what the problem is |
| 20:22 |
est31 |
ah i see he says "singleplayer is unplayable for me" |
| 20:23 |
hmmmm |
we can't know one without the other |
| 20:23 |
est31 |
what would be the next step in your opinion |
| 20:23 |
est31 |
here |
| 20:24 |
est31 |
I have fixed some of the bugs I deemed as critical |
| 20:24 |
est31 |
in the past |
| 20:25 |
|
rubenwardy joined #minetest-dev |
| 20:25 |
hmmmm |
hmm |
| 20:25 |
hmmmm |
well it depends on wayward or pilzadam showing up |
| 20:25 |
hmmmm |
!seen pilzadam |
| 20:25 |
ShadowBot |
hmmmm: I saw pilzadam in #minetest-dev 2 weeks, 5 days, 7 hours, 10 minutes, and 21 seconds ago saying "Zeno`, what else could be done instead of bisecting?" |
| 20:25 |
hmmmm |
yikes.. |
| 20:25 |
hmmmm |
!seen wayward_one |
| 20:25 |
ShadowBot |
hmmmm: I saw wayward_one in #minetest-dev 1 week, 2 days, 7 hours, 14 minutes, and 38 seconds ago saying "celeron55: it seems the FAQ on forum.minetest.net returns a 403 error" |
| 20:25 |
est31 |
#2942 |
| 20:25 |
ShadowBot |
https://github.com/minetest/minetest/issues/2942 -- Fix srp default password by est31 |
| 20:25 |
VanessaE |
!seen wayward_tab |
| 20:25 |
ShadowBot |
VanessaE: I saw wayward_tab in #minetest-dev 5 weeks, 5 days, 19 hours, 29 minutes, and 3 seconds ago saying "The minimap doesn't seem to work for android atm. I just git pulled and built, and added the respective lines to minetest.conf, to no avail." |
| 20:25 |
est31 |
#2915 |
| 20:25 |
ShadowBot |
https://github.com/minetest/minetest/issues/2915 -- Shift-Click in Minetest Game replicates items |
| 20:25 |
est31 |
this is my regression |
| 20:25 |
VanessaE |
feh. |
| 20:26 |
est31 |
#2904 |
| 20:26 |
ShadowBot |
https://github.com/minetest/minetest/issues/2904 -- No sound on Android dev build |
| 20:26 |
est31 |
not mine |
| 20:26 |
est31 |
#2847 |
| 20:26 |
ShadowBot |
https://github.com/minetest/minetest/issues/2847 -- Font characters broken |
| 20:27 |
est31 |
so dont tell me i dont fix bugs |
| 20:27 |
hmmmm |
you do |
| 20:27 |
hmmmm |
i never said you didn't... |
| 20:27 |
est31 |
ok then |
| 20:27 |
|
rubenwardy1 joined #minetest-dev |
| 20:29 |
est31 |
our current master isn't much buggier than 0.4.12 is, no? |
| 20:29 |
sfan5 |
est31: should i already make a build of git HEAD? |
| 20:29 |
est31 |
so what about the following: we do a release, but we stay in freeze, and fix the bugs, and make a bugfix 0.4.14? |
| 20:29 |
sfan5 |
also what about the translation import? |
| 20:30 |
|
OldCoder joined #minetest-dev |
| 20:30 |
est31 |
ok, will do |
| 20:30 |
VanessaE |
est31: fix the very worst of the worst bugs, releast 0.4.13, stay in freeze, fix everything else, release 0.4.14, then release the freeze. |
| 20:31 |
VanessaE |
release* |
| 20:31 |
VanessaE |
s/release the/end the/ |
| 20:32 |
|
RealBadAngel joined #minetest-dev |
| 20:32 |
VanessaE |
and that means RBA's fancy features would have to wait until post-0.4.14 |
| 20:32 |
VanessaE |
oh hi RBA ;) |
| 20:33 |
rubenwardy1 |
#3025 |
| 20:33 |
ShadowBot |
https://github.com/minetest/minetest/issues/3025 -- Fix segfaults caused by the Environment not being initialized yet by rubenwardy |
| 20:33 |
sfan5 |
thanks est31 |
| 20:34 |
est31 |
3025 looks good |
| 20:39 |
est31 |
ok, further reviews for 3025? |
| 20:40 |
sfan5 |
looks good |
| 20:40 |
nrzkt |
okay for me, but why m_env is not initalized and those calls are done ? |
| 20:40 |
nrzkt |
how can m_env not being initialized when those calls are performed |
| 20:41 |
est31 |
well its people not using the engine properly, and the engine doesnt give a clear message "you dont use me properly" |
| 20:41 |
nrzkt |
m_env = new ServerEnvironment(servermap, m_script, this, m_path_world, m_area_mgr); |
| 20:41 |
est31 |
so its bad |
| 20:41 |
RealBadAngel |
est31, have you logged on skyblock today? |
| 20:41 |
nrzkt |
sorry for the extra arg |
| 20:41 |
est31 |
no |
| 20:41 |
nrzkt |
it's in the Server constructor |
| 20:41 |
RealBadAngel |
est31, give it a shot, theres something that should be visible from your base |
