Minetest logo

IRC log for #minetest-dev, 2016-07-21

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

All times shown according to UTC.

Time Nick Message
00:00 Megaf the chunks that the server send to client are too big
00:00 Megaf the mapblocks
00:00 Megaf it should be much more, granular, fine
00:00 Megaf and dynamic
00:00 paramat heh
00:00 Megaf even with fog at 0.4 I can see the map appearing in a very non smooth way
00:01 Megaf building and trees sudenly pop into view instead of a smooth fade in from the fog
00:01 Megaf making the fog useless
00:01 paramat yeah fog is just a colour change
00:01 Megaf fog should also giver performance gains since theres less visable stuff
00:02 paramat nope, with fog closer there's the same rendering distance
00:04 paramat you might like 0.85. anyway it's my fault, i insisted RBA kept it at 0.9 :]
00:04 Megaf I quite like 0.4
00:05 Megaf in my server it looks great with all the bridges and highways
00:05 Megaf it blends quite ok
00:20 paramat Minetest 0.4 is now 4 years old
00:22 paramat i arrived 3 months later
00:28 Megaf paramat: I mean, 0.4 as value for fog in game.cpp
00:29 paramat i know :]
00:30 paramat i just saw Calinou's comment in the other channel
00:31 Megaf what channel?
00:34 paramat #minetest
00:37 paramat will merge #4339 in a few hours if no objections. mapgen bugfix
00:37 ShadowBot https://github.com/minetest/minetest/issues/4339 -- CavesNoiseIntersection: Place riverbed node under river water by paramat
01:15 yang2003 joined #minetest-dev
01:49 yang2003 joined #minetest-dev
01:56 TheReaperKing joined #minetest-dev
02:02 STHGOM joined #minetest-dev
02:04 paramat left #minetest-dev
02:05 Void7_ joined #minetest-dev
02:05 Miner_48er joined #minetest-dev
02:07 jin_xi joined #minetest-dev
02:17 Erebus_ joined #minetest-dev
02:26 Megaf joined #minetest-dev
02:44 STHGOM_ joined #minetest-dev
02:47 STHGOM joined #minetest-dev
03:06 OldCoder joined #minetest-dev
03:20 yang2003 joined #minetest-dev
03:29 yang2003 joined #minetest-dev
03:30 Void7_ joined #minetest-dev
03:37 paramat joined #minetest-dev
03:43 paramat merging #4318 #4339
03:44 ShadowBot https://github.com/minetest/minetest/issues/4318 -- lua_api.txt: Remove tooltip checkbox[] element by everamzah
03:44 ShadowBot https://github.com/minetest/minetest/issues/4339 -- Mgvalleys / cavegen: Place riverbed nodes under river water by paramat
03:56 Megaf joined #minetest-dev
04:07 paramat done
04:08 paramat left #minetest-dev
05:06 Hunterz joined #minetest-dev
05:50 nrzkt joined #minetest-dev
06:18 OldCoder Recently, worlds seem to lock-up during announce to server list. Is this a known issue?
06:21 hmmmm paramat:  who signed off on 4339?
06:54 DonBatman joined #minetest-dev
06:55 Darcidride joined #minetest-dev
07:32 nrzkt joined #minetest-dev
07:38 Krock joined #minetest-dev
07:59 Amaz joined #minetest-dev
08:05 tenplus1 joined #minetest-dev
08:06 tenplus1 hi folks... could a kindly dev check out pull game#965 tested and ready to be added
08:06 ShadowBot https://github.com/minetes​t/minetest_game/issues/965 -- Ability to Disable bones, drop items or keep inventory. by tenplus1
08:09 tenplus1 left #minetest-dev
08:35 raingloom joined #minetest-dev
09:00 Amaz joined #minetest-dev
09:41 edgrey joined #minetest-dev
10:14 Fixer joined #minetest-dev
10:33 Krock joined #minetest-dev
11:10 Lunatrius` joined #minetest-dev
11:14 Lunatrius` joined #minetest-dev
11:43 Krock2 joined #minetest-dev
12:01 Krock joined #minetest-dev
12:25 STHGOM joined #minetest-dev
12:54 est31 joined #minetest-dev
13:02 proller joined #minetest-dev
14:18 Void7_ joined #minetest-dev
14:24 edgrey joined #minetest-dev
14:51 edgrey joined #minetest-dev
14:54 edgrey joined #minetest-dev
14:56 edgrey joined #minetest-dev
15:31 Void7_ joined #minetest-dev
15:33 Grandolf joined #minetest-dev
15:33 halt_ joined #minetest-dev
15:41 hmmmm joined #minetest-dev
15:48 KaadmY joined #minetest-dev
15:55 proller joined #minetest-dev
