Config
Log for #openttd on 29th June 2021:
Times are UTC Toggle Colours
00:02:44  *** Tirili has quit IRC
00:02:53  *** Gustavo6046 has quit IRC
00:12:01  *** Tirili has joined #openttd
00:37:17  <FLHerne> haha
00:39:10  *** Gustavo6046 has joined #openttd
00:41:05  *** Gustavo6046 has quit IRC
00:54:48  *** Gustavo6046 has joined #openttd
02:34:30  *** glx has quit IRC
02:52:34  *** Wuzzy has quit IRC
02:53:56  *** D-HUND has joined #openttd
02:57:19  *** debdog has quit IRC
03:05:05  *** WormnestAndroid has joined #openttd
03:13:07  *** WormnestAndroid has quit IRC
03:13:57  *** WormnestAndroid has joined #openttd
03:47:31  *** Tirili has quit IRC
03:59:25  *** roadt__ has joined #openttd
04:06:32  *** roadt_ has quit IRC
04:15:56  *** Flygon has joined #openttd
04:48:24  *** Gustavo6046 has quit IRC
04:50:43  *** Gustavo6046 has joined #openttd
05:54:46  *** andythenorth has joined #openttd
05:55:35  *** sla_ro|master has joined #openttd
06:11:16  *** andythenorth has quit IRC
06:44:02  *** andythenorth has joined #openttd
06:59:11  *** andythenorth has left #openttd
07:20:02  *** gelignite has joined #openttd
07:21:00  *** kgarza_ has joined #openttd
07:28:33  *** kgarza has quit IRC
08:03:29  <TrueBrain> that moment you rebase a patch from before fmt, and it errors out on fmt .. for no apparent reason other than to say: fuck youuuuuuuu
08:03:29  <TrueBrain> lol
08:25:35  *** grossing has quit IRC
08:25:40  *** Elouin has quit IRC
08:41:45  *** tokai|noir has quit IRC
08:45:02  <TrueBrain> so I did the rebase patch by patch, compiling each one fixing issues .. and the fmt error I got, I no longer get .. wth .. weird compilers being weird
08:48:19  *** tokai has joined #openttd
08:48:19  *** ChanServ sets mode: +v tokai
08:52:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JOk7C
08:53:02  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JOk7C
09:04:40  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #9404: Fix f7e390bd: freeaddrinfo() is not guaranteed to handle a nullptr graceful https://git.io/Jcmup
09:05:25  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcmzJ
09:06:46  *** Gustavo6046 has quit IRC
09:08:16  *** Gustavo6046 has joined #openttd
09:08:32  <TrueBrain> milek7: if you can confirm #9404 to fix the issue, that would be appreciated :D
09:09:16  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #9404: Fix f7e390bd: freeaddrinfo() is not guaranteed to handle a nullptr graceful https://git.io/JcmzA
09:09:39  <TrueBrain> wow, quick review :o
09:11:10  <TrueBrain> "_network_game_info.join_key = _network_game_info.join_key;" <- that moment you know you did some C++-ification, but did it in an automated way ..
09:11:17  <TrueBrain> I love having useless statements in my code, don't you?
09:11:28  <LordAro> eh, compiler will get rid of it for you
09:11:34  <TrueBrain> exactly!
09:11:57  <TrueBrain> I was testing deepsource.io on some Python repos of mine .. it made some .. wrong assumptions
09:12:06  <TrueBrain> I had a line of code: "mydict[key]"
09:12:11  <TrueBrain> it says: that is a useless statement
09:12:14  <TrueBrain> I say: you are dead-wrong
09:12:31  <TrueBrain> in Python the act of accessing a key can have a result .. especially as this was a defaultdict
09:12:45  <TrueBrain> so "if key in mydict" is True after that statement
09:13:03  <LordAro> mm
09:13:11  <LordAro> though i would start to question the quality of the code :p
09:13:14  <LordAro> bit of a code smell
09:13:16  <TrueBrain> you got to love the nuances of different programming languages :D
09:13:26  <TrueBrain> owh, it does have a comment in front of it explaining why it is needed, don't worry :D
09:13:30  <TrueBrain> as for sure it is unusual
09:13:32  <LordAro> :D
09:13:42  *** gelignite has quit IRC
09:13:44  <TrueBrain> but for a linter to say: "useless statement" ... you better be sure that is a correct claim :)
09:13:55  <TrueBrain> if it would have said: "please don't do this crap", I agree :D
09:14:50  <TrueBrain> euhm ... my PR now fails on something brilliant ... "Check Annotations" fails ...
