| Time |
Nick |
Message |
| 00:05 |
diemartin |
yay no more guests |
| 00:05 |
VanessaE |
and the world is once again at peace. :) |
| 00:06 |
rubenwardy |
hate to burst your bubble(s) |
| 00:06 |
rubenwardy |
but all the Android clones we have will probably keep it |
| 00:07 |
diemartin |
better than nothing |
| 00:07 |
VanessaE |
um, wut? "VanessaE, be patient!" |
| 00:08 |
diemartin |
no_guests will take care of the rest |
| 00:13 |
paramat |
now merging |
| 00:14 |
VanessaE |
paramat: care to explain ^^^ |
| 00:14 |
paramat |
oh that :] |
| 00:22 |
|
Tmanyo joined #minetest-dev |
| 00:23 |
paramat |
merged. closed a 3 year old issue and a performance optimisation, nice |
| 00:23 |
VanessaE |
cool |
| 00:23 |
VanessaE |
now merge #5186 :) |
| 00:23 |
ShadowBot |
https://github.com/minetest/minetest/issues/5186 -- Soft node overlay by juhdanad |
| 00:23 |
VanessaE |
(NOW you can take a pot shot :P ) |
| 00:25 |
paramat |
Vanessa no big issue, a few days after i started the mese post light PR you were being impatient on IRC that i hadn't completed it yet, so i wrote that |
| 00:25 |
VanessaE |
:P |
| 00:25 |
paramat |
no problem though |
| 00:30 |
paramat |
wow it got merged, thanks |
| 00:31 |
VanessaE |
? |
| 00:36 |
sofar |
I just merged a few minor things |
| 01:14 |
|
TC02 joined #minetest-dev |
| 01:35 |
|
csijdlw joined #minetest-dev |
| 01:37 |
|
TheReaperKing joined #minetest-dev |
| 01:52 |
|
lordfingle joined #minetest-dev |
| 01:53 |
|
ssieb joined #minetest-dev |
| 02:05 |
|
Miner_48er joined #minetest-dev |
| 02:12 |
|
octacian_ joined #minetest-dev |
| 02:38 |
|
betterthanyou710 joined #minetest-dev |
| 02:59 |
|
TC04 joined #minetest-dev |
| 03:44 |
|
DI3HARD139 joined #minetest-dev |
| 04:30 |
|
fling joined #minetest-dev |
| 06:00 |
|
octacian joined #minetest-dev |
| 06:33 |
|
troller joined #minetest-dev |
| 06:33 |
|
DI3HARD139 joined #minetest-dev |
| 06:35 |
|
diemartin joined #minetest-dev |
| 06:39 |
|
nerzhul joined #minetest-dev |
| 06:40 |
Sokomine |
er. did mod security get restricted even more? |
| 06:44 |
|
AntumDeluge joined #minetest-dev |
| 06:50 |
Sokomine |
security is important. i want to be able to enable mod security and use it. but if normal operations like reading and writing a file with config values for a mod no longer works, mod security seems kind of obsolete - because it cannot be used in a productive way. i'm certainly missing something here. please tell me how to use it properly or where i can find up-to-date information on how it can be us |
| 06:51 |
Sokomine |
ed (apart from disabling entirely) |
| 06:53 |
nerzhul |
Sokomine, you nolonger need reading or writing files, now you have mod metadata for that |
| 06:53 |
|
Hunterz joined #minetest-dev |
| 06:53 |
nerzhul |
call your metadata container at mod load and use it to read/store what you need |
| 06:55 |
Sokomine |
ah. mod metadata sounds like a good idea |
| 06:55 |
Sokomine |
can you point me to a summary of the new feature? |
| 06:56 |
nerzhul |
https://github.com/minetest/minetest/commit/ef6feca501fcf0d5a1fd2021f1d4df96a4533f65 |
| 06:56 |
nerzhul |
it's fresh |
| 06:57 |
nerzhul |
mod metadata is exact same API than nodemeta and itemstack meta, unified by ruben |
| 06:57 |
Sokomine |
thanks. it really ought to be mentionned somewhere...and it will break i'd say all of my old local worlds |
| 06:57 |
nerzhul |
Sokomine, the mod security enable is not my fact, i don't know how you can migrate your world with it enabled |
| 06:58 |
nerzhul |
mod metadata is not a response to server side modding, it was created at first for client side modding which never call io as mod security is strengher on it |
| 06:58 |
nerzhul |
but having it server side permit to solve the io security problems with mods |
| 06:59 |
Sokomine |
