| Time |
Nick |
Message |
| 00:10 |
|
Alias joined #minetest-dev |
| 00:10 |
|
erle joined #minetest-dev |
| 01:06 |
|
pgimeno_ joined #minetest-dev |
| 01:49 |
|
AliasAlreadyTake joined #minetest-dev |
| 04:00 |
|
MTDiscord joined #minetest-dev |
| 06:31 |
|
calcul0n joined #minetest-dev |
| 07:20 |
|
srifqi joined #minetest-dev |
| 07:24 |
srifqi |
thank you, sfan5 |
| 07:25 |
srifqi |
i don't have the permission on Weblate |
| 08:20 |
srifqi |
oh, if let's say there are errors on the English string/text that are just found, how will it affect the translation workflow? |
| 08:30 |
|
Noisytoot joined #minetest-dev |
| 08:49 |
|
calcul0n joined #minetest-dev |
| 09:07 |
sfan5 |
translators will translate the mistakes, if we fix it on next update then the translations have to be re-done |
| 09:12 |
srifqi |
so, is it possible that multiple translation update happening in a short time? |
| 09:14 |
sfan5 |
we can do them whenever we want |
| 09:14 |
sfan5 |
the only downside is more clutter in the commit history |
| 09:15 |
srifqi |
i see. i asked it because i think i found some error (mistype) in the English strings while translating |
| 09:30 |
MTDiscord |
<rollerozxa> it might not be necessary to have to translate some strings twice with different capitalisation |
| 09:30 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/747163566800633906/1165220607454089286/image.png?ex=65460f41&is=65339a41&hm=3cbea9a968556e115ac18b60d2c1c83445ff704f712729f38ff47977bae9faef& |
| 09:30 |
MTDiscord |
<rollerozxa> (K is capitalised in the pause menu but not in the main menu settings dialog) |
| 09:32 |
sfan5 |
that's indeed not very useful |
| 09:45 |
|
appguru joined #minetest-dev |
| 09:48 |
srifqi |
oh, i just noticed that Javanese was added (with no translated strings) |
| 09:51 |
MTDiscord |
<rollerozxa> also looks like generate_from_settingtypes doesn't exclude the temporary settings category so there's a bunch of setting names and descriptions that don't show up anywhere in the translations |
| 10:12 |
erle |
what's the oldest ubuntu/debian that minetest is targeting? |
| 10:14 |
ROllerozxa |
assumedly anything that contains GCC 7.5+ |
| 10:15 |
ROllerozxa |
workflows run on ubuntu 20.04 as the oldest, with some version of GCC 7 |
| 10:15 |
erle |
ROllerozxa then consider the following https://github.com/minetest/minetest/issues/13899#issuecomment-1773729151 |
| 10:15 |
erle |
i doubt ubuntu 20.04 will have a new enough libsdl2-dev |
| 10:15 |
erle |
because my debian does not have it |
| 10:15 |
ROllerozxa |
well, hopefully you will have updated by the time SDL2 becomes default then |
| 10:17 |
erle |
ROllerozxa should i open an issue about the build process not verifying the version of SDL before starting the build? |
| 10:19 |
ROllerozxa |
sounds reasonable. SDL2 also has preprocessor constants to check what version is being compiled with, so things requiring newer version could be disabled or circumvented during compilation if that's possible |
| 10:25 |
ROllerozxa |
I just checked my debian 11 (oldstable?) build slave VM and it has SDL2 2.0.14 as the newest, so I suppose that should be the minimum and anything requiring newer versions should be gated by preprocessor directives |
| 10:30 |
erle |
yes |
| 10:30 |
erle |
that's the issue here |
| 10:30 |
erle |
i also have debian 11 |
| 10:30 |
erle |
ROllerozxa you open an issue? |
| 10:31 |
erle |
i have to do a thing |
| 10:37 |
srifqi |
since we have the new settings, shouldn't the phrase, "Set to true to ...," be changed? |
| 10:40 |
srifqi |
btw, how should we fix these problems? should we postpone the fixes for 5.8.1 to have more time for checking all English strings/texts? |
| 10:51 |
erle |
the build process of minetest is still so bad that it wants to rebuild everything when switching between branches :( |
| 10:51 |
erle |
how do you solve this besides getting a faster computer? |
| 10:52 |
erle |
(ccache has false positives) |
| 10:55 |
[MTMatrix] |
<localhost> eh... this is future... 100500GB ram, 10Tgz cpu for start small things... |
| 10:55 |
ROllerozxa |
(solving it without getting a faster computer? well, borrowing a faster computer I suppose...) |
| 10:56 |
erle |
well the last time i proposed to address the problems of the build problems the only ppl interested were … authors of build systems |
| 10:56 |
erle |
the problems of the build process i mean |
| 10:56 |
erle |
hehe |
| 10:56 |
erle |
the problems have their own problems |
| 10:56 |
erle |