09:14:54  <TrueBrain> warning: String name 'STR_NETWORK_SERVER_LIST_ENTER_IP' does not exist in master file
09:15:07  <TrueBrain> since when do we pick up strgen warnings :D
09:15:33  <TrueBrain> I want to let eints fix this problem ;)
09:19:24  <peter1138> Hmm, do I change my UUIDs to match another system, or add another reference field...
09:19:58  <TrueBrain> the first sounds bad :P
09:20:40  <peter1138> I think you're right.
09:20:49  <TrueBrain> \o/
09:20:56  <peter1138> Imagine if someone comes along and adds a third system...
09:21:10  <TrueBrain> omg
09:21:11  <TrueBrain> no
09:21:11  <TrueBrain> no
09:21:13  <TrueBrain> please god no
09:21:28  <TrueBrain> (I hope you are imagining the right The Office meme now)
09:21:49  <peter1138> I've never watched The Office, so... no.
09:21:52  <TrueBrain> peter1138: I do have to ask .. do you make it a point to write proper english on IRC these days? Or did you always do that? :P
09:22:13  <TrueBrain> as in, full sentences, starting with a capital, ending with a dot, etc :P
09:23:53  <peter1138> It's not a concious effort right now, although I do remember years ago a guy getting totally wound up by me doing so.
09:24:11  <TrueBrain> I noticed it a few weeks ago, and I am surprised how consistent you are with it :D
09:24:19  <TrueBrain> once you notice, it is hard not to notice :)
09:36:55  <TrueBrain> yippie ... you can now reuse your join-key over and over and over and over again, as long as you know the secret that belongs to it :D
09:36:55  <TrueBrain> sweet
09:39:17  *** Gustavo6046 has quit IRC
09:47:20  *** Gustavo6046 has joined #openttd
09:50:04  <LordAro> TrueBrain: dependabot alerts have got longer
09:50:13  <TrueBrain> few more days :P
09:50:28  <TrueBrain> been testing dependabot to open PRs as they come
09:50:33  <TrueBrain> it is not as bad as I imagined
09:50:45  <TrueBrain> might want to try it on a few OpenTTD repos too
09:50:57  <TrueBrain> but I think it would be good to mute it reporting on IRC, as it is noisy
09:51:08  <TrueBrain> either way, why I mention this .. those PRs are often easier to merge
09:51:14  <TrueBrain> as it is not "everything at once" concept
09:51:26  <LordAro> feel like they'll get forgotten if they're not announced here :p
09:54:24  <TrueBrain> as if they aren't forgotten if they are ;)
09:54:36  <TrueBrain> I mean, there has been .. how many up for a whole month now :D
09:54:50  <LordAro> oh no
09:54:53  *** tokai|noir has joined #openttd
09:54:53  *** ChanServ sets mode: +v tokai|noir
09:55:06  <TrueBrain> but we can try with announcing, I am fine with that
09:55:23  <TrueBrain> more people can judge a lot easier if a single one can be merged I am sure :) Just some need more attention
10:00:36  <FLHerne> TrueBrain: Eh, just make it auto-merge them :p
10:00:43  <TrueBrain> FLHerne: yeah ... no :P
10:00:51  <TrueBrain> GitHub also made that nearly impossible, which is a good thing :D
10:00:53  <FLHerne> People can always complain if it breaks
10:01:35  *** tokai has quit IRC
10:01:35  <TrueBrain> breaking is not what worries me :D But too often these days a project gets hijacked and someone uploads a .. weird new version :)
10:01:49  <TrueBrain> GitHub now made it that dependabot PRs cannot access ANY secrets
10:02:09  <TrueBrain> as people were exfiltrating them by updating things like NodeJS deps with code to do that :D
10:02:54  <LordAro> isn't security fun?
