Minetest logo

IRC log for #minetest-dev, 2014-08-25

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

All times shown according to UTC.

Time Nick Message
00:09 EvergreenTree joined #minetest-dev
00:11 VanessaE https://forum.minetest.net/viewtopic.php?f=11&t=2041&start=400#p152081
00:13 VanessaE shouldn't builds for windows have openal as an optional dep?
00:26 sapier if they're compiled against you need it ... wonder why it's missing
00:27 VanessaE http://kcat.strangesoft.net/openal.html  this seems to be the preferred version?
00:27 VanessaE after some googling (which frankly I don't expect an average user to figure out)
00:31 ShadowNinja VanessaE: No, my limit was never merged, other people didn't want message limiting.  sapier: No, it passes through that exception handler, although it could be expanded.
00:32 VanessaE ShadowNinja: well that explains that. ok.
00:32 sapier ShadowNinja: nope didn pass, but kahrl found a solution for the primary issue, didn't test if there are different issues
00:32 ShadowNinja sfan5: I can raise the !tell per-user message limit if you want.  There really were a lot of messages waiting though.
00:42 Tesseract joined #minetest-dev
00:42 Tesseract joined #minetest-dev
00:48 not^v joined #minetest-dev
00:57 sapier left #minetest-dev
00:59 Miner_48er joined #minetest-dev
01:30 Eater4 ShadowNinja: I think that luarocks is the only thing that will get the irc mod working on unbuntu
01:31 VanessaE --> #minetest
02:34 sol_invictus joined #minetest-dev
04:20 zat joined #minetest-dev
04:47 jin_xi joined #minetest-dev
05:49 Hunterz joined #minetest-dev
05:52 Selat joined #minetest-dev
06:00 blaise for what it's worth, here's a crash.. I was exploring when it happened
06:00 blaise http://pastebin.com/ajAKT8zd
08:32 sapier joined #minetest-dev
08:36 blaise sapier: good morning?
08:42 restcoser joined #minetest-dev
08:55 sapier good morning blaise how may I help you?
09:34 casimir joined #minetest-dev
09:44 proller joined #minetest-dev
09:48 iqualfragile joined #minetest-dev
09:48 rubenwardy joined #minetest-dev
10:28 nore joined #minetest-dev
10:59 Jordach joined #minetest-dev
10:59 khonkhortisan joined #minetest-dev
11:30 Garmine joined #minetest-dev
11:30 ImQ009 joined #minetest-dev
11:33 proller joined #minetest-dev
11:46 restcoser joined #minetest-dev
11:53 casimir joined #minetest-dev
12:36 selat joined #minetest-dev
13:08 AnotherBrick joined #minetest-dev
13:40 NakedFury joined #minetest-dev
13:40 CraigyDavi joined #minetest-dev
14:17 hmmmm joined #minetest-dev
14:31 Ritchie_ joined #minetest-dev
14:33 hax404 joined #minetest-dev
14:34 rmilan joined #minetest-dev
14:39 Hunterz joined #minetest-dev
14:46 Hunterz1 joined #minetest-dev
14:47 Jordach joined #minetest-dev
14:51 rmilan joined #minetest-dev
15:01 PilzAdam joined #minetest-dev
15:29 jin_xi joined #minetest-dev
15:34 Calinou joined #minetest-dev
16:10 exio41 joined #minetest-dev
16:11 Hunterz joined #minetest-dev
16:27 nore joined #minetest-dev
16:31 proller joined #minetest-dev
16:40 ShadowNinja blaise: You're either using too big of a range or the minumum is bigger than the maximum.
16:40 ShadowNinja (Although assert(0) is definitely the wrong way to handle that)
16:47 ShadowNinja sapier: Where's core.get_us_time() documented?  And it should use the unit's name rather than a character that looks like it's abreviation character, such as get_micro_time() or get_microsecond_time().
16:51 exio4 joined #minetest-dev
16:52 VanessaE didn't I already say that? :P
16:53 VanessaE late to the game again, ShadowNinja  ;)
16:53 VanessaE [08-23 18:35] <VanessaE> that should be renamed before it becomes too entrenched.