nerzhul: i'm afraid migrating old worlds isn't possible anymore without disabling mod security alltogether |
| 06:59 |
nerzhul |
you can request insecure env i think |
| 06:59 |
nerzhul |
permitting to migrate data |
| 06:59 |
nerzhul |
and after 1 or 2 release, just break the migration model and keep the secure env |
| 07:00 |
nerzhul |
merging #5193 thanks sofar |
| 07:00 |
Sokomine |
hmmm. based on my experience with the trusted environment, i wouldn't trust its functionality too far. i doubt something considered as criminal as reading a mod config file would be allowed that way.... |
| 07:00 |
ShadowBot |
https://github.com/minetest/minetest/issues/5193 -- Change default nodetimer_interval to 0.2s. by sofar |
| 07:22 |
|
jordan4ibanez joined #minetest-dev |
| 07:23 |
red-002 |
huh when was io broken in mod security?? |
| 07:26 |
Sokomine |
red-002: i don't know when exactly. it is, for me, now. reading a file in the folder of the current world no longer works. and that is what quite a lot of my mods do in order to store and retrieve their data |
| 07:27 |
red-002 |
well that's a bug then |
| 07:28 |
nerzhul |
red-002, in mod_security then |
| 07:28 |
nerzhul |
red-002, are we okay with CSM branch since my yesterday squash ? |
| 07:28 |
red-002 |
it looks ok to me |
| 07:30 |
red-002 |
well that will need fixing before release |
| 07:33 |
Sokomine |
ah. it seems at least read access can be regained by adding the world path in front of the file name. that used to be default afaik.... |
| 07:37 |
Sokomine |
regarding mod_security: reading seems to be possible with minetest.get_worldpath()..'/name_of_file' - while writing works only with 'name_of_file' |
| 07:38 |
Sokomine |
and that's probably because i add the worldpath when saving. it's still strange that reading the file without worldpath works without mod security and doesn't with it beeing enabled |
| 07:42 |
Sokomine |
so it works again. as long as the full filename is given |
| 08:11 |
Zeno` |
hmm |
| 08:28 |
sofar |
I have the weirdest problem with node sounds |
| 08:28 |
sofar |
somehow my client only plays the first footstep sound |
| 08:29 |
sofar |
and then it stays quiet |
| 08:29 |
sofar |
when I jump, I hear footstep sounds each jump |
| 08:29 |
sofar |
but not when I walk |
| 08:38 |
sofar |
ohhh, it's plainly playing the sound from a node 1m below my feet, not directly below! |
| 08:38 |
sofar |
what the ... how does that just happen? |
| 08:39 |
nore |
sofar: does this happen with normal mintest_game? |
| 08:39 |
nore |
=e |
| 08:39 |
nore |
grr |
| 08:39 |
nore |
*+e |
| 08:39 |
sofar |
no |
| 08:40 |
sofar |
I gotta get sleep, way too late again |
| 08:40 |
sofar |
later |
| 08:40 |
nore |
bye |
| 08:41 |
nore |
for the logs: could it be the player model that is different, so a different collision box size or something related? |
| 08:41 |
red-002 |
yeah footstep sounds seem to work in strange ways |
| 08:43 |
sofar |
mtg's model actually |
| 08:45 |
|
nrzkt joined #minetest-dev |
| 08:47 |
|
ssieb joined #minetest-dev |
| 08:59 |
|
Krock joined #minetest-dev |
| 08:59 |
|
Krock joined #minetest-dev |
| 09:13 |
|
ssieb joined #minetest-dev |
| 09:38 |
|
nrzkt joined #minetest-dev |
| 09:40 |
|
ssieb joined #minetest-dev |
| 11:05 |
|
troller joined #minetest-dev |
| 11:18 |
|
lisac joined #minetest-dev |
| 11:38 |
|
Player_2 joined #minetest-dev |
| 12:07 |
|
Fixer joined #minetest-dev |
| 12:16 |
|
YuGiOhJCJ joined #minetest-dev |
| 12:30 |
|
Krock joined #minetest-dev |
| 12:30 |
|
proller joined #minetest-dev |
| 12:55 |
|
TC02 joined #minetest-dev |
| 12:57 |
|
nrzkt joined #minetest-dev |
| 13:10 |
|
rubenwardy joined #minetest-dev |
| 13:26 |
|
diemartin joined #minetest-dev |
| 14:37 |
|
diemartin joined #minetest-dev |
| 14:40 |
|
octacian joined #minetest-dev |
| 14:43 |
|
