Time |
Nick |
Message |
00:11 |
|
shadowzone joined #minetest-dev |
00:16 |
ShadowNinja |
est31: AFAIK such a thing never existed. If you mean node metadata, that was never discontinued. |
00:17 |
est31 |
ShadowNinja, I'll try to dig up the place where I read about it |
00:17 |
* est31 |
searches for the code pickaxe |
00:29 |
|
ShadowNinja joined #minetest-dev |
01:04 |
|
manax joined #minetest-dev |
01:05 |
|
proller joined #minetest-dev |
02:09 |
|
rickmcfarley joined #minetest-dev |
02:18 |
|
Zeno` joined #minetest-dev |
02:30 |
|
shadowzone joined #minetest-dev |
02:33 |
est31 |
VanessaE, what is the exact problem with #118? |
02:33 |
ShadowBot |
https://github.com/minetest/minetest/issues/118 -- New drawtype requests |
02:35 |
VanessaE |
see Calinou's last post. that is the only reason I never closed that issue. |
02:43 |
|
domtron joined #minetest-dev |
02:58 |
|
Guest62948 joined #minetest-dev |
03:00 |
* Zeno` |
looks at ShadowNinja quizzically |
03:01 |
Zeno` |
https://github.com/minetest/minetest/blob/master/src/util/string.cpp#L65 and then line 75 of the same file |
03:01 |
Zeno` |
in your handwriting! |
03:01 |
est31 |
leeeeeeaaaaakkkkk |
03:01 |
Zeno` |
this is a job for Inspector Gadget! |
03:02 |
est31 |
? |
03:02 |
Zeno` |
go go gadget keyboard! |
03:05 |
ShadowNinja |
Zeno`: Oops. |
03:05 |
ShadowNinja |
Zeno`: Feel free to fix that, I'm working on the database. |
03:06 |
Zeno` |
how does this look: https://gist.github.com/Zeno-/5af815db7fa2da0768fe |
03:06 |
Zeno` |
I can't compile or test, but *shrug* |
03:06 |
est31 |
you @ work? |
03:07 |
Zeno` |
nah, just don't have an Android build environment |
03:07 |
est31 |
it only affects android? |
03:08 |
Zeno` |
it's wrapped in ifdef _ANDROID #endif pair, so yep |
03:08 |
est31 |
ah ok |
03:12 |
Zeno` |
est31, I have tested in my brain though |
03:12 |
est31 |
wow, an android build environment in your brain? |
03:12 |
Zeno` |
yep |
03:13 |
Zeno` |
anyway, merging it :) |
03:13 |
Zeno` |
in 10 minutes (trivial change) |
03:14 |
Zeno` |
lines 3 and 4 are not really necessary I suppose |
03:15 |
Zeno` |
actually, I'll change lines 3 & 4 to make it a bit closer to how the man page says it works |
03:15 |
Zeno` |
should work( |
03:15 |
Zeno` |
* |
03:16 |
Zeno` |
ShadowNinja, https://gist.github.com/Zeno-/5af815db7fa2da0768fe |
03:16 |
Zeno` |
nah that's not right... stop confusing me, est31! |
03:17 |
Zeno` |
fixed |
03:18 |
ShadowNinja |
Zeno`: Seems like it will compile and not leak, although I don't understand that function. |
03:18 |
Zeno` |
well, it does the same as before without the leak so *shrug* |
03:18 |
Zeno` |
:D |
03:19 |
Zeno` |
and I have sort of tested in (in normal C) just not on Androud |
03:23 |
Zeno` |
actually those 3 functions are... stupid |
03:25 |
|
leat joined #minetest-dev |
03:29 |
|
domtron joined #minetest-dev |
03:34 |
Zeno` |
going to merge #2377 as well and get it out of the way before it gets clobbered and needs rebasing again |
03:34 |
ShadowBot |
https://github.com/minetest/minetest/issues/2377 -- Wrap lines longer than 80 characters in lua_api.txt by rubenwardy |
03:39 |
Zeno` |
I am beginning to hate lua_api.txt... the most minor changes make merging a nightmare :/ |
03:44 |
ShadowNinja |
It should really be a directory with a bunch of md/rst/something files in it. |
03:45 |
ShadowNinja |
The only downside is that it's harder to search through, but you could just cat them together. |
03:48 |
Zeno` |
possibly |
03:49 |
est31 |
grep -r |
03:49 |
est31 |
c55 has authorized me to split the file up, but haven't found time yet |
03:49 |
Zeno` |
it's a pity doxygen doesn't output markup... if it did est31 could write doxygen input for all of them |
03:50 |
Zeno` |
although it does output XML... I wonder if there is an XML to markup util |
03:53 |
est31 |
c55 didnt want a manual step involved |
03:54 |
Zeno` |
est31, manual in what way? running doxygen? |
03:55 |
est31 |
manual as in "you have to execute a step before you can read the docs" |
03:55 |
est31 |
so yes |
03:56 |
Zeno` |
well, doxygen is easy enough to read. :) |
03:56 |
Zeno` |
just as easy as the current format and sometimes easier heh |
03:56 |
est31 |
the current format is at one place |
03:57 |
Zeno` |
the doxygen comments don't have to be in the source file |
03:57 |
est31 |
also doxygen was dismissed as many lua functions are defined in lua |
03:57 |
Zeno` |
"defined in lua"? |
03:57 |
est31 |
like table.copy |
03:57 |
est31 |
builtin/* |
03:58 |
Zeno` |
I don't understand why that's an issue... |
03:58 |
est31 |
so with doxygen, you can write separate docs? |
03:58 |
Zeno` |
yeah |
03:59 |
est31 |
btw, doxygen can output to markdown |
03:59 |
est31 |
I have to recall what c55 really wanted, I'm not sure what consensus was. |
03:59 |
est31 |
v |
03:59 |
est31 |
http://www.stack.nl/~dimitri/doxygen/manual/markdown.html |
03:59 |
est31 |
/v/d |
04:00 |
Zeno` |
that's for markdown in the actual doxygen comment I think |
04:02 |
Zeno` |
I guess the disadvantage of having all the Lua functions in a file of their own, instead of in the code, is that it's no different to maintaining the separate lua_api.txt. However the advantage would be that we could generate html as well |
04:04 |
est31 |
markdown has been designed to be converted into html |
04:04 |
Zeno` |
true |
04:04 |
est31 |
and there is this: http://rubenwardy.com/minetest_modding_book/lua_api.html |
