| Time |
Nick |
Message |
| 00:03 |
|
DuCake joined #minetest-dev |
| 00:30 |
garywhite |
Are any devs on rn? |
| 00:31 |
garywhite |
Cause I may have found the issue for #4528 |
| 00:31 |
ShadowBot |
https://github.com/minetest/minetest/issues/4528 -- minetest 0.4.14 64bit crashes at startup on Windows 10 Pro 64bit: debug.txt permissions/path problem |
| 00:38 |
|
emunand joined #minetest-dev |
| 00:47 |
garywhite |
Please look at #4528 I think I found the issue |
| 00:47 |
ShadowBot |
https://github.com/minetest/minetest/issues/4528 -- minetest 0.4.14 64bit crashes at startup on Windows 10 Pro 64bit: debug.txt permissions/path problem |
| 01:00 |
est31 |
garywhite: awesome, wait until Zeno is around again :) |
| 01:00 |
garywhite |
Is he in charge of that issue? |
| 01:00 |
|
emunand joined #minetest-dev |
| 01:00 |
est31 |
sort of yes |
| 01:01 |
garywhite |
ok |
| 01:05 |
|
proller joined #minetest-dev |
| 01:22 |
|
kaeza joined #minetest-dev |
| 01:40 |
|
lordfingle joined #minetest-dev |
| 01:47 |
|
Player_2 joined #minetest-dev |
| 02:31 |
|
Void7 joined #minetest-dev |
| 02:57 |
|
paramat joined #minetest-dev |
| 03:00 |
paramat |
so the 2 mod security PRs can be merged now? will do so in a few hours if no objections |
| 03:01 |
paramat |
or someone else feel free to |
| 03:04 |
est31 |
#4849 didnt get two approvals |
| 03:04 |
ShadowBot |
https://github.com/minetest/minetest/issues/4849 -- Mod security: Allow read-only access to all mod paths by ShadowNinja |
| 03:04 |
paramat |
but the author's counts as one |
| 03:06 |
ShadowNinja |
est approved, I approve, and sfan5 seems generally positive although he didn't actually approve it. |
| 03:14 |
est31 |
ok |
| 03:14 |
est31 |
paramat: still, the author has to express that approval |
| 03:14 |
paramat |
ok yes |
| 03:14 |
est31 |
but SN just did, so its fine to merge |
| 03:15 |
paramat |
this is why i prefer that the author's approval is actually added to labels |
| 03:15 |
paramat |
but this tends to upset people :] |
| 03:16 |
est31 |
well i guess merging a pr with implicit author approval upsets those people as well |
| 03:16 |
est31 |
plus it upsets me :) |
| 03:16 |
paramat |
yeah |
| 03:42 |
paramat |
sofar if you're around https://github.com/minetest/minetest/issues/4810#issuecomment-268142713 |
| 03:51 |
paramat |
wow project channel is just a bunch of greetings :D |
| 04:24 |
|
Hunterz joined #minetest-dev |
| 04:26 |
|
octacian joined #minetest-dev |
| 04:57 |
garywhite |
paramat are you active? |
| 05:02 |
paramat |
i will be in 30 mins |
| 05:02 |
garywhite |
ok |
| 05:21 |
|
Zeno` joined #minetest-dev |
| 05:23 |
garywhite |
Hi Zeno` |
| 05:23 |
garywhite |
Oops, sorry |
| 05:32 |
Zeno` |
? |
| 05:32 |
Zeno` |
hi |
| 05:33 |
|
est31 joined #minetest-dev |
| 06:11 |
|
Hunterz joined #minetest-dev |
| 06:12 |
paramat |
gary how can i help? |
| 06:14 |
garywhite |
I figured out that it's not an installer issue for the Win10 issue...Can't remember the issue# |
| 06:14 |
paramat |
#4528 |
| 06:14 |
ShadowBot |
https://github.com/minetest/minetest/issues/4528 -- minetest 0.4.14 64bit crashes at startup on Windows 10 Pro 64bit: debug.txt permissions/path problem |
| 06:15 |
garywhite |
It's a permissions issue, MT has to be able to have administrator privs to install to Program Files |
| 06:15 |
garywhite |
Cause Windows' User Account Control from Vista onwards requires administrator access to write to anything in Program Files |
| 06:16 |
paramat |
yeah saw your comment, Zeno` is the dev to talk to about this |
