Time |
Nick |
Message |
00:01 |
|
proller joined #minetest |
00:04 |
|
Extex joined #minetest |
00:07 |
|
TheAltHacker joined #minetest |
00:54 |
MTDiscord |
<Jonathon> https://forum.minetest.net/viewtopic.php?f=14&t=27980&p=409042#p409042 oh look who it is |
01:15 |
|
specing_ joined #minetest |
01:23 |
|
RhineDevil joined #minetest |
01:34 |
|
RhineDevil joined #minetest |
02:13 |
|
Yad joined #minetest |
02:28 |
erle |
here is my minetest-servers tool https://forum.minetest.net/viewtopic.php?f=14&t=27984&p=409054 |
03:00 |
|
mrgreymatter joined #minetest |
03:20 |
|
crabycowman123 joined #minetest |
03:50 |
|
Extex joined #minetest |
03:52 |
|
Extex joined #minetest |
03:53 |
|
Extex joined #minetest |
04:00 |
|
MTDiscord joined #minetest |
04:13 |
|
toulene joined #minetest |
04:50 |
|
Lesha_Vel joined #minetest |
05:01 |
|
riff-IRC joined #minetest |
06:01 |
paradust |
\window 4 |
06:02 |
paradust |
oops, mistype |
06:02 |
|
calcul0n joined #minetest |
06:25 |
|
debiankaios joined #minetest |
06:40 |
|
lemonzest joined #minetest |
06:57 |
|
definitelya joined #minetest |
07:39 |
|
wing joined #minetest |
08:23 |
|
sobkas joined #minetest |
09:05 |
|
Fixer joined #minetest |
09:28 |
|
est31 joined #minetest |
09:58 |
|
hn joined #minetest |
09:58 |
hn |
hhhhi... |
09:58 |
hn |
hope here wont get a ban from age you think is incorrect.... |
09:59 |
hn |
(but was correct, i just know discord situations... not my first tbh...) |
10:00 |
sfan5 |
? |
10:00 |
hn |
nothing nvm |
10:01 |
hn |
just moved out of discord |
10:03 |
ROllerozxa |
I don't think liberachat has a minimum age requirement like discord does |
10:03 |
hn |
yea... |
10:04 |
hn |
just yesterday peeps thought im undreraged sp wanted to leave to avoid drama |
10:04 |
hn |
even if im aged |
10:04 |
hn |
(enough) |
10:05 |
hn |
(and if you think i'll come back - you have 0 chanes for it) |
10:08 |
hn |
anyways - something was happening at night? |
10:15 |
hn |
any hackathons or smth in mt? |
10:16 |
|
grouinos joined #minetest |
10:17 |
erle |
hn there was a game jam last christmas |
10:18 |
erle |
i mean it was timed so the games got released then |
10:18 |
hn |
k |
10:18 |
hn |
fine |
10:19 |
hn |
just was asdkin |
10:21 |
erle |
hn, you can find the games on contentdb. one of the best is “alter”. |
10:21 |
hn |
ikr? just wanted to fill empty channel |
10:21 |
hn |
lol |
10:25 |
erle |
hn i have a friend in rome. do you know his name? |
10:25 |
hn |
no? |
10:26 |
erle |
sussus |
10:26 |
hn |
lol |
10:27 |
hn |
(yes this actually was funni) |
10:27 |
erle |
his name is sussus amogus |
10:27 |
hn |
xD |
10:27 |
definitelya |
h |
10:27 |
hn |
? |
10:28 |
erle |
h ist one fourth haha |
10:28 |
hn |
idk |
10:29 |
definitelya |
hn: h |
10:29 |
hn |
o ok |
10:29 |
hn |
nicks you meant? |
10:29 |
ROllerozxa |
I don't get the joke and I'm too afraid to ask what "sussus amogus" has to do with rome |
10:30 |
hn |
it was ne guys in politics |
10:30 |
hn |
R0llerozxa ↑↑↑ |
10:32 |
erle |
ROllerozxa there is an old monty python sketch i believe you should watch to understand it: https://yewtu.be/watch?v=yzgS61zgPEg |
10:32 |
definitelya |
hn: Are you saying Sussus Amogus was an infuential politician of Ancient Rome? The more you know! |
10:33 |
hn |
ye |
10:33 |
hn |
lol |
10:34 |
erle |
definitelya watch the youtube video |
10:34 |
hn |
this one : https://en.wikipedia.org/wiki/Valentinian_II |
10:34 |
hn |
googled. first result |
10:35 |
definitelya |
Yes I've seen it, thank. |
10:35 |
hn |
;0 |
10:35 |
hn |
;) |
10:35 |
definitelya |
Those guys were hilarious. |
10:35 |
hn |
ye lol |
10:36 |
hacknorris |
fine cuz forgor |
10:36 |
erle |
my friend sussus, he has a wife you know |
10:36 |
erle |
they call her |
10:36 |
erle |
impostoria |
