Time Nick Message 00:01 rubenwardy you could search for it in builtin/mainmenu 00:01 MTDiscord padding allows you to remove vrobs arbitrary limits, however due to rounding errors, you probably should set it slightly negative 00:02 rubenwardy looks like window_width isn't used at all 00:02 rubenwardy another property on that table is shown in the About tab - render info 00:02 MTDiscord so basically use padding to remove vrobs arbitrary limits, then figure out some conversion from screen_width to formspec units, and set that in size[] 00:03 MTDiscord using -0.01 should be good enough, from my testing at least 00:24 MTDiscord coredevs, Im complaining that there is no way to do animation in the mainmenu; there is no analogue to minetest.after or minetest.register_globalstep that I can see. If you know of a way tho? maybe smth with the async functions? 00:25 MTDiscord you cant use a animated texture? 00:26 MTDiscord isnt the download button in content animated? 00:26 MTDiscord yes you can use an animated texture, but that isnt the same thing as animating formspec elements 00:27 MTDiscord sliding movement of elements 00:27 MTDiscord maybe I want the minetest logo to slide in from the top, for example 00:27 MTDiscord well, that can be hacked in 00:27 MTDiscord but, there are many other usecases that cant 00:28 MTDiscord depending on what you want to do, you can use button pressed hacks, see pipeworks toggled, basically you have two buttons and pressed moves the texture half way, to give a animated feel 00:33 MTDiscord https://cdn.discordapp.com/attachments/749727888659447960/960336150667591760/unknown.png 00:33 MTDiscord @MisterE ^ 12:18 BuckarooBanzai Krock: about your latest additions to the `unified_inventory` mod: is the backwards-compat break intended? For example in `5.1.1` the button-group is displayed on the top of the ui: https://pandorabox.io/images/8/81/Unified_inventory_in_minetest_5_1_1.png 12:26 MTDiscord BuckarooBanzai: https://github.com/minetest-mods/unified_inventory/blob/master/mod.conf#L8 12:27 BuckarooBanzai ok, its intentional then ? 12:27 MTDiscord has been 5.4 for a long time 12:28 MTDiscord since VE did the update for formspec v4 12:28 MTDiscord try out fs51 for backwards compat formspecs 12:28 BuckarooBanzai Never had an issue myself, i tend to keep my client up to date, that was a screenshot of another player, not everybody updates stuff regularly :) 12:42 tech_exorcist hello, is it possible to move chests with their content without emptying them, destroying them, placing them where desired, and filling them up again? 12:47 BuckarooBanzai tech_exorcist: use the `technic` wrench or WE ;) 12:48 tech_exorcist BuckarooBanzai: how do I use it without creative mode? 12:49 BuckarooBanzai hm, you don't, afaik it has to explicitly enabled and is only available with `/give` 12:49 BuckarooBanzai i'm assuming you are on the "stock" technic mod, right? 12:50 tech_exorcist I looked up "technic" in the contents tab in MT (5.5.0) and installed the first one 12:51 tech_exorcist the name is "technic" in the list of installed content 12:51 BuckarooBanzai there was some effort in the `mt-mods` technic fork to make them stable and usable within a survival context 12:51 BuckarooBanzai you are playing locally in singleplayer, right? 12:52 tech_exorcist yes 12:53 tech_exorcist in this world in particular I decided to only use resources that can be farmed, such as trees, wheat, cotton, flowers 12:53 BuckarooBanzai can't you just `/give` yourself a few wrenches then? You don't really have an advantage over anything with them IMO 12:53 tech_exorcist sure I can, but what's the technical name to use with /give? 12:54 tech_exorcist sorry, I'm bad at explaining myself 12:56 BuckarooBanzai `/give wrench:wrench` should work 12:56 BuckarooBanzai Sorry: `/giveme wrench:wrench` should work 12:56 tech_exorcist it works 12:57 tech_exorcist and now the furnace and the chest have changed aspect 12:57 tech_exorcist but the technical name is still the same in the debug view 12:58 tech_exorcist when I remove `technic` from the world the chest and furnace should go back to normal, right? (btw, the furnace is the only thing for which I mined non-renewable resources) 12:59 BuckarooBanzai