04:04 |
est31 |
so it is already possible |
04:04 |
Zeno` |
how does he generate that? |
04:05 |
Zeno` |
the format looks familiar |
04:05 |
est31 |
dunno, the dude that invented markdown wrote a md->html generator |
04:05 |
est31 |
and many others have followed |
04:05 |
est31 |
the advantage however would be linking, and a more machine readable format |
04:06 |
Zeno` |
it almost looks like the output from http://sphinx-doc.org/ |
04:12 |
est31 |
c55 has proposed creating multiple md files |
04:12 |
est31 |
http://irc.minetest.ru/minetest-dev/2015-02-12#i_4152443 |
04:12 |
est31 |
Zeno` ^ |
04:13 |
Zeno` |
Well, that's fine I think. I do think that autogenerating html is important as well, though (whatever the solution ends up being) |
04:18 |
|
selat joined #minetest-dev |
04:23 |
|
selat joined #minetest-dev |
04:31 |
|
selat joined #minetest-dev |
04:39 |
|
selat joined #minetest-dev |
04:42 |
|
chchjesus joined #minetest-dev |
04:44 |
|
Hunterz joined #minetest-dev |
06:00 |
Zeno` |
I've made a small fix to loadPlayer: https://github.com/Zeno-/minetest/commit/d731fd1ff7842567a69943e0834cad0e7ef8c024 |
06:01 |
Zeno` |
But now I don't really understand why it's done that way at all. Under what circumstances could (should?) getPlayer() return a player if player does not already exist? |
06:05 |
Zeno` |
i.e. I am guessing that assert(player == NULL) would (or should) always be true |
06:11 |
Zeno` |
looking at server.cpp lines 3357-3376 the issue becomes even less clear |
06:13 |
Zeno` |
kahrl_, ShadowNinja ^ |
06:13 |
Zeno` |
ShadowNinja, you refactored it maybe you know :p |
06:16 |
ShadowNinja |
Zeno`: If it's already loaded reload its config, if it isn't create it. |
06:17 |
ShadowNinja |
(The old system kept every player loaded at all times) |
06:17 |
ShadowNinja |
Could probably make it a no-op if it's already loaded now. |
06:17 |
Zeno` |
ShadowNinja, but looking at server.cpp on the lines I specified that... hmm |
06:18 |
Zeno` |
anyway, I'll leave it and just fix the memory leak (which I've done) |
06:18 |
ShadowNinja |
Comments may be outdated. |
06:18 |
Zeno` |
I'm reading the codes, not the comments :) |
06:18 |
* ShadowNinja |
is trying to remove the server requirement from migration. |
06:19 |
Zeno` |
I will be committing https://github.com/Zeno-/minetest/commit/d731fd1ff7842567a69943e0834cad0e7ef8c024 at some point though to make sure the leak cannot occur |
06:30 |
ShadowNinja |
#2395 |
06:30 |
ShadowBot |
https://github.com/minetest/minetest/issues/2395 -- Clean up database API and save the local map on an interval by ShadowNinja |
06:38 |
|
Hunterz joined #minetest-dev |
06:55 |
Zeno` |
ShadowNinja, The client doesn't start a server now! ? |
06:55 |
ShadowNinja |
Zeno`: Yep. It started a server for local map saving. |
06:56 |
Zeno` |
even if local map saving was disabled? |
06:59 |
Zeno` |
oh I see |
06:59 |
Zeno` |
wow |
07:00 |
Zeno` |
I thought you were talking about the db (oops). I see what you're saying now |
07:30 |
|
kilbith joined #minetest-dev |
07:36 |
ShadowNinja |
So... I just made migrating over 89 times faster... |
07:36 |
ShadowNinja |
http://pastebin.ubuntu.com/10443765/ |
07:36 |
ShadowNinja |
Yes, that's 4.8 minutes vs 3 seconds. |
07:37 |
ShadowNinja |
Biggest MT optimization ever? |
07:41 |
VanessaE |
wow, nice |
07:42 |
VanessaE |
you should throw VE-Survival's map at it |
07:42 |
VanessaE |
see how it handles that sucker ;) |
07:42 |
ShadowNinja |
VanessaE: git fetch shadowninja database-cleanup |
07:44 |
VanessaE |
ok, let's give it a shot. |
07:44 |
VanessaE |
gimme a few. |
07:51 |
|
ImQ009 joined #minetest-dev |
07:58 |
VanessaE |
ShadowNinja: FAIL |
07:58 |
VanessaE |
http://pastebin.ubuntu.com/10444020/ |
07:59 |
ShadowNinja |
VanessaE: Remove the trailing comma in the statement above. |
08:01 |
VanessaE |
after BLOB ? |
08:02 |
ShadowNinja |
Yep. |
08:02 |
ShadowNinja |
Or re-pull. |
08:02 |
ShadowNinja |
I also fixed some other minor things. |
08:02 |
VanessaE |
ok |
08:03 |
VanessaE |
re-pulled, rebuilding... |
08:04 |
ShadowNinja |
VanessaE: It took 56.7 seconds for it to get to createDatabase()? |
08:04 |
VanessaE |
ShadowNinja: I guess so, yeah |
08:05 |
VanessaE |
ok, deleted the map.sqlite that was created. trying again. |
08:07 |
VanessaE |
it's migrating now |
08:07 |
ShadowNinja |
VanessaE: Sure there's no :(){:|:&};: running in the background consuming resources? ;-) |
08:07 |
VanessaE |
15% completed. |
08:07 |
VanessaE |
25% |
08:08 |
|
MattJ joined #minetest-dev |
08:09 |
VanessaE |
50% (further messages suppressed :P ) |
08:11 |
ShadowNinja |
VanessaE: What's a "normal" migration time for this map? |
08:12 |
VanessaE |
hours. |
08:12 |
VanessaE |
around 3 hours last time I did it, I think |
08:12 |
VanessaE |
a redonkulously long time anyway |
08:12 |
ShadowNinja |
Oh, it probably takes loads of RAM still, right? |
08:12 |
ShadowNinja |
Should save every second or so. |
08:12 |
VanessaE |
http://pastebin.ubuntu.com/10444241/ |
08:13 |
VanessaE |
done. |
08:15 |
VanessaE |
I didn't notice the RAM usage, it finished before I thought to check it |
08:16 |
ShadowNinja |
VanessaE: Can you do that? Mine finishes too fast :-) |
08:16 |
VanessaE |
haha sure |
08:16 |
VanessaE |
gimme a sec |
08:18 |
ShadowNinja |
This patch makes it write every second, it's quite a bit slower on my computer though (adds 25-50%): http://ix.io/gAy (also, haha, paste ID) |