10:36 |
hacknorris |
lmao |
10:37 |
ROllerozxa |
ah I get it now lmao |
10:37 |
hacknorris |
ye xd |
10:48 |
|
Flabb joined #minetest |
11:47 |
|
lumidify joined #minetest |
11:54 |
|
grouinos joined #minetest |
12:10 |
|
erle joined #minetest |
13:07 |
|
LibreLikeGNU joined #minetest |
13:15 |
|
specing_ joined #minetest |
13:23 |
|
RhineDevil^ joined #minetest |
13:30 |
|
grouinos joined #minetest |
13:46 |
hacknorris |
anyways - someone already made parkour other than "inside the box" and isnt me? just asking... |
14:01 |
hacknorris |
dead chat |
14:19 |
erle |
hacknorris yes |
14:19 |
erle |
go to minigame server |
14:19 |
erle |
there is a parkour section |
14:38 |
hacknorris |
? |
14:38 |
hacknorris |
ah minigame. probly smol. there are no big ones? |
14:50 |
|
RhineDevil joined #minetest |
14:57 |
hacknorris |
dead chat |
14:58 |
|
jason101 joined #minetest |
14:59 |
jason101 |
i'm fascinated how i rename the wor |
15:01 |
hacknorris |
? |
15:10 |
|
grouinos joined #minetest |
15:16 |
|
hn joined #minetest |
15:19 |
hn |
so is there any bigger parkuor server in minetest except of "inside the box" ? |
15:32 |
|
Talkless joined #minetest |
15:50 |
|
Gustavo6046 joined #minetest |
15:53 |
|
Gustavo6046 joined #minetest |
16:01 |
|
Verticen joined #minetest |
16:02 |
|
Wuzzy joined #minetest |
16:06 |
Wuzzy |
I am slightly confused by the register_decoration noise handling |
16:07 |
Wuzzy |
so the docs say that decorations are placed per "square division" but what does that mean exactly? |
16:07 |
Wuzzy |
and at which point are decorations placed, anyway? whenever the Perlin noise value >= 0? |
16:07 |
hacknorris |
x^2/y ? or smth... maths |
16:07 |
hacknorris |
idk |
16:08 |
hacknorris |
ah, decorations? |
16:08 |
hacknorris |
that their on one square at surface and making X on it probly... |
16:08 |
hacknorris |
or smth |
16:19 |
|
kamdard joined #minetest |
16:36 |
MTDiscord |
<AFCM> excuse me but which it the AuthMechanism used with modern servers? |
16:38 |
MTDiscord |
<SX> SRP? |
16:38 |
erle |
SRP, why? |
16:38 |
erle |
AFCM are you making a non-minetest client? |
16:39 |
MTDiscord |
<AFCM> I mean |
16:39 |
MTDiscord |
<AFCM> switch (chosen_auth_mechanism) { case AUTH_MECHANISM_FIRST_SRP: { // send srp verifier to server std::string verifier; std::string salt; generate_srp_verifier_and_salt(getPlayerName(), m_password, &verifier, &salt); NetworkPacket resp_pkt(TOSERVER_FIRST_SRP, 0); resp_pkt << salt << verifier << (u8)((m_password.empty()) ? 1 : 0); |
16:39 |
MTDiscord |
Send(&resp_pkt); break; } case AUTH_MECHANISM_SRP: case AUTH_MECHANISM_LEGACY_PASSWORD: { u8 based_on = 1; if (chosen_auth_mechanism == AUTH_MECHANISM_LEGACY_PASSWORD) { m_password = translate_password(getPlayerName(), m_password); based_on = 0; } std::string playername_u = lowercase(getPlayerName()); m_auth_data = |
16:39 |
MTDiscord |
srp_user_new(SRP_SHA256, SRP_NG_2048, getPlayerName().c_str(), playername_u.c_str(), (const unsigned char *) m_password.c_str(), m_password.length(), NULL, NULL); char *bytes_A = 0; size_t len_A = 0; SRP_Result res = srp_user_start_authentication( (struct SRPUser *) m_auth_data, NULL, NULL, 0, (unsigned char **) &bytes_A, &len_A); |
16:39 |
MTDiscord |
FATAL_ERROR_IF(res != SRP_OK, "Creating local SRP user failed."); NetworkPacket resp_pkt(TOSERVER_SRP_BYTES_A, 0); resp_pkt << std::string(bytes_A, len_A) << based_on; Send(&resp_pkt); break; } case AUTH_MECHANISM_NONE: break; // not handled in this method } |
16:39 |
erle |
stop doing this |
16:40 |
MTDiscord |
<AFCM> ? |
16:40 |
erle |
pasting lines of code |
16:40 |
erle |
use a pastebin like https://mister-muffin.de/paste |
16:40 |
MTDiscord |
<SX> Code blocks does not look good on IRC side... |
16:40 |
MTDiscord |