| 20:41 |
nrzkt |
then, why m_env is called before in its initialization in Server constructor ? |
| 20:41 |
nrzkt |
this is the good question and the good bugfix. |
| 20:42 |
nrzkt |
to do* |
| 20:43 |
est31 |
the environment isn't initialized at init tim |
| 20:43 |
est31 |
time* |
| 20:43 |
est31 |
look at server.cpp line 299 |
| 20:43 |
nrzkt |
it's in server constructor. What do you call init time ? |
| 20:43 |
est31 |
the mods are loaded, then only we init the environment |
| 20:44 |
nrzkt |
we should init m_env before mods no ? |
| 20:44 |
est31 |
perhaps its more proper to not return, but give a nice proper "sorry i cannot do that dave" error |
| 20:44 |
est31 |
why should we |
| 20:44 |
est31 |
perhaps m_env depends on info from mods |
| 20:44 |
nrzkt |
because mods depend on m_env |
| 20:44 |
est31 |
i dont know the side effects of moving the change |
| 20:45 |
est31 |
no, they dont have to |
| 20:45 |
est31 |
of moving the m_env = call* |
| 20:45 |
nrzkt |
m_env doesn't do anything in its constructor |
| 20:46 |
|
rubenwardy joined #minetest-dev |
| 20:46 |
est31 |
rubenwardy, seen the log? we are discussing your pr right now |
| 20:46 |
sfan5 |
est31: I'll delay the builds for now (made a mistake w/ version number anyway), tell me if you have decided whether to merge rubenwardy's PR |
| 20:47 |
sfan5 |
s/if/when/ |
| 20:47 |
nrzkt |
if we could move m_env call before in Server constructor it's better for runtime |
| 20:47 |
est31 |
but i dont know the side effects |
| 20:47 |
est31 |
it seems too easy |
| 20:47 |
est31 |
and its too short before a release to risk such a change |
| 20:47 |
nrzkt |
we only need to look at m_env tests everywhere. |
| 20:48 |
est31 |
amongst the argument that the side effects are not known |
| 20:48 |
nrzkt |
Release could be delayed. Stability is better than give shit in users repo for undefined time |
| 20:48 |
nrzkt |
i will not release for android with such bugs. |
| 20:48 |
est31 |
which bug? |
| 20:48 |
rubenwardy |
if there is no environmrnt, there'll be no players so a return is okay ( |
| 20:48 |
nrzkt |
the sound bug ? |
| 20:49 |
est31 |
it is fixed |
| 20:49 |
est31 |
can you test latest master? |
| 20:49 |
nrzkt |
the memleak bug ? which will cause many problems on android |
| 20:49 |
est31 |
its older than 0.4.12 |
| 20:49 |
est31 |
so it already causes :) |
| 20:49 |
rubenwardy |
if there is no env, there are no players, so returning is fine. Although they are still using it wrong. |
| 20:49 |
est31 |
for the chat_send_players yes |
| 20:49 |
nrzkt |
i don't have time tomorrow to test it, it's late and my build env has a little bit changed since 0.4.12 release i need to fix |
| 20:49 |
est31 |
there its ok |
| 20:50 |
nrzkt |
but please, don't give a release to have a release |
| 20:50 |
est31 |
nrzkt, fine then, we have released android 0.4.12 independently from android too |
| 20:51 |
rubenwardy |
I don't like the 0.4.13 then 0.4.14 idea. |
| 20:51 |
nrzkt |
i also noticed my server was also stuck , maybe because of the emerge problem noticed, but it's very recent, since 1 month max... |
| 20:51 |
rubenwardy |
seems a release for the sake of release |
| 20:51 |
nrzkt |
+1 |
| 20:51 |
VanessaE |
rubenwardy: what would you do instead? |
| 20:51 |
est31 |
well right now i am motivated to do things, perhaps once i am not motivated to do anything |
| 20:51 |
nrzkt |
i am motivated to release a good minetest :) |
| 20:52 |
est31 |
point me to an actual regression which was worse than 0.4.12 |
| 20:52 |
rubenwardy |
feature freeze can last another week. If the bugs remaining aren't so common, |
| 20:53 |
rubenwardy |
omg, the backspace is too close to send :( |
| 20:53 |
est31 |
in one week the really serious bugs wont be fixed either |
| 20:53 |
rubenwardy |
so commonly found or vital, i guess they could be missed. |
| 20:54 |
est31 |
i just want to release something so that players can enjoy the many fixes and additions that happened since 0.4.12 |
| 20:54 |
est31 |