08:18 |
VanessaE |
about 1.8 GB while it was getting to the createDatabase() part... |
08:19 |
VanessaE |
in fact about that amount all the way through |
08:19 |
VanessaE |
at least to the 30% mark |
08:20 |
VanessaE |
and it's only pushing ~50 MB/sec write speed (on a SSD) so you could multithread this for further gains I'll bet |
08:20 |
ShadowNinja |
VanessaE: Hmmm, compare to my latest push. |
08:20 |
ShadowNinja |
SQLite3 doesn't really do multiprocessing well. IDK about the other DBs. |
08:20 |
VanessaE |
note: Ctrl-C does not actually cancel the migration process |
08:21 |
VanessaE |
^C03:20:50: INFO: sigint_handler(): Ctrl-C pressed, shutting down. |
08:21 |
VanessaE |
Migrated 8536635 blocks, 58.7887% completed. |
08:21 |
VanessaE |
(and it keeps on going) |
08:21 |
ShadowNinja |
--migrate dummy wroks now too, which is usefull for testing I guess. |
08:21 |
ShadowNinja |
I'll add a check for *kill. |
08:22 |
VanessaE |
lemme get that pull, sec. |
08:22 |
VanessaE |
s/pull/push/ |
08:25 |
VanessaE |
ok, running it agaimn |
08:25 |
VanessaE |
-m |
08:26 |
VanessaE |
still uses a bit over 1.8 GB, and the write speed seems a bit slower. |
08:27 |
VanessaE |
I'll let it run to completion. |
08:31 |
|
Player_2 joined #minetest-dev |
08:31 |
VanessaE |
http://pastebin.ubuntu.com/10444487/ |
08:31 |
VanessaE |
interesting. |
08:31 |
VanessaE |
it uses less CPU and is faster than the last run |
08:32 |
VanessaE |
(the CPU meter showed it operating in distinct "packets" this time around) |
08:33 |
ShadowNinja |
Huh, odd. |
08:34 |
VanessaE |
maybe just variations in other stuff going on on my machine, or stuff that's cached perhaps |
08:35 |
VanessaE |
at any rate, 6-7 mins is a hell of a lot faster than 3+ hours :D |
08:35 |
VanessaE |
let's just see if the resultant world runs :) |
08:39 |
VanessaE |
nope. |
08:40 |
VanessaE |
the client hangs at "item definitions" |
08:40 |
VanessaE |
there's a few server messages but it never actually "gets into" the game as far as thje ... |
08:40 |
VanessaE |
wait |
08:40 |
VanessaE |
there it goes |
08:40 |
VanessaE |
that took a good long while |
08:40 |
VanessaE |
03:40:49: ERROR[ServerThread]: ServerEnv: Trying to store id=344 statically but block (273,-1,-37) already contains 354 objects. Forcing delete. |
08:41 |
VanessaE |
threw a few of those. that's probably why |
08:45 |
VanessaE |
"Well, everything's here, right down to the last hitchin' post." |
08:45 |
VanessaE |
lgtm |
08:48 |
VanessaE |
the only noticeable thing was how ridiculously long the server took to startup |
08:48 |
VanessaE |
I don't recall my production servers taking that long, but I could be wrong about that. |
08:49 |
VanessaE |
(I pretty much never actually run this map in singleplayer) |
08:49 |
|
nrzkt joined #minetest-dev |
08:50 |
VanessaE |
normal startup time for a Dreambuilder world is only about, oh, 20-25 seconds on this machine. this one took about 1 minute. |
08:50 |
* VanessaE |
checks her test world |
08:50 |
VanessaE |
yep, that seems normal |
08:51 |
VanessaE |
must be a mod loading issue, probably technic slowing things down. |
08:51 |
nrzkt |
hello :) |
08:51 |
VanessaE |
hi nrz |
08:52 |
kilbith |
slt |
08:52 |
nrzkt |
kilbith: failed :p |
08:55 |
ShadowNinja |
nrzkt: Any news on my security patch? |
08:56 |
ShadowNinja |
Also, can you check ^ ? |
08:56 |
nrzkt |
ShadowNinja, very busy this week ShadowNinja, i'll look at it this afternoon, i haven't got time for MT, but i will work on it this afternoon |
08:57 |
ShadowNinja |
Alright nrzkt, NP. And with that I'm off. |
09:00 |
nrzkt |
ShadowNinja: i like your database improvements :) |
09:17 |
VanessaE |
bbl. bned |
09:17 |
VanessaE |
bed |
09:17 |
nrzkt |
ShadowNinja: for your security mod, i think we must merge it into dev-0.5 not master |
09:46 |
|
Zeno` joined #minetest-dev |
10:10 |
Zeno` |
hmm |
10:11 |
Zeno` |
mapgen problems :( |
10:14 |
kilbith |
show us ? |
10:17 |
Zeno` |
1) http://i.imgur.com/5mR9Ehe.jpg 2) http://i.imgur.com/e3ilYGn.jpg |
10:17 |
Zeno` |
very ugly |
10:18 |
Zeno` |
seed = 51351515151515 |
10:18 |
Zeno` |
mgv6 |
10:19 |
Zeno` |
http://dpaste.com/2ANTEFG |
10:21 |
Zeno` |
can someone confirm? |
10:21 |
|
proller joined #minetest-dev |
10:21 |
Zeno` |
new world with map_meta set to what I pasted? |
10:24 |
|
selat joined #minetest-dev |
10:41 |
|
SopaXorzTaker joined #minetest-dev |
10:44 |
Zeno` |
kilbith, can you test with master? |
10:44 |
Zeno` |
co-ords -95, 30, 1990 |
10:45 |
kilbith |
yes |
10:45 |
kilbith |
a moment |
10:45 |
Zeno` |
thank you (test with no mods if that's ok) |
10:46 |
Zeno` |
I will merge #2376 once you can confirm that the ugly terrain is present without it |
10:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/2376 -- Add find_surface_nodes_in_area Lua call which permit to only get the nodes which are under air nodes by nerzhul |
10:52 |
kilbith |
Zeno`, yes i confirm : https://lut.im/MBE5i7Pu/NUV317gj |
10:52 |
kilbith |
but you use a custom biome api |
10:53 |
Zeno` |
can you open the issue? |
10:53 |
Zeno` |
(sorry, I'm working on 3 things at once lol) |
10:53 |
kilbith |
preferably you, since you have a better acknowledgement of that issue |
10:53 |
Zeno` |
hmm ok |
10:54 |
Zeno` |
I would have confirmed and tagged it as soon as it was opened if it was you :P |
10:55 |
kilbith |
it's not shocking for me anyways |
10:58 |
Zeno` |
I think it looks terrible :( |