| 06:16 |
paramat |
i don't understand it at all :] |
| 06:17 |
garywhite |
Lemme put it in more...Linux-y terms |
| 06:17 |
garywhite |
Since most of the people here use Linux |
| 06:17 |
paramat |
ok i'll merge the 2 mod security PRs in a moment, so they get some last minute testing #4849 #4919 |
| 06:17 |
ShadowBot |
https://github.com/minetest/minetest/issues/4849 -- Mod security: Allow read-only access to all mod paths by ShadowNinja |
| 06:17 |
ShadowBot |
https://github.com/minetest/minetest/issues/4919 -- Fix RemoveRelativePathComponents (again) and fix mod security with paths starting with ".." by ShadowNinja |
| 06:19 |
garywhite |
Basically, the MInetest installer file installs to the main program folder (idk what that is in Linux). When you try to launch it, the computer wants MT to have su privs to be able to write to debug.txt, or anything in the program folder, thus crashing the program. |
| 06:20 |
garywhite |
Oh shoot... |
| 06:21 |
garywhite |
paramat: Zeno` may have a bigger problem than at first |
| 06:21 |
paramat |
chat to him about it, i'm busy merging |
| 06:22 |
garywhite |
He's not on |
| 06:22 |
|
lumidify joined #minetest-dev |
| 06:24 |
garywhite |
This means that if it cannot write to Program Files, then no mods or anything can be added without administrator permission for the same reasons |
| 06:28 |
garywhite |
I'm heading off for the night then. gn all (dunno if that's allowed tbh) |
| 06:29 |
paramat |
that's ok |
| 06:30 |
paramat |
you can see in the username list he is around |
| 06:31 |
paramat |
4919 doesn't apply after 4849, so i will merge 4849 only now |
| 06:32 |
garywhite |
Oh, he's voiced...I was looking in regular |
| 06:32 |
Zeno` |
yeah it's a big problem |
| 06:33 |
Zeno` |
which is why kaeza and I are suggesting the workaround for now |
| 06:33 |
Zeno` |
and installing in a place where the user has full write permissions |
| 06:33 |
Zeno` |
to fix it properly will require a lot of editing in the C++ source and I think better left until the next release |
| 06:34 |
garywhite |
Anywhere else would work, I can install to D:\Program Files (I have 2 partitions) and it works |
| 06:34 |
garywhite |
I'd have to agree with Zeno`, use the workaround |
| 06:34 |
Zeno` |
yeah. Then we have lots and lots of time after release to work on fixing it "properly" |
| 06:36 |
paramat |
now merging 4849 |
| 06:42 |
paramat |
merged |
| 06:44 |
paramat |
ShadowNinja 4919 did not apply on merge, see https://github.com/minetest/minetest/pull/4919#issuecomment-268166092 |
| 06:44 |
|
YuGiOhJCJ joined #minetest-dev |
| 06:45 |
|
OldCoder joined #minetest-dev |
| 06:50 |
Zeno` |
how did that not get tagged as controversial? |
| 06:50 |
Zeno` |
too late now I guess. Oh well |
| 06:52 |
paramat |
hm i see, it should have been |
| 06:52 |
Zeno` |
what's done is done :) |
| 06:55 |
Zeno` |
I guess it's not too bad |
| 06:56 |
Zeno` |
if someone has a private Lua mod (not open source) they should make sure they don't install mods that allow it to be stolen |
| 07:00 |
paramat |
yeah doesn't seem so bad to me either |
| 07:00 |
paramat |
reading through |
| 07:00 |
Zeno` |
I'm not overjoyed by it though hehe |
| 07:00 |
Zeno` |
but *shrug* |
| 07:01 |
Zeno` |
protocol has been followed and I had my chance to object before it was merged |
| 07:01 |
paramat |
the alternative is turning off mod security for this release |
| 07:01 |
Zeno` |
nah |
| 07:01 |
Zeno` |
this is ok |
| 07:01 |
paramat |
which is worse than this |
| 07:01 |
Zeno` |
yes, I agree |
| 07:02 |