and after the release, we can get better than any release before, regarding bugs |
| 20:54 |
est31 |
because this seems to be the demand here |
| 20:55 |
rubenwardy |
maybe switch to semver now, so we can issue patches |
| 20:58 |
est31 |
ok let's not merge rubenwardy's pr right now, but do what nrzkt suggested, it doesn't seem that wrong |
| 20:58 |
est31 |
s/right now// |
| 20:59 |
est31 |
nrzkt, can you make a commit? |
| 20:59 |
nrzkt |
yes i looked at the code and didn't see a reason to not move the m_env init before |
| 20:59 |
nrzkt |
if you thing i don't see for this issue, please tell me :) |
| 20:59 |
est31 |
I didnt see a reason either |
| 21:01 |
rubenwardy |
hmmmm agrees with my methid |
| 21:02 |
est31 |
silently ignoring the error? |
| 21:02 |
est31 |
i've thought he likes to crash, and grind the server to a halt |
| 21:02 |
nrzkt |
silently ignore error instead of fixing is not the better way if we can fix it easily |
| 21:02 |
nrzkt |
est31: assert in debug mode permit to find unhandled cases which couldn't occurs |
| 21:02 |
rubenwardy |
no, not initialising the environment first this close to release |
| 21:03 |
rubenwardy |
#3025 |
| 21:03 |
ShadowBot |
https://github.com/minetest/minetest/issues/3025 -- Fix segfaults caused by the Environment not being initialized yet by rubenwardy |
| 21:03 |
nrzkt |
s/couldn't/shouldn't/ |
| 21:03 |
nrzkt |
rubenwardy: if m_env is initialized before mods in Server::Server this solve the problem too |
| 21:04 |
nrzkt |
it's because some mods are executing some API code at server loadMod |
| 21:04 |
est31 |
Either we should do it as right now, by crashing, or we should perhaps give a nicer error msg |
| 21:04 |
est31 |
and then either crash, or resume |
| 21:04 |
est31 |
explaining that we are initializing |
| 21:04 |
nrzkt |
ignore m_env remove the behavious which could permit to do something on environnement at serverload |
| 21:04 |
est31 |
but silently ignoring things will get modders file this as bug again |
| 21:04 |
nrzkt |
behaviour*. This could be interesting |
| 21:05 |
est31 |
it wont remove any behaviour |
| 21:05 |
est31 |
because right now we crash at serverload |
| 21:05 |
nrzkt |
atm it seems we could do lua calls when we load a mod |
| 21:05 |
nrzkt |
this could be interesting to load env before to let mods do something too |
| 21:06 |
est31 |
they cant do that interesting things |
| 21:06 |
nrzkt |
atm no, i know |
| 21:07 |
nrzkt |
but if we load env before mods: 1. we fix the crash 2. we permit mods to do something at server load |
| 21:07 |
est31 |
rubenwardy, does hmmmm still talk to you in pm? why doesnt he talk here? |
| 21:07 |
est31 |
2. isnt needed, and 1. isnt a problem if the error message has a backtrace to the mod |
| 21:07 |
est31 |
and 2. is dangerous |
| 21:08 |
est31 |
because many things e.g. aliases are still existing in incomplete form |
| 21:08 |
rubenwardy |
Server::setTimeOfDay is a bad idea to silently fail |
| 21:08 |
rubenwardy |
all the others are okay as they're player related |
| 21:08 |
rubenwardy |
the roll back one was a mistake, i didnt mean to add it |
| 21:09 |
nrzkt |
est31: exact |
| 21:09 |
rubenwardy |
I'd be surprised if rollback could be run at load time |
| 21:09 |
est31 |
the particle spawner ones give return values too. |
| 21:09 |
nrzkt |
est31: i don't understand why mods are loaded soon in Server. They should be loaded last no ? |
| 21:09 |
est31 |
if you remove the ones you mentioned, we can merge it imo |
| 21:10 |
est31 |
nrzkt, some mods may depend on others |
| 21:10 |
est31 |
classical example is the craft guide |
| 21:10 |
rubenwardy |
so silently ignoring is absolutely fine, except for Server::setTimeOfDay |
| 21:10 |
rubenwardy |
it shouldn't crash |
| 21:10 |
nrzkt |
est31: ok |
| 21:10 |
est31 |
no it isnt |
| 21:10 |
est31 |
for player related yes |
| 21:11 |
est31 |
but for the spawner things no |
| 21:11 |
est31 |
but there you don't ignore silently |
| 21:11 |
est31 |
you give actual return values |
| 21:12 |
rubenwardy |