10:03:41  <TrueBrain> the years of neglect is finally catching up with us :D
10:03:48  <TrueBrain> for Python it is not that bad, as we have rather few dependencies
10:03:51  <TrueBrain> but if you check a NodeJS project
10:03:57  <TrueBrain> how .. ever .... has that gone well for so long
10:04:36  *** Gustavo6046 has quit IRC
10:06:04  <TrueBrain> GitHub is reporting a security alert to upgrade to aiohttp 3.7.4 or higher .. where we run 3.7.4.post0
10:06:10  <TrueBrain> I think some parsing goes wrong at GitHub's side there :P
10:10:09  *** Samu has joined #openttd
10:29:44  <DorpsGek> [OpenTTD/OpenTTD] fsimonis opened pull request #9405: Reduce allocations in string handling https://git.io/Jcmd2
10:31:32  *** Gustavo6046 has joined #openttd
10:32:05  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #9405: Reduce allocations in string handling https://git.io/JcmdN
10:32:53  <TrueBrain> you are quick today LordAro :D
10:34:15  <TrueBrain> really cool way to analyze OpenTTD, that is nice
10:35:00  <peter1138> I don't those "allocations" need to be avoided. And the string stack change seems odd.
10:36:01  <peter1138> Maybe fine, I don't really know what that is doing.
10:36:53  <peter1138> I don't THINK those...
10:38:54  <peter1138> They will be fixed "stack" allocations so I'm not sure there's any performance hit with them.
10:40:00  <peter1138> Or is it heap? I dunno. I guess I can shut up :D
10:40:12  <TrueBrain> it is a local variable, so I would guess they are on the stack too
10:40:15  <TrueBrain> but .. it is 2048 bytes long
10:40:35  <TrueBrain> so I would guess that it is not that great for performance, to trash your stack with 2048 every freaking time :P
10:40:53  <TrueBrain> maybe compilers put it on the heap exactly because of that, dunno :D
10:40:53  <peter1138> Trash?
10:41:06  <TrueBrain> sorry, cache-line trashing :D
10:41:32  <TrueBrain> well, it is within a page .. who knows
10:42:24  <TrueBrain> godbolt says it stays on the stack
10:42:31  <TrueBrain> for what-ever that is worth :D
10:42:46  <TrueBrain> "sub     rsp, 1928"
10:42:47  <TrueBrain> lol
10:43:45  <peter1138> As it becomes non-threadsafe that could be more of an issue than a few stack "allocations".
10:44:12  <TrueBrain> I am surprised that a heapallocator would pick this up
10:44:21  <TrueBrain> I wonder what is going on there exactly
10:44:46  <TrueBrain> but yeah, I think LordAro correctly nudge the author to that part .. thread safety is a thing :D
10:45:12  <TrueBrain> it most likely "is fine" in the current design, as these functions are "most likely" only called from the draw-thread :P
10:45:43  <LordAro> you'd hope so, really
10:45:43  <peter1138> It's not impossible that all of these calls are only done in a GUI context, so not an issue.
10:45:45  <LordAro> but...
10:45:50  <peter1138> Hah.
10:46:05  <TrueBrain> this doesn't happen often that we are all 3 in agreement like this :P
10:46:09  <TrueBrain> I got to frame this now :D
10:46:23  <TrueBrain> anyway, I have been wondering before, why GetString works like this
10:47:14  <TrueBrain> I am just waiting for Rb to C++-ify it :P
10:47:51  <peter1138> Well building a string in a preallocated space is easier for memory management, I guess.
10:48:56  <TrueBrain> shit, lunchtime has come and gone .. lets try to catch up
10:50:31  <peter1138> "The Marketplace has extensions that can help with '.cs~stashed changes' files"
10:50:38  <peter1138> That seems... no thanks.