the changed "aspects" for the furnace and chest come most likely from the pipeworks mod i guess 13:01 BuckarooBanzai but yes: removing the mod should revert that but all nodes from the mod you placed will become "unknown nodes" then 13:01 BuckarooBanzai *all the technic-nodes 13:01 tech_exorcist I see, I didn't place any `technical` blocks so I should be good to go (sorry if I mistranslated "aspects" btw) 13:02 BuckarooBanzai no worries :) 17:37 gpcf hi, I got a question regarding formspecs and tabbed UIs 17:38 gpcf is there any relatively lag-free way to create a tabbed UI, other than reinventing the wheel and implementing buttons to pass the pages? 19:00 MTDiscord I tried you solution and maybe I did it wrong, but the chat seems not affected by the change at all. Do I need to have a certain value in the minetest.conf? 19:00 MTDiscord Shouldn't. It occurs to me though that hilariously I was distracted while compiling that patch and never actually finished testing it myself either... 19:04 MTDiscord @Bastrabun sorry, this is the patch that actually works, not the one I posted earlier (back that one out). http://sprunge.us/Ffysfj 19:05 MTDiscord It just changes the startup state of chat to hidden. You can still manually enable it, in case you need to do debuggy stuff. 19:05 MTDiscord Cool, thank you ? Will test that immediately 19:06 MTDiscord Actually, you can see how you can change the initial state of a lot of things in the surrounding lines. Might be really nice if those could be driven from config instead of hard-coded ? 19:17 MTDiscord I'd like that, too. At the moment way too much is hardcoded. Especially the death formspec being hardcoded bugs me these days. 19:18 MTDiscord The second solution worked, the client now started with chat (and hud) disabled ? Thank you very much! 19:18 erle Bastrabun if you want to have more toggle-able-options for stuff hardcoded in vanilla minetest you can look at how waspsaliva does it 19:20 MTDiscord This client is a special case, I use it for streaming to twitch 24/7, it autoreconnects after my ISP throws me out every 24 hours. Yay, German ISP. For more general-purpose stuff I need it to be in the default MT client, so that everyone has it. Waspsaliva heavily uses CSMs, right? 19:21 erle yeah, the idea is to make extra CSM APIs and then have CSMs for it. it is probably a good i dea to use it for streaming if you like stuff like tracers (the stuff that minecraft streamers use). 19:21 erle i used tracers for a while to not get surprised by other placers 19:22 erle there are more evil things of course 19:22 erle like a CSM that records where you saw a player last 19:22 erle it's unfortunately quite easy to build a lava prison there 19:23 erle Bastrabun my fav ws CSM is the one that does “put the node in hand 1 layer thicc on all nodes with the same name that you can reach” 19:23 erle because that way you only need to layout the foundations of your castle 19:24 erle and can basically move along the walls having it build up 19:24 erle making towers etc. much less tedious 19:24 erle and it's such a simple mechanism! 19:24 MTDiscord As long as CSMs are used mainly for cheating and automating stuff (instead of actually playing the game), I can't enable CSMs on the server. Which unfortunaltey rules out all the legitimate uses for CSMs too. I haven't found a solution except waiting for server side CSMs. 19:25 erle wait how is this cheating? 19:25 erle it's doing the same thing that you'd do manually, you still need to have the materials 19:25 erle just saving you clicks 19:25 MTDiscord It's an afk mechanic - I don't want those 19:25 erle wdym afk? people use it to skip grinding 19:26 erle and since the reach is like 5 nodes or so, you can't really to much afk with it 19:27 erle well, i guess this is another “virtual legos are a neoliberal utopia and you only *honest* work should be rewarded, automating grinding is not honest” conversation, in which case, i am done here. 19:27 erle oh wait, i am not 19:27 MTDiscord I think that's a gamedesign decision, not a political one. 19:28 erle i greatly admire the developers of crawl (the roguelike), they have one button for “go to the nearest location you have not yet explored“ and one for “hit the closest enemy” 19:28 erle then they designed their game around that hehe 19:28 erle so basically … you can hit those buttons a lot, but you'll not get far 19:28 MTDiscord Entirely legit. Rmember the autofire buttons on joysticks? 