recently at work i wanted to talk about build process issues (i have given a talk about it on CCC camp) and how people solve them and it was the only workshop that day that ZERO people signed up to. |
| 10:57 |
erle |
turned out that a) almost everyone is aware of them b) people find the topic *extremely* annoying and rather get a faster computer than think about it |
| 10:57 |
erle |
(faster computer does not help with miscompiles though) |
| 10:58 |
[MTMatrix] |
<localhost> (imagine compile chrome...) |
| 10:58 |
erle |
the thing is, the use case “switch branch and recompile” is so common |
| 10:58 |
erle |
anyway, if i get no input on how others do this, i'll drop the topic again |
| 10:59 |
erle |
i'm only trying to see what happens if i set this SDL mouse wheel constant myself right now |
| 10:59 |
erle |
and if it has to be 1 or 0 |
| 10:59 |
erle |
cargo culting at its best |
| 11:21 |
erle |
it is *still* compiling and i changed a *single* line |
| 11:22 |
erle |
i should really work on the dofiles again |
| 11:22 |
erle |
and provide them externally ig |
| 11:25 |
erle |
srifqi about the translations, should i file bugs about untranslated strings or not? |
| 11:25 |
erle |
i found some |
| 11:25 |
erle |
and if so, where? |
| 11:25 |
rubenwardy |
if it's in the settings menu then that's a known issue, it's a checkbox in the mega issue |
| 11:25 |
erle |
yeah it is |
| 11:26 |
sfan5 |
srifqi: IMO we can fix them just as we have time, no need to delay anything |
| 11:26 |
erle |
i guess this means “no don't file an issue” then |
| 11:26 |
rubenwardy |
correct yeah |
| 11:26 |
rubenwardy |
providing it's the text of a setting - like a dropdown value |
| 11:27 |
rubenwardy |
https://github.com/minetest/minetest/issues/13476 |
| 11:27 |
erle |
i'll have to check that |
| 11:27 |
erle |
it was not the text of a setting though, let me verify after the compile |
| 11:27 |
erle |
since i am trying out the SDL build as grorp suggested, what's the reason that it is not default besides requiring a fairly new SDL2? |
| 11:27 |
rubenwardy |
there's still bugs |
| 11:27 |
rubenwardy |
it will be default soon :tm: |
| 11:29 |
erle |
since the plus input bug actually is part of a larger bug that ISO_Level3_Shift can be bound (and it should not, since some layouts need it to input ”normal” characters), could someone tell me where to look for the code that defines which keys are bindable and which are modifier keys (so the character is bindable) or are *only* keys bindable now? |
| 11:30 |
erle |
because if only keys are bindable, i get that there are issues: after all, shift is used for sneaking, so it has to be bindable. but i am sure neo2 is not the only layout where you get problems if you can not input or bind specific key combinations. |
| 11:31 |
erle |
i must say i did not think about this at all: after all, it *does* make sense to treat “+” and ”keypad +” separately … right up until you encounter a layout where + is a key combination thing |
| 11:36 |
erle |
aaaaand i may have found a miscompile again :( |
| 11:56 |
erle |
i found the funniest mistranslation: binding the right alt key in German just says “Linke Taste” in the interface. |
| 11:56 |
erle |
(it means “left key”) |
| 12:17 |
erle |
i made an issue for that #13904 |
| 12:17 |
ShadowBot |
https://github.com/minetest/minetest/issues/13904 -- Keys are either mis-detected or mis-labeled in German localization of keybinding dialog |
| 12:28 |
[MTMatrix] |
<Zughy> Thanks erle for your umpteenth unrequested stream of consciousness |
| 12:28 |
erle |
ah, sarcasm, that's creative |
| 12:29 |
erle |
zughy i'll take the hint and put my thoughts into issues instead |
| 12:29 |
erle |
(until i forget again) |
| 12:40 |
erle |
anyone knows how to reach v-rob best? (i remember we used to chat about the keyboard thing) |
| 14:00 |
srifqi |
for the English texts, here is what I found using a spell checker: #13907 |
| 14:00 |
ShadowBot |
https://github.com/minetest/minetest/issues/13907 -- Text issues in the Minetest's English texts/strings |
| 14:56 |
|
appguru joined #minetest-dev |
| 15:50 |
|
imi joined #minetest-dev |
| 18:48 |
|
srifqi joined #minetest-dev |
| 18:55 |
srifqi |
hello. merging #13883 in 5 minutes |
| 18:55 |
ShadowBot |
https://github.com/minetest/minetest/issues/13883 -- Change some keys to be triggered once every key press by srifqi |
| 19:00 |
srifqi |
done |
| 22:09 |
|
diceLibrarian joined #minetest-dev |
| 22:33 |
|
panwolfram joined #minetest-dev |