STHGOM joined #minetest-dev |
| 15:11 |
|
betterthanyou710 joined #minetest-dev |
| 15:19 |
|
proller joined #minetest-dev |
| 15:46 |
|
twoelk joined #minetest-dev |
| 15:50 |
|
troller joined #minetest-dev |
| 15:52 |
|
rubenwardy joined #minetest-dev |
| 15:53 |
rubenwardy |
#4966 |
| 15:53 |
ShadowBot |
https://github.com/minetest/minetest/issues/4966 -- Use tree to list mods rather than textlist by rubenwardy |
| 15:56 |
|
proller joined #minetest-dev |
| 16:34 |
|
rubenwardy joined #minetest-dev |
| 16:57 |
|
twoelk|2 joined #minetest-dev |
| 17:56 |
|
Darcidride joined #minetest-dev |
| 18:10 |
|
numzero joined #minetest-dev |
| 18:39 |
|
Fixer_ joined #minetest-dev |
| 18:44 |
|
rubenwardy joined #minetest-dev |
| 18:54 |
|
ssieb joined #minetest-dev |
| 19:00 |
|
STHGOM joined #minetest-dev |
| 19:13 |
|
nnnn20431 joined #minetest-dev |
| 19:17 |
|
Taoki joined #minetest-dev |
| 19:24 |
|
STHGOM_ joined #minetest-dev |
| 19:26 |
|
QwertyDragon joined #minetest-dev |
| 19:29 |
|
octacian_ joined #minetest-dev |
| 19:30 |
|
STHGOM joined #minetest-dev |
| 19:30 |
|
STHGOM joined #minetest-dev |
| 20:09 |
Hijiri |
sofar: does it matter whether you are above or below y=0? |
| 20:09 |
sofar |
Hijiri: I was below 0 |
| 20:10 |
sofar |
can't try right now ... at work |
| 20:10 |
sofar |
but, I was thinking about that as well |
| 20:34 |
|
octacian__ joined #minetest-dev |
| 20:38 |
|
blaze joined #minetest-dev |
| 20:59 |
|
octacian joined #minetest-dev |
| 21:00 |
|
numZero joined #minetest-dev |
| 21:10 |
|
octacian_ joined #minetest-dev |
| 21:54 |
|
proller joined #minetest-dev |
| 21:56 |
red-002 |
#5194 seems to be a bad idea |
| 21:56 |
ShadowBot |
https://github.com/minetest/minetest/issues/5194 -- per client code, anti grief method, and maybe more. |
| 21:58 |
rubenwardy |
it has been discussed before and rejected |
| 21:59 |
rubenwardy |
it would help against casual griefer |
| 21:59 |
rubenwardy |
*ers |
| 21:59 |
red-002 |
yeah I suppose |
| 21:59 |
rubenwardy |
but not hackers |
| 21:59 |
red-002 |
still how would a code like this be created |
| 22:00 |
red-002 |
it can't reveal any personal info so device name or user account name is out of the question |
| 22:01 |
red-002 |
plus those are not unique |
| 22:03 |
red-002 |
If mt was closed source it wouldn't be that bad of an idea |
| 22:04 |
red-002 |
The thing is that people will start to trust these id's to identify a user |
| 22:04 |
red-002 |
and that will get messy |
| 22:15 |
|
DI3HARD139 joined #minetest-dev |
| 22:23 |
|
DI3HARD139 left #minetest-dev |
| 22:24 |
|
DI3HARD139 joined #minetest-dev |
| 22:31 |
|
octacian joined #minetest-dev |
| 22:32 |
|
Foz joined #minetest-dev |
| 22:49 |
nore |
however, a possibility would be to ban by mac address |
| 22:49 |
sofar |
those can be changed trivially |
| 22:49 |
nore |
mac adress can be changed easily? |
| 22:50 |
sofar |
yup |
| 22:50 |
nore |
I thought it was hardware and impossible to change without changing the hardware |
| 22:50 |
* nore |
is confused |
| 22:50 |
sofar |
https://www.google.com/webhp?sourceid=chrome-instant&ion=1&espv=2&ie=UTF-8#q=change+mac+address |
| 22:50 |
VanessaE |
why not pipe lshw output through some encryption protocol and use the fingerprint from it? |
| 22:51 |
* sofar |
wrote linux network drivers at one point, fairly sure you can trust his advice :) |
| 22:51 |
VanessaE |
seems to me that's about as unique-but-non-personal as it can get |
| 22:51 |
sofar |
VanessaE: can be spoofed by running in a restricted environment of container |
| 22:51 |
VanessaE |
hm |
| 22:52 |
nore |
VanessaE: anything like that can be spoofed |
| 22:52 |
sofar |
there is a way around this |
| 22:52 |
nore |
you need something that *can't* be spoofed even with modifying the client |
| 22:52 |
sofar |