10:58 |
Zeno` |
if you fly around a bit you will see it more |
11:05 |
|
FR^2 joined #minetest-dev |
11:27 |
Zeno` |
ok, for #2376 I am going to rename the function to minetest.find_nodes_in_area_under_air |
11:27 |
ShadowBot |
https://github.com/minetest/minetest/issues/2376 -- Add find_surface_nodes_in_area Lua call which permit to only get the nodes which are under air nodes by nerzhul |
11:27 |
Zeno` |
and then merge |
11:28 |
Zeno` |
The bug VanessaE noticed is unrelated |
11:43 |
kilbith |
Zeno`, don't forget to close the PR |
11:43 |
Zeno` |
I won't |
11:43 |
Zeno` |
just doing some stuff |
11:49 |
Zeno` |
ok, now I suppose I must report this mapgen issue since kilbith is too lazy :p |
11:51 |
kilbith |
inb4 paramat deviates the bug in "feature" ;) |
11:55 |
Zeno` |
heh |
11:59 |
Zeno` |
there done |
11:59 |
Zeno` |
can you confirm? :P |
12:00 |
kilbith |
done |
12:19 |
|
MinetestForFun joined #minetest-dev |
12:28 |
|
selat joined #minetest-dev |
12:30 |
|
leat joined #minetest-dev |
12:30 |
Zeno` |
bbl |
12:32 |
|
leat joined #minetest-dev |
12:33 |
|
leat joined #minetest-dev |
12:35 |
|
leat joined #minetest-dev |
12:42 |
|
kilbith joined #minetest-dev |
12:45 |
|
Player_2 joined #minetest-dev |
13:08 |
|
nrzkt joined #minetest-dev |
13:15 |
|
est31 joined #minetest-dev |
13:31 |
|
shadowzone joined #minetest-dev |
13:31 |
|
crazyR_ joined #minetest-dev |
13:31 |
|
crazyR_ joined #minetest-dev |
13:51 |
|
est31 joined #minetest-dev |
13:51 |
est31 |
nrzkt, SHA-1 is broken where collision resistance matters |
13:51 |
est31 |
for password hashing collisions are irrelevant |
13:51 |
est31 |
only important thing are second preimage attacks |
13:52 |
est31 |
but there isnt even an efficient one for MD5 |
13:53 |
est31 |
what matters for hashing algorithms is how slow they are |
13:53 |
est31 |
password hashing algs * |
13:53 |
est31 |
the slower the better |
13:54 |
est31 |
because then its more expensive to crack them |
13:54 |
est31 |
Thats however *only* to prevent passwords getting weaker because of moore's law |
13:57 |
est31 |
I don't say that SHA-256 wouldn't be *better* |
13:57 |
est31 |
even git uses SHA-1 and nobody has a problem. |
13:59 |
est31 |
but thats bikeshedding |
13:59 |
est31 |
lets wait what that guy codes ... |
14:00 |
sfan5 |
<est31> even git uses SHA-1 and nobody has a problem. |
14:00 |
sfan5 |
*sigh* |
14:00 |
sfan5 |
git could use md4 and everything would be fine |
14:04 |
nrzkt |
est31, you have requested a protocol bump if i could remember ? |
14:05 |
est31 |
yes |
14:05 |
nrzkt |
and also it's you which have created the serverlist update? |
14:05 |
est31 |
yes |
14:05 |
nrzkt |
i noticed a bug on the last question |
14:05 |
nrzkt |
s/question/point |
14:05 |
est31 |
ok which |
14:05 |
nrzkt |
if the server is not in master serverlist, the server is marked as incompatible |
14:06 |
nrzkt |
but we don't know if it's or not because it isn't in server list |
14:06 |
est31 |
so how do you access the server? from the favorites? |
14:06 |
nrzkt |
can you provide a patch to disable the version check if the server which i want to connect is not in public serverlist (because i don't know its compat) |
14:06 |
nrzkt |
yes |
14:06 |
est31 |
or by entering its ip and port? |
14:09 |
est31 |
usually if the fields are nil the server is allowed |
14:09 |
est31 |
so they are set because of some reason |
14:10 |
|
BlockMen joined #minetest-dev |
14:10 |
BlockMen |
nrzkt, https://github.com/minetest/minetest/commit/3b1ac5b889bc84fe3f73eee583228cef10d945c1 |
14:10 |
BlockMen |
wtf?! |
14:11 |
nrzkt |
look at the issue in the commit msg |
14:11 |
nrzkt |
we have bugs with this PR |
14:11 |
BlockMen |
that gives YOU the right to just revert something? |
14:11 |
BlockMen |
where is the discussion on irc? |
14:12 |
nrzkt |
this issue is opened since 2 weeks and cause problem to user and me too. Then the it's valid for a revert. |
14:12 |
nrzkt |
It's impossible to provide a patchfix to it, else i would ask to you :) |
14:12 |
BlockMen |
its 2 DAYS |
14:12 |
nrzkt |
days sorry, miss tip |
14:12 |
BlockMen |
furthermore you cant just revert what you want |
14:12 |
BlockMen |
this is not even your maintainer area |
14:13 |
est31 |
If you provide me with a way to reproduce your issue nrzkt I'll try to fix it |
14:13 |
BlockMen |
so you would have needed another to agree |
14:13 |
nrzkt |
i don't revert what i want, i revert a bug in master that i can confirm |
14:13 |
BlockMen |
and just for the protocoll, this is a bug that has been introduced by the node highlighting |
14:13 |
BlockMen |
so this patch just showed more things to fix |
14:13 |
BlockMen |
not to hide everything by reverting my commit |
14:13 |
|
Zeno` joined #minetest-dev |
14:14 |
nrzkt |
what do we must fix then ? |
14:14 |
BlockMen |
i will provide a patch for that next days |
14:14 |
BlockMen |
just stop to do what you want here |
14:14 |
nrzkt |
provide them together please. |
14:14 |
BlockMen |
we have rules, accept them or leave |
14:14 |
Zeno` |
err, what's being discussed? |
14:15 |
nrzkt |
est31: i have 127.0.0.1 port 30000 with my pseudo on favorites . I double click, no connection, it said the protocol is 13 |
14:15 |
nrzkt |
https://github.com/minetest/minetest/commit/3b1ac5b889bc84fe3f73eee583228cef10d945c1 |
14:15 |
BlockMen |
Zeno`, nrzkt reverts things he dont have a clue of withput any discussion |
14:15 |