<AFCM> ok |
16:41 |
erle |
if the user does not exist, you get AUTH_MECHANISM_FIRST_SRP, otherwise AUTH_MECHANISM_SRP, correct? |
16:42 |
MTDiscord |
<AFCM> but why is AUTH_MECHANISM_SRP just after AUTH_MECHANISM_LEGACY_PASSWORD |
16:43 |
MTDiscord |
<AFCM> Are AUTH_MECHANISM_SRP and AUTH_MECHANISM_LEGACY_PASSWORD handled the same? |
16:43 |
MTDiscord |
<AFCM> (I don't know much about cpp) |
16:43 |
erle |
AFCM try examining the packets on the wire using the wireshark dissector |
16:43 |
erle |
you will see how the SRP works |
16:44 |
erle |
again, what are you trying to do? |
16:44 |
MTDiscord |
<SX> I've not looked at that bu I would guess if client used legacy password auth then server accepts that |
16:44 |
MTDiscord |
<AFCM> web authentification service |
16:45 |
MTDiscord |
<SX> Just auth directly against DB, if you want SRP for web pages then browser javascript implementation for SRP is needed... |
16:45 |
erle |
AFCM explain? |
16:45 |
erle |
i don't get it, you want a web service to log in? |
16:46 |
erle |
or to register? |
16:46 |
MTDiscord |
<SX> Otherwise you'll be doing "crippled SRP" like what I suppose "legacy password auth" is doing in engine code |
16:46 |
MTDiscord |
<SX> (password sent to server and server does "SRP", not real SRP anymore but same thing basically) |
16:47 |
|
grouinos joined #minetest |
16:48 |
MTDiscord |
<AFCM> you have a website allowing users to create an account then they can connect to minetest servers with their credentials and the minetest server will verify them using the web API |
16:48 |
MTDiscord |
<SX> I mean functions same but loses security features, client password is exposed |
16:49 |
MTDiscord |
<SX> You want your own auth backend service which directly authenticates user without minetest engine... |
16:49 |
MTDiscord |
<SX> I guess there's at least few ready made pieces around git... somewhere... |
16:50 |
|
lsab joined #minetest |
16:51 |
lsab |
hello. I'm using i3 inventory, as I'm adm with creative privs I can't see recipes, is there some way to see that? |
16:53 |
MTDiscord |
<AFCM> I would use both bcrypt (for the website/API) and SRP (for the API endpoint allowing minetest to compare it with the encrypted password given by clients) |
16:54 |
|
Hawk777 joined #minetest |
16:55 |
MTDiscord |
<SX> Why you would use both if you're planning to use SRP anyway? |
16:56 |
erle |
split-brain account db? |
16:56 |
MTDiscord |
<SX> (any hash based auth takes away benefits of SRP) |
16:58 |
MTDiscord |
<SX> Well there's Buckaroo's https://hub.docker.com/r/minetestauth/minetest-auth-proxy / https://github.com/minetest-auth-proxy which does password auth using engine (through mod) |
17:01 |
MTDiscord |
<SX> But if building custom solution from scratch I would recommend going through direct db connection instead of using mod as a proxy, mod makes things a lot easier / faster to get working though. |
17:01 |
MTDiscord |
<SX> (easier because you can completely skip actual authentication implementation) |
17:10 |
|
est31 joined #minetest |
17:10 |
erle |
AFCM did you see this? might be of interest https://forum.minetest.net/viewtopic.php?f=14&p=409075#p409075 |
17:12 |
|
Verticen joined #minetest |
17:14 |
MTDiscord |
<AFCM> I basically want to encode a string the same way its encoded then its passed to the minetest auth API https://github.com/minetest/minetest/blob/2d8eac4e0a609acf7a26e59141e6c684fdb546d0/doc/lua_api.txt#L8937 |
17:16 |
|
est31 joined #minetest |
17:26 |
chaverinho |
. |
17:28 |
|
v-rob[m] joined #minetest |
18:26 |
MTDiscord |
<SX> You can't. Well, you can but it is auth implementation detail that depends on other things. Basically here's "default" encoding: https://github.com/minetest/minetest/blob/2d8eac4e0a609acf7a26e59141e6c684fdb546d0/src/util/auth.cpp#L29-L47 |