spawners are player related |
| 21:12 |
est31 |
you should document that if id is -1 the addition failed |
| 21:13 |
est31 |
and the return value for the deletion should be documented too |
| 21:13 |
rubenwardy |
that was already returned |
| 21:13 |
rubenwardy |
see the code below my addition |
| 21:13 |
nrzkt |
a thing for next release, if we could use std::map instead of std::vector for hud client side and server side it could be good to preserve some rare case of client desync on player HUD id |
| 21:13 |
est31 |
still it should be documented |
| 21:13 |
rubenwardy |
but yes, whilst I am there i should add comments |
| 21:14 |
rubenwardy |
i also made a mistake, all the comments i add should be "may" not "will" (only the first is correct) |
| 21:16 |
est31 |
will is ok for me |
| 21:16 |
rubenwardy |
nrzkt, how could a desync happen? |
| 21:16 |
|
Amaz joined #minetest-dev |
| 21:16 |
nrzkt |
i don't know but i notice this bug when using client in debug mode, the assertion can sometimes happen |
| 21:18 |
rubenwardy |
est31 it's may as the environment could be non-null whilst initialised |
| 21:18 |
rubenwardy |
but then again those functions won't be called other than mods in the init stage, afaik. |
| 21:19 |
rubenwardy |
does the hud-add packet give an id to the client, to make the client have the same id? |
| 21:19 |
est31 |
well it will be null sometime during initialisation |
| 21:20 |
est31 |
whether thats the same time the method is called during initialisation isnt required |
| 21:20 |
rubenwardy |
true |
| 21:21 |
|
kilbith joined #minetest-dev |
| 21:21 |
|
Niebieski joined #minetest-dev |
| 21:28 |
Niebieski |
Hey, I'm trying to add Minetest into list of open-source games on Wikipedia. |
| 21:28 |
Niebieski |
https://en.wikipedia.org/wiki/List_of_open-source_video_games |
| 21:28 |
sfan5 |
do we have an article? |
| 21:28 |
est31 |
nope |
| 21:28 |
sfan5 |
i thought we didn't have one |
| 21:28 |
est31 |
and we dont get one |
| 21:28 |
est31 |
unless we get more "notable" |
| 21:28 |
Niebieski |
We don't have to have an article to add into this list. |
| 21:29 |
Niebieski |
I guess... |
| 21:29 |
sfan5 |
est31: don't tell me Terasology has an article |
| 21:29 |
Niebieski |
Refresh now I saved it. |
| 21:29 |
sfan5 |
please say no |
| 21:29 |
rubenwardy |
we need notable sources |
| 21:29 |
kilbith |
we had an article on the french page but deleted because "MT was not enough important amongst dozens of clones" |
| 21:29 |
sfan5 |
Niebieski: are you sure we don't need an article? literally every other entry has one |
| 21:30 |
VanessaE |
sfan5: there doesn't appear to be a terasology article. |
| 21:30 |
rubenwardy |
we're marked now, Wikipedia deletionists are watching our page |
| 21:30 |
sfan5 |
there is a german minetest article though https://de.wikipedia.org/wiki/Minetest |
| 21:30 |
est31 |
Niebieski, just try to add it, wiki people will remove it if they dont like it |
| 21:30 |
Niebieski |
No I'm not sure, but I added it anyway. |
| 21:30 |
est31 |
and you arent related with us |
| 21:30 |
Niebieski |
Hahahaha. xD |
| 21:30 |
est31 |
seems wiki people dont like people related to projects editing pages about them |
| 21:31 |
est31 |
s/with/to/ |
| 21:31 |
est31 |
lol omg :D |
| 21:31 |
Niebieski |
I'm telling you guys because I don't know what to add in the description ? |
| 21:31 |
sfan5 |
est31: uploaded the mingw builds, based on current HEAD because there seems no conseus on whether to merge rubenwardy's PR |
| 21:31 |
VanessaE |
Niebieski: "a C++ open source block sandbox game similar to Minecraft." |
| 21:31 |
est31 |
sfan5, no problem |
| 21:32 |
sfan5 |
Niebieski: "Open-Source, Voxel-based sandbox game similar to Minecraft" |
| 21:32 |
est31 |
Right now I'm editing a slightly modified version of his pr |
| 21:32 |
Niebieski |
Guys the list is about open-source games only. |
| 21:32 |
sfan5 |
oh right |
| 21:32 |
sfan5 |
drop the open-source then |
| 21:33 |
Niebieski |
also no one has talked about technical specs of the game like coded in.. etc. |