paramat |
me too :] |
| 07:04 |
Zeno` |
you agree with your own comment. How unconventional :p |
| 07:11 |
Zeno` |
I can merge #4919 if you like |
| 07:12 |
ShadowBot |
https://github.com/minetest/minetest/issues/4919 -- Fix RemoveRelativePathComponents (again) and fix mod security with paths starting with ".." by ShadowNinja |
| 07:12 |
paramat |
please do if you can solve the conflict |
| 07:12 |
Zeno` |
ok making new patch now |
| 07:17 |
Zeno` |
paramat, merge now? |
| 07:17 |
paramat |
i see the conflict now, it's not a conflict in the diffs, but one in the lines surrounding the diff that github uses to locate the diffs |
| 07:17 |
paramat |
+1 |
| 07:18 |
Zeno` |
I'm going to leave as 2 commits |
| 07:18 |
Zeno` |
they both seem quite large |
| 07:18 |
paramat |
fine |
| 07:19 |
paramat |
thanks |
| 07:20 |
Zeno` |
np (all done) |
| 07:25 |
paramat |
looks like 4923 may not be a blocker https://github.com/minetest/minetest/issues/4923#issuecomment-268172786 |
| 07:27 |
paramat |
map gen limit was never properly implemented, it's a bit of a mess |
| 07:31 |
paramat |
to solve the problems it may have to be added to mapgen params, unfortunately, i wish we had never added the feature now |
| 07:32 |
Zeno` |
I was looking at that yesterday and could reproduce haha |
| 07:32 |
Zeno` |
I think I have to somehow get a mob to spawn |
| 07:32 |
Zeno` |
and it all go too hard :P |
| 07:32 |
paramat |
limiting a world should have been left to mods |
| 07:32 |
Zeno` |
well to an extent |
| 07:33 |
Zeno` |
we have to have it fit in s16 and reserve a bunch of space around the edges to make handling edge cases (no pun intended) easier to handle |
| 07:33 |
Zeno` |
the reserved area, as I understand it, makes handling those edge cases entirely unnessessary |
| 07:35 |
Zeno` |
although why lua is allowing boats outside the map is a mystery |
| 07:35 |
Zeno` |
maybe it's a ghost ship! |
| 07:35 |
Zeno` |
get it? |
| 07:35 |
* Zeno` |
waits for the audience to laugh |
| 07:35 |
Zeno` |
ok, I'm going to my corner now |
| 07:41 |
hmmmm |
boats saling off the map are explorers |
| 07:41 |
hmmmm |
i just hope for their sake mr. columbus was right |
| 07:42 |
paramat |
hmmmmm what do you think about mapgen limit being added to mapgen params? to be per-world |
| 07:42 |
hmmmm |
it would be a shame for them to fall off and get deleted... or something |
| 07:42 |
hmmmm |
yeah, we're going to have to eventually do that |
| 07:42 |
paramat |
someone is preparing a PR to do this |
| 07:43 |
Zeno` |
here be dragons |
| 07:44 |
Zeno` |
apparently old maps never actually wrote that in unexplored places |
| 07:44 |
Zeno` |
maybe we could put dragons on out-of-map areas |
| 07:44 |
|
nrzkt joined #minetest-dev |
| 07:44 |
hmmmm |
sounds like something for a mod |
| 07:45 |
hmmmm |
i'm upset there haven't been world boundary mods that do something interesting |
| 07:45 |
paramat |
i keep having an idea to add a few mapchunks of air around the world edge, so that you can actually fall off or build off-edge |
| 07:45 |
hmmmm |
iirc all they do are add undestructable node walls |
| 07:45 |
paramat |
in core. not sure if a good idea |
| 07:45 |
hmmmm |
i doubt it |
| 07:45 |
hmmmm |
there's so much that could go wrong with that idea |
| 07:46 |
paramat |
very specialised too, maybe a mod thang |
| 07:48 |
Zeno` |
well it's all a boring grey |
| 07:48 |
Zeno` |
we spent 9 months digging to the centre of the world to see... grey |
| 07:51 |
Zeno` |
there could at least have been some kind of evil fog there |
