Time Nick Message 02:15 NoInternet Hello? 04:54 Guest485 merging #9756 04:54 ShadowBot https://github.com/minetest/minetest/issues/9756 -- Add PostgreSQL authentication backend by nerzhul 04:58 Guest485 merging #9759 04:58 ShadowBot https://github.com/minetest/minetest/issues/9759 -- Various cleanups by sfan5 05:02 Guest485 merging #9748 05:02 ShadowBot https://github.com/minetest/minetest/issues/9748 -- Script: Enforce type checks if not nil by SmallJoker 06:35 Guest63951 sfan5 #9719 is now ready, your latest comments has been taken in account 06:35 ShadowBot https://github.com/minetest/minetest/issues/9719 -- Add MetricsBackend with prometheus counter support by nerzhul 06:49 nerzhul_ it seems my bridge has some weird issues, why my nick is not available :o 09:27 Guest71878 merging #9762 09:27 ShadowBot https://github.com/minetest/minetest/issues/9762 -- Forbid object:attach(obj, ...) by nerzhul 10:47 Wuzzy Is it possible that hte output of minetest.get_position_from_hash will change in future versions? 10:50 sfan5 I'm sure that'd break some mods, so no 11:16 ANAND !tell Krock #8817 wants to be your friend 11:16 ShadowBot ANAND: O.K. 11:21 ANAND I'm in need of some inputs regarding #8755 11:21 ShadowBot https://github.com/minetest/minetest/issues/8755 -- [NO SQUASH] ItemDefinition: Add wield_offset and wield_rotation by ClobberXD 11:21 ANAND Pls halp 11:27 ANAND If possible, I'd also like some help with fixing the merge conflict in #7924 - I've tried to fix it thrice but I ain't getting anywhere :) 11:27 ShadowBot https://github.com/minetest/minetest/issues/7924 -- [NO SQUASH] Allow binding dig, place actions to keys; remove LMB/RMB hardcoding by ClobberXD 12:26 ANAND sfan5: Are you planning on reviewing #8271? :D 12:26 ShadowBot https://github.com/minetest/minetest/issues/8271 -- Fix breath statbar scaling; defer breath_bar hiding by one second by ClobberXD 12:31 sfan5 not sure if that's the ideal solution really 13:26 ANAND Ok, what do you have in mind? 14:10 sfan5 I don't know 14:16 ANAND Mhm 14:16 ANAND So what part of the PR do you think feels wrong? 14:21 sfan5 the part where hiding the bar is delayed by 1s because it's broken otherwise 14:21 sfan5 wait no 14:22 sfan5 that's not the intention here 14:23 sfan5 however doesn't this PR mean that the player has 1 less "breath" when entering water? 14:24 ANAND It does, in a way 14:25 ANAND One sec 14:28 ANAND I assume it's supposed to be 10, but it was increased to 11 just so that all 10 bubbles are visible before the statbar vanishes 14:28 sfan5 yes 14:30 ANAND This PR does reduce max_breath by 1, because the reason it was extra before has been fixed 14:31 sfan5 what happens if you regain the breath bar while the 1s to hide it has not expired yet? 14:31 ANAND To preserve the current behaviour, we could keep max_breath at 11, and only merge the fix 14:32 sfan5 nah I don't think that is necessary 14:33 ANAND I'll check that scenario now 14:40 ANAND sfan5: It looks like the same statbar is being used again, if the breath starts decreasing before the statbar disappears 14:42 ANAND Been dipping my head in and out of water for more than 2 mins. Seems consistent; nothing seems broken. 14:42 sfan5 I'd have expected a second statbar 14:48 ANAND Tbh, I expected the same thing, but the code is smart enough to cache the HUD element IDs and update them accordingly 15:15 ANAND So, sfan5, what do you think? 15:15 sfan5 the pr is fine 15:17 ANAND Good to know, thanks 15:22 Wuzzy What do i have to do to get DevTest merged? #9450 15:22 ShadowBot https://github.com/minetest/minetest/issues/9450 -- Replacement for Minimal Development Test (PR) by Wuzzy2 15:23 rubenwardy post it in -dev... oh wait 15:23 sfan5 I think ruben wanted to review it ;) 15:23 rubenwardy I have a few times 15:23 rubenwardy haven't got back to looking at it again 15:39 sfan5 merging #9749 in 10m 15:39 ShadowBot https://github.com/minetest/minetest/issues/9749 -- Fix detection of in-place path_locale when RUN_IN_PLACE=0 by sfan5 15:58 Wuzzy rubenwardy: look again, please 15:59 Wuzzy Or do you do a full review now? 16:03 rubenwardy done 16:12 nerzhul sfan5 i wait for you on the prom PR, i want it to be closed before working on other topics which may require prom instrumentation 16:30 Guest46840 i'm think again about our nice Server class reduction, and one first thing i want to change, is the ClientInterface. I want to rename it to SessionMgr, as it's just a Session manager for our minetest client sessions 16:42 nrzkt i will take a look after on this rename, first i will try to cleanup the server constructor 16:49 sfan5 nrzkt: yes I'll look at it after I'm done with this 16:49 nrzkt fine 16:51 Krock . 