| 21:33 |
Niebieski |
I would say "voxel-base sandbox game similar to Minecraft". |
| 21:34 |
Niebieski |
based* |
| 21:34 |
sfan5 |
sounds good |
| 21:36 |
est31 |
ok what bout https://github.com/est31/minetest/commit/91ecbd7bdb0b2bf4c8504cfd089d02d17e92cd66 |
| 21:36 |
est31 |
we dont need the new it seems |
| 21:37 |
est31 |
pushing that in 5 minutes |
| 21:38 |
sfan5 |
seems ok |
| 21:38 |
est31 |
also tested it, the error works |
| 21:38 |
est31 |
even shows where exactly the error occured |
| 21:38 |
est31 |
which lua line |
| 21:39 |
rubenwardy |
check that -1 is in fact returned to lua |
| 21:39 |
est31 |
yes |
| 21:39 |
rubenwardy |
as server.deletepartile != l_env::l_delete_particle |
| 21:39 |
rubenwardy |
okay |
| 21:41 |
nrzkt |
why throw on delete but not in add ? |
| 21:42 |
rubenwardy |
because right below it fails silently on no player |
| 21:43 |
est31 |
you mean the deletion? |
| 21:43 |
est31 |
agreed, that should be fixed too |
| 21:44 |
rubenwardy |
no, in add there's if (player) |
| 21:44 |
est31 |
it does return -1; there |
| 21:44 |
est31 |
thats not silent |
| 21:44 |
est31 |
it gives the error to lua |
| 21:45 |
est31 |
there /is/ a silent fail in deleteParticleSpawner |
| 21:45 |
est31 |
but dunno if we should throw an error there |
| 21:45 |
est31 |
if a player doesnt exist, it usually means they have logged off |
| 21:47 |
rubenwardy |
if there is no environment, then there are no players. |
| 21:47 |
est31 |
agreed |
| 21:47 |
rubenwardy |
although the mod is still using the api wrong |
| 21:47 |
est31 |
see what i have pushed http://git.io/v3tqV |
| 21:49 |
est31 |
nrzkt, you still around? |
| 21:49 |
|
Miner_48er joined #minetest-dev |
| 21:51 |
nrzkt |
est31: not for a long time, i will go to bed. What do you want ? |
| 21:51 |
est31 |
is the android version code 13 accurate? |
| 21:51 |
hmmmm |
wow lots of talk |
| 21:51 |
hmmmm |
I wanted to talk to rubenwardy in PM about delete_area specifically to reduce crosstalk and noise |
| 21:52 |
nrzkt |
i look at the console to be sure |
| 21:52 |
hmmmm |
I also approve of the way things dependent upon the environment being active are handled |
| 21:52 |
est31 |
version code 13 for minetest 0.4.13 |
| 21:52 |
hmmmm |
if you guys don't like the silent error thing, then it's trivial, just change return type to a bool, fail by returning false, throw LuaError if the server method returns false |
| 21:53 |
hmmmm |
"Environment not initialized yet" |
| 21:53 |
nrzkt |
current play store version is 11 |
| 21:53 |
hmmmm |
moving the environment creation before the mod initialization might be a mistake |
| 21:53 |
est31 |
ok then, its larger :) |
| 21:53 |
hmmmm |
there are VERY subtle side effects with changing the ordering of initialization of things |
| 21:53 |
est31 |
^ |
| 21:53 |
nrzkt |
yes, for releasing i need at least 12 :) |
| 21:53 |
est31 |
that was my main concern too. |
| 21:53 |
hmmmm |
I screwed up one time by moving something around |
| 21:54 |
hmmmm |
besides, setTimeOfDay or whatever would get overwritten when the env meta is loaded |
| 21:54 |
hmmmm |
so it's kinda useless |
| 21:54 |
est31 |
yea |
| 21:55 |
nrzkt |
recent comment: Crashed on me First I spawned In a lagoon and died. Than It crashed on me |
| 21:55 |
nrzkt |
The interface of this game needs some work... This Mobile Version loses too much FPS, also the HUD, Crafting Menu, Pause Menu, the Main Menu needs some work to be Mobile Friendly. :) |
| 21:55 |
est31 |
agreed |
| 21:56 |
nrzkt |
Unplayable for me Tapped the button to increase the UI size multiple times when I first played. The taps took like 10 seconds to register. Of course I tapped too many times. Now the UI is fucked up and unusable. I did try to clear data/cache/reinstall but nothing changed. |
| 21:56 |
est31 |
clearing data doesnt work |
| 21:56 |
est31 |
its the "proper paths" problem |
| 21:56 |
est31 |
look into your sdcard folder |
| 21:56 |
nrzkt |
i see, we are using a stupid data path |
| 21:57 |
est31 |