| 07:51 |
Zeno` |
or walls of animated flames |
| 07:51 |
Zeno` |
or some cheeseburgers |
| 07:52 |
Zeno` |
flying toasters or turtles might be nice as well |
| 07:59 |
paramat |
yes best i do this as a mod, simple too, just a bunch of air added |
| 07:59 |
Zeno` |
mist would be better |
| 07:59 |
Zeno` |
but swirling mist |
| 07:59 |
Zeno` |
I guess that'd best be done with shaders though :( |
| 08:00 |
Zeno` |
why not a galaxy/space scene? :D |
| 08:12 |
hmmmm |
because it all takes time and effort |
| 08:12 |
hmmmm |
minetest is ultimately just a game we work on in our spare time |
| 08:12 |
Zeno` |
but we have heaps of spare time |
| 08:12 |
hmmmm |
you do |
| 08:12 |
Zeno` |
</sarcasm> |
| 08:12 |
hmmmm |
ok have fun coding up a galaxy/space scene :P |
| 08:12 |
hmmmm |
sigh |
| 08:12 |
Zeno` |
lol :) |
| 08:14 |
Zeno` |
I don't even really have time to be on IRC, but if I don't have distractions I'd go insane |
| 08:14 |
hmmmm |
same here |
| 08:14 |
Zeno` |
more insane* |
| 08:14 |
hmmmm |
IRC is one of those things that don't seem to take much time or attention but it actually does build up |
| 08:15 |
hmmmm |
if i didn't have ADD right now i would've had my workie work done |
| 08:15 |
hmmmm |
for example what just happened to me was... |
| 08:16 |
hmmmm |
i finished up this new interface for some thing, and now i go to replace the invocation for it in main(), and i notice a problem |
| 08:17 |
hmmmm |
now i branch off execution like if (i_can_use_thing()) { ReallyGoodThing rgt; rgt.LoadThingMabobbers(); } else { OldHackyWorkaround workaround; workaround.DoStuff(); } ... |
| 08:17 |
hmmmm |
oh crap, can't do that because it's gonna fall out of scope! |
| 08:17 |
hmmmm |
so just move them out of the scope |
| 08:17 |
hmmmm |
oh crap, constructors do work in them among other stuff! |
| 08:18 |
hmmmm |
so at this point i back myself into a hole |
| 08:18 |
hmmmm |
do I change the interface to have some kind of Initialize() function for both? |
| 08:18 |
hmmmm |
do I change it to allocate/initialize using new instead? |
| 08:19 |
hmmmm |
and instead of making a solid decision and going forward, i go take a shit/eat a snack/etc. |
| 08:19 |
hmmmm |
come back and nothing's changed, i still haven't figured out what i want to do |
| 08:19 |
hmmmm |
:( |
| 08:20 |
hmmmm |
i envy people who are able to make fast solid decisions on interface design |
| 08:22 |
Zeno` |
geez :( |
| 08:22 |
|
nrzkt joined #minetest-dev |
| 08:22 |
celeron55 |
just use C++ and throw in new and unique_ptr |
| 08:22 |
celeron55 |
no need to think |
| 08:23 |
hmmmm |
well that was just an example of the types of things i get stuck on |
| 08:24 |
celeron55 |
try to imagine how much i care about those kinds of decisions when i set out to get things done |
| 08:24 |
celeron55 |
maybe that makes you feel less bad |
| 08:27 |
hmmmm |
i should start taking inspiration from you and Nike |
| 08:27 |
celeron55 |
just stick with the first thing that works without 10s of lines of spaghetti code; people will care more about what you will do next than what you did there |
| 08:28 |
celeron55 |
altough i kind of disqualify myself from saying any of this because none of my latest work has ended up upstream yet |
| 08:28 |
Zeno` |
you should see the code my peers write. It's 99% throwaway code all of the time |
| 08:28 |
nrzkt |
Zeno`: java style yeah yeah :D |
| 08:28 |
Zeno` |
mainly because it's not our focus, but still |
| 08:28 |
Zeno` |
nrzkt, worse :D |
| 08:29 |
Zeno` |
and most of it Fortran. (although these days Python is taking over) |
