Minetest logo

IRC log for #minetest-dev, 2013-03-05

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:09 hmmmm vanessa, it wasn't totally regenerated?  there was still some of your stuff there?  too bad you didn't have a screenshot of exactly what it was
00:09 VanessaE correct
00:09 VanessaE sorry I didn't think to snap a screen of it
00:09 VanessaE but I can describe it further if you want
00:10 hmmmm wait, it was just one 16x16x16 block that was totally regenerated, and the rest of the mess was caused in the surrounding area
00:10 hmmmm right?
00:10 VanessaE mmmm...
00:10 hmmmm yea?
00:10 VanessaE you know that might account for it
00:10 VanessaE but,
00:11 VanessaE there's a house that stands some 30m tall....  which had dirt poured on top of it.  But there's no way in hell for that dirt to have been there.
00:11 hmmmm well this is a multi-block chunk based generator, so the 80x80x80 sized area was regenerated
00:11 VanessaE (my brick mansion near the spawn)
00:11 hmmmm starting from 0,0,0 to 80,80,80
00:11 VanessaE if it happens again I'll take a screenshot
00:12 hmmmm that one block that failed to load was initially emerged from disk as containing all CONTENT_IGNORE
00:12 hmmmm so all of the nodes that it attempted to place there were
00:14 VanessaE yeah but placing nodes on top of a building, way higher than they could ever have been?
00:14 hmmmm but the rest of it was not content_ignore, so it didn't place any new terrain or anything.  what it did instead was generate caves right in the middle of your already-generated areas, perhaps some trees, and then on_generate was called again, placing the ores
00:14 VanessaE (as in, the land where the building stands was never that high)
00:16 VanessaE that's about what it looked like to me as well, so I assumed it was a standard mapgen run in response to the alleged missing block
00:16 VanessaE the question of course is why did it bug out in the first place..
00:56 jin_xi joined #minetest-dev
01:21 jojoa1997 joined #minetest-dev
02:29 jojoa1997 left #minetest-dev
02:42 ecube joined #minetest-dev
03:02 ecube_ joined #minetest-dev
03:13 ecube- joined #minetest-dev
03:28 ShadowNinja hmmmm: https://github.com/minetest/minetest/pull/525
03:47 hmmmm great
03:47 hmmmm i'll put that in next
03:47 VanessaE ew.
03:47 hmmmm yuck
03:47 VanessaE using \( to escape [ just feels...dirty.
03:48 hmmmm it's a stopgap for until i rip out the current tokenization and put something better in
03:48 VanessaE fair enough
03:48 hmmmm well
03:49 hmmmm sorta wonder if i should even bother doing it later.  i might do it all at once
03:49 hmmmm i mean what i actually need to do is skip to the next [ until the previous char isn't \\
03:50 hmmmm and same with ]
03:51 ShadowNinja VanessaE: agreed, just an alternative for ow
03:51 ShadowNinja *now
04:37 ffoxin joined #minetest-dev
04:38 hmmmm woah getting sapier's stuff done is proving to be a real challenge
04:38 hmmmm when i cherry picked from his branch it screwed up l_span_tree()
04:39 hmmmm now i'm scared that it might've messed up other code randomly that i won't be able to catch as easily
04:39 hmmmm gulp
04:52 ffoxin joined #minetest-dev
05:12 VanessaE btw, that issue with newly placed blocks occasionally replacing that which is already there only happens when the server is lagging.
05:12 VanessaE the worse the lag, the more likely the place event will replace a node rather than add next to/on top of it
05:16 hmmmm interesting.. hey vanessa, are you not supposed to be able to mine cacti with a stone pick?
05:17 VanessaE not as far as I know
05:17 VanessaE been a while since I last tried
05:17 hmmmm wait you should be able to or shouldn't
05:17 VanessaE but I *think* you're supposed to only cut them down with an axe or sword? (of course a mese pick cuts anything)
05:18 hmmmm can't take anything for granted after i saw how badly the merge went
05:20 VanessaE understood
05:23 VanessaE ugh, mapgen lag is killing me today (yes, new terrain, so says the verbose log)
05:38 hmmmm wonder why
05:38 VanessaE heh
05:46 hmmmm whaddya guys think  https://github.com/kwolekr/minetest/commits/master
05:47 VanessaE holy smokes
05:47 VanessaE tl;dr
05:48 hmmmm this was messier than it should've been.
05:48 VanessaE care to squash them?
05:48 VanessaE might be easier to read anyway
05:48 hmmmm it's only three
05:48 hmmmm besides if i squash them, wouldn't sapier's authorship be dropped?