there is a directory "Minetest" |
| 21:57 |
est31 |
remove or rename it |
| 21:57 |
nrzkt |
not very user friendly |
| 22:01 |
rubenwardy |
the black screen on start up isn't very good also |
| 22:01 |
est31 |
it is neccessary right now, but will be removed |
| 22:01 |
est31 |
(mostly) |
| 22:02 |
rubenwardy |
there are a lot of items on screen, hide them if they didn't have the privs, and maybe move some to pause |
| 22:02 |
est31 |
? |
| 22:02 |
rubenwardy |
pressing the back button breaks flow, I'd like a close button on screen. it's close not back |
| 22:03 |
rubenwardy |
the fly/noclip/view all/fast/debug icons |
| 22:03 |
est31 |
its the android way |
| 22:03 |
est31 |
about the back button |
| 22:03 |
rubenwardy |
it doesn't feel right |
| 22:03 |
est31 |
for me it does |
| 22:04 |
est31 |
and besides we dont have it on normal minetest either |
| 22:04 |
est31 |
e.g. on desktop |
| 22:04 |
VanessaE |
back (out of formspec) --> back (out of the active play) --> back (out of the back, to the home screen) |
| 22:04 |
VanessaE |
seems logical to me |
| 22:05 |
VanessaE |
out of the app* |
| 22:05 |
rubenwardy |
also, creative inventory - back doesn't go back a page |
| 22:05 |
VanessaE |
why would it? |
| 22:05 |
rubenwardy |
because it's back |
| 22:06 |
VanessaE |
that ain't how it works in android :) |
| 22:06 |
rubenwardy |
it took me a few minutes to work out back = pause menu |
| 22:06 |
VanessaE |
it's not "page back", it's "previous item in history". |
| 22:07 |
rubenwardy |
my back button vibrates and takes half a second to work |
| 22:07 |
rubenwardy |
which is probably the main cause of breaking |
| 22:08 |
rubenwardy |
(breaking flow) |
| 22:08 |
VanessaE |
turn off haptic feedback then :) |
| 22:08 |
est31 |
btw whats the default mapgen? |
| 22:08 |
rubenwardy |
also, I would like to see ab official Minetrst play account, nrz is unprofessional |
| 22:08 |
est31 |
v7 or v6? |
| 22:08 |
VanessaE |
(I honestly never understood the need for it in many situations where it's used. that usual "pwop!" sound seems to be enough for me) |
| 22:08 |
VanessaE |
est31: v6 |
| 22:09 |
est31 |
good |
| 22:09 |
rubenwardy |
turn it off for all apps? |
| 22:09 |
est31 |
ok then, seems I can make the release commit |
| 22:09 |
VanessaE |
rubenwardy: I guess so. this is OS-level stuff isn't it? |
| 22:09 |
rubenwardy |
it still doesnt solve the half second lag. |
| 22:10 |
rubenwardy |
could you click outisde of a formspec to close? |
| 22:10 |
rubenwardy |
that would solve the problen |
| 22:10 |
nrzkt |
rubenwardy: i don't see why i'm unprofesionnal, it's a gratis sentence, but i'm out, and if you have 100$ to give to google, go. |
| 22:11 |
est31 |
nice, mushrooms |
| 22:11 |
rubenwardy |
isn't it £10? |
| 22:11 |
* VanessaE |
throws a badger at est31 |
| 22:11 |
rubenwardy |
lol |
| 22:11 |
est31 |
rofl |
| 22:11 |
rubenwardy |
it doesnt matter that much though |
| 22:12 |
est31 |
what are the spores for? |
| 22:12 |
rubenwardy |
for children etc (looking at the reviews) it isn't that obvious to press the back buttons |
| 22:13 |
VanessaE |
so you can farm more mushrooms |
| 22:13 |
est31 |
hrmm, there seems to be a bug |
| 22:13 |
rubenwardy |
but if i can press the bg, I'm fine :P |
| 22:13 |
VanessaE |
rubenwardy: it also isn't obvious to them that this isn't minecraft, either |
| 22:13 |
est31 |
when I mine a mushroom, i get a mushroom plus some spores |
| 22:13 |
VanessaE |
I've lost count of the number of times someone's written game[ ]mode[ ]1 |
| 22:13 |
est31 |
now when I place that mushroom I got, I get again a mushroom plus some spores |
| 22:13 |
est31 |
we need a mushroom_without_spores node |
| 22:14 |
VanessaE |
est31: sounds like drops aren't configured right |
| 22:14 |
VanessaE |
there's supposed to be a way to only drop one OR the other |
| 22:14 |
est31 |
ok |
| 22:14 |
est31 |
can you fix this? |
| 22:15 |
VanessaE |
er...where's the code? |
| 22:15 |
est31 |
flowers mod |
| 22:16 |
VanessaE |
wat |
| 22:16 |
VanessaE |
