| Time |
Nick |
Message |
| 00:21 |
|
ch98 left #minetest-dev |
| 00:38 |
kahrl |
wondering how to best choose an available request_id in httpfetch |
| 00:39 |
kahrl |
my current plan is to keep a table that maps a request_id to the callback function |
| 00:39 |
kahrl |
then use lua_objlen()+1 to get a free id |
| 00:39 |
kahrl |
that should work, yes? |
| 00:40 |
kaeza |
kahrl, this may come in handy: http://www.lua.org/manual/5.1/manual.html#luaL_ref |
| 00:41 |
kahrl |
interesting |
| 00:44 |
kahrl |
ah, it uses lua_objlen in conjunction with a free list |
| 00:45 |
kaeza |
according to the guys at #lua, luaL_ref has some optimization under LuaJit |
| 00:46 |
kaeza |
(though in this case may not be particularly critical) |
| 00:47 |
kahrl |
yeah |
| 00:48 |
kahrl |
I'm mostly looking for something that is less than linear, in case some mod screws up and keeps adding requests to the queue |
| 00:49 |
kahrl |
(but then you're screwed anyway, because httpfetch serves the requests first-come-first-serve) |
| 04:08 |
|
Miner_48er joined #minetest-dev |
| 06:16 |
|
Ritchie joined #minetest-dev |
| 06:23 |
|
neko259 joined #minetest-dev |
| 06:56 |
|
Akien joined #minetest-dev |
| 07:01 |
|
iqualfragile joined #minetest-dev |
| 07:01 |
|
Calinou joined #minetest-dev |
| 07:10 |
|
emptty joined #minetest-dev |
| 07:29 |
|
iqualfragile joined #minetest-dev |
| 08:08 |
|
proller joined #minetest-dev |
| 08:13 |
|
darkrose joined #minetest-dev |
| 08:13 |
|
darkrose joined #minetest-dev |
| 08:43 |
|
iqualfragile joined #minetest-dev |
| 09:16 |
|
Taoki[laptop] joined #minetest-dev |
| 09:31 |
|
jin_xi joined #minetest-dev |
| 09:46 |
|
neko259 joined #minetest-dev |
| 10:02 |
|
Calinou_ joined #minetest-dev |
| 10:36 |
|
Jordach joined #minetest-dev |
| 10:57 |
|
Calinou joined #minetest-dev |
| 11:37 |
|
Zeitgeist_ joined #minetest-dev |
| 12:07 |
|
kaeza joined #minetest-dev |
| 13:39 |
|
jojoa1997 joined #minetest-dev |
| 14:32 |
|
mrtux joined #minetest-dev |
| 14:36 |
|
PilzAdam joined #minetest-dev |
| 14:43 |
|
hmmmm joined #minetest-dev |
| 15:02 |
|
Calinou joined #minetest-dev |
| 15:03 |
|
jojoa1997 joined #minetest-dev |
| 15:11 |
|
Akien joined #minetest-dev |
| 15:25 |
|
diemartin joined #minetest-dev |
| 15:51 |
|
BlockMen joined #minetest-dev |
| 16:56 |
|
Miner_48er joined #minetest-dev |
| 16:59 |
|
kaeza joined #minetest-dev |
| 17:06 |
VanessaE |
https://github.com/minetest/minetest/issues/885 <--- this has the potential to break LOTS of mods |
| 17:11 |
|
nyuszika7h joined #minetest-dev |
| 17:22 |
|
iqualfragile joined #minetest-dev |
| 17:29 |
PilzAdam |
thexyz, when do you plan to rebase and push the weblate commits? |
| 17:35 |
thexyz |
oh, right |
| 17:36 |
thexyz |
guess I'll do that tomorrow |
| 17:39 |
thexyz |
apparently, new weblate versions have rebase support |
| 17:50 |
|
sapier joined #minetest-dev |
| 18:07 |
|
sapier joined #minetest-dev |
| 18:16 |
|
sapier joined #minetest-dev |
| 18:20 |
|
init joined #minetest-dev |
| 18:41 |
|
BlockMen left #minetest-dev |
| 19:10 |
|
Ritchie joined #minetest-dev |
| 19:34 |
|
iqualfragile joined #minetest-dev |
| 20:15 |
|
iqualfragile_ joined #minetest-dev |
| 20:20 |
|
ecube joined #minetest-dev |
| 20:31 |
|
iqualfragile joined #minetest-dev |
| 20:54 |
|
iqualfragile joined #minetest-dev |
| 20:55 |
PilzAdam |
VanessaE, https://github.com/minetest/minetest/issues/885 |
| 20:55 |
VanessaE |
strange |
| 20:56 |
VanessaE |
well I fixed the affected recipes already, but something ain't right |
| 20:56 |
PilzAdam |
are the actual nodes defined when you register your recipes with the aliases? |
| 20:56 |
VanessaE |
maybe it only affects items, not nodes' |
| 20:57 |
VanessaE |
they should be - the mod that uses them has the other mod as an optional dep |
| 20:57 |
PilzAdam |
have you checked for typos? |
| 20:57 |
VanessaE |
(gloopblocks from yesterday or older, vs. glooptest) |
| 20:58 |
VanessaE |
the recipes were fine - UI's crafting guide says the recipe was valid. |