16:53 VanessaE [08-23 18:35] <VanessaE> core.get_clock_microseconds() or something
16:54 ShadowNinja VanessaE: Oh, I've lost a lot of backlog from transitions and resets, and I haven't been reading all of it before now due to being busy.
16:54 VanessaE no worries.
16:54 VanessaE that new core working out okay?
16:57 ShadowNinja Yep.  But let's not get off-topic here.
16:58 VanessaE just checking.
17:26 iqualfragile joined #minetest-dev
17:37 zat joined #minetest-dev
17:42 VanessaE ShadowNinja: any ETA on ShadowBot?
17:42 Zefram_Fysh it is normal practice in ASCII-only media to use "u" in place of mu in metric unit symbols.  there is no ambiguity in doing so.  "get_us_time" looks fine to me.  though "get_micro_time" and "get_microsecond_time" also look fine
17:43 VanessaE shit, wrong channel.  that shoulda been on #minetest.
17:43 casimir joined #minetest-dev
17:43 VanessaE Zefram_Fysh: to a US user, it really does look like "get US time".
17:43 Zefram_Fysh oh, haha
17:43 exio4 joined #minetest-dev
17:43 Zefram_Fysh I hadn't thought of that interpretation
17:44 VanessaE it's merely the word order that causes that interpretation :)
17:51 Miner_48er joined #minetest-dev
17:57 ImQ009 joined #minetest-dev
17:59 Calinou get_micro_time
17:59 Calinou is likely the best
18:23 kaeza joined #minetest-dev
18:42 RealBadAngel imho it should be get_time_(miliseconds, microseconds, etc)
18:49 celeron55 probably, but that's slower
18:50 VanessaE slower
18:50 VanessaE ?
18:50 celeron55 wait, i was thinking that meant a parameter to the function
18:50 celeron55 but obviously not because _
18:50 Zefram_Fysh functions with longer names take longer to call, as any fule kno
18:52 kahrl except when they don't: http://news.php.net/php.internals/70691
19:00 celeron55 anwyay, i'm voting get_time_us()
19:01 celeron55 i even remember using that in something i wrote some day
19:01 celeron55 so it must be the best one!
19:08 ShadowNinja kahrl: strlen() would still take longer, but LOL.  I'm fine with get_micro(second)_time or get_time_micro(second).  But not us because Lua doesn't allow the real micro character in identifiers, and it's hard to type on most keyboard layouts (or at least the US one) anyway.
19:12 sapier why can't ppl just mention their concerns prior things get merged ... I'll add the docs which got lost by some rebase. I'll not rename it this time if you want a different name change it yourself
19:13 ShadowNinja sapier: It's internal-only currently, right?  Because we can't rename it after it becomes public without adding cmpatability code.
19:13 sapier I'm annoyed about renaming things always a lot of work and as ppl don't comment when it's still a pull it's done multiple times
19:14 ShadowNinja sapier: What pull was that in?
19:14 sapier you can rename it without any problems as it's never been within a stable version
19:15 sapier the only mod using it is mobf, since about a year ago
19:17 sapier the pull was profiling pull request
19:17 sapier because there I needed us precision
19:23 ShadowNinja sapier: I was away for two weeks and didn't see that.
19:24 VanessaE seems simple enough to rename the core function and provide a compat wrapper
19:25 sapier well celeron was here too and didn't mention it by now too ... I'll not wait 4 months for each commit to be checked by noone anyway.
19:25 VanessaE and to be fair, I believe I did raise my concern about the name when I first heard about it, which was not a year ago :)
19:25 sapier true, if I remember correct that was AFTER merge too ;-)
19:26 VanessaE yeah, I didn't have a chance to veto it beforehand :P
19:26 VanessaE (as if it would have stopped the merge spree :P )
19:26 sapier that one was days before
19:26 VanessaE ok so s/spree//
19:29 ShadowNinja sapier: Still indented wrong: https://github.com/minetest/minetest/commit/0c94f2fb64ff54edca790a7106ec43bbfb884058#diff-c6d551e431c1fb79f5cf714da9b6c604R20
19:29 iqualfragile joined #minetest-dev
19:29 ShadowNinja sapier: It should be register_x("foo", {\n\t...\n\t})
19:29 yummy_ginger joined #minetest-dev
19:30 yummy_ginger any one with some sence online now?