19:28 erle no, because i doubt i ever had one 19:29 erle i'd use keyboard macros 19:29 erle i did once make a fishing bot that just rightclicked a lot lol 19:29 erle proving that fishing is a stupid mechanic 19:33 MTDiscord I had all kinds of discussions with people who wanted to automate things. I'm ok with thumb-savers, or stuff that enables people that lost a finger or have poor eyesight. I'm not ok with people downloading a schematic and then have it placed into the world while they sleep. Ohter may want it the other way round. I want art and poetry, not copy and recitation. 19:40 erle Bastrabun, i think cheating is not purely automation, but stuff that takes the player out of the game loop 19:40 definitelya_ SSCSM sounds a bit scary to me; could you get hacked really easily with it, if you join a malicious server? 19:40 erle i.e. autoforward: okay. autofly: okay. autoforward that avoids ravines/lava/mobs: cheating. 19:41 definitelya_ nvm, this is one of those neverending discussions I'm entering I think lol. 19:41 erle definitelya_, SSCSM is an idea that no one knows how to implement securely. my proposal for a compromise is that minetest should come with a bunch of CSMs (like mumble and rumble and chatlog) that can be activated or contentdb can have vetted CSMs that are Definitely Not Cheating™. 19:42 erle so far no one said they find it good 19:42 erle which probably means everyone thinks it sucks ^^ 19:42 erle but i'm not the only person with an opinion, so … :D 19:42 definitelya_ erle: See my concerns above. 19:43 definitelya_ of course! 19:43 erle they should definitely not be server-sent 19:43 erle that's a recipe for disaster 19:45 Hawk777 How is server-sent Lua in a Minetest client any qualitatively different, security-posture-wise, from server-sent Javascript in a web browser? 19:46 erle not at all 19:46 erle and server-sent javascript in a web browser is an unworkable security nightmare 19:46 definitelya_ Maybe we could mandate it all be open source at least. 19:46 Hawk777 Your opinion is very very niche. It is, generally speaking, an accepted phenomenon, regardless of your personal feelings on the matter. 19:46 erle giant companies spend millions of eurodollars on it and many caffeinehours and they can't get it right 19:46 erle uh, it's not a feeling thing 19:47 erle it's a *fact* thing 19:47 erle definitelya_, does not matter, the complexity matters 19:47 Hawk777 The *fact* is that Javascript in web browsers is an accepted thing in the general world today. 19:47 erle Hawk777 go read Security Applications of Formal Language Theory, Len Sassaman, Meredith L. Patterson, Sergey Bratus, Michael E. Locasto, Anna Shubina [Dartmouth Computer Science Technical Report TR2011-709], published in IEEE Systems Journal, Volume 7, Issue 3, Sept. 2013 19:47 Hawk777 Why? 19:47 erle https://langsec.org/papers/langsec-tr.pdf 19:48 erle because then you will understand why a) we have to live with the javascript b) you should not create another abomination like that 19:48 erle basically, once you switch to a stupid system, you can't really undo it 19:48 definitelya_ erle: Yes, but if an SSCSM is auditable, it can be vetted and accepted officially, maybe. 19:48 erle also i want to sleep 19:49 definitelya_ gn 19:49 erle definitelya_ no, not even then 19:49 Hawk777 Meh, I don’t have time to read a 31-page paper. Bye. 19:49 erle look just read that PDF, it explains in very clear terms why executing arbitrary code is not a good idea 19:49 definitelya_ xD 19:49 erle i.e. computers were a mistake 19:50 Hawk777 My opinion doesn’t matter anyway, I’m not making any of the decisions. 19:50 definitelya_ fire was a mistake. 19:50 definitelya_ go back to monke. 