| 21:01 |
VanessaE |
sokomine's colormachine vs. andersje's fork of the stained_glass mod had the same issue, and for the same reason |
| 21:05 |
VanessaE |
nodes were renamed and aliased back to their original names, but the recipes ceased to work after that. |
| 21:07 |
kahrl |
you have to alias the original name to the new name |
| 21:08 |
VanessaE |
kahrl: minetest.register_alias("gloopores:kalite_lump", "glooptest:kalite_lump") |
| 21:09 |
VanessaE |
(glooptest: is the correct namespace now) |
| 21:09 |
kahrl |
that's right then |
| 21:09 |
VanessaE |
that's the alias that was affected. the recipe called for gloopores:kalite_lump, sticks, and coal |
| 21:10 |
VanessaE |
I had to change it to use glooptest:kalite_lump in order to make it work |
| 21:14 |
PilzAdam |
VanessaE, can you find the commit that breaks it for you? |
| 21:15 |
VanessaE |
not right now, no |
| 21:15 |
VanessaE |
I'm not even sure how long it's been broken, this is the first I'd seen of it |
| 21:15 |
|
iqualfragile joined #minetest-dev |
| 21:22 |
|
SpeedProg joined #minetest-dev |
| 21:30 |
VanessaE |
PilzAdam: can you take a look at https://github.com/GloopMaster/glooptest vs. https://github.com/VanessaE/gloopblocks/commit/2b90ff9185543703331e5b8a709b5aa8cdff43a3\ |
| 21:30 |
VanessaE |
er lose the \ |
| 21:31 |
VanessaE |
install these, back gloopblocks down to 2b90ff91, and see what it does |
| 21:33 |
PilzAdam |
VanessaE, what recipe should I test? |
| 21:33 |
VanessaE |
the one for the evil stick |
| 21:33 |
VanessaE |
one kalite lump, one regular stick, one lump of coal |
| 21:34 |
VanessaE |
(shapeless) |
| 21:35 |
VanessaE |
https://github.com/VanessaE/gloopblocks/blob/2b90ff9185543703331e5b8a709b5aa8cdff43a3/crafts.lua#L58 |
| 21:35 |
VanessaE |
there's the recipe as it existed before I fixed it to avoid the alias. |
| 21:37 |
PilzAdam |
ERROR[main]: mod "glooptest" has unsatisfied dependencies: "glooptest" |
| 21:37 |
|
iqualfragile joined #minetest-dev |
| 21:37 |
VanessaE |
so install glooptest :P |
| 21:37 |
PilzAdam |
thats your branch |
| 21:37 |
|
Miner_48er joined #minetest-dev |
| 21:37 |
VanessaE |
it's odd because, depends.txt clearly has "glooptest?" in it. |
| 21:39 |
PilzAdam |
Name gloopblocks:rainbow_block does not follow naming conventions: "modname:" or ":" prefix required |
| 21:39 |
VanessaE |
eh? |
| 21:39 |
kahrl |
an optional dependency means: if the depended on mod is enabled in the world configuration (or exists as a game/world mod), make it an actual dependency; otherwise drop it |
| 21:39 |
VanessaE |
PilzAdam: that isn't happening here |
| 21:39 |
PilzAdam |
oh, your repo is gloopblocks and not a fork of glooptest |
| 21:39 |
VanessaE |
PilzAdam: correct - you need both. |
| 21:40 |
|
VanessaE joined #minetest-dev |
| 21:40 |
VanessaE |
oops. |
| 21:41 |
VanessaE |
gloopblocks uses glooptest (formerly gloopores) to get the kalite from. |
| 21:42 |
PilzAdam |
default:stick + default:coal_lump + Kalite lump recipe (shapeless) does not work in master of gloopblocks |
| 21:42 |
VanessaE |
exactly. |
| 21:42 |
VanessaE |
it works in gloopblocks HEAD because I changed it to avoid the aliased item. |
| 21:43 |
|
iqualfragile joined #minetest-dev |
| 21:43 |
PilzAdam |
ehm, whats the difference between master and HEAD? |
| 21:43 |
sapier |
https://github.com/minetest/minetest/pull/887 fix infinite loop in pathfinder |
| 21:44 |
VanessaE |
well the link I gave you was one commit behind HEAD, before that fix. |
| 21:44 |
sapier |
I guess noone except mobf used pathfinder by now ;-P |
| 21:44 |
VanessaE |
it won't work in that state because of this bug or whatever it is. |
| 21:44 |
kahrl |
what does 'cat debug.txt | grep "setting alias"' say? |
| 21:44 |
VanessaE |
surely you know master = the name of the brach, HEAD = the position in that repo |
| 21:44 |
PilzAdam |
gloopblocks/crafts.lua:65 is "glooptest:kalite_lump" for me, and the recipe does not work |
| 21:44 |
kahrl |
(after deleting debug.txt + running the game once) |