that's a weird place to put it |
| 22:16 |
est31 |
https://github.com/minetest/minetest_game/blob/master/mods/flowers/init.lua |
| 22:16 |
VanessaE |
found it. |
| 22:16 |
rubenwardy |
not the floral debate again :( |
| 22:17 |
rubenwardy |
anyway, gtg |
| 22:17 |
est31 |
bye |
| 22:17 |
VanessaE |
wait |
| 22:17 |
VanessaE |
est31: add "max_items = 1" to the drop def for that node |
| 22:17 |
VanessaE |
right before line 148 |
| 22:18 |
VanessaE |
that should ensure that the user never gets both. |
| 22:18 |
VanessaE |
oh it was ruben leaving. fine. gtfo ;) |
| 22:19 |
est31 |
hrmmm, can you also drop stacks? |
| 22:19 |
est31 |
because it would be good to drop multiple mushrooms |
| 22:19 |
est31 |
err seeds |
| 22:19 |
VanessaE |
I've never tried |
| 22:19 |
VanessaE |
but I think so |
| 22:19 |
VanessaE |
it's probably just a matter of appending a space and the stack count |
| 22:19 |
VanessaE |
(as in any other itemstack usage) |
| 22:19 |
est31 |
yea |
| 22:20 |
VanessaE |
now as for where that code belongs, frankly I'd have preferred it to be in farming mod. |
| 22:21 |
hmmmm |
hum :/ |
| 22:21 |
hmmmm |
I can't replicate an error in error handling with a triple-nested setup |
| 22:21 |
VanessaE |
mushrooms are not plants, let alone flowers, and can be farmed, more or less |
| 22:25 |
est31 |
agreed |
| 22:26 |
est31 |
but i dont know minetest_game policy well enough to do such a change before a release |
| 22:26 |
VanessaE |
well it's not a bug, and changing it now would break nodes |
| 22:27 |
VanessaE |
(because they're all prefixed "flowers:" and soon I'll have to adapt plantlife's mod to use them) |
| 22:27 |
VanessaE |
so I guess it stays. |
| 22:28 |
* VanessaE |
looks at the number of merge conflicts between dreambuilder and upstream minetest_game and cries a little |
| 22:40 |
hmmmm |
I think we should take a very close, hard look at travelnet |
| 22:40 |
hmmmm |
this may be the cause of error in error handling |
| 22:40 |
VanessaE |
well, I DO use it on my servers |
| 22:40 |
* VanessaE |
pokes Sokomine |
| 22:40 |
hmmmm |
I cross correlated with thatgraemeguy, cornernote, tchncs, and now you |
| 22:41 |
* hmmmm |
goes to install travelnet |
| 22:41 |
est31 |
VanessaE, what bout http://pastebin.com/HayDBFmh |
| 22:41 |
hmmmm |
that is literally the only shared mod between all people who have the problem :) |
| 22:41 |
VanessaE |
est31: that seems right to me. |
| 22:42 |
VanessaE |
hmmmm: that said, it's a fairly popular mod. might be a good candidate for inclusion into minetest_game (after the current problem is sorted and any bugs therein are fixed) |
| 22:43 |
est31 |
it does change gameplay |
| 22:43 |
est31 |
before, you had to walk, now you can take the travelnet |
| 22:43 |
VanessaE |
yes, but not in an essential way, after all a singleplayer user can just /teleport if they want |
| 22:45 |
VanessaE |
and most servers have some kind of limited-teleport mod installed also. |
| 22:46 |
VanessaE |
hmmmm: got any particular failure in mind in that mod, btw? |
| 22:48 |
hmmmm |
not yet |
| 22:48 |
hmmmm |
but whatever it is |
| 22:48 |
hmmmm |
it is NOT the same thing as hybriddog posted |
| 22:49 |
VanessaE |
hybrid dog uses a "hacked" client. |
| 22:49 |
hmmmm |
? what do you mean |
| 22:50 |
VanessaE |
various client modifications, for example he can place a nodee by writing something like "please place 1 2 3" or so. |
| 22:50 |
VanessaE |
I don't know what all. |
| 22:50 |
hmmmm |
oh yeah but that has nothing to do with this |
| 22:50 |
hmmmm |
i think he's been pretty helpful tbh |
| 22:50 |
hmmmm |
i like him |
| 22:50 |
VanessaE |
well just saying to keep it in mind. |
| 23:06 |
est31 |
hrmm, mushrooms are very easy to farm as it seems |
| 23:06 |
est31 |
easier than farming crops |
| 23:07 |
est31 |
well, thats not that important |
| 23:07 |
VanessaE |
btw let's resolve game#611 before release |
| 23:07 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/611 -- Junglewood texture too light, boring |
| 23:08 |
est31 |