| 08:30 |
Zeno` |
fortunately I don't have to look at that a lot of the time because the code I look at is for embedded stuff (i.e. in C) |
| 08:30 |
Zeno` |
and embedded code has its own issues... haha |
| 08:30 |
Zeno` |
mainly due to stack |
| 08:32 |
hmmmm |
embedded code is some of the sloppiest shit i've seen for w/e reason |
| 08:33 |
Zeno` |
well yeah |
| 08:34 |
Zeno` |
to "be safe" just make the variable global |
| 08:34 |
Zeno` |
so everything is global and large functions without structure are common as well |
| 08:34 |
hmmmm |
:D |
| 08:34 |
Zeno` |
just to be safe |
| 08:34 |
hmmmm |
gotta love the large functions without structure |
| 08:34 |
hmmmm |
at least the globals can be compartmentalized |
| 08:36 |
hmmmm |
one leads to another though, if you have lots of globals you tend to have lots of large blob functions |
| 08:36 |
hmmmm |
because you don't need to think about what things each sub-section of the code is dependent upon |
| 08:36 |
hmmmm |
by splitting off functions into smaller functions you get a better idea of what the inputs/outputs are due to the forced separation |
| 08:37 |
Zeno` |
yeah |
| 08:37 |
Zeno` |
the funny thing is... our embedded stuff isn't life threatening or "mission critical" or anything |
| 08:38 |
hmmmm |
it's sloppy regardless of whether it's life threatening or not, guaranteed |
| 08:38 |
Zeno` |
if one of the remote sensors fails we can just reboot |
| 08:39 |
Zeno` |
with no harm done at all. It's weird |
| 08:39 |
hmmmm |
the sloppiest 'professional' code i've seen has got to be the display for a piece of digital aircraft instrumentation |
| 08:39 |
Zeno` |
oh, we're in good hands why we fly then? |
| 08:39 |
hmmmm |
:D |
| 08:39 |
hmmmm |
nah you'd never get a chance to fly in that kind of plane |
| 08:39 |
hmmmm |
but if the rumors are true, avoid Boeing jets, they have serious firmware issues |
| 08:40 |
Zeno` |
that's why I only fly in SR71s |
| 08:40 |
hmmmm |
like remember just a few years ago the engines dropped power because some internal timer rolled over 42 days |
| 08:40 |
hmmmm |
lmao :( |
| 08:40 |
Zeno` |
:D |
| 08:40 |
Zeno` |
dinner time |
| 08:40 |
Zeno` |
bbiab |
| 08:40 |
hmmmm |
I forget what model that was |
| 08:40 |
Zeno` |
the umm something bird |
| 08:40 |
hmmmm |
lol |
| 08:40 |
paramat |
blackbird |
| 08:41 |
Zeno` |
high speed/altitude spy plane |
| 08:41 |
Zeno` |
yeah blackbird |
| 08:41 |
hmmmm |
no no I mean the model of the boeing passenger plane with the issue i mentioned |
| 08:41 |
hmmmm |
i think it was the 777 dreamliner |
| 08:41 |
paramat |
ah |
| 08:42 |
paramat |
i removed 'blocker' from #4923 |
| 08:42 |
ShadowBot |
https://github.com/minetest/minetest/issues/4923 -- map_generation_limit trigger server crash |
| 08:44 |
paramat |
looks like we have incoming workarounds for #4528 and #4638 |
| 08:44 |
ShadowBot |
https://github.com/minetest/minetest/issues/4528 -- minetest 0.4.14 64bit crashes at startup on Windows 10 Pro 64bit: debug.txt permissions/path problem |
| 08:44 |
ShadowBot |
https://github.com/minetest/minetest/issues/4638 -- Don't ship with broken languages |
| 08:44 |
paramat |
that leaves #4810 which doesn't seem critical |
| 08:44 |
ShadowBot |
https://github.com/minetest/minetest/issues/4810 -- Unittests crash on Release build |
| 08:46 |
paramat |
and #4888 is not essential apparently |
| 08:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/4888 -- Handle multiple deserialization of a block's entities by Rogier-5 |
| 08:50 |
paramat |