10:50:40  <Rubidium> GetString is also in the network code
10:54:59  *** Gustavo6046 has quit IRC
11:17:15  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9405: Reduce allocations in string handling https://git.io/JcYeB
11:17:19  <TrueBrain> lets ask the author if he can enlighten us :)
11:26:10  *** whatsthetime has quit IRC
11:35:49  *** Gustavo6046 has joined #openttd
12:20:27  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JOk7C
12:20:43  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9234: UDP query of game script https://git.io/JcYqU
12:21:42  *** glx has joined #openttd
12:21:42  *** ChanServ sets mode: +v glx
12:23:43  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcYqg
12:24:49  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JOk7C
12:24:53  <TrueBrain> stupid MSVC with its rightful complaints
12:25:30  <glx> it's weird how compiler complain differently :)
12:27:19  <TrueBrain> yup
12:27:45  <TrueBrain> glx: I have an issue with the Check Annotation you might know what to do with .. it fails on strgen stuff that will be fixed by eints after the PR is merged
12:28:04  <glx> but strgen warnings should not be picked
12:28:44  <TrueBrain> they are now ...
12:28:55  <TrueBrain> I also thought they weren't, but .. :D
12:29:01  <glx> they don't match the problem matcher
12:29:12  <glx> unless matcher changed
12:30:08  <TrueBrain> or GitHub added their own?
12:30:09  <TrueBrain> dunno :P
12:33:29  <glx> only gcc/clang it seems
12:33:49  <TrueBrain> indeed
12:34:03  <TrueBrain> well, Check Annotation doesn't prevent merging, so I guess it is okay
12:34:16  <glx> https://github.com/ammaraskar/gcc-problem-matcher/commit/a141586609e2a558729b99a8c574c048f7f56204 <-- recent change
12:34:17  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JOk7C
12:34:40  <TrueBrain> :D
12:36:22  <glx> guess we can make a small change to strgen so it doesn't match :)
12:36:47  <TrueBrain> feels silly
12:36:54  <TrueBrain> I think it is okay that it tells us something is worth checking out
12:37:00  <TrueBrain> but it is also fine to ignore these warnings :P
12:37:53  <glx> main issue is it only reports 10 first warnings for each run
12:40:18  <glx> so expected strgen warnings will hide unexpected compile warnings if not checking build logs manually
12:40:22  <TrueBrain> true
12:41:59  <TrueBrain> either way, STUN patch is done \o/ Fixed every bug people reported to me .. pretty sure there are more bugs hiding, but only one way to find those :D :D
12:50:51  <TrueBrain> glx: maybe a good solution is to change "Warning" into "Hint" or something
12:50:57  <TrueBrain> as clearly it is not that big of a deal if it is for the non-english
12:51:13  <TrueBrain> or "Info"
12:52:15  <TrueBrain> would also means that warnings triggered by the base language are reported
12:52:36  <glx> I can look at it
12:52:43  <TrueBrain> \o/
12:53:08  <glx> and --warning switch should still warn (as it's explicit)
12:58:30  *** snail_UES_ has joined #openttd
13:00:32  *** Kitrana has quit IRC
13:08:08  <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcYnl
13:09:22  <glx> https://github.com/OpenTTD/OpenTTD/blob/c539343ba914d68a9ab560bf09ed1854ab6d43f6/src/strgen/strgen_base.cpp#L652-L656 but https://github.com/OpenTTD/OpenTTD/blob/c539343ba914d68a9ab560bf09ed1854ab6d43f6/src/strgen/strgen_base.cpp#L821
13:09:59  <glx> so basically the check is useless
13:12:00  *** Kitrana has joined #openttd
13:14:18  *** Kitrana has quit IRC
13:15:32  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcYcy
13:26:57  <Samu> im using AIGroup.GetCurrentUsage for the first time
13:27:34  <Samu> it's helping me manage the number of ships on a route so far
13:33:55  <Samu> hmm, if ships end up stacking, there will be periods of current usage being ~ 100%
13:35:11  *** WormnestAndroid has quit IRC
13:35:13  <Samu> i think that's still okay, it means there may be more passengers waiting at a dock
13:35:25  <Samu> so it's okay to add an extra ship
13:36:41  *** WormnestAndroid has joined #openttd
13:42:28  *** nielsm has joined #openttd
14:00:16  <Samu> hmm not doing good in the long run
14:00:34  <Samu> who would have thought ship stacking would ruin management
14:02:19  <Samu> if one dock provides tons of passengers and the other doesn't, i get a bouncing usage of > 90% and < 10%
14:02:55  <Samu> ship stacking ruins the usage of AIGroup.GetCurrentUsage
14:03:04  <Samu> must think of another method
14:13:07  <Samu> > 66% doing better than > 50%
14:15:11  <glx> TrueBrain: oh in theory msvc matcher should also pick strgen warnings, but output format is wrong ;)
14:17:06  <glx> matcher expects "file(line)" but strg outputs "file (line)"
14:17:59  <glx> msvc itself understands "file (line)" btw
14:33:43  *** WormnestAndroid has quit IRC
14:51:54  *** sla_ro|master has quit IRC
15:16:24  <glx> ah no it's not matched because there's no code, the space before '(' doesn't matter
15:39:55  *** WormnestAndroid has joined #openttd
16:22:36  *** HerzogDeXtEr has joined #openttd
16:32:48  *** Taschi has joined #openttd
16:41:46  *** gelignite has joined #openttd
16:44:39  *** grossing has joined #openttd
16:45:46  <peter1138> Hmm, well...
