| Time |
Nick |
Message |
| 00:16 |
|
proller joined #minetest-dev |
| 00:42 |
|
Alias2 joined #minetest-dev |
| 02:48 |
|
v-rob joined #minetest-dev |
| 03:02 |
|
behalebabo_ joined #minetest-dev |
| 03:18 |
|
lhofhansl joined #minetest-dev |
| 03:29 |
|
queria joined #minetest-dev |
| 03:33 |
|
queria joined #minetest-dev |
| 04:44 |
|
jordan4ibanez joined #minetest-dev |
| 05:00 |
|
MTDiscord joined #minetest-dev |
| 05:17 |
|
Pexin joined #minetest-dev |
| 06:06 |
|
v-rob joined #minetest-dev |
| 06:36 |
|
erlehmann joined #minetest-dev |
| 07:44 |
|
v-rob joined #minetest-dev |
| 07:57 |
|
Baytuch joined #minetest-dev |
| 07:57 |
Baytuch |
hello |
| 07:58 |
Baytuch |
guys, pls, fix cert of the servers.minetest.net |
| 07:58 |
Baytuch |
used expired cert DST Root CA X3 |
| 07:59 |
Baytuch |
2 s:/C=US/O=Internet Security Research Group/CN=ISRG Root X1 |
| 07:59 |
Baytuch |
i:/O=Digital Signature Trust Co./CN=DST Root CA X3 |
| 07:59 |
MTDiscord |
<Sublayer plank> your problem |
| 07:59 |
MTDiscord |
<Sublayer plank> update your root certificates |
| 08:01 |
Baytuch |
https://letsencrypt.org/docs/dst-root-ca-x3-expiration-september-2021/ |
| 08:01 |
|
Baytuch left #minetest-dev |
| 08:09 |
MTDiscord |
<MNH48> yes, just like what's said in the article, client need to update root certificate or update their browser or update openssl from their side (depending on what the client is actually using) |
| 08:12 |
|
jordan4ibanez joined #minetest-dev |
| 08:41 |
|
appguru joined #minetest-dev |
| 08:57 |
|
Fixer joined #minetest-dev |
| 09:41 |
|
erlehmann joined #minetest-dev |
| 09:53 |
|
calcul0n_ joined #minetest-dev |
| 10:53 |
|
erlehmann joined #minetest-dev |
| 11:51 |
MTDiscord |
<Warr1024> Oof, I haven't seen a site that has TLS 1.0 or 1.1 enabled in a while ':-) |
| 12:51 |
|
tekakutli joined #minetest-dev |
| 13:00 |
|
erlehmann joined #minetest-dev |
| 13:07 |
|
proller joined #minetest-dev |
| 14:37 |
|
HuguesRoss joined #minetest-dev |
| 14:40 |
|
HuguesRoss joined #minetest-dev |
| 14:46 |
|
HuguesRoss joined #minetest-dev |
| 14:48 |
|
HuguesRoss joined #minetest-dev |
| 14:49 |
|
HuguesRoss joined #minetest-dev |
| 15:00 |
|
Taoki joined #minetest-dev |
| 15:12 |
|
erlehmann joined #minetest-dev |
| 15:16 |
|
Toothless joined #minetest-dev |
| 16:10 |
|
appguru joined #minetest-dev |
| 16:16 |
|
Toothless joined #minetest-dev |
| 16:56 |
|
lhofhansl joined #minetest-dev |
| 17:21 |
|
v-rob joined #minetest-dev |
| 17:39 |
erlehmann |
i have a somewhat trivial thing, some test nodes for devtest https://github.com/minetest/minetest/pull/11978 |
| 17:39 |
erlehmann |
i have already used the test nodes to debug texture errors in minetest |
| 17:40 |
erlehmann |
wait. oops, the ones i used were slightly different. |
| 17:40 |
erlehmann |
if this gets accepted, i want to do the same for PNG and JPEG |
| 17:41 |
lhofhansl |
Added "One Approval", need a second one. |
| 17:41 |
erlehmann |
i am pretty sure that there is something wonky going on with transparency rendering in inventory vs world btw |
| 17:42 |
erlehmann |
because the nodes with alpha channel are black in world and transparent in inventory and hand |
| 17:42 |
erlehmann |
lhofhansl any idea why that could happen? |
| 17:42 |
erlehmann |
i can give a screenshot |
| 17:42 |
lhofhansl |
Not offhand, no :( |
| 17:43 |
erlehmann |
i think adding devtest nodes is very important because i have several times created textures that did not load |
| 17:43 |
erlehmann |
not only TGA, PNG too |
| 17:43 |
lhofhansl |
I agree. |
| 17:43 |
erlehmann |
the generated PNG nodes from hecks test only one specific mode of PNG (32bpp RGBA) |
| 17:43 |
rubenwardy |
You need to do use_texture_alpja |
| 17:43 |
erlehmann |
rubenwardy oh. then i will add that! |
| 17:43 |
erlehmann |
sorry |
| 17:44 |
erlehmann |
