| Time |
Nick |
Message |
| 00:12 |
|
Cornelia joined #minetest-dev |
| 00:22 |
p_gimeno |
what makes double quotes preferable to single? My personal convention is to use single quotes for strings that don't represent translatable texts, and double for strings that do. |
| 00:27 |
rubenwardy |
it comes from C and such |
| 00:27 |
rubenwardy |
double are strings, single are chars |
| 00:30 |
paramat |
sfan, no need, trivial maintenance. merging trivial #8724 |
| 00:30 |
ShadowBot |
https://github.com/minetest/minetest/issues/8724 -- Silence invalid clang format warnings in guiScrollBar.cpp by paramat |
| 00:41 |
|
ANAND joined #minetest-dev |
| 01:14 |
|
Cornelia joined #minetest-dev |
| 01:35 |
|
FullTimeLinux joined #minetest-dev |
| 01:48 |
|
SolarArray joined #minetest-dev |
| 02:05 |
|
Cornelia joined #minetest-dev |
| 02:06 |
|
SolarArray joined #minetest-dev |
| 02:14 |
|
Cornelia joined #minetest-dev |
| 02:27 |
|
Cornelia joined #minetest-dev |
| 02:27 |
|
Lone_Wolf joined #minetest-dev |
| 02:48 |
|
Cornelia joined #minetest-dev |
| 03:26 |
|
Foz joined #minetest-dev |
| 05:42 |
|
loggingbot_ joined #minetest-dev |
| 05:42 |
|
Topic for #minetest-dev is now Minetest core development and maintenance. Minetest 5.0.1 released! Chit-chat goes to #minetest. http://irc.minetest.net/minetest-dev/ http://dev.minetest.net/ |
| 05:50 |
|
ssieb joined #minetest-dev |
| 06:59 |
|
sofar joined #minetest-dev |
| 08:45 |
sfan5 |
~tell paramat the "trivial" rule applies to small patches, 263 line changes is not a small patch |
| 08:45 |
ShadowBot |
sfan5: O.K. |
| 10:46 |
|
proller joined #minetest-dev |
| 11:54 |
|
Fixer joined #minetest-dev |
| 12:33 |
|
thecow275 joined #minetest-dev |
| 12:35 |
|
calcul0n joined #minetest-dev |
| 12:36 |
|
thecow275 joined #minetest-dev |
| 12:42 |
|
Cornelia joined #minetest-dev |
| 12:52 |
|
Cornelia joined #minetest-dev |
| 12:59 |
|
pauloue joined #minetest-dev |
| 13:16 |
|
Cornelia joined #minetest-dev |
| 13:30 |
|
Cornelia joined #minetest-dev |
| 13:31 |
|
Lia joined #minetest-dev |
| 13:34 |
ANAND |
How does MT serialize a vector of strings? Is this implemented in the code? I couldn't find any function in src/util/serialize.{h|cpp} that does this. |
| 13:41 |
sfan5 |
have you checked the networkpacket code? |
| 13:49 |
ANAND |
Ah, I haven't checked that part yet, thanks :) |
| 13:53 |
ANAND |
I still can't find code that serializes a vector of strings, but I'm getting a couple of ideas by looking at how a single std::string object is serialized |
| 13:57 |
sfan5 |
where do you see vector serialization being used? |
| 14:02 |
ANAND |
I'm in the process of implementing wielditem meshes, which requires adding a table of textures like in the entity definition (i.e. textures = { character.png } ) to the item definitions. I'm planning to use a vector of string to represent this table in the C++-side. |
| 14:02 |
ANAND |
vector of strings* |
| 14:03 |
ANAND |
Any inputs much appreciated :) |
| 14:06 |
sfan5 |
oh |
| 14:06 |
sfan5 |
u16 count and then serialize each string |
| 14:08 |
ANAND |
Yes, I figured it out. Thanks anyway :) |
| 14:17 |
|
Taoki joined #minetest-dev |
| 14:20 |
ANAND |
#8725 |
| 14:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/8725 -- Add support for mesh items by ClobberXD |
| 14:21 |
ANAND |
WIP ofc |
| 14:22 |
|
Wuzzy joined #minetest-dev |
| 14:40 |
ANAND |
Trivial PR #8526 |
| 14:40 |
ShadowBot |