nrzkt |
Zeno` |
14:15 |
BlockMen |
that is against any rules we have |
14:15 |
nrzkt |
i only revert two lines of code which causes problems since 2 days in master |
14:15 |
Zeno` |
ok, let's just calm down and see what was reverted |
14:15 |
Zeno` |
link? |
14:15 |
nrzkt |
https://github.com/minetest/minetest/commit/3b1ac5b889bc84fe3f73eee583228cef10d945c1 |
14:16 |
Zeno` |
ok, and how does that cause problems BlockMen? |
14:16 |
BlockMen |
Zeno`: what do you mean? |
14:16 |
Zeno` |
I mean how does reverting those 2 lines cause a problem? |
14:17 |
* Zeno` |
looks at logs |
14:17 |
BlockMen |
first it breaks another thing again, second it is not ok because he breaks the rules |
14:17 |
BlockMen |
this is a COMMUNITY project |
14:17 |
BlockMen |
not a nrzkt on man show |
14:17 |
BlockMen |
he is not allowed to do what he wants |
14:18 |
sfan5 |
it doesn't even matter what those 2 lines break or do not break |
14:18 |
kilbith |
the bug seems occurs on OpenGL 4.4, all is fine on 2.1 |
14:18 |
Zeno` |
still reading logs |
14:18 |
nrzkt |
yes. and i solve a problem in master. And be quiet, i was absent since 1 week for work. I don't do what i want i work on my things and i solve issues. If a trivial commit causes issues, it's simple: revert. the commit isn't lost but we fix the problem temporary since a new workaround |
14:18 |
sfan5 |
nobody is allowed to do something out of their maintainer area without consent of another maintainer |
14:19 |
sfan5 |
for small fixes it's enough to announce it |
14:19 |
Zeno` |
ok, so from what I see the only mistake nrzkt made was not to announce it |
14:19 |
nrzkt |
yes sorry, i was on the train. then no IRC possible |
14:19 |
sfan5 |
huh? |
14:19 |
Zeno` |
It's trivial... a simple "I'm going to revert xxxxxx in 10 minutes" is all that would have been needed. Right? |
14:20 |
sfan5 |
how can you push something to github but not use irc on the train? |
14:20 |
sfan5 |
Zeno`: yes |
14:20 |
BlockMen |
Zeno`, no the draworder is messed again. hiding other bugs by just reverting is no solution |
14:20 |
nrzkt |
sfan5: IRC connection not stable, unlike a simple HTTPS request :( |
14:20 |
nrzkt |
Paris data in trains is... shit |
14:20 |
sfan5 |
¯\_(ツ)_/¯ |
14:20 |
Zeno` |
BlockMen, of course it's not. But something that simple does not need approval. Just announcement. |
14:20 |
sfan5 |
then don't push at all and wait |
14:21 |
Zeno` |
fair enough nrzkt? |
14:21 |
BlockMen |
Zeno`, also since it did not break anything important ans since its just since TWO days there was no reason to revert it that fix |
14:21 |
nrzkt |
yeah, okay then. and Zeno` don't be so speed on my PRs |
14:21 |
kilbith |
btw, the revert rule is not even stated : http://dev.minetest.net/Organisation |
14:21 |
BlockMen |
he could have wait for statement by me or anything |
14:21 |
BlockMen |
kilbith, revert = commit |
14:21 |
sfan5 |
kilbith: everything you do w/ git that changes something is a commit |
14:21 |
Zeno` |
ok, I think this is a mountain out of a molehill personally :) nrzkt has said he will announce in future |
14:22 |
nrzkt |
okay, sorry then. Now, you talked about a workaround you find, what is it ? |
14:22 |
nrzkt |
Zeno` why do you rename my function in the PR ? |
14:22 |
Zeno` |
nrzkt, so that it can be generalised in the future without breaking existing Lua API |
14:23 |
Zeno` |
nrzkt, and also because it's a better description of what the function actually does ;) |
14:23 |
BlockMen |
nrzkt, since you dont even get what is going on in draworder i wont talk with you about that |
14:23 |
BlockMen |
just dont break our rules and do your stuff |
14:23 |
nrzkt |
BlockMen, maybe if the team talk about its project it could work better ? |
14:23 |
BlockMen |
^ lol |
14:24 |
Zeno` |
nrzkt, after my name change (of the function) it's possible to keep the new change and as well extend it |
14:24 |
BlockMen |
says the ego guy |
14:24 |
nrzkt |
Zeno` okay, i see, no problem for me |
14:24 |
BlockMen |
nrzkt, just fuck yourself |
14:24 |
nrzkt |
wtf |
14:24 |
nrzkt |
can you be polite please ? |
14:24 |
est31 |
wow this channel is clearly productive today |
14:24 |
kilbith |
^ celeron55 |
14:25 |
Zeno` |
BlockMen, I understand where you're coming from, but I also understand where nrzkt is coming from. Are you sure this particular issue is such a big deal? |
14:27 |
Zeno` |
But, yeah. nrzkt you were ok to commit that change, BUT you forgot to say you were going to. Minor issue |
14:27 |
kilbith |
strangely the insults are not sanctionned today... |
14:27 |
sfan5 |
kilbith: or maybe celer​on55 is just not here right now |
14:28 |
BlockMen |
Zeno`, its about he acts. Rushing through every thing is not ok. When there is no important reason its not ok to revert things just because he THOUGHT it breaks something. |
14:28 |
Zeno` |
insults "not being sanctioned" would mean that they're ok :P |
14:28 |
kilbith |
c55 is not the only OP... |
14:28 |
BlockMen |
the issue was opend TWO days ago |
14:28 |
sfan5 |
kilbith: anyway, what do you expect an op to do about this? |
14:28 |
nrzkt |
yeah i reverted your commit after testing the issue, i take a little time to reproduce the problem. Because master must be as stable as possible as celeron55 i reverted it because of maintainers which pulls on master |
14:29 |
kilbith |
sfan5: the same thing that you did earlier to me ? |
14:29 |
kilbith |
well, doesnt matter |
14:29 |
