| Time |
Nick |
Message |
| 00:34 |
|
YuGiOhJCJ joined #luanti-dev |
| 02:53 |
|
YuGiOhJCJ joined #luanti-dev |
| 03:34 |
|
Panquesito7 joined #luanti-dev |
| 04:00 |
|
MTDiscord joined #luanti-dev |
| 05:23 |
|
Panquesito7_ joined #luanti-dev |
| 05:30 |
|
Panquesito7 joined #luanti-dev |
| 10:12 |
|
SpaceManiac joined #luanti-dev |
| 10:57 |
Krock |
sfan5: that milestone PR fixes an issue which is also in the milestone... would be nice to have it fixed though I do not use smooth scrolling, thus little interest. |
| 10:57 |
Krock |
about texture loading priorities: the first texture is overwritten by whatever is found next |
| 10:58 |
Krock |
it might be more efficient to search in opposite direction (return on first match). I see that the docs can be confusing. |
| 11:10 |
sfan5 |
quiz question: we have both of these functions, what's the difference between them? |
| 11:10 |
sfan5 |
std::vector<std::basic_string<T>> str_split(const std::basic_string<T> &str, T delimiter) |
| 11:10 |
sfan5 |
std::vector<std::basic_string<T>> split(const std::basic_string<T> &s, T delim) |
| 11:14 |
Krock |
does one trim, and the other doesn't? |
| 11:15 |
sfan5 |
wrong |
| 11:16 |
Krock |
ah. std::getline vs hand-rolled |
| 11:17 |
sfan5 |
not really |
| 11:17 |
sfan5 |
split() has an escape character, str_split() does not |
| 11:17 |
Krock |
time to rename split to split_unescape or so |
| 11:18 |
sfan5 |
% rg '\bsplit\(' src | wc -l |
| 11:18 |
sfan5 |
96 |
| 11:18 |
sfan5 |
no thanks |
| 11:19 |
Krock |
did a similar thing with the deSerializeLongString or so a while back because it was not obvious whether u16 or u32 was used |
| 11:19 |
Krock |
it's worth it in the long run |
| 11:21 |
sfan5 |
we also have no unittests for either of them, wonderful |
| 11:29 |
Krock |
any other changes you'd like to include before the RC build? I've got the network protocol version bump soon ready, where I'd then use its builds. |
| 11:30 |
sfan5 |
i'm currently working on some bug fixes but nothing essential |
| 11:37 |
Krock |
unittests passed. Will push https://pastebin.com/raw/JWN2SjJm in about 15 minutes and write the RC forum post and github release in the meantime |
| 12:03 |
Krock |
~tell SFENCE Would you please be so nice to provide a macOS build based on commit ce321b6 ? Thank you! |
| 12:03 |
ShadowBot |
Krock: OK. |
| 12:09 |
Krock |
publishing RC build in 15 minutes. any objections? |
| 12:13 |
sfan5 |
go ahead |
| 12:28 |
Krock |
Would someone please be so nice to update the pinned topics in https://forum.luanti.org/viewforum.php?f=18 ? |
| 12:33 |
rubenwardy |
done |
| 12:34 |
rubenwardy |
I'll start working on the blog posrt |
| 12:36 |
Krock |
thanks |
| 12:57 |
sfan5 |
man I don't like this filesys code |
| 12:58 |
sfan5 |
I hope nobody asks me to explain it during review |
| 12:58 |
Krock |
Hey Grok |
| 13:44 |
rubenwardy |
do we support rebinding buttons on gamepads? |
| 14:01 |
Sendein |
ye |
| 15:37 |
|
Panquesito7 joined #luanti-dev |
| 15:45 |
sfan5 |
well apparently https://github.com/microsoft/vcpkg/issues/53394 is breaking the msvc jobs |
| 15:47 |
sfan5 |
vcpkg does make an "we just glued some shit together" impression to me. and some people say it's the best c++ package manager... |
| 15:48 |
rubenwardy |
I am also not impressed with vcpkg |
| 15:49 |
rubenwardy |
annoying how async functions can't reference local funcs in the same file |
| 15:57 |
sfan5 |
>unescape_string |
| 15:57 |
sfan5 |
>it just blindly deletes all backslashes |
| 15:57 |
sfan5 |
??? |
| 15:57 |
Krock |
sfan5: it's probably the best c++ package manager for Windows |
| 15:59 |
sfan5 |
bonus: imagesource.cpp has a method of the same name that does it correctly |
| 16:02 |
sfan5 |
apparently "Conan 2" is an alternative |
| 16:03 |
sfan5 |
nevermind maybe unescape_string is not actually broken |
| 16:05 |
Krock |
in a better world we'd have unittests for all functions |
| 16:06 |
rubenwardy |
unit testing pure utils is like the bare minimum |
| 16:07 |
rubenwardy |
the test in minetest did not stand for unit test |
| 16:10 |
sfan5 |
it stands for "let's see if this works somehow" |
| 16:12 |
[MatrxMT] |
<Zughy> Wait, isn't Minetest for mining crypto? |
| 16:24 |
sfan5 |
the implementation in imagesource.cpp actually had some bugs |
| 16:24 |
sfan5 |
this happens if you don't unit test |
| 16:52 |
Krock |
rubenwardy: https://github.com/luanti-org/contentdb/issues/672#issuecomment-5282253514 2x the same git hash FYI |
| 17:55 |
rubenwardy |
fixed |
| 18:16 |
Krock |
another low-handing fruit: https://pastebin.com/raw/RCLx9e5B ... |
| 18:17 |
Krock |
except for the remark on "this". Copy&paste mistake. |
| 19:09 |
|
ivanbu joined #luanti-dev |
| 21:00 |
sfan5 |
btw if #17442 I can obviously pick only the non-refactor commits and push them to master |
| 21:00 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/17442 -- [no squash] Misc fixes and filesystem stuff by sfan5 |
| 21:00 |
sfan5 |
+is approved |
| 21:01 |
sfan5 |
it would also be good to run the unit tests on windows but alas the msvc build is broken |
| 22:34 |
|
panwolfram joined #luanti-dev |
| 23:44 |
|
MTDiscord joined #luanti-dev |