19:50 erle Hawk777 if you want to debate this, you need to have a solid foundation 19:50 erle regardless of if your opinion matters, you'll ask smarter questions if you can answer the dumber ones yourself 19:51 definitelya_ Hawk777: ditto 19:51 erle and maybe figure out how a system that *would* work well would look like 19:51 erle (hint: entirely unlike javascript lol) 19:52 Hawk777 Fine, fine, you’ve poked at me enough to add that paper to my reading list. Gonna be months before I get to it though, possibly years. My reading list is long. 19:52 erle Hawk777 if you read it, follow it up with The Seven Turrets of Babel: A Taxonomy of LangSec Errors and How to Expunge Them, Falcon Darkstar Momot, Sergey Bratus, Sven M. Hallberg, Meredith L. Patterson, in IEEE SecDev 2016, Nov. 2016, Boston. 19:53 erle Hawk777 http://langsec.org/papers/langsec-cwes-secdev2016.pdf 19:53 Hawk777 And since you don’t care about my opinion if I haven’t read it, I obviously have nothing more to say to you. 19:53 definitelya_ Hawk777: Don't do tomorrow what you could complete today! /s 19:53 erle Hawk777 you'd be well-advised to read both of those papers before designing a new data format or network protocol in any case 19:54 erle i mention this just in case your reading list is sorted by urgency 19:54 erle or applicability 19:54 Hawk777 It isn’t, and even if it were, it probably wouldn’t be ordered by $random_internet_person’s idea of urgency. 19:55 Hawk777 But I am interested enough to read them sometime. 20:49 kabou why are formspecs placed in node meta instead of node definitions? 20:50 kabou what am I missing about the need to have a different formspec for same blocks at different map positions? 20:50 kabou would it not make more sense (and save map space) to put formspecs inside the node definitions? 20:57 MTDiscord furnaces? 20:57 MTDiscord You can't tell how much fuel is in a furnace, or how far along it is in the process, at node registration time. 21:00 kabou yeah furnaces 21:01 kabou but the amount of fuel etc isn't that kept in the inventory lists of the node meta? 21:01 MTDiscord The formspec needs to be updated to change the arrow/flame graphics 21:02 MTDiscord you could, in principle, register a function in the node def to calculate what the updated formspec should be ... but that's basically on_timer already :-) 21:03 MTDiscord The alternative is that you'd have to have every formspec element need to have some way to pull its parameters from an inventory list or other meta field. At that level of complexity, it's better to just pull the entire formspec out of one meta field... 21:06 kabou i still find it weird.  the formspec could be simply a hud thing that is updated by timers and other callbacks defined in the node definition 21:08 kabou state that is pertininent to specific nodes at specific map positions can be put in the node meta (like the node timers are) without having to put all of the basically static formspec definition copied in every map pos 21:09 sfan5 without lua callbacks or prerendered formspecs you can barely convert state into the matching formspec 21:10 sfan5 callbacks were probably avoided for client-side prediction reasons 21:10 sfan5 but many nodes don't really need or want this so a formspec in the nodedef would make sense for them, I have implementing this on my TODO list 21:12 kabou well I am not familiar with the whole history of the thing and the design choices that made it into what it is, so I'll have to go with what you guys say 21:12 kabou it just appeared to me to be somewhat weird how it is when  I looked at the furncaes implementation 21:14 kabou one of the problems I noticed is that when the formspec gets changed or updated in newer versions of the code, you have to update all the node metas with an lbm or some other trick 21:21 MTDiscord Being able to pull it from node meta instead of having to painstakingly set it in the node every time you on_construct one (and bloating mapblock sizes) would be helpful, yeah, even if it's not a 100% substitute. 21:22 MTDiscord I could see a use-case for supporting both (check meta first, fallback to def) where the def one just provides an initial default, and then formspec interactions could choose to replace it on a per-node basis. 21:23 kabou agree, that would be a neat feature to have a per map node override 21:24 kabou while providing the basic ability to not repeat the same data unnecessarily 21:25 kabou and allowing seamless upgrades of game/mod code without the need to check every map node