16:00 paramat joined #minetest-dev
16:00 Hunterz joined #minetest-dev
16:20 SloanOnLinux joined #minetest-dev
16:20 whitephoenix joined #minetest-dev
16:21 Darcidride joined #minetest-dev
16:26 nrzkt joined #minetest-dev
16:43 GnomeKris joined #minetest-dev
16:43 Hunterz joined #minetest-dev
16:52 blaze joined #minetest-dev
17:05 halt_ joined #minetest-dev
17:06 paramat hmmmmm i announced 4339 twice 2 days ago, highlighted you and asked for reviews but no-one did. since it was a borderline trivial, simple and obvious mapgen bugfix to my own code i decided to announce merge and give 3 hours for objections
17:12 proller joined #minetest-dev
17:38 hmmmm that doesn't fit the definition of trivial though
17:38 hmmmm trivial is like adding a piece of documentation or changing around a comment
17:38 hmmmm i was completely gone doing other stuff the past 2 weeks, there has to be some other developer to do a review
17:39 hmmmm 2 days is not a long time by any means... many prs stay around for months even
17:41 paramat yeah i admit i took some initiative here, but it was reasonable
17:47 paramat engine devs are very absent currently, even sofar seems busy
17:49 hmmmm calling that change of code functionality 'reasonably trivial' creates a slippery slope for what could be called trivial down the road
17:49 hmmmm you're going to erode the contribution process if you continue doing this
17:49 hmmmm maybe to you it is trivial, but what if you made a mistake and there was a security vulnerability due to it?
17:50 hmmmm and then there's a bunch of people all running minetest who get a virus because of your code change not being peer reviewed
17:53 paramat it's simple mapgen code i understand completely, my own code, in an area i have previously been given authority over. if i had any doubt of it's safety i would not have merged it
17:55 hmmmm but the point of code review is not self-confidence
17:55 hmmmm you might've overlooked something that if you had known about, you would have a doubt then
17:56 hmmmm calling a code functionality change like that simple is a matter of opinion
17:56 paramat it was borderline trivial, and i admit probably not trivial, but considering the other circumstances this was ok, i even announced and waited 3 hours
17:56 hmmmm we drew the hard line of what 'trivial' is defined as with documentation changes or code changes that do not affect functionality or significantly impact structure
17:57 hmmmm if a code change is not trivial in that sense, then it needs to get approval
17:57 hmmmm some of my changes i think are very simple and i have all the confidence in the world in them, but i still don't go ahead and merge it on my own
17:57 hmmmm i wait for others to review it, even if it takes a week
17:57 paramat you are ultra-strict about this for others, but relaxed about it when you do it yourself
17:58 hmmmm how so
17:58 hmmmm what code changes have i defined as trivial that aren't trivial by that definition
17:58 paramat because you've taken the initiative in a reasonable circumstance yourself before
17:58 paramat which was fine by me
17:59 hmmmm do you mean this?  https://github.com/minetest/minetest/commi​t/27aff22a9b68044d3ea51db731597834336effa3
18:00 hmmmm now this is an example of a commit that i would consider trivial even though it does change code
18:00 hmmmm i don't remember if it had approval or not before merging
18:01 paramat well maybe not that, i just know you've done it and est has complained about you taking initiative :]
18:01 hmmmm i know that more than a year ago i used to be a lot looser about what trivial is
18:01 hmmmm but not since we came up with actual definitions
18:01 paramat nevermind about slippery slope, just consider what i actually did, and the circumstances
18:01 hmmmm i think it was earlier this year when we spoke with celeron
18:01 paramat no you've done it more recently
18:02 hmmmm and est31 complained about it?
18:02 hmmmm you must be talking about the PR i merged of somebody else's that i reviewed