05:48 VanessaE three really LONG commits :-)
05:48 VanessaE actually I dunno if it would or not
05:49 VanessaE git commit --author="Original Author <email@server>"
05:49 VanessaE after the squash
05:49 VanessaE if I read it right
05:49 VanessaE http://stackoverflow.com/questions/3305218/how-to-merge-branch-and-retain-the-authorship
05:50 hmmmm ah author switch
05:51 VanessaE fear my google-fu ;-)
05:52 hmmmm yeah a lot less happened today than i wanted
05:52 VanessaE eh, well, you only have two hands and just so much time to get shit done.
06:01 hmmmm mm https://github.com/kwolekr/minetest/commit/bdbdeab0053d9ebbaffea17effeba777b710d390
06:01 hmmmm crazy
06:03 VanessaE is it just me, or are you adding a metric assload of new code here?
06:04 hmmmm eh no
06:04 hmmmm that's just you
06:04 VanessaE :P
06:04 hmmmm there's a little bit of additional code added like license comments and include directives, but it's not bad when you consider the improvement of it not being all in one file
06:05 VanessaE right
06:05 VanessaE I'm just looking at this one commit, that's what was throwing me off
06:05 hmmmm did you miss the "823 additions, 6894 deletions not shown " part?
06:05 VanessaE in fact, I did
06:05 VanessaE ok, 'nuff said
06:06 hmmmm PHEW
06:06 VanessaE blew right past that file while skimming through
06:07 hmmmm why do large files like this crop up in the first place?  why don't the people writing them stop and say "hmm, it takes a ridiculous amount of time to scroll through this file, perhaps it's a good idea to split some related things into a separate file"
06:08 VanessaE could be many things; surely some laziness in there, but I think a lot just has to do with the expected expenditure of time vs. perceived benefit
06:08 hmmmm takes no time at all if you do it while you're coding it up and not all at once
06:08 VanessaE you know, an extension of "if it ain't broke, don't fix it"
06:08 VanessaE true truew
06:09 hmmmm the facedir rotation stuff
06:09 hmmmm let's do that now
06:09 VanessaE on the other hand, I've had people bitch at me for following that ideal too closely - for splitting files up when they think they shouldn't be yet
06:09 VanessaE wait, I think RBA is still fighting that out
06:09 hmmmm above 1500 LoC is when it gets uncomfortable
06:10 VanessaE ugh
06:10 VanessaE homedecor/crafts.lua ...  1758 lines.
06:10 VanessaE what was that I was saying about reasons not to split up a file? :)
06:12 hmmmm i dunno, but when it's 5000, or 6000, or 7500 like scriptapi, it closely resembles the thought patterns of the same kinds of people who don't cut their fingernails
06:12 VanessaE in that particular instance, I never split the file up because all of the code therein seemed too closely related..  it feels somehow "wrong" to do so, but I can't really say why
06:12 hmmmm it's just as gross too
06:27 iqualfragile joined #minetest-dev
06:51 VanessaE joined #minetest-dev
08:59 Calinou joined #minetest-dev
09:59 jin_xi joined #minetest-dev
10:03 proller joined #minetest-dev
10:38 Taoki[laptop] joined #minetest-dev
10:46 Taoki[laptop] joined #minetest-dev
10:47 Taoki[laptop] joined #minetest-dev
10:51 Taoki[laptop] joined #minetest-dev
11:43 darkrose joined #minetest-dev
11:43 darkrose joined #minetest-dev
12:29 sfan5| joined #minetest-dev
14:19 ShadowNinja joined #minetest-dev
14:33 PilzAdam joined #minetest-dev
14:38 PilzAdam How is the state of realeasing?
14:55 Topic for #minetest-dev is now Minetest core development and maintenance. Chit-chat goes to #minetest. Consider this instead of /msg celeron55. http://irc.minetest.ru/minetest-dev/ http://dev.minetest.net/
14:55 thexyz PilzAdam: see git log
14:55 PilzAdam shouldnt we create a tag for 0.4.5 in minetest and minetest_game?
14:55 thexyz i think the only thing we left is to update links at minetest.net download page
14:55 thexyz celeron55: ^
14:56 thexyz PilzAdam: we probably should
14:56 * PilzAdam will do
14:57 proller tag needed for freebsd port
15:03 PilzAdam done
15:05 PilzAdam thexyz, also push to the stable_0.4 branch?
15:07 thexyz guess so
15:10 PilzAdam 31 commits for minetest_game and 219 commits for the engine
15:46 celeron55 oh
15:47 celeron55 so shall we use http://minetest.ru/builds/minetest-0.4.5.zip as the official windows build?
15:49 PilzAdam has it gettext and ttf support?