is that a 0.4.12 "regression"? |
| 23:08 |
VanessaE |
not exactly as such |
| 23:08 |
VanessaE |
it's a style regression that frankly looks terrible, and apparently a lot of people agree |
| 23:09 |
est31 |
I agree too |
| 23:09 |
VanessaE |
I figured we could hash out the two textures being discussed (the jungle tree top and jungle wood planks) and get that in for the release. |
| 23:10 |
VanessaE |
this wood, https://github.com/minetest/minetest_game/issues/611#issuecomment-128830359 |
| 23:10 |
VanessaE |
and whatever we decide on for the top (I think my last post is pretty good) |
| 23:11 |
est31 |
like your last one |
| 23:11 |
est31 |
do you want to make a pr? |
| 23:11 |
VanessaE |
I can do that but I don't wanna take credit for kilbith's wood image |
| 23:11 |
est31 |
you can make two commits, and I'll change the author for the wood |
| 23:11 |
VanessaE |
ok. |
| 23:11 |
VanessaE |
moment. |
| 23:12 |
est31 |
I like your last trunk image btw |
| 23:12 |
est31 |
stilll a slight bit too coffe-ish, but overall good |
| 23:16 |
VanessaE |
https://github.com/minetest/minetest_game/pull/618 |
| 23:19 |
VanessaE |
wait, lemme fix lazyj's credit. |
| 23:19 |
est31 |
well, part of it is your texture too |
| 23:20 |
VanessaE |
nah |
| 23:20 |
VanessaE |
all I did was combine his + default |
| 23:20 |
VanessaE |
fixed. |
| 23:23 |
VanessaE |
huh, see kilbith's comment on that PR? O_o |
| 23:25 |
est31 |
o_O |
| 23:26 |
VanessaE |
maybe I coulda turned the color sat down a bit but I thought it looked good |
| 23:27 |
|
kilbith joined #minetest-dev |
| 23:27 |
kilbith |
ok, wtf is that commit ? https://github.com/minetest/minetest_game/commit/6a64ec4dac8c92b8435107fb6ab1bc6bb7ce1820 |
| 23:27 |
kilbith |
is that a mistake or are you just breaking the damn rules ? |
| 23:28 |
est31 |
removed it, was against the rules |
| 23:28 |
kilbith |
same for that : https://github.com/minetest/minetest_game/commit/39de13fe9e816446c2a54ead7166863d2d7656e8 |
| 23:28 |
kilbith |
and the previous one |
| 23:29 |
est31 |
The mushrooms had to be fixed, otherwise people get too much on servers, etc. |
| 23:29 |
est31 |
about trunks, we already had a regressed release |
| 23:30 |
kilbith |
do you realize that you have 50% of chance to not get a mushroom in your inv ? |
| 23:30 |
kilbith |
are you experienced in modding ? |
| 23:31 |
est31 |
i have read the algorithm, and this seemed satisfying to me |
| 23:31 |
kilbith |
you just added a new feature |
| 23:31 |
est31 |
you have 50% chance to not get a mushroom |
| 23:31 |
kilbith |
without getting 2 approvals |
| 23:32 |
kilbith |
this is so stupid, you dig a mushroom, you get it |
| 23:32 |
est31 |
its a bug, i wanted to fix it |
| 23:32 |
kilbith |
the bug was for the spores |
| 23:33 |
est31 |
perhaps the proper fix would have been to register mushroom_without_spores |
| 23:33 |
est31 |
then have an abm to add the spores |
| 23:33 |
est31 |
but it gives terrible gameplay |
| 23:33 |
kilbith |
no discussion, nothing.. |
| 23:34 |
VanessaE |
kilbith: btw, real strangler fig wood: http://urbanforestrecycling.com/_Media/sany0003-4-1_med.jpeg (so mine/lazyj's image is actually pretty close to the darker parts of those boards) |
| 23:34 |
est31 |
you can still make it easier, but making it harder is hard on public servers if people have chests full of it |
| 23:34 |
est31 |
imagine if nyan cats were as common if stones |
| 23:35 |
est31 |
as* |
| 23:35 |
est31 |
and then you make them super rare |
| 23:35 |
|
alket joined #minetest-dev |
| 23:35 |
est31 |
nobody will be believed they added it |
| 23:36 |
est31 |
we have RC commits set for what the release should be |
| 23:36 |
est31 |
s/added it/dug it after the super rare event/ |
| 23:36 |
est31 |
so, fine, the commits are tagged |
| 23:37 |
est31 |
tomorrow sb can write the changelog, and the release can be made "official" |
| 23:50 |
Hijiri |
est31: just make it require chests full of it |
| 23:50 |
Hijiri |
maybe 5 |
| 23:50 |
est31 |
lol |
| 23:53 |
|
wischi joined #minetest-dev |