https://github.com/minetest/minetest/issues/8526 -- Move HTTP request logging to infostream by ClobberXD |
| 14:40 |
ANAND |
oops |
| 14:41 |
ANAND |
#8726 |
| 14:41 |
ShadowBot |
https://github.com/minetest/minetest/issues/8726 -- ItemDefinition: Fix inventory_overlay and wield_overlay not being (de)serialized by ClobberXD |
| 14:48 |
|
twoelk joined #minetest-dev |
| 14:54 |
|
Taoki joined #minetest-dev |
| 14:55 |
|
Jordach joined #minetest-dev |
| 15:54 |
|
Soni joined #minetest-dev |
| 15:54 |
Soni |
can we get cratera support? |
| 15:55 |
Soni |
just have cratera support in minetest core |
| 15:55 |
Soni |
except without cratera itself, just the support for it |
| 15:55 |
Soni |
all cratera really needs is carefully constructed table structures |
| 15:55 |
Soni |
cratera: https://marc.info/?l=lua-l&m=156458156026332&w=2 |
| 15:56 |
sfan5 |
no, we can't have support for your homemade programming toy |
| 15:58 |
Soni |
can we have traits then? |
| 15:58 |
Soni |
and a patched luajit with full support for traits? |
| 15:59 |
Soni |
traits are all that's missing from lua's OOP collection |
| 15:59 |
sfan5 |
nothing prevents you from using a patched luajit |
| 16:00 |
Soni |
the point of Cratera Compiler v1.0.1 is that it doesn't require a patched lua |
| 16:00 |
Soni |
(or luajit or w/e) |
| 16:00 |
Soni |
but really, can you just give us node[Inventory] = Inventory and things like that? |
| 16:01 |
Soni |
where Inventory just defers to the old inventory system, whatever that is |
| 16:02 |
Soni |
(you don't need Cratera Compiler to implement that) |
| 16:26 |
ANAND |
Re: #8726 - The PR is invalid as `inventory_overlay` and `wield_overlay` are indeed already (de)serialized, but just a little further down the method after a version check. |
| 16:26 |
ShadowBot |
https://github.com/minetest/minetest/issues/8726 -- ItemDefinition: Fix inventory_overlay and wield_overlay not being (de)serialized by ClobberXD |
| 16:27 |
sfan5 |
that PR wouldn't be good anyway since you have to add compatibility code when you change the serialization like that |
| 16:28 |
sfan5 |
merging #8723 |
| 16:28 |
ShadowBot |
https://github.com/minetest/minetest/issues/8723 -- Allow toolcaps to override the built-in times for dig_immediate by sfan5 |
| 16:32 |
ANAND |
Agreed |
| 16:33 |
ANAND |
Hm, that means I can't simply (de)serialize extra fields added in the mesh items PR. I'd have to add version checks... |
| 16:34 |
ANAND |
I guess that PR wasn't completely useless, after all ;) |
| 16:37 |
ANAND |
How do I check for the protocol version before (de)serializing the extra fields? Do I even need to check for the version if I serialize the fields right at the end? |
| 16:38 |
ANAND |
I suppose clients only deserialize what they know, and omit the rest of the data |
| 16:38 |
sfan5 |
if the fields are at the end a try { } catch (SerializationError&) {} would work |
| 16:39 |
ANAND |
Oh nice |
| 16:41 |
ANAND |
https://github.com/minetest/minetest/blob/master/src/itemdef.cpp#L218 |
| 16:41 |
ANAND |
:D |
| 16:42 |
ANAND |
Is it worth adding some sort of a version check before serializing, to prevent serializing stuff that's going to be discarded while deserializing anyway? |
| 16:44 |
|
Krock joined #minetest-dev |
| 18:13 |
|
MillersMan joined #minetest-dev |
| 18:27 |
sfan5 |
ANAND: no |
| 18:41 |
Krock |
will merge web#184 in 10 minutes (trivial) |
| 18:41 |
ShadowBot |
https://github.com/minetest/minetest.github.io/issues/184 -- Corrected the Lua API Link in the footer by mmattes |