sfan5 |
kilbith: it doesn't make sense to kick people for one insult |
14:29 |
Zeno` |
If it's such a huge issue just add the 2 lines back :/ |
14:30 |
kilbith |
it's deeper than that, you should dig out the logs |
14:30 |
Zeno` |
(personally I think the z-order thing does cause the reported bug though... but if it needs to be added back for peace then so be it) |
14:31 |
nrzkt |
BlockMen your commit was present since 1 week, then it's very long for maintainers, hopefully one has report the issue (i haven't tested it before because i'm working on my own branch which is a little bit outdated). I approved your commit 1 week ago because it was trivial and seems good, but some openGL engines have problems with it (and the previous engine i tested doesn't have problem with it). |
14:32 |
nrzkt |
then, we are somes on the channel, we can vote for the integration of this change, if you want, now we are alive :p |
14:32 |
BlockMen |
Zeno` the issue are not the two lines. the issue is to think to know everything better and "fix" something just besides. that he broke the git rules is just a bonus |
14:34 |
nrzkt |
people can do errors. I reverted your commit and it's not the good mean. You insult me and break MT on some opengl engines :p |
14:34 |
BlockMen |
nrzkt, you still dont get it, great |
14:37 |
nrzkt |
it's not a menace problem, like you i worked all the week at work, it's friday, then now we can be cool, calm and work on MT without having issue in the team. Insults are generally used everywhere, then i don't care. |
14:38 |
Zeno` |
BlockMen, I think he's accepted that the "rule" was broken |
14:38 |
kilbith |
yes, and expressed his apologize already |
14:38 |
nrzkt |
and then i'm calm. okay i reverted a problematic PR, okay it was yours. Then now we need to move forward and fix it properly |
14:38 |
Zeno` |
A trivial change can be merged without agreement if it's announced in the channel and nobody object within a short period |
14:39 |
Zeno` |
Now, should this commit be reversed? |
14:39 |
Zeno` |
the commit that reversed the change I mean |
14:40 |
Zeno` |
Is the bug report still open? |
14:40 |
nrzkt |
i closed it after the revert |
14:40 |
celeron55 |
did nrzkt ask from another core developer before committing that revert? |
14:40 |
nrzkt |
no celeron55 |
14:40 |
celeron55 |
that's wrong then |
14:41 |
nrzkt |
and you are right |
14:41 |
Zeno` |
we've all accepted that :) |
14:41 |
Zeno` |
so, let's put it back (the stuff that was reverted) and move on |
14:42 |
* Zeno` |
smashes the molehill |
14:42 |
nrzkt |
while this conversation occurs i add some commits to #2327, like the damage fix PR. Can someone look at this before i merge it to dev-0.5 ? |
14:42 |
Zeno` |
stupid damn ants and moles! |
14:42 |
ShadowBot |
https://github.com/minetest/minetest/issues/2327 -- [Patch 3/4] Network rework: Packet fixes & cleanups by nerzhul |
14:43 |
celeron55 |
i agree with reverting it though; it seems like a regression that has to be solved before the actual commit can be used |
14:44 |
|
psedlak joined #minetest-dev |
14:44 |
celeron55 |
i won't agree with unreverting the revert |
14:44 |
Zeno` |
fair enough. The lesson has been learned by the nrzkt I am guessing |
14:44 |
nrzkt |
i'm an object ? :p |
14:44 |
Zeno` |
lol |
14:45 |
Zeno` |
yes |
14:45 |
Zeno` |
I malloced you |
14:45 |
nrzkt |
i want to break free ! |
14:46 |
nrzkt |
what about #2374 ? |
14:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/2374 -- Update minetest.po (Japanese) by Rui914 |
14:46 |
Zeno` |
rui has always maintained that translation afaik |
14:46 |
Zeno` |
so without knowing Japanese and not looking at it I'd say it's ok |
14:47 |
nrzkt |
then i can merge ? |
14:47 |
Zeno` |
yes, it seems fine |
14:48 |
Zeno` |
they've just changes some em-dashes and stuff by the looks of it |
14:49 |
nrzkt |
yes. And for #2391 ? turkish add ? |
14:49 |
ShadowBot |
https://github.com/minetest/minetest/issues/2391 -- Turkish language(Fix) by Rui914 |
14:50 |
Zeno` |
hmm |
14:50 |
Zeno` |
is he a native Turkish speaker? |
14:50 |
nrzkt |
no but the original committer seems to be |
14:50 |
|
CraigyDavi joined #minetest-dev |
14:51 |
Zeno` |
oh I see |
14:51 |
Zeno` |
well, I think it would be ok if the author was changed when committing |
14:51 |
Zeno` |
Rui seems to have just fixed the kind of dumb way that Mahmut presented the PR :/ |
14:52 |
nrzkt |
i fixed the commit header, ok for merge ? |
14:53 |
est31 |
nrzkt, I can't reproduce |
14:53 |
nrzkt |
(in the patch i'll provide, not in the commit) |
14:53 |
Zeno` |
Author is now Mahmut Elmas mahmutelmas06gmail.com? |
14:53 |
nrzkt |
Zeno` Author: mahmutelmas06 <mahmutelmas06users.noreply.github.com> |
14:53 |
Zeno` |
ok, I agree to that |
14:53 |
est31 |
I cant reproduce your issue nrzkt, on commit 38e6280552 which is after yours |
14:53 |
|
psedlak joined #minetest-dev |
14:54 |
est31 |
s/yours/my change/ |
14:54 |
nrzkt |
est31: PR 3/4 is merged in your testing build ? |
14:54 |
nrzkt |
you need it |
14:54 |
est31 |
ah I should merge it |
14:54 |
nrzkt |
yes, because it's due to the default version set. Without merge, version is not increased :p |
14:54 |
nrzkt |
ok, merged Zeno` |
14:55 |
est31 |
ah it sets a default version, yea remember that |
14:55 |
est31 |
dammit you are right |
14:56 |
nrzkt |
i think we must disable filtering for servers which are not in public server list :) |
14:59 |