| 21:45 |
VanessaE |
12:54:31: VERBOSE[main]: ItemDefManager: setting alias gloopores:kalite_lump -> glooptest:kalite_lump |
| 21:46 |
PilzAdam |
that recipe doesnt work, even without the alias |
| 21:46 |
VanessaE |
the one that uses glooptest:kalite_lump works fine for me, I tested that early this morning. |
| 21:47 |
VanessaE |
the one that uses gloopores:kalite_lump doesn't. |
| 21:47 |
kahrl |
can you try making it shaped? |
| 21:47 |
PilzAdam |
oops, forgot to enable gloopblocks |
| 21:47 |
VanessaE |
in a little while, sure. watching a movie. |
| 21:48 |
VanessaE |
PilzAdam: derp :) |
| 21:48 |
|
iqualfragile joined #minetest-dev |
| 21:49 |
PilzAdam |
kahrl, it works with the alias if its shaped |
| 21:49 |
VanessaE |
ok so you can confirm it doesn't work if it's shapeless? |
| 21:49 |
PilzAdam |
yea |
| 21:49 |
VanessaE |
ok good. |
| 21:49 |
VanessaE |
so I'm not crazy then :) |
| 21:49 |
VanessaE |
stained_glass btw is also shapeless. |
| 21:49 |
PilzAdam |
no worries, you are still crazy |
| 21:49 |
VanessaE |
that's probably why it was b0rk3d there too |
| 21:49 |
VanessaE |
:P |
| 21:50 |
PilzAdam |
why does generating mapblocks take so damn long with glooptest? |
| 21:50 |
VanessaE |
dunno |
| 21:51 |
VanessaE |
maybe the treasure chests |
| 21:51 |
PilzAdam |
more than 20 seconds per mapblock |
| 21:51 |
VanessaE |
I have those disabled on my maps |
| 21:51 |
kahrl |
apparently CraftDefinitionShaped uses craftGetItemNames to follow aliases, but CraftDefinitionShapeless doesn't |
| 21:52 |
VanessaE |
leave it to me to spot the corner cases and obscure bugs :D |
| 21:52 |
VanessaE |
(if this can truly be called a bug :) ) |
| 21:56 |
kahrl |
I guess this should fix it: https://gist.github.com/kahrl/6330631 |
| 21:58 |
PilzAdam |
kahrl, yep, that works |
| 21:59 |
VanessaE |
any idea how long this state has existed? |
| 21:59 |
kahrl |
at least since group support for crafting has been added, I think |
| 21:59 |
kahrl |
maybe longer |
| 22:00 |
VanessaE |
naw, can't be that long ago |
| 22:00 |
VanessaE |
(could it?) |
| 22:00 |
PilzAdam |
https://github.com/minetest/minetest/commit/0c91a0d59db70f3f502004d4c37fecd4e10c9401 |
| 22:00 |
VanessaE |
huh. |
| 22:05 |
|
iqualfragile joined #minetest-dev |
| 22:09 |
|
kaeza joined #minetest-dev |
| 22:20 |
kahrl |
sapier, that pathfind fix looks reasonable, but I don't have a clue about the pathfinding code, does anyone else? |
| 22:21 |
|
iqualfragile joined #minetest-dev |
| 22:22 |
sapier |
no |
| 22:23 |
sapier |
I guess I'm still the only one who at least somehow understands it |
| 22:23 |
sapier |
and noone seems to have used it by now too |
| 22:23 |
PilzAdam |
AFAIK jin_xi has used it for a streets/path mod |
| 22:24 |
sapier |
I wonder why he didn't run into that bug mobf triggers it very soon |
| 22:24 |
sapier |
-soon + quick |
| 22:26 |
sapier |
the problem is once one of those continue situations happens it's never cleared as direction isn't updated |
| 22:32 |
|
iqualfragile_ joined #minetest-dev |
| 22:39 |
|
iqualfragile joined #minetest-dev |
| 22:42 |
sapier |
https://github.com/minetest/minetest/pull/889 should fix issue 857 |
| 22:45 |
kaeza |
it's interesting that the paste on #857 referenced tons of other issues :P |
| 22:45 |
sapier |
https://github.com/minetest/minetest/pull/884 "low_priority_enhancement" you made a hell out of modmgr layout beeing that bad and now it's classified that way? |
| 22:57 |
Miner_48er |
what is num_emerge_threads compatible with in the code? only map generation? |
| 22:58 |
PilzAdam |
Miner_48er, that setting is only for mapgen, yes |
| 23:31 |
|
jojoa1997 joined #minetest-dev |
| 23:31 |
kahrl |
sapier: btw, a while ago a pushed support for the {engine,minetest}.httpfetch api to my httpfetch branch |
| 23:32 |
kahrl |
was too tired to continue doing the rest of the plan |
| 23:32 |
VanessaE |
push to upstream and let the world test it :-) |
| 23:32 |
kahrl |
heh, not yet |
| 23:56 |
|
iqualfragile joined #minetest-dev |