removed 4888 from milestone |
| 09:07 |
Zeno` |
Rogier-5 should be added to contributers |
| 09:09 |
Zeno` |
I can't see him there unless I'm blind |
| 09:10 |
hmmmm |
maybe you should apply a gaussian blur filter to the screen |
| 09:12 |
Zeno` |
lol |
| 09:12 |
Zeno` |
I gotta hand it to you. That was fkn funny |
| 09:15 |
Zeno` |
hey, it helped! |
| 09:16 |
hmmmm |
'night |
| 09:16 |
Zeno` |
sleep well |
| 09:16 |
Zeno` |
paramat, https://github.com/minetest/minetest/pull/4934 |
| 09:16 |
Zeno` |
one more for you to deal with ;) |
| 09:18 |
paramat |
ok |
| 09:20 |
Zeno` |
maybe leave it to last before you add the release tag |
| 09:21 |
Zeno` |
I should have left the previous PR open longer |
| 09:21 |
* Zeno` |
eats blue cheese for... dessert |
| 10:49 |
|
paramat joined #minetest-dev |
| 11:24 |
|
proller joined #minetest-dev |
| 11:25 |
|
jin_xi joined #minetest-dev |
| 11:46 |
|
lumidify joined #minetest-dev |
| 11:46 |
|
eidy joined #minetest-dev |
| 12:06 |
|
Fixer joined #minetest-dev |
| 12:07 |
|
Human_G33k joined #minetest-dev |
| 12:20 |
|
proller joined #minetest-dev |
| 12:27 |
paramat |
#4936 |
| 12:27 |
ShadowBot |
https://github.com/minetest/minetest/issues/4936 -- Objectpos_over_limit: Use max map gen limit by paramat |
| 12:56 |
nrzkt |
paramat: why remove the g_settings->getU16("map_generation_limit") ? if map_generation_limit < real limit it's used |
| 12:56 |
nrzkt |
can you explain a little bit |
| 12:58 |
paramat |
it's complex but explained in the issue linked |
| 12:59 |
paramat |
however i think that PR is incomplete |
| 13:00 |
paramat |
i doubt this PR can be ready for release, the map gen limit setting has created complex problems |
| 13:00 |
nrzkt |
paramat: ofc |
| 13:04 |
sfan5 |
paramat: unfortunately a "simple workaround" for the broken languages thing is not possbiel |
| 13:04 |
sfan5 |
gettext loads the language files by itself |
| 13:04 |
sfan5 |
afaik it also detects the language by itself |
| 13:05 |
paramat |
ok |
| 13:06 |
paramat |
i just seem to remember discussion of a hacky workaround for release |
| 13:08 |
paramat |
"just move the translation files somewhere else is easiest probably" |
| 13:09 |
sfan5 |
indeed |
| 13:10 |
paramat |
sorry, obviously i don't understand the situation |
| 13:10 |
|
turtleman joined #minetest-dev |
| 13:19 |
paramat |
ok i'm tired so back on the 21st. 4923 seems unfixable in time. 4810 not critical. 4638 and 4528 hopefully have workarounds coming. i'll leave it to others now |
| 13:30 |
sfan5 |
pushing http://pastebin.com/raw/Z52tLBT2 in 10 minutes |
| 14:06 |
|
twoelk joined #minetest-dev |
| 14:21 |
|
blaze joined #minetest-dev |
| 14:31 |
|
Fixer joined #minetest-dev |
| 14:31 |
|
rubenwardy joined #minetest-dev |
| 15:09 |
|
Hunterz joined #minetest-dev |
| 15:16 |
|
DuCake joined #minetest-dev |
| 15:43 |
|
proller joined #minetest-dev |
| 15:44 |
|
johnnyjoy joined #minetest-dev |
| 15:48 |
|
lumidify joined #minetest-dev |
| 16:11 |
|
domtron joined #minetest-dev |
| 16:13 |
|
hmmmm joined #minetest-dev |
| 16:16 |
|
Void7 joined #minetest-dev |
| 16:22 |
|
domtron joined #minetest-dev |
| 16:43 |
|
octacian_ joined #minetest-dev |
| 16:47 |
|
Hunterz joined #minetest-dev |
| 17:08 |
|
Krock joined #minetest-dev |
| 17:08 |
|
nrzkt joined #minetest-dev |
| 17:13 |
|
twoelk joined #minetest-dev |
| 17:34 |
|
red-001 joined #minetest-dev |
| 17:38 |
red-001 |
huh couldn't the installer just grant read-write permissions to all users for that minetest install location? |
| 17:38 |
red-001 |
for the* |