est31 |
it was not me doing that default versions... |
14:59 |
celeron55 |
i added #5 here due to blockmen's request: http://dev.minetest.net/Git_Guidelines#Upstream_commit_rules |
14:59 |
est31 |
but no problem I fix it |
14:59 |
celeron55 |
it's probably generic enough to be worth being there |
14:59 |
celeron55 |
and it's what we've been doing since forever |
15:00 |
celeron55 |
nrzkt: ^ |
15:00 |
nrzkt |
thanks est31. I must do others things, i'll be back later, please provide a PR i'll test it and merge it this week-end |
15:00 |
est31 |
ok nrzkt |
15:01 |
Zeno` |
what got added, celeron55? |
15:02 |
Zeno` |
oh I see. nvm |
15:02 |
nrzkt |
yes celeron55. But i don't rush. I have tested the bug before revert, it's not a rush. I have a very high priority to user datas, because i'm working on such environments at works, not in R&D but production, and the data consistency is my priority in every project. Security of the datas and the network |
15:02 |
nrzkt |
s/works/work |
15:04 |
celeron55 |
it's rushing if one reverts a commit made by someone else when the commit is not doing anything harmful in the long term |
15:04 |
celeron55 |
if it was corrupting map data, then of course you would revert it |
15:04 |
celeron55 |
but this is doing like nothing |
15:04 |
celeron55 |
commits are not equal |
15:05 |
nrzkt |
no problem celeron55 :). ofc each commit can have different impact |
15:06 |
nrzkt |
i added protocol bump on #2372, because the serialized object passed over network change. This commit mustn't be merged into master, only in dev-0.5 |
15:06 |
ShadowBot |
https://github.com/minetest/minetest/issues/2372 -- Fix animation restarts, implement support for non-looped animations by MirceaKitsune |
15:06 |
Zeno` |
wait... there is a new dev branch? |
15:06 |
* Zeno` |
looks |
15:07 |
nrzkt |
yes, as celeron55 requested 1 week ago |
15:07 |
Zeno` |
lol, I didn't notice :3 |
15:08 |
nrzkt |
it's used for incompatible change with master, atm, because we keep master for 0.4.x, because we must tell maintainers to switch to other branch and let them switch to it. And when dev-0.5 will be okay, we will merge it into master |
15:08 |
Zeno` |
I remember the conversation but I think that was the night I was "at the disco" |
15:08 |
nrzkt |
:p |
15:09 |
Zeno` |
i.e. blind and rotten drunk |
15:09 |
Zeno` |
so it escaped my memory |
15:10 |
nrzkt |
i need to go, see you. est31, good luck with the fix. When it was valid and i test it will be merged and i merge PR 3/4 in dev-0.5 |
15:10 |
Zeno` |
fine, nrzkt. Don't say bye to me, or BlockMen or sfan5! |
15:10 |
est31 |
nrzkt, do you accept other network breaking changes in that branch too? |
15:11 |
sfan5 |
? |
15:11 |
nrzkt |
dev-0.5 is for that est31, all incompatible changes accepted by core dev will go to it, like colored chat and some other features :) |
15:11 |
Zeno` |
sfan5, being pedantic. His comment did say bye to all really |
15:11 |
nrzkt |
lol Zeno` :p |
15:20 |
est31 |
celeron55, why is the dev branch rebased onto master and not merged? I think merging is better because pull works then. |
15:21 |
est31 |
in here: http://dev.minetest.net/Git_Guidelines |
15:23 |
Zeno` |
huh? |
15:23 |
est31 |
not now, some day |
15:25 |
est31 |
Pasting relevant sentences: "Freature freezes take effect in the master branch, and while the freeze is active, a separate development branch ("dev", or some other suitable name) is made into which pull requests are merged. It is then *rebased* onto master once the freeze is over." |
15:28 |
Zeno` |
kind of odd |
15:31 |
est31 |
Its nothing important, but the more breaking git setup is, the less people will use latest dev branch, and test it same time. that makes searching for causes harder when the change was some time ago |
15:33 |
Zeno` |
IMO the *freeze* should be in the separate branch |
15:35 |
Zeno` |
stable should be rebased to the freeze branch and master left alone |
15:35 |
Zeno` |
that's how most projects work anyway |
15:36 |
est31 |
rebases aren't nice |
15:36 |
est31 |
I'm indifferent on what is better |
15:36 |
est31 |
stable on master or dev |
15:36 |
est31 |
but stuff should be merged |
15:36 |
Zeno` |
I'm talking more about the branches than rebasing vs. merging |
15:38 |
Zeno` |
merging (can) results in a non-linear history so I can kind of see how that is good |
15:38 |
Zeno` |
but the branch management for minetest needs improvement IMO |
15:39 |
Zeno` |
it's all pretty simple... stable branch may not included everything in master |
15:40 |
Zeno` |
all dev should always go to master |
15:40 |
Zeno` |
(whether feature freeze is in effect or not) |
15:40 |
est31 |
there are projects which do it this way and others which do it that way |
15:41 |
|
fz72 joined #minetest-dev |
15:41 |
est31 |
but fact is cmake process currently requires to have version tags *inside* history |
15:41 |
Zeno` |
Most projects do not make a new branch just because there is a feature freeze though |
15:41 |
est31 |
but freezes are blocking and no fun for those which dont want to freeze |
15:42 |
Zeno` |
yeah, that's why a freeze should not be a "branch" |
15:42 |
Zeno` |
and stuff *should* continue to be pushed to master |
15:42 |
Zeno` |
the "freeze" should be in the stable branch |
15:43 |
Zeno` |
and commits cherry picked back to that until freeze is thawed |
15:43 |
est31 |
I dont think nonlinear history is that bad |
15:43 |
est31 |