15:50 celeron55 it's thexyz's build, http://forum.minetest.net/viewtopic.php?id=4240
15:50 thexyz it has everything
15:50 thexyz i also removed -minetest.ru suffix for this one
15:51 PilzAdam then its ok, I think
15:54 celeron55 site updated; i guess i'll post a release notice on c55.me/blog as that is what some people follow
15:54 PilzAdam should I write a forum topic?
15:55 celeron55 i'll also copy the 0.4.4 release notice from the website to there, as the news section wasn't going to become persistent
15:55 thexyz minetest.net still has some "Minetest-c55"
15:56 celeron55 (because it lacks everything that you would expect from a proper news feed)
16:02 PilzAdam http://forum.minetest.net/viewtopic.php?pid=73978#p73978
16:08 thexyz > The main changes are a new mese system (mese is now a crystal), a new public server list and many new translations.
16:08 thexyz ohh, so *those* are the main changes
16:08 celeron55 main changes? lol
16:08 PilzAdam fixed
16:09 thexyz lol
16:12 celeron55 http://c55.me/blog/?p=1436
16:12 thexyz oh, yes, someone should write a proper changelog
16:13 thexyz because at the moment we only have two improper ones
16:13 celeron55 if somebody finishes it, i'll copy it over to the blog
16:13 celeron55 but as of now it seems to be messy
16:16 celeron55 thexyz: let's apply the 0.3 fix now?
16:16 celeron55 do you have it around somewhere
16:16 thexyz um.. what fix?
16:16 celeron55 the password exploit
16:17 thexyz oh
16:17 celeron55 well, i'll find the 0.4 commit somewhere, it should be trivially backportable
16:18 celeron55 i can include the notice about it on that blog post and everyone else can include it too in their posts
16:23 celeron55 ...where the hell is it
16:28 celeron55 it's like the whole thing didn't even exist in history
16:29 celeron55 i can't find the fix in any version of 0.4...
16:29 celeron55 ha, now i found it
16:31 Jordach joined #minetest-dev
16:31 Jordach joined #minetest-dev
16:38 celeron55 hmm, this also fails to build with irrlicht 1.8
16:38 celeron55 well, i'll not care for that now; i have irrlicht 1.7.1 too for cases like this 8)
16:41 Calinou joined #minetest-dev
16:47 celeron55 things that are fucking perfect: 1) a program built with irrlicht 1.7.1 headers will actually work just fine with a 1.8 shared library, 2) minetest 0.3.2 loads the relevant things from 0.4.5 config just fine
16:47 celeron55 8)))
16:52 hmmmm thexyz, the changelog i made wasn't ever finished and it was supposed to be finished by somebody who has knowledge of the changes in lua stuff, and also someone willing to put the time into typing up all the bugfixes
16:55 rubenwardy joined #minetest-dev
16:56 hmmmm argh
16:56 hmmmm speaking of changelog, people are taking the inclusion of BiomeDefManager as "there are biomes now being generated"
16:57 hmmmm i'm going to have to take the time to expand on specifically what each feature does
16:57 hmmmm perhaps the dev wiki changelog is too technical
16:58 Exio people? jojoa? :P
16:58 PilzAdam Calinou's changelog lists more the user-visible things
16:59 Exio what about a "technical stuff" section with a kwolekr-changelog saying what was added, removed, fixed, changed, in the core and other for the "user-visible" things?
17:00 celeron55 0.3.3 with the fix pushed; PilzAdam: see the note in my blog post and edit that addition to your forum post too
17:02 celeron55 that should get enough coverage to it... though i'll still need to see if i can poke some activity to the debian/ubuntu package maintainers
17:02 hmmmm yeah but his changelog is not organized at all and is completely wrong on many of the contributions
17:03 PilzAdam celeron55, will there be an official win build for 0.3.3?
17:04 hmmmm yikes, people still use 0.3.x?
17:04 hmmmm i don't believe that.  it's just celeron alone
17:05 PilzAdam redcrab and gameboom have 0.3 servers IIRC
17:05 Jordach redcrab has a 0.3 server
17:06 PilzAdam is there a forum topic for it?
17:07 celeron55 PilzAdam: there isn't an official build for even 0.3.2 8)
17:07 celeron55 (which had some missing things added to the creative inventory)
17:07 PilzAdam same as 0.4.0
17:08 celeron55 what the fuck man: https://travis-ci.org/minetest/minetest/builds/5259190
17:08 PilzAdam 0.4.1 was released 20 hours after 0.4.0
17:09 celeron55 why does that try to build a version that doesn't have those files that imply that it should even try
17:10 celeron55 also: launchpad's stable recipes are borked: https://code.launchpad.net/~minetestdevs/+recipe/minetest-c55-stable
17:10 celeron55 it just says "Oops! Sorry, something just went wrong in Launchpad. We’ve recorded what happened, and we’ll fix it as soon as possible. Apologies for the inconvenience." 8)
17:10 celeron55 going to try that sometime later i guess
17:11 PilzAdam would it be possible to call it "Minetest" instead of "Minetest-c55" in launchpad too?
