Time  Nick       Message
02:52 Hijiri     what is wuzzy referring to here by "fine pointing"? https://forum.minetest.net/viewtopic.php?p=273554#p273554
02:53 cheapie    Hijiri: I think that's the "player pointed at (X,Y) position on this face of this node" instead of just "player pointed at this face of this node"
02:54 cheapie    (if that makes any sense)
03:08 Hijiri     cheapie: what is it called in the API?
09:56 SG9kb3I    I have just one question. Why is my nick the same as my password? Ugh
09:58 Raven262   maybe cause you accidentally typed your password instead of your nick?
10:13 SG9kb3I    oh well
12:59 blue-002   do formspecs elements get uids?
12:59 blue-002   or just ids
13:19 blue-002   #5847
13:19 ShadowBot  https://github.com/minetest/minetest/issues/5847 -- fix dropdown menu selection by red-001
13:55 Wuzzy      Is this bug considered a blocker?: https://github.com/minetest/minetest/issues/4095
13:59 Calinou    this isn't a new bug, I remember doing that in 2014
13:59 Calinou    but I never reported it for some reason
13:59 Calinou    I suggest outputting a warning to console if the path is invalid, and fallback to the default value
14:57 Wuzzy      "it's an old bug, so no need to fix it" :D
14:58 Wuzzy      1 year later: "these 1000 bugs are very old, no need to fix them". XD
14:58 red-001    the font path bug?
14:58 Wuzzy      yeah
14:58 Wuzzy      if font_path etc. get botched for some reason, Minetest segfauls on start
14:59 Wuzzy      the only way to recover is to fix minetest.conf :(
14:59 Wuzzy      I suppose on Windows this is especially nasty because it just insta-crashes without error dialog, if I am not mistaken
14:59 red-001    oh great my PR has a heisenbug
15:00 Wuzzy      oh wait. windows has that default segfault window, right?
15:00 red-001    yes I think so
15:01 Wuzzy      I am going to test Minetest this week. :-) I already reported some issues/bugs.
15:05 Calinou    all it has is "Minetest has stopped working"
15:05 Calinou    it's everything but descriptive to the user :P
15:05 Calinou    but at least you know it crashed, instead of looking like it exits normally (many Linux distributions still do that in their default setups)
15:06 nerzhul    Wuzzy, i think we shuld call C:\$MFT\Minetest to ensure windows is crashing too
15:07 Wuzzy      Calinou: ikr :-( I hate it if a software just ... disappears. this is the worst.
15:18 Sokomine   hi. anyone got any slightly deeper knowledge of pathfinding in mt?
15:25 Wuzzy      Sokomine might have find a bug in pathfinding, not sure. Sokomine will explain.
15:26 red-001    create an issue?
15:26 red-001    iirc there was someone on github that was working on rewriting pathfinding
15:27 Sokomine   yes, burli did. his pathfinding worked slightly better than the one in the core but had the same issue
15:29 nerzhul    what is the issue ?
15:32 Sokomine   the path found is by no means the shortest one
15:33 Sokomine   see my screenshots in https://forum.minetest.net/viewtopic.php?f=9&t=17406
15:34 Sokomine   the blue rails mark the path the engine provided. i did experiment a bit with parameters now, and setting searchdistance to a lower value (to 1 instead of 10) seems to help a bit in one case. still, the path found is by no means the shortest
15:37 Sokomine   a higher value of maxsearchdistance leads to the path leading around the house first instead of straight through the front door. also the path starts to crisscrossing multiple times. it would be helpful to be able to talk to someone who's more aware of how pathfinding works and is implemented
15:52 nerzhul    Sokomine, it means the PF is not good yes
15:59 Sokomine   nerzhul: that could well be. i've now experimented a bit more with burlis algorithm. seems that one really does perform better in my case. it can also be taught to go through doors
16:00 nerzhul    it's nice
16:27 Krock      red-001, how exactly does #5847 fix the reported issue?