also it helps confirming the hashes |
15:43 |
est31 |
(unimportant side note) |
15:43 |
Zeno` |
well, non-linear is not that bad, but stopping development in master is |
15:44 |
est31 |
as long as it continues somewhere |
15:44 |
est31 |
perhaps your idea is better perhaps c55's |
15:45 |
est31 |
If you decided to move back to other model, #2357 should be reopened. |
15:45 |
ShadowBot |
https://github.com/minetest/minetest/issues/2357 -- Make git version discovery work on master by est31 |
15:46 |
|
shadowzone joined #minetest-dev |
15:49 |
Zeno` |
it's not "my idea". It's the most common git workflow |
15:50 |
Zeno` |
It's probably the most common because it's the most sane :) |
15:50 |
|
proller joined #minetest-dev |
15:51 |
Zeno` |
you never rebase or even "merge" the freeze branch... you cherry-pick to it |
15:52 |
est31 |
if you do that you will need to change your tools |
15:52 |
Zeno` |
what tools? |
15:52 |
est31 |
git describe is used in a way it cant be used anymore |
15:53 |
est31 |
it searches the last tag |
15:53 |
Zeno` |
so I would need to change from using git to... ? |
15:53 |
est31 |
see #2357 |
15:53 |
ShadowBot |
https://github.com/minetest/minetest/issues/2357 -- Make git version discovery work on master by est31 |
15:53 |
est31 |
git would stay |
15:53 |
est31 |
just git describe wouldnt be possible to use anymore |
15:54 |
Zeno` |
that's not a tool issue |
15:54 |
est31 |
just noting that |
15:55 |
Zeno` |
this is a: https://github.com/minetest/minetest/pull/2357/files#diff-af3b638bc2a3e6c650974192a53c7291R26 issue |
15:55 |
est31 |
? |
15:56 |
Zeno` |
you've put the last commit has into CMakeLists.txt and think that's normal? |
15:56 |
Zeno` |
s/has/hash |
15:56 |
est31 |
not last commit hash |
15:56 |
est31 |
only last release commit hash |
15:56 |
Zeno` |
last release hash, I know |
15:56 |
est31 |
there are already two forced commits every release |
15:57 |
est31 |
one for bumping version, one for saying "we're on dev again" |
15:57 |
est31 |
so the second commit can also update that hash |
15:57 |
est31 |
its manual tagging |
15:57 |
Zeno` |
that's only because branches are being used incorrectly |
15:57 |
est31 |
thats because git describe doesnt fit the branching use you propose |
15:58 |
Zeno` |
it doesn't have to |
15:58 |
Zeno` |
seriously, think about it. And we'll chat more tomorrow :) |
15:58 |
Zeno` |
Right now I am going to idle :) |
15:58 |
est31 |
happy idling |
15:58 |
Zeno` |
tu |
16:12 |
|
shadowzone_ joined #minetest-dev |
16:15 |
|
celeron55 joined #minetest-dev |
16:32 |
|
roniz joined #minetest-dev |
16:40 |
|
rubenwardy joined #minetest-dev |
16:47 |
|
ImQ009 joined #minetest-dev |
17:04 |
|
rubenwardy joined #minetest-dev |
17:08 |
|
AnotherBrick joined #minetest-dev |
17:33 |
|
Hunterz joined #minetest-dev |
17:36 |
|
est31 joined #minetest-dev |
17:42 |
|
Calinou joined #minetest-dev |
18:04 |
|
Krock joined #minetest-dev |
18:05 |
|
roniz joined #minetest-dev |
18:06 |
|
est31 joined #minetest-dev |
18:06 |
|
proller joined #minetest-dev |
18:34 |
est31 |
so is it true that minetest logs doesn't record when you /setpassword for somebody? |
18:34 |
est31 |
so it records every single block that has been dug, but no password changes? |
18:35 |
est31 |
I'm a bit angry to be honest |
18:36 |
est31 |
this got prio #1 on my to-fix list now |
18:38 |
VanessaE |
don't you just LOVE minetest's priorities? ;) |
18:38 |
est31 |
yea diggydiggydig |
18:41 |
|
alket joined #minetest-dev |
18:52 |
celeron55 |
the logging hasn't really been designed for admins but for developers; not that that would be the way it should be, but that's how you can understand it |
19:06 |
|
Calinou_ joined #minetest-dev |
19:22 |
|
proller joined #minetest-dev |
19:36 |
|
selat joined #minetest-dev |
19:56 |
|
gravgun joined #minetest-dev |
19:58 |
|
neoascetic joined #minetest-dev |
20:04 |
|
AnotherBrick joined #minetest-dev |
20:06 |
|
AnotherBrick joined #minetest-dev |
20:12 |
est31 |
celeron55, thanks for the explanation. I'll do a PR. |
20:13 |
est31 |
I'll leave the particular hashes out. they might be sensitive. |
20:32 |
|
roniz_ joined #minetest-dev |
20:49 |
|
gravgun joined #minetest-dev |
21:08 |
|
proller joined #minetest-dev |
21:12 |
|
SudoAptGetPlay joined #minetest-dev |
21:13 |
|
BlockMen left #minetest-dev |
21:16 |
|
roniz joined #minetest-dev |
21:35 |
|
chchjesus joined #minetest-dev |
21:40 |
|
MattJ joined #minetest-dev |
21:42 |
|
roniz_ joined #minetest-dev |
21:47 |
|
Miner_48er joined #minetest-dev |
21:51 |
|
proller joined #minetest-dev |
22:05 |
est31 |
wow 2400 issues at minetest |
22:05 |
est31 |
anyway #2400 is out |
22:05 |
ShadowBot |
https://github.com/minetest/minetest/issues/2400 -- Add /setpassword and /clearpassword logging by est31 |
22:06 |
est31 |
btw more than a half issue per commit ;9 |
22:25 |
|
luizrpgluiz joined #minetest-dev |
22:26 |
|
luizrpgluiz left #minetest-dev |
22:44 |
|
proller joined #minetest-dev |
22:44 |
|
Taoki joined #minetest-dev |
23:14 |
|
shadowzone joined #minetest-dev |
23:32 |
|
proller joined #minetest-dev |
23:42 |
|
proller joined #minetest-dev |
23:45 |
ShadowNinja |
#2402 |
23:45 |
ShadowBot |
https://github.com/minetest/minetest/issues/2402 -- Clean up and tweak build and versioning systems by ShadowNinja |
23:53 |
est31 |
ShadowNinja, it fails |
23:53 |
est31 |
travis |
23:56 |
|
proller joined #minetest-dev |