17:12 celeron55 maybe, but it requires changes in multiple places and possibly deleting/recreating things
17:13 celeron55 it consists of bzr mirrors of stuff and a project and whatever things like that
17:13 celeron55 all having c55 in them
17:14 celeron55 so you can't just press a button and expect it to be done 8)
17:14 PilzAdam maybe ask the launchpad people for help and let them do the work? ;-)
17:15 celeron55 they aren't going to do any work, but i could ask "the minetest launchpad guy" if he's interested
17:15 PilzAdam and who is that?
17:16 celeron55 juhani numminen (mentioned on the launchpad project page or somewhere); i'll throw him an e-mail
17:17 celeron55 (it's always fun to write about minetest development in finnish, as this is the only case ever where i can do that)
17:23 celeron55 done and out ->
17:34 sfan5 proller: https://github.com/minetest/minetest/pull/526/files#L0R34
17:34 sfan5 you can't just copy it from ogre cmake files
17:35 sfan5 ... ${OGRE_SOURCE_DIR}/Dependencies/lib/release/libGLESv2.lib ...
17:35 proller randomize lines& 8)
17:35 proller ok
17:35 proller i test it in linux, and it works
17:35 sfan5 also     IF (NOT OGREKIT_BUILD_ANDROID)
17:36 proller maybe remove all win32 shit ?
17:36 sfan5 and what the fuck is "GLESL"  https://github.com/minetest/minetest/pull/526/files#L0R107
17:37 proller openGL ES Lib
17:38 proller i will remove all non linux
17:38 sfan5 that sounds stupid to me
17:39 proller GLESL or removing ?
17:42 Exio ehm
17:43 Exio http://dpaste.com/1013207/ in debian6
17:45 Exio what i did; git clone; mkdir -p minetest/build; cd minetest/build; cmake ..; make
17:50 VanessaE 12:49:31: ACTION[ServerThread]: Server: Player with an empty name tried to connect from 194.166.73.103
17:50 VanessaE 12:50:17: ACTION[ServerThread]: Server: Player with an invalid name tried to connect from 194.166.73.103
17:50 VanessaE oh nice.
18:01 iqualfragile joined #minetest-dev
18:05 Exio other user in debian 6 reproduced that, it doesn't irrlicht 1.7 or what?
18:05 Exio (the last commit)
18:05 thexyz why do you want to build inside minetest/build?
18:08 celeron55 to keep the source directory clean; i think such should be supported while at least i am generally way too lazy to do such
18:11 Exio as c55 said
18:16 Calinou joined #minetest-dev
18:34 proller joined #minetest-dev
18:59 hmmmm speaking of libs, we should at some point either move away totally from jthread, rename it, or change the version so that the compiler will never attempt to use a system installed version of jthread
19:00 hmmmm otherwise the build would break
19:07 sfan5 what about something like "#ifndef _JTHREAD_SPECIALLY_FOR_MINETEST \n #error other jthread versions are not supported \n #endif"
19:11 hmmmm well i don't want to have them uninstall the system jthread to build it
19:14 sfan5 well... IIRC ''#import "jthread.h"'' searchs in the current direcotry first
19:15 sfan5 *#include
19:15 hmmmm your python is showing
19:15 PilzAdam java?
19:16 hmmmm ah yeah java too
19:32 proller sfan5, now better? https://github.com/minetest/minetest/pull/526/files
19:32 ffoxin_entry joined #minetest-dev
19:33 sfan5 proller: ${SOURCE_DIR}/Dependencies/lib/release/libGLESv2.lib is wrong and BUILD_ANDROID also
19:33 sfan5 but its better
19:33 proller remove?
19:33 proller maybe it helps in future
19:38 hmmmm the hell is this  https://github.com/minetest/minetest/blob/master/src/strfnd.h#L140
19:38 hmmmm ugh
19:39 hmmmm i'll deal with that whole file sometime late
19:39 hmmmm later
20:15 thexyz suggestion: switch to ustring (http://irrlicht.sourceforge.net/forum/viewtopic.php?t=37296)
20:18 BackupCoder joined #minetest-dev
20:19 Exio hm, my keyboard got "glitched" (only 's' working and 'a' working as 's'), is this known or something?
20:19 Exio removed focus, retried and working
20:30 VanessaE hmmmm: ew.
20:30 VanessaE someone never heard of caching. :-/

| Channels | #minetest-dev index | Today | | Google Search | Plaintext