19:30 sfan5 as you can see there is nobody online
19:30 ShadowNinja sapier: And you use globals.  Also, no spaces around { and }.
19:30 iqualfragile yummy_ginger: wtf is a sence?
19:30 yummy_ginger ok so your not here either lol
19:31 sapier I'll not discuss about any (still draft) coding style guides today
19:31 iqualfragile there was a bit of discussion about enabling mods to paralelize tasks, can somebody update me on that?
19:32 ShadowNinja sapier: https://github.com/minetest/minetest/commit/0c94f2fb64ff54edca790a7106ec43bbfb884058#diff-1305560bd8befb32862f0feeefabd02eR1538
19:32 sapier iqualfragile: quite a lot of work to go
19:32 ShadowNinja sapier: It's the coding style used everywhere else.
19:33 sapier if you feel it's wrong fix it I won#t ... ahh there's the doc I added :-(
19:33 ShadowNinja And that's how they'll stay until some other devs start discussing it productively.
19:35 sapier exactly ... guess noone has any interest in them because except of you and me noone is regulary writing (larger potions of) lua code
19:35 ShadowNinja sapier: https://github.com/minetest/minetest/commit/0c94f2fb64ff54edca790a7106ec43bbfb884058#diff-7a58683d8f04f389d439e6abb8a88b19R7   This should be stored seperately so that a mod named total can't mess it up.
19:36 sapier fix it if you don't like it
19:36 ShadowNinja (Also, use .total instead of ["total"] since it's a valid identifier.
19:37 sapier If you spent half the time complaining about some coding issues for finding the player deletion bug you triggered we might have a solution by now
19:37 ShadowNinja sapier: No, it's a bug in your code, you should fix it.
19:37 sapier ok I'm gonna fix it
19:37 ShadowNinja Um... https://github.com/minetest/minetest/commit/0c94f2fb64ff54edca790a7106ec43bbfb884058#diff-7a58683d8f04f389d439e6abb8a88b19R154
19:38 ShadowNinja Use local ret = {toregister(...)}  And return unpack(ret).
19:39 sapier Everything you want
19:39 sapier right after you found and fixed the player deletion bug
19:40 ShadowNinja sapier: We're asynchronous...  But can you ink to a detailed description of the bug?
19:41 sapier quite simple have a lot of entities in one block and you'll trigger player object to be deleted causing player to be unloaded resulting in player beeing stall
19:41 ShadowNinja sapier: Oh, I'm pretty sure you broke that.
19:42 sapier yes for sure I did add the patch unloading the players
19:43 ShadowNinja VanessaE: Can you test https://github.com/minetest/minetest/issues/1425 before and after the commit mentioned in the issue?
19:43 sapier https://github.com/minetest/minetest/commit/7e6db1b80344a519e53a9967a159c8d3585a9b9d this is the commit causing the issue
19:44 VanessaE ShadowNinja: impossible for me to test now
19:44 ShadowNinja Hmmm, it's probably cause by that player<->SAO link.
19:44 VanessaE that commit predates other commits which I now rely on.
19:44 sapier get down fron that altitude ShadowNinja there's obviously to low o2 there
19:47 sapier ShadowNinja: if there's no player object there's no way to keep that player alive disconnect is the only sane action to do ... and yes actually that's never supposed to happen but you guys don't like it if I add assert for things that are not allowed to happen ;-P
19:48 * ShadowNinja digs into his old IRC log database looking for something usefull that sapier said.
19:50 sapier << leaves prior having to kill ShadowNinja
19:51 sapier left #minetest-dev
19:51 iqualfragile cake anyone?
19:52 VanessaE the cake is a lie.