| 17:39 |
garywhite |
Zeno`: I don't think it's just altering the C++ code to fix the Windows issue |
| 17:45 |
|
Gael-de-Sailly joined #minetest-dev |
| 17:47 |
Zeno` |
it's not |
| 17:48 |
Zeno` |
it's a combination of thing, but the engine code *will* need to be changed |
| 17:48 |
red-001 |
Are you talking about the installer isssue? |
| 17:48 |
Zeno` |
red-001, yes |
| 17:49 |
Zeno` |
red-001, to fix it properly we need to change default paths (for the installer version) in C++ |
| 17:49 |
Zeno` |
it's probably going to be messy, which is why the workaround is proposed for this release |
| 17:50 |
|
Darcidride joined #minetest-dev |
| 17:50 |
Zeno` |
we have to make sure that we don't break existing installations and only affect installations done with the installer |
| 17:51 |
Zeno` |
it sounds easy at first, but when you look at the code... it's not |
| 17:53 |
Zeno` |
the other issue is we have limited resources at the moment to test Windows builds |
| 17:54 |
twoelk |
if working on windows paths anyways, could we make more of them configurable? (still dreaming of a portableapps.com version) |
| 17:56 |
garywhite |
Zeno`: Let me run a test by turning off my User Account Control & see if it allows |
| 17:57 |
* twoelk |
could also test windows builds at work right now (boss just went home) |
| 17:58 |
red-001 |
Wouldn't you still not have premissions to write there? |
| 17:58 |
red-001 |
I though that was a file system thing |
| 17:58 |
garywhite |
idk... |
| 17:59 |
garywhite |
It's not User Account Control issue |
| 18:01 |
garywhite |
+1 for workaround |
| 18:04 |
Krock |
I also could build for Win32 but currently I haven't got much free time to develop and doing tests |
| 18:08 |
Calinou |
I once got a win64 build working but it was hours of work |
| 18:08 |
Calinou |
and some things were statically linked, others dynamically |
| 18:08 |
garywhite |
Zeno`: http://superuser.com/questions/1031275/windows-allow-normal-non-elevated-programs-to-write-to-program-files-director |
| 18:09 |
Calinou |
it's a terrible idea to want to write to Program Files |
| 18:09 |
Calinou |
just use %APPDATA% |
| 18:09 |
garywhite |
I don't think it'd be possible to write to Program Files without elevating Minetest to run as an administrator |
| 18:10 |
garywhite |
Thanks to non-removable features introduces with UAC in Windows Vista in 2006 |
| 18:10 |
Calinou |
it's a good thing, program installation path should not be writeable, it encourages bad practice |
| 18:10 |
Calinou |
it's the same on Linux btw |
| 18:11 |
nrzkt |
i agree with Calinou , it's not because you use a shitty OS you should do shit :p |
| 18:11 |
garywhite |
Calinou: MT has to be able to write to the debug file |
| 18:11 |
est31 |
but the debug file shouldnt reside in the same place as the executable |
| 18:11 |
Calinou |
garywhite: then write it to appdata if it's not writable in cwd |
| 18:12 |
garywhite |
est31: The debug & the EXE both reside in /bin |
| 18:14 |
garywhite |
So basically Zeno` there is no way, I think, that this issue can be fixed outside of not installing to Program Files |
| 18:14 |
garywhite |
Or the workaround |
| 18:20 |
Fixer |
just use %APPDATA%/minetest * |
| 18:21 |
garywhite |
Fixer: It has to be a place where the user can access it to install mods & subgames |
| 18:22 |
Fixer |
right |
| 18:22 |
garywhite |
%APPDATA% is hidden to my knowledge |
| 18:23 |
Calinou |
garywhite: it is hidden but we can have a shortcut created in Start Menu |
| 18:23 |
Calinou |
"Open user data folder" |