17:10:52  *** andythenorth has joined #openttd
17:11:04  <andythenorth> TrueBrain if STUN is done, goes it ship 2.0
17:11:09  <andythenorth> out-of-cycle release
17:15:21  <TrueBrain> why not just 1.11.3 :P
17:15:30  <andythenorth> where is the glamour
17:16:02  <glx> ok I think I have a working hack
17:20:49  <glx> test running on https://github.com/glx22/OpenTTD/actions/runs/983544863
17:22:59  <DorpsGek> [OpenTTD/OpenTTD] spnda commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOLu
17:24:19  <Taschi> everybody knows that transferring information between two computers is an unsolved problem in computer sciences
17:24:37  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOL6
17:27:30  <Taschi> sheesh, i make ONE joke about how computer networking is complicated and kind of annoying and my soccer livestream chooses that exact moment to fall apart
17:27:55  <DorpsGek> [OpenTTD/OpenTTD] spnda commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOLx
17:29:54  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOtm
17:31:42  <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOtW
17:32:19  *** Progman has joined #openttd
17:32:46  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOt0
17:33:20  <glx> so yeah, there's magic involved
17:36:36  <TrueBrain> also I just copy/pasted existing entries :P
17:37:18  <glx> no annotations on my 9017 copy
17:39:49  <Taschi> what exactly does the "annotation" checker do?
17:39:54  <Taschi> check for compiler warnings?
17:40:28  <glx> yes it "fails" the CI if there's any warning (else they are forgotten)
17:40:54  <Taschi> it fails if there's a warning from *any* compiler?
17:41:16  <glx> exactly
17:41:42  <glx> we don't want any warning in master
17:42:05  <Taschi> this is one of those moments where i'm very happy to work in Java and with just one compiler, but to be fair, we just ignore warnings anyway
17:42:08  <glx> without using Werror because that's stops compilation early
17:45:38  *** Flygon has quit IRC
17:49:38  <DorpsGek> [OpenTTD/OpenTTD] glx22 opened pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOq5
17:52:26  *** Taschi has quit IRC
18:10:11  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOYD
18:14:26  <DorpsGek> [OpenTTD/OpenMSX] ConiKost opened issue #22: No tests (make check) is possible https://git.io/JcOOG
18:22:36  *** sla_ro|master has joined #openttd
18:28:02  *** Elouin has joined #openttd
18:41:05  *** Wolf01 has joined #openttd
18:49:35  <LordAro> https://arstechnica.com/gadgets/2021/06/hackers-exploited-0-day-not-2018-bug-to-mass-wipe-my-book-live-devices/
18:49:57  <TrueBrain> owh boy
18:50:47  <TrueBrain> that is just ...
18:50:47  <TrueBrain> wth
18:51:02  <LordAro> "oops"
18:51:27  <TrueBrain> owh, lines of code with a // are NOT working?
