Time  Nick       Message
05:58 nerzhul    hello, merging #6306
05:58 ShadowBot  https://github.com/minetest/minetest/issues/6306 -- Fix OSX builds (closes #6289, fixes #6270) by neoascetic
05:58 nerzhul    tricky include place on osx, but now CI will become happy
06:28 nerzhul    merging #6302
06:28 ShadowBot  https://github.com/minetest/minetest/issues/6302 -- Network cleanup by nerzhul
06:30 nerzhul    merging #6127 too
06:30 ShadowBot  https://github.com/minetest/minetest/issues/6127 -- Safe digging and placing by bendeutsch
06:31 nerzhul    and #6187
06:31 ShadowBot  https://github.com/minetest/minetest/issues/6187 -- Fix Android node selection distance by juhdanad
06:31 nore       nerzhul: could you review again #5274 when you have the time, btw?
06:31 ShadowBot  https://github.com/minetest/minetest/issues/5274 -- [WIP] Clientside translate by Ekdohibs
06:31 nerzhul    nore i should do it :p but i will go to work in a few minutes, i will look at work
06:32 nore       ok, thanks :)
06:32 nerzhul    can you rebase, to make travis happy, we now have macosx build fix ?
06:32 nerzhul    i have an idea to find the correct language on android
06:33 nore       hmmmm it says there are no conflicts
06:33 nerzhul    currently i'm working on JNI for a android java app to make some parts fasters using C++ and i know how to make C++ and java interact, i have an idea how to fetch the client language and give it to minetest, i should show a POC at a moment :p
06:33 nore       do we have a way to force travis to rebuild?
06:33 nerzhul    nore, it's just to make travis build macosx and make the PR green, a macosx build fix on travis has beeen merged
06:34 nerzhul    login into travis, you are a dev you can launch a rebuild
06:39 nerzhul    i wrote some more comments
06:40 nore       maybe I should remove the testing commit now
06:41 nore       or at least push it to a different branch and have this one clean
06:41 nore       doing that now
06:42 nore       nerzhul: why should translate_ext be static?
06:42 nore       model_ext, sound_ext, etc. are not
07:40 nerzhul    sfan5, can you remember me what i should change on #6293 please ?
07:40 ShadowBot  https://github.com/minetest/minetest/issues/6293 -- Verify compiler version before building anything by nerzhul
07:41 nerzhul    for me gcc 4.9 is the minimum c++11 compatible version due to move constructor support (not working in 4.8)
07:48 nerzhul    merging #6218 in ~10 mins
07:48 ShadowBot  https://github.com/minetest/minetest/issues/6218 -- Make entity selection and collision boxes independently settable by stujones11
08:02 nerzhul    merging #6307 too, another macosx build failure
08:02 ShadowBot  https://github.com/minetest/minetest/issues/6307 -- Fix OSX builds caused by __WORDSIZE again by neoascetic
08:54 sfan5      nerzhul: it should only require certain versions for ENABLE_SYSTEM_JSONCPP=0
08:55 nerzhul    sfan5, and if i set minimum as 4.8 just for C++11 because it's the minimum known compiler version for MT ? and 4.9 when using embedded jsoncpp ?
08:58 sfan5      sounds fine
08:58 nerzhul    okay i'm working on a such modification
09:04 nerzhul    sfan5, #6293 what do you think about that ?
09:04 ShadowBot  https://github.com/minetest/minetest/issues/6293 -- Verify compiler version before building anything by nerzhul
09:22 nerzhul    if you can take 5 min to check heh :) sfan5
09:40 celeron55  just imagine what this would have been like back when people first started suggesting c++11
09:40 celeron55  a disaster
09:40 celeron55  looks like it's working so-so now
09:41 celeron55  which is nice
09:41 nerzhul    compilers were not ready when first people suggest that :p but now it's the C++ norm for everybody, many lbis are switching to it
09:41 nerzhul    libs*
09:43 nerzhul    it's a long work, on MT we prepared some work, last week, using clang-tidy i migrated many things, but w eare not really using all the powerness of C++11 heh, constexpr string switch case in core can be nice (formspec handler for example), lambda functions (i don't see a usecase on MT)
09:44 nerzhul    some programs are using C++14, but we don't have any interest to be c++14 compliant, it's a minor evolution of C++ and we are not feature blocked, the best feature in C++14, for me, is the constexpr function used by constexpr variable which permits for example to generate binary hash tables at compilation time, using algorithm
11:15 nore       nerzhul: the build seems to be failing on OSX with the declaration of a std::map<std::wstring, std::wstring>, do you know how to fix that please?