if anyone can tell me how to make gimp generate A1R5G5B5 images, i can add that |
| 17:44 |
erlehmann |
otherwise i'll do it later |
| 17:45 |
erlehmann |
this is without use_texture_alpha https://mister-muffin.de/p/S4rv.png |
| 17:45 |
erlehmann |
i will add it |
| 17:46 |
erlehmann |
rubenwardy use_texture_alpha has an effect in the world, but not in inventory cubes? |
| 17:46 |
erlehmann |
that seems like a bug to me |
| 17:47 |
MTDiscord |
<Benrob0329> It's probably missing code more than broken code :P |
| 17:47 |
erlehmann |
well i guess then the test nodes already helped lol |
| 17:50 |
erlehmann |
i wish i remember which PNG modes were broken |
| 17:55 |
erlehmann |
rubenwardy lhofhansl i updated the branch, thanks! seems i found an alpha bug already … maybe. the black-and-white texture is weirdly interpreted. https://mister-muffin.de/p/bVur.png |
| 17:57 |
erlehmann |
theory: what if minetest interprets 16bpp textures that are meant as 8 bit grayscale + 8 bit alpha channel … as A1R5G5B5 |
| 17:57 |
|
Thomas-S joined #minetest-dev |
| 17:57 |
|
Thomas-S joined #minetest-dev |
| 17:57 |
erlehmann |
but well |
| 17:57 |
erlehmann |
i'll debug that when the test nodes are accepted ig |
| 17:58 |
erlehmann |
bc then i am going to make more test nodes |
| 17:58 |
erlehmann |
for more stuff |
| 17:59 |
erlehmann |
rubenwardy do you see any other possible improvements? |
| 18:02 |
erlehmann |
i have added a picture to the PR |
| 18:38 |
MTDiscord |
<luatic> erlehmann: I have a bug report for this |
| 18:38 |
MTDiscord |
<luatic> It boils down to use_texture_alpha being applied to the node but not the wielditem |
| 18:39 |
erlehmann |
luatic is yes i figured. also the yellowish tint. |
| 18:39 |
MTDiscord |
<luatic> See #11714 |
| 18:39 |
ShadowBot |
https://github.com/minetest/minetest/issues/11714 -- use_texture_alpha = "opaque" does not apply to wielditem |
| 18:39 |
erlehmann |
thank you! |
| 18:39 |
erlehmann |
luatic do you have any existing issue for the yellowish tint? |
| 18:51 |
erlehmann |
lhofhansl, rubenwardy what drawtype would you like for the texture test nodes? i just realized i can look into the world wallhack-style because i forgot to set that. glasslike? |
| 18:52 |
MTDiscord |
<luatic> erlehmann: I don't, feel free to open one |
| 18:53 |
erlehmann |
luatic any drawtype advice though for the wallhack nodes? |
| 18:53 |
MTDiscord |
<luatic> Wallhack? |
| 18:54 |
|
appguru joined #minetest-dev |
| 18:54 |
erlehmann |
luatic if you look at the screenshot here you can see that it is possible to look behind the world https://github.com/minetest/minetest/pull/11978 |
| 18:54 |
erlehmann |
the transparency |
| 18:55 |
erlehmann |
i know possible workarounds (set the drawtype), but i do not know what would be best for devtest |
| 18:55 |
|
kilbith joined #minetest-dev |
| 18:57 |
kilbith |
why is CBD always switching back to my OS language |
| 18:57 |
kilbith |
there must be no cookies there |
| 18:57 |
kilbith |
* CDB |
| 18:57 |
MTDiscord |
<luatic> kilbith: yep, it's annoying |
| 18:57 |
|
queria joined #minetest-dev |
| 19:00 |
erlehmann |
i just noticed that it seems that the ordering of items in inventory seems to have changed between 5.4.1 and 5.5-dev. any pointers to why? |
| 19:00 |
erlehmann |
basically, i added my test nodes to a branch based on 5.4.1 and they were in another order in devtest chest of everything |
| 19:01 |
MTDiscord |
<luatic> erlehmann: The ordering is most likely done by the Lua side of things |
| 19:02 |
MTDiscord |
<luatic> Let me have a look at what it uses, but my first guess would be that if it uses something like pairs which is hash-map based instead of sorting a table, this can very much happen |
| 19:02 |
erlehmann |
oh okay |
| 19:02 |
erlehmann |
it is just a bit weird |
| 19:02 |
erlehmann |
because they are close together |
| 19:06 |
MTDiscord |