18:51:33  <TrueBrain> I was thinking it was for higher privileges
18:52:52  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/JcOn3
18:52:53  <DorpsGek>   - Update: Translations from eints (by translators)
18:57:25  <DorpsGek> [OpenTTD/OpenMSX] glx22 commented on issue #22: No tests (make check) is possible https://git.io/JcOOG
19:02:36  <DorpsGek> [OpenTTD/OpenTTD] fsimonis commented on pull request #9405: Reduce allocations in string handling https://git.io/JcOCk
19:03:53  <TrueBrain> so now the real issue pops up :D
19:03:59  <LordAro> uh
19:04:10  <LordAro> pretty sure that still counts as a compiletime constant
19:04:28  <TrueBrain> godbolt agrees with you
19:05:10  <DorpsGek> [OpenTTD/OpenMSX] ConiKost commented on issue #22: No tests (make check) is possible https://git.io/JcOOG
19:26:39  <DorpsGek> [OpenTTD/OpenTTD] glx22 updated pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOq5
19:37:23  *** gelignite has quit IRC
19:38:21  *** gelignite has joined #openttd
19:44:00  <DorpsGek> [OpenTTD/website] dadadom opened pull request #217: Fix link to COPYING.md https://git.io/JcO0s
19:45:30  <DorpsGek> [OpenTTD/website] TrueBrain approved pull request #217: Fix link to COPYING.md https://git.io/JcO0w
19:51:07  <DorpsGek> [OpenTTD/website] TrueBrain merged pull request #217: Fix link to COPYING.md https://git.io/JcO0s
19:51:38  <DorpsGek> [OpenTTD/website] TrueBrain created new tag: 1.4.32 https://git.io/JcOEA
19:53:00  <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOuq
19:54:14  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOuE
19:56:23  *** nielsm has quit IRC
20:00:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOzX
20:01:18  <TrueBrain> sorry glx, I could have looked before writing a comment, would have been easier .. sorry :)
20:01:35  <glx> oh strgen is not easy to follow
20:04:40  <glx> I think only 4 strgen_warning calls are only for translation (not counting the ones for -w)
20:08:34  <TrueBrain> I noticed also stuff like utf-8
20:08:41  <TrueBrain> but that might need to become an error :D
20:09:21  <TrueBrain> anyway, maybe good to find a better name for that function not to confuse future-us
20:10:23  <TrueBrain> the more I look at it
20:10:30  <TrueBrain> the more I just realise it is already a mess
20:10:49  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcO2X
20:11:15  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcO2D
20:12:05  <glx> yeah and utf-8 thing is not related to lang files (I think it can't come from them), it's from explicit calls with SCC_, or via the EmitSingleChar table, so really an strgen bug if it happens
20:14:01  <glx> so the 2 important warnings should indeed be errors
20:14:19  <TrueBrain> not for this PR :P
20:14:20  <andythenorth> yo
20:15:17  <glx> but yeah strgen is already a mess (like the macro so MSVC reports error without stopping build)
20:15:45  <andythenorth> so newgrf flag to disable the desert / above snowline town growth behaviour?
20:17:40  <glx> oh and of course strgen_warning (and similar) are different when building openttd :)
20:19:27  <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #9406: Change: [strgen] 'warnings' for translations are now 'infos' unless invoked with -w or -t https://git.io/JcOq5
20:19:51  <glx> let's hide the mess under the carpet now :)
20:20:01  <TrueBrain> :D
20:20:09  <glx> it works, don't fix
20:20:25  <TrueBrain> sometimes words to live by
20:21:15  <andythenorth> so can I build my fork on the official infra?