19:54 RentedMule joined #minetest-dev
19:55 ShadowNinja ''tell sapier select sender, message from backlog join sender on backlog.senderid=sender.senderid where bufferid=208;  -->  http://pastebin.ubuntu.com/8143576/  I remember you mentioning a better place to relete the player, somewhere in the connection code.  But I don't know where it is.
19:55 HLuaBot Unexpected nested command closer.
19:55 ShadowNinja ''tell sapier select sender, message from backlog join sender on backlog.senderid=sender.senderid where bufferid=208; --\> http://pastebin.ubuntu.com/8143576/ I remember you mentioning a better place to relete the player, somewhere in the connection code. But I don't know where it is.
19:55 HLuaBot I'm already holding too many messages for that user.
19:56 iqualfragile relete?
19:56 iqualfragile delete?
19:58 ShadowNinja Limit upped and sent in PM.  Currious that he has so many waiting though...
19:59 VanessaE I believe it has been mentioned that the bot is not delivering messages in a timely manner
20:00 ShadowNinja Yes, it seems to be failing somehow.
20:00 ShadowNinja It's causing an error().
20:04 HLuaBot joined #minetest-dev
20:05 ShadowNinja Hopefully it'll work now.  Forgot to create a table before using it.
20:05 VanessaE oops.
20:22 Ritchie joined #minetest-dev
20:32 proller joined #minetest-dev
20:40 cg72 joined #minetest-dev
20:41 cg72 did sapier leave?
20:43 VanessaE yeah for now
20:43 VanessaE he'll be back tomorrow
20:49 alexxs joined #minetest-dev
21:07 proller joined #minetest-dev
21:26 ImQ009 joined #minetest-dev
21:35 cg72 ok i think we all need to sit down and get some written guide lines on what the style of code should be as bits of it are totally different and as this is ONE project we need to have uniformity to some extent
21:35 cg72 just some standards so its not a free-for-all
21:35 Calinou mod consistency guidelines would be useful
21:35 Calinou on the forums
21:35 Calinou for modders to use
21:36 Calinou we have Lua code style guidelines for builtin and Code style guidelines for C++ already, on dev wiki
21:36 Calinou mod consistency isn't only about coding, it's about naming of blocks, but also texturing, packaging, configuration files and all that
21:37 Zefram_Fysh left #minetest-dev
21:37 cg72 lol
21:37 Calinou someone with time can make a forum topic (a guide)
21:38 cg72 well if i told them to do it my way it may not be up to everyones standards, so i think it needs to be agreed on. i have one init and tons of lua files, some mods only have an init :)
21:40 VanessaE look at the way homedecor does it.
21:40 VanessaE multiple individual mods, some of which (like the main homedecor mod) consist of exactly that - an init.lua that loads a bunch of additional Lua files.
21:43 cg72 but VanessaE is that the way we all can agree upon :)
21:43 cg72 i do it that way
21:43 VanessaE it's a good style, because code-wise it's easy to maintain, and from a user's standpoint it's easier to disable code segments you don't want.
21:43 VanessaE rm somefile.lua ; touch somefile.lua
21:44 VanessaE is a lot easier than trying to grep and sed the code.
21:50 cg72 mt users and rm or touch rotf
21:50 VanessaE think server owners
21:51 VanessaE I figure if you've got "enough brain cells to form a synapse" as one person on Slashdot put it the other day, you can figure out rm and touch :)
21:51 VanessaE but asking a server owner to master sed, grep, and awk is asking too much
21:52 VanessaE (I can barely form a regex when I need to, forget the more complex matches some code would need)
21:56 VanessaE http://xkcd.com/723/  <--- I don't know whether this represents a sad state of affairs of the world, or just really good tech? :)
21:57 alexxs joined #minetest-dev
22:06 alexxs joined #minetest-dev
22:08 Jordach_ joined #minetest-dev
22:08 iqualfragile joined #minetest-dev
22:11 Anchakor_ joined #minetest-dev
22:22 OldCoder joined #minetest-dev
22:28 VanessaE joined #minetest-dev
22:36 Exio4 joined #minetest-dev
23:21 rmilan joined #minetest-dev
23:39 alexxs joined #minetest-dev

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