12:00 nerzhul    nore i think <wstring> is not included
12:00 nore       nerzhul: I just included <string> which wasn't before, waiting for the build
12:00 nore       (it's painfully long)
12:05 nerzhul    i optimised build time (~10% less time) but macosx built is very long
12:05 nerzhul    can you use std::unordered_map instead ?
12:05 nerzhul    if you don't care about record order
12:05 nore       hmmm I think it is already unordered_map actually
12:05 nore       and that's why it failed, because it couldn't find a hash function
12:06 nerzhul    there is no hash function for std::wstring ?
12:06 nore       well if you don't include <string>, there is none :)
12:07 nore       that's why I'm trying again including it, we'll seen quite soon
12:07 nore       https://travis-ci.org/minetest/minetest/jobs/267935673
12:14 nore       \o/ success
12:14 nore       there are quite a lot of "ld: warning: object file (/usr/local/lib/libIrrlicht.a(Irrlicht.o)) was built for newer OSX version (10.12) than being linked (10.11)" btw
12:21 nerzhul    yeah it's not related to cmake build
13:46 stujones11 I think #6199 can be closed now :)
13:46 ShadowBot  https://github.com/minetest/minetest/issues/6199 -- Disable pointability for entities
13:47 nerzhul    done
13:48 nerzhul    882 issues remaining
13:48 nore       that's.... a lot
13:49 nerzhul    close issues with no activity since 1 year, 3 issues remianing :D
14:07 shalmezad  #4616 can be closed as well since you did #pragma once
14:07 ShadowBot  https://github.com/minetest/minetest/issues/4616 -- Header guards use identifiers reserved for the compiler/library implementation
14:29 nerzhul    exact
15:13 nerzhul    merging #6308 MSVC fix
15:13 ShadowBot  https://github.com/minetest/minetest/issues/6308 -- MSVC: Dont define min/max macros in minwindef.h by adrido
15:13 nerzhul    (trivial)Ã
15:14 nerzhul    sfan5, can you take 1 min to review #6293 please ? :)
15:14 ShadowBot  https://github.com/minetest/minetest/issues/6293 -- Verify compiler version before building anything by nerzhul
15:14 nerzhul    we are near 140 PR, yeah
15:15 sfan5      looks good to me
15:15 sfan5      you can merge if you tested it
15:15 nerzhul    it's tested
15:15 nerzhul    i setted clang version to 47.5 and it refuse to compile :D
15:16 nore       could we merge #5274 now that the build works on OSX as well btw? or are there more things to do?
15:16 ShadowBot  https://github.com/minetest/minetest/issues/5274 -- [WIP] Clientside translate by Ekdohibs
15:16 nerzhul    remove the WIP tag ? :D
15:16 nerzhul    merged, ty sfan5
15:17 nore       done :p
15:18 nerzhul    you missed a static
15:18 sfan5      nore: btw the wool example you game is not good practice
15:18 nore       nerzhul: hmmm, the thing is, the other lines above are not static
15:18 nore       I could make them so, though
15:18 sfan5      it wouldn't work if the word order changes for different colors
15:19 nore       hmmm
15:19 nore       translations are tricky, indeed... do you have a language where this is the case?