16:27 ShadowBot  https://github.com/minetest/minetest/issues/5847 -- fix dropdown menu selection by red-001
16:28 Krock      it still iterates through the whole dropdown list and returns NULL when nothing was found
16:28 red-001    uses the most recently added drop downlist that has the correct name
16:29 red-001    see my comment on the issue
16:32 red-001    updated the comment on the PR
17:09 VanessaE   nerzhul, et al., I know we're in feature freeze, but is there any chance my suggestions in #5768 (regarding simplifying the item string for a colorized recipe) can be implemented before release?
17:09 ShadowBot  https://github.com/minetest/minetest/issues/5768 -- Item and node colorization PRs in engine and game possibly essential for release
17:10 nerzhul    no chance
17:10 VanessaE   :(
17:11 nerzhul    rules are rules, they are there to ensure all works as intended
17:11 VanessaE   the exchange between juhdanad and I is going too slowly to suit the timing of the freeze
17:11 VanessaE   did you read my comments there at all?
17:11 VanessaE   what I propose has no chance of breaking anything (other than the proposal itself perhaps just not working right)
17:12 nerzhul    i read , but rules are rules
17:12 VanessaE   rules are meant to be broken ;)
17:12 nerzhul    i have also PR which doesn't have any chance to break anything, but they are not merged :)
17:12 nerzhul    no
17:12 VanessaE   well that sucks balls.
17:12 nerzhul    not mine as it seems
17:12 nerzhul    :p
17:12 VanessaE   :P
17:12 VanessaE   however,
17:12 VanessaE   to be fair
17:13 VanessaE   the discussion started before freeze
17:13 VanessaE   (well the issue opened before)
17:13 nerzhul    yeah like many pr, but sorry
17:14 VanessaE   well then let's make plans to make a "point release" shortly after 0.4.16 goes out.
17:14 VanessaE   to contain that, and your PR, and whatever else missed the 0.4.16 deadline
17:20 nerzhul    in fact, i don't know how the 0.5 can be managed with 0.4, i think it's quite impossible as we are not enough in the project for this, maybe some PR and bugfix can be backported to stable 0.4-stable but only if the branch doesn't require too many maintenance, but i don't know. I want to shorten intervals between release but i know it will not be possible before 0.5.0 release because 0.5.0 will require a lots of effort thigns we
17:20 nerzhul    want to break and enhance and the 0.5.0 cycle itself should take 6 months, but after i hope we can have 4 month release cycles with 10 feature freeze days or 2 weeks if there was many modifications
17:21 VanessaE   the only reason I want my suggestions in before release (or in the next point release if we do that) was just so that I don't have to use that convoluted method juhd* proposes - I have like a dozen mods that would need it.
17:22 VanessaE   and as you well know, code duplication (which this would surely be) is a bad thing.
17:22 VanessaE   plus it would make double work for me to revise those mods later to use a more simplified method.
17:22 nerzhul    i know, like some minetest parts needs rewrite or refactor but all those should wait for release
17:23 nerzhul    and you have ever done like this before, waiting a little more can be done and it will permit to perfect this PR after release because maybe breaking version can help it to be better
17:23 VanessaE   you remember my arguments from the other day about this
17:24 VanessaE   it isn't so much about waiting (which I'm willing to do).  it's about what I have to tell my users to get. :)
17:52 Krock      red-001, what I don't understand is why there can be multiple dropdowns with the same value, taking the advanced settings menu as example
17:53 red-001    it happens when the drop down list is changed but the name isn't
17:53 red-001    e.g. the formspec is updated with new data
17:54 Krock      this means there's garbage in the m_dropdowns variable
17:55 red-001    which breaks the advance settings menu as it uses the same name for all drop downs
17:55 red-001    Krock, yes
17:55 red-001    a better fix would also get rid of that issue
17:56 Krock      that's why I was so confused why this should fix the problem. It would be indeed much a better solution to clear the dropdown menus when the formspec is updated
17:56 Krock      s/why this/how this/
17:57 Krock      do you think this is doable? (haven't looked at this part of code much yet)
17:57 red-001    thats a good point
17:58 Krock      I'd rather not use this ugly workaround
17:58 red-001    wait now that I think about it this issue shouldn't even occur
17:58 red-001    isn't a new formspec generated every time the main menu is updated?