| 18:50 |
|
DS-minetest joined #minetest-dev |
| 18:50 |
Krock |
merging... |
| 18:52 |
|
ensonic joined #minetest-dev |
| 18:53 |
|
twoelk|2 joined #minetest-dev |
| 19:24 |
Soni |
can we have minimal cratera support |
| 19:24 |
|
Edgy1 joined #minetest-dev |
| 19:33 |
|
proller joined #minetest-dev |
| 19:40 |
|
MillersMan_ joined #minetest-dev |
| 19:57 |
|
Exe_Virus joined #minetest-dev |
| 19:58 |
Exe_Virus |
Hello, can anyone point me to a simplish mapgen mod that starts out with singlenode as the mapgen? I'm trying to create a custom mapgen |
| 19:58 |
|
fwhcat joined #minetest-dev |
| 20:01 |
Krock |
Lua? |
| 20:02 |
Krock |
https://github.com/SmallJoker/flatgen/blob/affe55b0/init.lua#L69-L145 |
| 20:04 |
Exe_Virus |
Krock: thank you, that will work. Does that overwrite the normal mapgen (say I set it to fractal or something) |
| 20:04 |
Krock |
https://github.com/paramat/overgen/blob/master/init.lua |
| 20:04 |
Krock |
no, you'll have to configure your world as singlenode |
| 20:05 |
Krock |
I did not try what happens with anything else, though |
| 20:05 |
Krock |
maybe thy would just merge |
| 20:05 |
Krock |
or one overwrites the other |
| 20:05 |
Exe_Virus |
Just read through the init.lua, it overwrites the mapgen settting to always singlenode. |
| 20:05 |
Krock |
as a safety in case someone did not use singlenode |
| 20:06 |
|
Edgy1 joined #minetest-dev |
| 20:06 |
Exe_Virus |
Makes sense, other question: is there an easy way to reimplement perlin noise? (I want to make a rolling hills mapgen so entities can walk without needing to jump using slabs) |
| 20:06 |
Krock |
oh wait https://github.com/SmallJoker/peoplecantlua/blob/master/peoplecantmapgen/init.lua |
| 20:07 |
Krock |
why re-implement? you can use API functions |
| 20:07 |
Exe_Virus |
API has perlin noise exposed to lua? |
| 20:07 |
Krock |
FYI: Doing self-advertising. My code might not be the best, but I know that it works :3 |
| 20:08 |
Krock |
yes |
| 20:08 |
Exe_Virus |
Krock: I appreciate the self-advertising:) |
| 20:08 |
Krock |
https://github.com/minetest/minetest/blob/master/doc/lua_api.txt#L4158 |
| 20:08 |
Krock |
https://github.com/minetest/minetest/blob/master/doc/lua_api.txt#L5539 |
| 20:08 |
Exe_Virus |
Yeah I see those API calls now. Assumed they were engine only for some reason. |
| 20:09 |
Krock |
everything in lua_api.txt may be used for modding |
| 20:09 |
Exe_Virus |
Okay, that should get me started. Hopefully I'll have some luck getting them to look good :) |
| 20:09 |
Krock |
but if you begin to make use of core.run_callbacks it might not be that stable for the future |
| 20:10 |
Krock |
or core.luaentities is also a rather unfortunate public variable |
| 20:10 |
Krock |
this channel is mainly for minetest and minetest_game development. Future questions might be better suited in #minetest or #minetest-hub |
| 20:11 |
Exe_Virus |
interesting, I'll make sure I stay away from core as much as possible, shouldn't be a problem. Thank you, I thought mod talk was done here |
| 20:16 |
Krock |
!next |
| 20:16 |
ShadowBot |
Another satisfied customer. Next! |
| 20:21 |
|
lisac joined #minetest-dev |
| 21:14 |
|
ssieb joined #minetest-dev |
| 22:09 |
|
twoelk joined #minetest-dev |
| 23:34 |
|
Edgy1 joined #minetest-dev |
| 23:47 |
|
Lymkwi joined #minetest-dev |
| 23:51 |
|
Cornelia joined #minetest-dev |
| 23:58 |
|
twoelk left #minetest-dev |