15:19 nerzhul    it's a const array, just make it static to prevent allocate/unallocate it each time you call the function
15:20 nerzhul    sfan5, maybe it's not the correct sample, but '%d chairs' is a better example :p
15:20 sfan5      that's also a bad example
15:20 sfan5      gettext has lots of complicated stuff to handle plurals correctly
15:20 sfan5      but nobody uses that
15:21 nore       yes
15:21 nore       and handling plurals correctly will require more complicated stuff, unfortunately
15:22 nore       so I decided to just skip it for now, shouldn't be too complicated to add it later if we want it
15:22 nore       sfan5: do you have a better example I could use instead?
15:22 sfan5      nah
15:22 sfan5      it's likely not a problem in practice anyway
15:24 nore       well, we'll see if we need plurals; if we to, we can just add another API function for that
15:26 sfan5      nerzhul: the compiler will probably move that array out of the function scope anyway, no need for static really
15:26 sfan5      all you're saving is two push instructions to assemble the array on the stak
15:26 sfan5      +c
15:27 nerzhul    sfan5, ok
15:29 nore       sfan5: btw, yes, it is limited to 9 arguments in a single translated string
15:29 nore       I think it is in the documentation btw
15:30 sfan5      that's okay then
15:31 nore       hmmm I put that the textdomain was there for disambiguation, but I'm not sure what to put more about it; any idea about what I should write?
15:32 sfan5      idk
15:32 sfan5      each mod already seems to have its own translation scope so i don't see why you absolutely need textdomains
15:32 sfan5      wait no it doesn't
15:32 sfan5      hm
15:33 nore       yeah, that's the problem
15:33 nore       for some translations, you might even want to use the textdomain of other mods
15:33 sfan5      why not have each mod be in its own textdomain by default
15:34 sfan5      because i doubt sharing translations between mods is *that* useful
15:34 nerzhul    can someone take a look at #6107, i added my approval
15:34 ShadowBot  https://github.com/minetest/minetest/issues/6107 -- Overlays for wield and inventory images by juhdanad
15:34 nore       hmmm, so should I default to "minetest.get_current_modname()" when initializing text domains?
15:34 nore       the thing is, I can't really do that for the translation file, however
15:35 nore       since it is a media resource, so the server will ignore the path it found the translation file in
15:35 sfan5      ugh
15:35 sfan5      then keep it this way
15:36 nore       ok, might add some more explaination about what it does
15:44 nore       sfan5: pushed some fixes, including more doc concerning textdomains, tell me if it's ok with you please :)
15:46 sfan5      see comments
15:46 sfan5      it's fine otherwise
15:48 nore       okay, thanks :)
15:48 nore       (pushed the fixes for the typos)
15:50 nore       could you please put a +1 on github so that I can merge it btw, if it's ok?
15:53 nore       thanks!
15:53 nore       squashing & merging then
15:53 nerzhul    yeah
15:53 nerzhul    139 PR opened, maybe at a point we can have less than 100 PR opened ? :D
15:54 nore       we need to merge more then :)
15:54 nore       (and faster)
15:56 nore       merged!
15:56 * nore     writes a forum post in the "News" section
16:03 nerzhul    it's difficult to merge faster the remaining PR they are not all finished
16:05 nore       yeah, unfortunately
16:06 nerzhul    but it's better than before 0.4.16 project has many interesting features, the 0.5.0 cleanup is not finished yet (i should re-audit the network packet protocol part to find and drop problems, upgrade the MIN_PROTO_vVERS to last)
16:16 nore       uh
16:16 nore       updatepo.sh doesn't add LANG_CODE to translations
16:16 nore       this is not normal
16:17 nerzhul    it's not the translations.cpp ?
16:19 nore       hmm the string is in network/clientpackethandler.cpp
16:19 nore       but it's still among the files xgettext is given
16:45 nore       pushing https://github.com/Ekdohibs/minetest/commit/b47c19c06f51269eef851be38a8d74f7d3190a93 in 10 minutes
16:45 paramat    nice small commit
16:46 nore       paramat: can I push it? :)
16:46 nore       (running updatepo.sh does quite a lot of operations, it seems...)