17:59 Krock      no, it's only generated when new formspec data is available
17:59 Krock      but it's updated frequently
17:59 red-001    but for this bug to happen new data has been be avaible
18:00 red-001    the formspec has to be changed
18:01 red-001    huh apprently not
18:02 Krock      GUIFormSpecMenu::regenerateGui  apparently doesn't contain  m_dropdowns.clear()  but clears any other fields
18:02 red-001    but thats also called when the formspec is resized
18:03 Krock      do checkboxes get unchecked on resize?
18:04 red-001    yes
18:04 red-001    it's a mess
18:04 Krock      uhm.. here they don't
18:04 red-001    nvm
18:04 Krock      nvm, the mainmenu directly stores a change in the .conf file
18:04 red-001    I was thinking of other input fields
18:05 red-001    e.g. password or username
18:07 Krock      well, either we have this garbage or the value will reset when the window is changed
18:09 red-001    well it would still get reset afaik even if we keep the grabage
18:10 Krock      lol, then clear the garbage :P
18:11 red-001    done
18:41 red-001    isn't there a script for creating a fake c++ file with all the setting from advance settings so gettext can work correctly?
18:41 red-001    RE: #5849
18:41 ShadowBot  https://github.com/minetest/minetest/issues/5849 -- Untranslatable advanced settings in 0.4.15-dev
19:19 nerzhul    red-001, strange, i translated those settings into weblate too
19:48 Wuzzy      nerzhul: oh
19:48 Wuzzy      these strings are in Weblate indeed. huh?
19:48 Wuzzy      let me check again
19:50 Wuzzy      nerzhul: Yep. I just re-downloaded 100% translated PO file. still not translated. o_O
19:53 Wuzzy      nvm
19:53 Wuzzy      forget what i just said
21:09 itubal     Hi Johnny Jo
21:11 itubal     How are you?
21:11 itubal     Bye Hunterz
21:16 itubal     by kilbith
21:18 red-001    hi itubal
21:33 nerzhul    itubal, hi but it's not hi channel but -dev channel
21:33 cheapie    Quick, someone go register ##minetest-hi :P
21:34 red-001    cheapie, done
21:35 Fixer      please no
21:36 Fixer      already too much forks of everything minetest
21:38 Fixer      including insanity like fun site frozen in 2012...
21:38 red-001    Hi I would like to annonce a new fork of minetest called minetest-hi, we here at minetest-hi understand that using chat can be complicated for new uses, so we decided to make it simple for everyone, In our version of minetest we replace all chat with a single word "hi"
21:40 itubal     And ... a Minetest fork that uses Python instead of Lua ??
21:40 itubal     jajaja!
21:40 itubal     A good idea.
21:42 red-001    also we are working on replacing the Lua scripting with our custom in-house version of x86 assembly in-order to make modding easier
21:43 itubal     Oh! i understand !!
21:45 itubal     this is a fork of "Comedy Central"
21:45 itubal     Comedy Central-Hi in assember code
21:46 red-001    well we considered it before we got a real nice letter from thier lawyers
21:46 red-001    apprently there is a thing called "trademarks" or something like that
21:46 red-001    not too sure on the details
21:47 red-001    as for python in minetest isn't there a mod for that yet?
21:47 itubal     No problem, called it ... "MineComedy"
21:47 red-001    adding it to core would be a lot of work
21:48 itubal     Free entrance forn newcomers.
21:48 rubenwardy erm, -> #minetest ?
21:48 Fixer      +1
21:50 itubal     Thankyou all for you funny words. Till tomorrow.