20:23:21  <glx> oh that's probably why the "should be errors" are warnings, because it's in code shared by squirrel
20:24:15  <TrueBrain> andythenorth: yup
20:24:34  <glx> s/by/with/
20:25:01  <andythenorth> 'assuming my fork builds at all' :P
20:25:32  <glx> should work, you just need to set the appropriate secrets I think
20:25:50  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9298: Change: move sensitive information to secrets.cfg and private information to private.cfg https://git.io/JGLne
20:27:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOwt
20:30:47  *** virtualrandomnumber has joined #openttd
20:31:03  *** virtualrandomnumber has quit IRC
20:35:29  *** Tirili has joined #openttd
20:37:16  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOr1
20:39:29  *** HerzogDeXtEr has quit IRC
20:42:31  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOoC
20:42:47  <TrueBrain> std::map vs std::unordered_map .. I read up on it the other day .. I ran away
20:42:51  <TrueBrain> I need frosch for guidance :P
20:43:05  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JOk7C
20:43:49  <TrueBrain> I could replace it for a std::vector<std::pair<>> and just ignore the problem, I guess :D
20:44:40  <TrueBrain> but a rb-tree vs bucket-linked-list .. that is a weird discussion to read up on on the Internet :D
20:48:51  <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOod
20:49:08  <Tirili> hey
20:49:11  <Rubidium> how big is the map generally going to be?
20:49:58  <milek7> std::map should be named std::sorted_map, imo :P
20:50:27  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOKJ
20:50:34  <TrueBrain> Rubidium: 2 items
20:50:35  <TrueBrain> :P
20:51:17  <TrueBrain> and another one closeby is one for every connection attempt .. so 1, possibly 2, if you are really weird: 3 :P
20:51:26  <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOKt
20:52:08  <TrueBrain> milek7: it is not so much the name, it is also that they are two completely different beasts
20:52:21  <TrueBrain> and on the internet ... that triggers a lot of opinions
20:52:23  <TrueBrain> A LOT OF THEM
20:52:49  <TrueBrain> in my use-case, performance-wise it will not be something we can measure
20:53:14  <TrueBrain> it becomes close to the pedantic side of it all .. "strictly seen" .. :D
20:53:36  <TrueBrain> milek7: would you have a chance to check https://github.com/OpenTTD/OpenTTD/pull/9404 on Haiku?
20:53:37  <Rubidium> TrueBrain: conceptually unordered_map might be better, but due to the small size the rb-tree will probably win out on both performance and size. So I'd keep it as is
20:54:00  <TrueBrain> Rubidium: k, tnx
20:54:13  <TrueBrain> would you mind adding that as a comment to the PR, so I can resolve the comment? :D
20:57:00  *** Samu has quit IRC
20:58:59  *** gelignite has quit IRC
21:00:57  <Tirili> Has anyone ever had the problem that some websites / ip-addresses are reachable (by ping for example) but others aren't?
21:01:19  <Tirili> Sorry for abusing this channel for offtopic!
21:01:47  <milek7> TrueBrain: maybe tomorrow
21:01:50  <milek7> but it did work before with this fix
21:01:57  <TrueBrain> haha, like we are always ontopic :D
21:01:58  <TrueBrain> k, tnx milek7
21:02:11  <TrueBrain> Tirili: IPv4 / IPv6 is most of the reason for me :P
21:02:15  <TrueBrain> and crappy ISPs in the past
21:02:28  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #9404: Fix f7e390bd: freeaddrinfo() is not guaranteed to handle a nullptr graceful https://git.io/Jcmup
21:02:29  <Tirili> ok
21:02:42  *** Wolf01 has quit IRC
21:05:20  <Tirili> There have been a lot of thunderstorms the last days, could that be related?
21:05:32  <TrueBrain> if you live in remote Alaska, sure
21:05:43  <Tirili> No Central Europe
21:05:59  <TrueBrain> but if you can reach some sites, and not others, it can have so many reasons
21:06:04  <TrueBrain> for example, if you are using the Tor network
21:06:08  <TrueBrain> that is fully expected to happen
21:06:19  <TrueBrain> if your ISP is being a bad boy, and Cloudflare / AWS / Azure / GCP / ... blacklisted you
21:06:20  <TrueBrain> sure, that happens to
21:06:22  <glx> broken peering too
21:06:25  <TrueBrain> a peering can be down
21:06:33  <TrueBrain> transit can be conguested
21:06:52  <TrueBrain> even satellites that are in the wrong orbit can be a cause
21:07:18  <Tirili> Maybe a thunderbolt has caused some infrastructure to break.