16:49 paramat    i don't understand this at all so can't judge it
16:50 paramat    but i trust you on this
16:51 nore       pushing then
16:57 nore       also pushing trivial https://github.com/Ekdohibs/minetest/commit/8221d3bc535b026c1bb06b5c76188a97c27c38cc needed for clientside translate to be usable in 10 minutes if it's ok
17:03 paramat    fine
17:04 nore       pushed
17:09 paramat    is #3493 useful enough to adopt? it has no support and is very old with no progress, close?
17:09 ShadowBot  https://github.com/minetest/minetest/issues/3493 -- Call lua on_punch if on_use for an item didn't result in an ItemStack by sapier
17:10 paramat    same for #3579
17:10 ShadowBot  https://github.com/minetest/minetest/issues/3579 -- Make INodeDefManager::getIds return a vector, not a set by kahrl
17:13 nerzhul    nore, we can close #5274 no ?
17:13 ShadowBot  https://github.com/minetest/minetest/issues/5274 -- Clientside translate by Ekdohibs
17:13 paramat    i'm going to close #4591, disapproved and no progress
17:13 ShadowBot  https://github.com/minetest/minetest/issues/4591 -- Infrastructure for sending server parameters to clients, overriding settings on the client by Rogier-5
17:13 nore       nerzhul: yep, doing that
17:13 nerzhul    i thought you merged through github but no :p
17:13 nerzhul    paramat, +1
17:14 paramat    closed 4591
17:14 nore       done
17:14 nerzhul    #3642 is interesting no ? maybe someone should adopt it
17:14 ShadowBot  https://github.com/minetest/minetest/issues/3642 -- Add error message when attempting to revoke privileges from server administrator by Splizard
17:15 paramat    nore opinion on #4916 ? is it worth adopting?
17:15 ShadowBot  https://github.com/minetest/minetest/issues/4916 -- Add a lua gettext variant for formspecs that splits text by Rogier-5
17:15 nerzhul    can we close #3818, #3810 and #3957; flagged as possible close ?
17:15 ShadowBot  nerzhul: Error: Delimiter not found in "Page is too big or the server took too much time to answer the request."
17:15 ShadowBot  https://github.com/minetest/minetest/issues/3810 -- Try to fix jumping at node edge again by ShadowNinja
17:15 ShadowBot  https://github.com/minetest/minetest/issues/3957 -- Set log level for field exist warning same as warning message by ShadowNinja
17:16 paramat    i thought i would give SN a month
17:16 nore       paramat: hmm, I'm not sure about it for now
17:17 paramat    ok close then? no progress
17:17 paramat    author disappeared
17:18 paramat    language-dependent newlines seems weird
17:18 nore       I guess we can still use SN's solution
17:20 paramat    i'd like to give SN a full month before closing his old PRs
17:20 nerzhul    rubenwardy, i like (and many others) #4713 and would see it in core, can you factorize the lua calls (and reduce indent level if possible it's crazy), and finish the PR ?
17:20 ShadowBot  https://github.com/minetest/minetest/issues/4713 -- Add on_grant and on_revoke callbacks by rubenwardy
17:20 nerzhul    some servers are now using 0.5.0-dev they are crazy
17:21 paramat    oh it's because some languages will alter line length i guess
17:21 nerzhul    sfan5, what do you think about adding a private mail contact address for server list server owners ?
17:22 nerzhul    never showing it but store it permitting devs to contact them ?
17:22 sfan5      what for
17:23 nerzhul    example: send a minetest release notice, or today (but they will not have a such thing before 0.5.0 release...) inform them they should not use 0.5.0-dev in production :p
17:25 paramat    #5451 is controversial, although MD might happen for lua_api i'm not sure this PR will go anywhere
17:25 ShadowBot  https://github.com/minetest/minetest/issues/5451 -- Rename doc/*.txt to *.md. by ensonic
17:26 paramat    plus it needs editing before changing to .md
17:27 nerzhul    README.md and client_lua_api are in MD , only server does resistance :p
17:35 paramat    hm well, i feel client lua api should not have been .md but consistent with lua api :]
17:39 paramat    ok, i've asked the author to improve the md formatting, which will be useful whether or not we change it to .md
17:40 paramat    if no progress in 1 month, close
17:42 paramat    so, anyone want to adopt #4916 ? nore shall i close it?