<luatic> Hmm, that'd be odd |
| 19:07 |
erlehmann |
like all the 8 nodes i added are next to each other |
| 19:08 |
erlehmann |
but in 5.5-dev they are ordered 1,2,3,10 |
| 19:09 |
erlehmann |
no |
| 19:09 |
erlehmann |
forget about it |
| 19:10 |
erlehmann |
apparently i changed some labels or something |
| 19:10 |
erlehmann |
they are always ordered 10,1,2,3 |
| 19:10 |
erlehmann |
luatic sorry! |
| 19:10 |
erlehmann |
i just tested it again to be sure |
| 19:10 |
erlehmann |
no bug here |
| 19:11 |
kilbith |
someone needs to put a duct tape on someone else' mouth |
| 19:11 |
kilbith |
channel has become way too noisy |
| 19:11 |
|
proller joined #minetest-dev |
| 19:11 |
kilbith |
call me rude but I'm sorry but I had to say it |
| 19:16 |
MTDiscord |
<luatic> erlehmann: it looks like nodes are indeed sorted, but the comparator is most likely bonkers |
| 19:16 |
MTDiscord |
<luatic> (not just nodes, all items in the devtest CoE) |
| 19:31 |
|
kilbith_ joined #minetest-dev |
| 19:32 |
|
appguru1 joined #minetest-dev |
| 19:33 |
|
proller joined #minetest-dev |
| 19:36 |
|
erlehmann joined #minetest-dev |
| 19:40 |
|
Toothless joined #minetest-dev |
| 19:51 |
appguru |
NVM the comparator isn't bonkers, just messy code |
| 19:51 |
appguru |
(IMO) |
| 20:15 |
erlehmann |
i updated #11978 to include 16bpp textures with “use_texture_alpha = "clip"” and it seems to me as if minetest just ignores the transparency bit in 16bpp textures |
| 20:15 |
ShadowBot |
https://github.com/minetest/minetest/issues/11978 -- Add TGA test nodes to devtest by erlehmann |
| 20:16 |
erlehmann |
as i said, cursed PNGs are next hehe |
| 20:21 |
rubenwardy |
Do cookies default to expiring? |
| 20:21 |
rubenwardy |
https://i.rubenwardy.com/cnZX1.png |
| 20:21 |
rubenwardy |
is the CDB locale cookie |
| 20:22 |
erlehmann |
why even |
| 20:23 |
erlehmann |
why have a cookie for locale? isn't there a header for that |
| 20:23 |
rubenwardy |
aha, it defaults to the session |
| 20:23 |
rubenwardy |
erlehmann: people want to change their languae |
| 20:23 |
erlehmann |
content negotiation this that |
| 20:23 |
rubenwardy |
I already support that header |
| 20:23 |
rubenwardy |
but I also have a dropdown allowing users to override that header |
| 20:23 |
erlehmann |
for the case that someone has their computer set to a language they do not want in minetest i guess? |
| 20:24 |
rubenwardy |
ContentDB is a website |
| 20:24 |
erlehmann |
yeah but what circumstances made the cookie necessary |
| 20:24 |
erlehmann |
is it common that people access it from computers where they can not change the locale? |
| 20:26 |
rubenwardy |
only the interface is translated, packages are still in English |
| 20:26 |
rubenwardy |
so they may wish to keep the interface in English |
| 20:26 |
erlehmann |
ah! |
| 20:27 |
rubenwardy |
luatic: fixed |
| 21:05 |
|
v-rob joined #minetest-dev |
| 21:11 |
|
Toothless joined #minetest-dev |
| 21:21 |
|
tekakutli joined #minetest-dev |
| 21:23 |
sfan5 |
rubenwardy: cookies are for the session if you don't specify expiry |
| 21:40 |
sfan5 |
appguru: I updated my MTG PR with docs |
| 21:45 |
appguru |
sfan5: Approved. |
| 22:13 |
|
v-rob joined #minetest-dev |
| 22:19 |
|
Toothless joined #minetest-dev |
| 22:43 |
|
Noisytoot joined #minetest-dev |
| 22:52 |
|
tekakutli joined #minetest-dev |
| 23:11 |
|
Toothless joined #minetest-dev |
| 23:36 |
MTDiscord |
<exe_virus> I'm starting to flesh out a PR for adding gltf (glb might be easier for our paradigm, honestly), and want to use the header-only tinygltf library. It's MIT licensed and uses jsonhpp. It's basically out of maintenance now, and so I'm wondering if I could bring it directly into our source codebase in the lib folder for the PR. And we add a credit to satisfy MIT. Any complaints with that plan? |
| 23:40 |
erlehmann |
why oh why not use a submodule |