| 18:23 |
Calinou |
alternatively, use the Windows "saved games" folder which is not hidden |
| 18:23 |
Calinou |
and also user-specific |
| 18:23 |
garywhite |
True |
| 18:24 |
garywhite |
That is easily accessible from Libraries in Windows Explorer (Not to be confused with Internet Explorer) |
| 18:24 |
Calinou |
yes |
| 18:26 |
garywhite |
Here's my proposal (and I would like all the devs to look at this please): Use Zeno`'s workaround for the time being, leave language as it is, and attempt to incorporate a fix build called 0.4.15a that adds multi-language support and fixes the issue for installing to Windows |
| 18:28 |
|
Thomas-S joined #minetest-dev |
| 18:36 |
|
AcidNinjaFWHR joined #minetest-dev |
| 18:38 |
|
lumidify joined #minetest-dev |
| 18:39 |
Fixer |
or in Documents/Minetest ? |
| 18:47 |
|
QwertyDragon joined #minetest-dev |
| 18:55 |
sfan5 |
garywhite: no |
| 18:55 |
sfan5 |
fixing multi-language support is not that easy |
| 18:55 |
sfan5 |
unless some dev is willing to dedicate themselves to fixing it no matter what this ain't gonna happen |
| 18:57 |
garywhite |
sfan5: I'd at least look into Spanish, that is, from my research, the 2nd most spoken language, English is 3rd. So I'd at least add Spanish into MT if possible |
| 18:57 |
sfan5 |
who said spanish isn't getting into the release? |
| 19:02 |
Krock |
don't we already ship spanish for a long time in mt already? |
| 19:03 |
sfan5 |
we do |
| 19:04 |
Krock |
good, then I wasn't wrong with that. garywhite, don't worry. The most used languages seem to be fine so far |
| 19:29 |
AcidNinjaFWHR |
But spanish is partially translated |
| 19:29 |
AcidNinjaFWHR |
btw how do we get translations from https://hosted.weblate.org/projects/minetest/minetest/ |
| 19:30 |
AcidNinjaFWHR |
to get merged into the github repo? |
| 19:31 |
sfan5 |
a coredev does that |
| 19:31 |
sfan5 |
garywhite: here's the updated list https://github.com/minetest/minetest/issues/4638 |
| 19:32 |
sfan5 |
as you can see there is just three that are definitely broken |
| 19:33 |
sfan5 |
!tell est31 I have updated and completed the list in issue 4638 (broken langs) |
| 19:33 |
ShadowBot |
sfan5: O.K. |
| 20:16 |
|
AcidNinjaFWHR joined #minetest-dev |
| 20:29 |
|
proller joined #minetest-dev |
| 21:05 |
|
eidy joined #minetest-dev |
| 21:19 |
|
DuCake joined #minetest-dev |
| 22:11 |
|
Human_G33k joined #minetest-dev |
| 22:16 |
|
est31 joined #minetest-dev |
| 22:24 |
|
proller joined #minetest-dev |
| 22:27 |
garywhite |
Someone correct me if I'm wrong, but don't devs usually make the MT builds? |
| 22:27 |
twoelk |
nope |
| 22:27 |
twoelk |
not all |
| 22:28 |
garywhite |
Cause someone made something called a snap of Minetest https://forum.minetest.net/viewtopic.php?f=42&t=16088 |
| 22:37 |
nrzkt |
garywhite, i'm doing the android build if it's the question |
| 22:37 |
|
Shara joined #minetest-dev |
| 22:45 |
|
DuCake joined #minetest-dev |
| 22:49 |
|
YuGiOhJCJ joined #minetest-dev |
| 23:03 |
|
rdococ joined #minetest-dev |
| 23:03 |
|
rdococ joined #minetest-dev |
| 23:05 |
|
rdococ joined #minetest-dev |
| 23:05 |
|
DuCake joined #minetest-dev |
| 23:11 |
|
twoelk left #minetest-dev |
| 23:23 |
|
minetester joined #minetest-dev |
| 23:36 |
|
minetester left #minetest-dev |
| 23:50 |
|
DuCake joined #minetest-dev |
| 23:50 |
|
lhofhansl joined #minetest-dev |
| 23:50 |
lhofhansl |
Interest in #4933? |
| 23:50 |
ShadowBot |
https://github.com/minetest/minetest/issues/4933 -- Improve occlusion culling accuracy and efficiency. by lhofhansl |