16:51 Krock ShadowBot: where's tell? 17:13 sfan5 apparently prometheus-cpp is not in the Arch repos or the AUR 17:14 sfan5 not really a good sign for software availability 17:18 nrzkt yep problem with c++ world and modern techs :( 17:18 nrzkt but it's the promoted one by prom 17:18 nrzkt https://prometheus.io/docs/instrumenting/clientlibs/ 17:19 sfan5 prometheus-cpp also pulls in https://github.com/civetweb/civetweb, wonder how many LOC that is 17:23 sfan5 hm PROMETHEUS_CPP_INCLUDE_DIR seems to be ignored, or I did something wrong 17:25 nrzkt you have a compilation issue ? 17:25 sfan5 nrzkt: did you think about moving the PrometheusMetricCounter, PrometheusMetricGauge and PrometheusMetricsBackend implementations to the .cpp file to avoid including prometheus headers everywhere? 17:26 nrzkt yep it can be nice to prevent embed prometheus object everywhere, it's right. Have you got time to amend this long time worked pr ? :) 17:27 sfan5 sure that's easy, I can do that 17:27 nrzkt i'm trying to remove the InventoryMgr composition on server, not so easy but should have a PR tomorrow :) 17:28 nrzkt spliting inventory manangement responsibility to a specific object 17:28 sfan5 also about the default port: 8080 is a common webserver port, I think e.g. 9100 would be better 17:28 sfan5 or do you think this isn't an issue? 17:28 nrzkt or put 30000 and remove the sed in docker ? :D 17:28 nrzkt like the MT UDP port ? :D 17:28 sfan5 hmm 17:29 sfan5 yeah that's be okay too 17:29 sfan5 that'd* 17:34 sfan5 how do I get the data to actually show up in the prometheus UI? 17:38 sfan5 oh wait 17:50 nrzkt curl localhost:30000/metrics 17:50 nrzkt you will have prometheus export 17:51 nrzkt if you want to see them in prometheus itself you need to configure a target in a prometheus storage server to collect metrics 17:51 rubenwardy I'll write a tutorial at some point 17:51 rubenwardy but I don't know how to interprete some of the stats in a useful manner and nerzhul keeps resolving my questions on it 17:52 sfan5 hm apparently std::atomic doesn't support double 17:53 nrzkt sfan5: hmm it compiled on my machine which is also a archlinux, wtf 17:54 sfan5 if you change it to use fetch_add I mean 17:54 sfan5 "m_counter = m_counter + number" is not thread-safe 18:02 nrzkt m_counter += you mean ? it deosn't work on atomic 18:02 sfan5 += would be safe but std::atomic doesn't implement that 18:02 nrzkt yeah :( 18:03 Krock when Atomic cannot help it's time for good old' Mutex 18:03 nrzkt yep 18:04 nrzkt can you comment the PR sfan5... i'm tired this to not be merged again, and i'm working on other code parts, i may forgot one info 18:20 sfan5 nrzkt: https://github.com/minetest/minetest/pull/9719/commits/97601cb0d4d7e33f21f3de5428935e49ee7a0e50 18:21 sfan5 when you look at the PR again check if it forgot anything 18:21 nrzkt is the brace style on the class you put in the cpp file proper ? 18:21 nrzkt very nice way to hide the implementation, cool :) 18:22 sfan5 no idea 18:22 nrzkt you can inline the constructor in the class def directly i think 18:22 nrzkt as all is in the cpp file and it's short constructors, i think it's acceptable 18:23 nrzkt appart that very nice patch thanks :D 18:25 nrzkt our inventory sending code is not very very nice. My PR is just to refactor the ownership of the inventory management but i think we may need to have more proper algo later 18:28 nrzkt ah the cleanup may be quite easy 18:28 sfan5 merging #9765, #9668 in 15m 18:28 ShadowBot https://github.com/minetest/minetest/issues/9765 -- Remove /LTCG from when compiling with clang-cl by Adriankhl 18:28 ShadowBot https://github.com/minetest/minetest/issues/9668 -- Collision information for Lua entities by sfan5 18:31 rubenwardy can you have TCP and UDP on the same port? 18:31 rubenwardy googled it, and yes 18:31 nrzkt yep it's not the same layer 4 protocol 18:40 nrzkt sfan5, i will take the refine on the prom part tomorrow 18:44 sfan5 sure 20:24 sfan5 Wuzzy: creative mode doesn't work in devtest, is this intentional? 