18:33 |
sfan5 |
you can take a plain text password, generate the SRP data and use this to reset an users password inside MT |
18:33 |
MTDiscord |
<SX> What password actually is depends on how it was created / updated, that's SRP detail. It is either SRP derived key (should be this in most cases) or actual password hash (for legacy) |
18:34 |
sfan5 |
also the code you linked is the legacy password |
18:37 |
sfan5 |
<AFCM> you have a website allowing users to create an account then they can connect to minetest servers with their credentials and the minetest server will verify them using the web API |
18:37 |
sfan5 |
^ if you want to do this I suggest writing directly to the database |
18:46 |
|
Desour joined #minetest |
19:04 |
MTDiscord |
<AFCM> what does "write to the database" exactly mean? |
19:08 |
sfan5 |
open auth.sqlite and write to the `auth` table, changing the `password` column |
19:08 |
MTDiscord |
<SX> Basically what said earlier: if building custom solution from scratch I would recommend going through direct db connection instead of using mod as a proxy That is, you connect to minetest user/auth db and do whatever is needed like authenticate user or create users |
19:10 |
MTDiscord |
<AFCM> but isn't sqlite only allowing a single client at a time? |
19:12 |
sfan5 |
only one client can write at a time |
19:12 |
MTDiscord |
<SX> btw will engine lock it, no? |
19:13 |
sfan5 |
sqlite will lock the file during writes |
19:16 |
MTDiscord |
<SX> well, some things take exclusive lock for software lifetime and prevents anyone else doing anything but that's legacy thing. Just I would not be surprised if there's some legacy involved :p |
19:26 |
erle |
sqlite is one writer many readers |
19:26 |
MTDiscord |
<SX> Ofc sqlite locking is not really that fool proof (because that's not possible with way how sqlite works) so better to have backups anyway |
19:28 |
erle |
last time a looked into sqlite code i was irritated to no end though |
19:28 |
MTDiscord |
<SX> (it could be better to go with postgres anyway for multiple concurrent db clients) |
19:29 |
erle |
[insert a rant about sqlite 1 based array indexing and UB here] |
19:30 |
erle |
does sqlite still try to use relational operators to compare pointers? :D |
19:54 |
|
mrgreymatter joined #minetest |
19:58 |
|
debiankaios joined #minetest |
20:27 |
fluxionary |
is there any weird logic for connected textures? trying to register some walls via `walls:register` in minetest_game, and get them to connect to the "regular" version of the node. e.g. a wall version of "default:brick". even after adding "default:brick" to the wall's connects_to list, and adding the "wall" group to "default:brick"'s definition, they still do not connect. |
20:30 |
fluxionary |
there's examples of something similar working in darkage, but darkage seems to have its own "register_wall" logic that doesn't rely on minetest_game |
20:33 |
fluxionary |
hm. seems to work if i use `override_item` instead of just mangling the connects_to list directly. *shrug*. |
20:39 |
Krock |
yes, you must push the definitions to Minetest again |
20:39 |
Krock |
i.e. minetest-register_item or override_item |
20:40 |
Krock |
minetest.registered_nodes modifications are Lua-side only |
20:44 |
|
Talkless joined #minetest |
21:06 |
|
FreeFull joined #minetest |
21:10 |
|
ronoaldo joined #minetest |
22:00 |
|
Sven_vB joined #minetest |
22:20 |
|
Verticen joined #minetest |
22:31 |
|
Gustavo6046 joined #minetest |
22:33 |
|
panwolfram joined #minetest |
22:33 |
|
Alias joined #minetest |
22:56 |
|
proller joined #minetest |
23:16 |
|
Wuzzy2 joined #minetest |
23:22 |
|
AliasAlreadyTake joined #minetest |
23:34 |
|
Lesha_Vel joined #minetest |