all you need to do is make accounts that can be verified by e.g. email |
| 22:53 |
sofar |
you can already do this with the http* api in minetest |
| 22:53 |
nore |
that was the idea between the mac address thing |
| 22:53 |
nore |
sofar: though, if you modify your mac address |
| 22:53 |
VanessaE |
sofar: redcrab had something like that. |
| 22:53 |
sofar |
just have the client log in *inside* the minetest client |
| 22:53 |
sofar |
and then verify *that* login |
| 22:53 |
nore |
do you still receive packets that are for the modified mac address? |
| 22:53 |
sofar |
nore: modifying the mac address is permanent until reboot, so everything works just fine |
| 22:54 |
nore |
okay |
| 22:54 |
nore |
I mean, I thought routers would not send you the packets, but I must be mistaken :) |
| 22:55 |
sofar |
unless you do strict mac filtering, it works |
| 22:55 |
sofar |
and nobody does that |
| 22:55 |
nore |
anyway, a per-device authentication would need to be something that can't be spoofed even if the client is modified |
| 22:55 |
nore |
so, that is, something you can't change without being unable to connect |
| 22:55 |
|
paramat joined #minetest-dev |
| 22:55 |
sofar |
hence, verifying an e-mail address works fine |
| 22:56 |
nore |
yeah |
| 22:56 |
sofar |
I know another great method |
| 22:56 |
sofar |
we should add 2fa |
| 22:56 |
sofar |
on top of that |
| 22:56 |
nore |
though if you have a subdomain, you have an infinite number of email addresses |
| 22:56 |
rubenwardy |
10 minute mail |
| 22:56 |
sofar |
banning entire subdomains is easy |
| 22:56 |
rubenwardy |
+ derives |
| 22:56 |
sofar |
especially if you regularly re-verify |
| 22:57 |
nore |
yeah, indeed |
| 22:57 |
sofar |
2fa will also deter many since it leaves a fingerprint |
| 22:57 |
nore |
though you must take care not to ban mail providers |
| 22:57 |
sofar |
but e.g. gmail makes sure people don't make thousands of addresses |
| 22:58 |
ShadowNinja |
celeron55: Re: Master-server; what do you mean by reusability? As for why it was rewritten: It used to be written in Perl -- it rather cryptic (as perl tends to be) and not many people seemed to be familiar enough with it to change it. |
| 22:58 |
nore |
2fa isn't easy; also many players do not have a second authentication factor |
| 22:58 |
* sofar |
raises hand, knows perl |
| 22:58 |
sofar |
duo mobile on android is fairly accessible |
| 22:59 |
nore |
yeah, but many do not have a smartphone, nor even a cell phone |
| 22:59 |
nore |
many of the players are children |
| 22:59 |
sofar |
or have e-mail |
| 22:59 |
ShadowNinja |
sofar: Well, here's what it looked like: https://github.com/minetest/minetest/blob/2336d21efd659d0f47abde26f906b5fbb26b1d65/util/master/master.cgi |
| 23:00 |
sofar |
myeah, that's your typical perl salad there |
| 23:00 |
sofar |
doesn't look horrible per se |
| 23:00 |
* nore |
doesn't know perl so finds this code rather cryptic |
| 23:00 |
rubenwardy |
seems to use punctuation as variables |
| 23:01 |
paramat |
#5199 |
| 23:01 |
ShadowBot |
https://github.com/minetest/minetest/issues/5199 -- Footsteps: Fix offset footstep sound bug by paramat |
| 23:01 |
sofar |
the new version seems to use whitespace as part of the language XD |
| 23:01 |
* sofar |
& python do not get along |
| 23:02 |
* nore |
actually likes python a lot |
| 23:02 |
red-002 |
pretty sure they added a way to fix that recently |
| 23:10 |
|
ZacToo joined #minetest-dev |
| 23:15 |
|
rom1504 joined #minetest-dev |
| 23:18 |
|
Darcidride joined #minetest-dev |
| 23:18 |
|
Calinou joined #minetest-dev |
| 23:22 |
|
Darcidride joined #minetest-dev |
| 23:25 |
|
garywhite joined #minetest-dev |
| 23:30 |
|
proller joined #minetest-dev |
| 23:32 |
|
Darcidride joined #minetest-dev |
| 23:38 |
|
Darcidride joined #minetest-dev |
| 23:50 |
|
DonBatman joined #minetest-dev |