17:42 ShadowBot  https://github.com/minetest/minetest/issues/4916 -- Add a lua gettext variant for formspecs that splits text by Rogier-5
17:44 paramat    #3739 is old with bugs and no progress, close? anyone else feel it's not worth dev time?
17:44 ShadowBot  https://github.com/minetest/minetest/issues/3739 -- Add support for Mumble positional audio by ShadowNinja
17:56 paramat    does anyone feel this would be useful? #4168
17:56 ShadowBot  https://github.com/minetest/minetest/issues/4168 -- Add type to get_objects_inside_radius
18:05 paramat    anyone feel this is useful? #5442 author has left MT so possible close
18:05 ShadowBot  https://github.com/minetest/minetest/issues/5442 -- Implement FFI to LuaJIT
18:17 paramat    closed #4916
18:17 ShadowBot  https://github.com/minetest/minetest/issues/4916 -- Add a lua gettext variant for formspecs that splits text by Rogier-5
18:30 Krock      Builtin item cleanup rebased: #4370
18:30 ShadowBot  https://github.com/minetest/minetest/issues/4370 -- Builtin item: Tidy up the code by SmallJoker
18:35 paramat    thanks will review
18:37 Krock      fine, thanks :)
18:51 nerzhul    Krock okay for me
18:57 nerzhul    Krock, can you review #4713 it sounds useful
18:57 ShadowBot  https://github.com/minetest/minetest/issues/4713 -- Add on_grant and on_revoke callbacks by rubenwardy
18:57 nerzhul    if rubenwardy can just fix the dup code i think it should be merged
18:58 Krock      was about to reply you that it's not a dupe
18:58 nerzhul    i misread ?
18:58 Krock      it differs: on_grant / on_revoke
18:59 nerzhul    i see dup not there but between chatcommands and auth.lua
18:59 Krock      or did you mean the part in chatcommmands.lua?
18:59 Krock      ah
19:00 nerzhul    :)
19:08 Krock      tested and it works. I don't see a solution to only run the callback once yet - so it'll have to be good enough for now
19:58 Dumbeldor  Hi, is it possible to have a review please ? :) #5793 Thanks
19:58 ShadowBot  https://github.com/minetest/minetest/issues/5793 -- [CSM] Add callback on open inventory by Dumbeldor
20:35 paramat    +1 for #4370
20:35 ShadowBot  https://github.com/minetest/minetest/issues/4370 -- Builtin item: Tidy up the code by SmallJoker
20:36 paramat    feel free to merge it
20:51 Krock      okay, thanks
20:56 Krock      merging #4370 in 10 minutes
20:56 ShadowBot  https://github.com/minetest/minetest/issues/4370 -- Builtin item: Tidy up the code by SmallJoker
21:39 nerzhul    merging #5793 in ~10 mins
21:39 ShadowBot  https://github.com/minetest/minetest/issues/5793 -- [CSM] Add callback on open inventory by Dumbeldor
21:39 nerzhul    good job on the PR merge/cleanup guys
21:39 nerzhul    ~12 PR closed/merged today, it's nice
22:06 rubenwardy Added comment to #5793
22:06 ShadowBot  https://github.com/minetest/minetest/issues/5793 -- [CSM] Add callback on open inventory by Dumbeldor
22:06 rubenwardy Seems odd that callbacks are run newest first
22:19 Dumbeldor  rubenwardy: thanks for review #5793 but i don't understand "This is inconsistent. Why ?" :(
22:19 ShadowBot  https://github.com/minetest/minetest/issues/5793 -- [CSM] Add callback on open inventory by Dumbeldor
22:59 Wuzzy      nore: My review of the new translation system is finished.
22:59 Wuzzy      https://forum.minetest.net/viewtopic.php?f=18&t=18349&p=290369#p290369