20:24 Wuzzy define "creative mode works" 20:25 sfan5 I place a "Light Source (14)" and it disappears from my inventory despite creative_mode = true 20:25 Wuzzy Creative Mode does not do anything 20:25 Wuzzy except engine magic 20:26 Wuzzy for infinite note placement, say "/infplace" in chat 20:26 sfan5 creative is handeled in builtin, you have to explicitly break it 20:26 Wuzzy Creative Mode does almost do nothing by default 20:26 Wuzzy the only thing it does is disable tool wear. the rest comes from game 20:27 sfan5 hm right 20:27 sfan5 it would be nice if infplace could default to core.is_yes(core.settings:get("creative_mode")) though 20:28 Wuzzy what's wrong with get_bool? 20:28 sfan5 nothing 20:28 Wuzzy sfan5: well its the Principle of Minimal Interference at work here 20:30 Wuzzy sfan5: infplace can be perma-enabled in settings tho 20:31 sfan5 are there supposed to be zero sounds in devtest? 20:32 Wuzzy there are sounds 20:32 sfan5 oh the eat sounds item makes one 20:32 sfan5 but walking is quiet everywhere 20:33 Wuzzy sounds are only used in the sound test nodes 20:34 sfan5 how do I get rid of the liquid test sound node? it's not buildable_to 20:34 Wuzzy bucket 20:34 sfan5 yeah I know but IMO keeping maybe grass, dirt + stone walk sounds would've been nice 20:34 sfan5 hm 20:35 Wuzzy original minimal only had dirt sound 20:36 sfan5 well the just that one 20:38 Wuzzy i only added 1 sound file to keep file size small. 20:40 Wuzzy sfan5 I'm currently playing. please collect all complaints and post them in the issue thanks 20:40 sfan5 sure 20:41 Wuzzy thx 20:42 Wuzzy but feel free to ping me IF its important 20:42 sfan5 I can't figure out what the formspec tester node and item do 20:42 sfan5 not sure if you consider that important 20:47 Wuzzy just dummy items for testing image formspec elements 21:26 sfan5 p_gimeno: are there any backwards compatibility considerations with https://github.com/minetest/minetest/pull/8468? 21:27 p_gimeno sfan5: not before 5.0 21:27 p_gimeno that broke with 5.0 21:27 sfan5 :/ 21:28 p_gimeno I need to go over it once again, to better remember the details 21:28 sfan5 would be appreciated 21:34 p_gimeno ok, so if my understanding is right (without launching MT and creating a test case to verify), current behaviour is that objects spin over a vertical axis regardless of how they are currently oriented (set_rot). 21:35 p_gimeno set_rot was added for 5.0, so it's possible that content has created that uses set_rot and spins. 21:35 p_gimeno has been* created 21:36 p_gimeno So there's that compatibility concern. 21:39 p_gimeno The probability isn't very big because making a rotated object spin over a vertical axis has very few use cases. 21:42 sfan5 If the old "feature" is unusably broken I can accept that as not a compatibility break 21:43 p_gimeno the only use case I can only think of is a hack to "repair" a badly made mesh 21:43 p_gimeno one that had the wrong axis pointing up, and is fixed by using set_rot to make it point up so it rotates properly 21:44 sfan5 oh and also, since that PR somehow changes behaviour, does lua_api.txt need to be updated? 21:45 p_gimeno maybe to document the new possibilities that are open 21:45 p_gimeno e.g. propellers are one example that can be done with relative rotation but not with the current absolute one 21:46 p_gimeno otherwise, I don't think it was ever specified how automatic_rotate spins, let me check 21:46 p_gimeno yeah it's very brief 21:46 p_gimeno automatic_rotate = 0, 21:46 p_gimeno -- Set constant rotation in radians per second, positive or negative. 21:46 p_gimeno -- Set to 0 to disable constant rotation. 21:47 p_gimeno maybe add: "Rotates over the local Y axis of the object" 21:48 sfan5 sounds good 21:53 p_gimeno there's a kind of issue that there's no control over the actual rotation, e.g. if you start the rotation and then stop it, it will stay at the angle it stopped, on top of what set_rot says 21:53 p_gimeno I thought that would not be a problem 21:54 p_gimeno things that start and stop spinning typically stay at the point where they stopped, like fan blades 21:55 p_gimeno that's purely client side; relogging would reset the rotation 22:18 p_gimeno hm, stujones11 has a point here about mesh animations: https://github.com/minetest/minetest/issues/8456#issuecomment-481415295 22:19 p_gimeno oh and itemframes pedestal should be tested with the PR 22:19 p_gimeno I think it should work, but it never hurts to check 23:11 Wuzzy sfan5: updated `devtest` repository, but not PR