18:02 paramat still, your strictness is appreciated, our quality has gone up
18:02 hmmmm the argument there was not about whether it was trivial or not, but whether it required 2 code reviews
18:03 paramat yeah maybe that
18:03 hmmmm and est argued that any api change should automatically require at least 2 code reviews
18:03 hmmmm that sounds reasonable to me, so that's what we go by now
18:03 hmmmm but that had nothing to do with the question of whether a commit was trivial or not
18:03 hmmmm it was an outside contribution from somebody else, and i did a code review, it's just that somebody thought that one wasn't enough
18:04 paramat sure, i admit my PR wasn't trivial, but was close
18:04 hmmmm which is why i'm not reverting it
18:04 paramat and i know that code better than anyone
18:04 hmmmm i'm just telling you watch it
18:04 paramat i will
18:04 hmmmm "knowing the code better than anyone" doesn't mean you can't make a mistake
18:05 paramat agreed
18:05 hmmmm i "know" the schematics code well and i made a mistake where instances never get dropped causing a slow memory leak
18:05 hmmmm but then again, i could argue that's also the fault of the developer for not tracking object handles
18:07 davisonio joined #minetest-dev
18:07 paramat anyway, i'm happy you're paying attention and noticed a suspicious merge
18:07 hmmmm the thing is, i wasn't paying attention
18:07 hmmmm like i said i've been busy all week and i just so happened to look on irc and saw that
18:07 hmmmm it was by chance
18:08 hmmmm this could've easily slipped by, and what if it had contained a security vulnerability
18:08 paramat that code obviously could not
18:09 hmmmm well how do you know that
18:09 paramat look at the code
18:10 hmmmm but nobody would've looked at the code without a code review and i hadn't looked at the channel by chance
18:12 paramat the code obviously cannot be harmful
18:12 hmmmm you'd be surprised at how non-obvious exploitable code can be
18:13 hmmmm it doesn't always need to involve "char foobar[10]; strcpy(foobar, boo);"
18:14 hmmmm it's also possible that a modification in logic could make a condition that was otherwise unexploitable now exploitable
18:15 paramat yeah. anyway even recently your strictness does vary a lot, depending on how relaxed you are, so watch it ;]
18:16 Void7_ joined #minetest-dev
18:20 * Krock wonders if paramat will add a self-destructing "One approval" label just for him
18:24 paramat heh
18:32 rain_gloom joined #minetest-dev
18:32 paramat i actually don't use the approve label for my own PRs anymore
18:35 davisonio joined #minetest-dev
18:37 Krock paramat, I was just a bit amused because sometimes it was approved, a day later not anymore
18:37 paramat yeah approval comes and goes frequently
18:39 Krock2 joined #minetest-dev
18:45 whitephoenix joined #minetest-dev
18:48 paramat nore sfan5 sofar please can anyone review game#965 ?
18:48 ShadowBot https://github.com/minetes​t/minetest_game/issues/965 -- Ability to Disable bones, drop items or keep inventory. by tenplus1
18:52 paramat 4 months old PR
18:58 Krock Oh.. looked at it so many times and can't find anything to correct in it
19:09 Miner_48er joined #minetest-dev
19:12 xunto joined #minetest-dev
19:26 Void7_ joined #minetest-dev
19:33 Fixer_ joined #minetest-dev
19:35 edgrey joined #minetest-dev
19:44 jomat joined #minetest-dev
19:48 Fixer joined #minetest-dev
20:03 jomat joined #minetest-dev
20:08 Fixer_ joined #minetest-dev
20:22 Amaz joined #minetest-dev
20:27 Void7_ joined #minetest-dev
20:44 SloanKloan joined #minetest-dev
20:53 Fixer joined #minetest-dev
20:54 Lunatrius joined #minetest-dev
21:03 troller joined #minetest-dev
21:30 Darcidride joined #minetest-dev
21:53 SloanOnLinux joined #minetest-dev
22:37 TheReaperKing joined #minetest-dev
23:08 Void7_ joined #minetest-dev
23:29 Tmanyo joined #minetest-dev
23:37 Player_2 joined #minetest-dev

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