21:07:30  <TrueBrain> rather unlikely that to be the cause, honestly
21:07:39  <TrueBrain> most internet exchanges are connected via fibre
21:07:50  <TrueBrain> and most sane ISPs have several peerings and transits
21:08:00  <TrueBrain> a lot has to go wrong before things become unreachable for a longer period
21:08:41  <Tirili> Cannot a lightning cause highly increased voltages for short times in the power network that finally destroys hardware?
21:08:57  <TrueBrain> sure, but that would mean you have NO internet :P
21:09:20  <glx> always possible, but yeah it's more like all or nothing in this case, not just some
21:09:22  <Tirili> Maybe just a few specific routes where affected.
21:09:28  <Tirili> ok
21:09:33  <TrueBrain> Internet is build to be redundant
21:09:36  <TrueBrain> if a route from A to B fails
21:09:42  <TrueBrain> it uses A -> C -> B to get to B
21:09:53  <TrueBrain> giving your ISP has the peerings or is willing to pay the transit to make that happen ofc
21:10:23  <Tirili> hm
21:10:48  <Tirili> Duckduckgo.org works, google.com doesn't.
21:11:15  <glx> google not reachable almost never happens :)
21:11:27  <glx> unless it's a DNS issue
21:11:51  <TrueBrain> you cannot ping a lot of domains in the world, if that is what you are trying
21:12:14  <glx> google is pingable
21:12:22  <Tirili> ping 8.8.8.8 does not work
21:12:25  <Tirili> here
21:12:46  <glx> works fine for me :)
21:13:05  <Tirili> Maybe I should give my ISP a call tomorrow.
21:15:13  <Tirili> Router restart doesn't help either.
21:15:41  <glx> traceroute may give a hint
21:15:52  <TrueBrain> http://weathermap.ovh.net/ <- just to give you an idea how for example OVH is interconnected
21:16:49  <TrueBrain> peering status of large exchanges used to be public, but these days that is no longer the case .. which is a bit sad
21:16:54  <TrueBrain> was fun to see how ISPs were doing
21:17:09  <glx> TrueBrain: still doesn't help when a datacenter is burnt ;)
21:17:19  <TrueBrain> glx: you could even see that on the weathermap
21:17:46  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOPR
21:18:07  <glx> my isp used to have a similar map, but they removed it a long time ago
21:18:15  <TrueBrain> tnx Rubidium !
21:18:32  <TrueBrain> glx: yeah ... and it used to be public but unpublished for a while too
21:18:34  <DorpsGek> [OpenTTD/OpenTTD] spnda commented on pull request #9017: Feature: play easier together with friends from behind home routers https://git.io/JcOPw
21:18:37  <TrueBrain> so if you knew the address, you could still see it
21:18:42  <TrueBrain> but even that ... is no longer the case
21:19:24  <TrueBrain> did he just add a graph without telling what the axis are? :D
21:20:11  <Tirili> TrueBrain, is that weathermap link a map of all the most important online routers?
21:20:24  <Tirili> in the world?
21:20:29  <TrueBrain> Tirili: that is OVH, and only OVH
21:20:52  <glx> now when I have an issue I just start by checking https://www.free-reseau.fr/ but it's not provided by my isp, it's a tool made by users
21:21:33  <Tirili> Do both of you live in France?
21:21:40  <TrueBrain> http://www.level3.com/~/media/files/maps/en-network-services-level-3-network-map.ashx <- this is one of the biggest backbones of the Internet (the one that supplies connectivity to the company your ISP buys it from)
21:21:48  <TrueBrain> but it is missing fancy realtime stats
21:21:52  <TrueBrain> just shows what is connected to what
21:22:15  <TrueBrain> it does show how crazy many redundant ways there are to get from A to B
21:22:43  <Tirili> Well I'd like to view it, but the god won't let me.
21:22:50  <Tirili> So thanks for the link..
21:22:58  <TrueBrain> sounds like your ISP didn't pay his bill :P
21:23:05  <TrueBrain> (I am joking)
21:23:18  <glx> IIRC they were touched by a destroyed submarine cable, but the result was just highly increased ping
21:25:30  <Tirili> It all started yesterday in the evening (early morning). First I could still check my e-mails, but now even that is impossible.
21:26:22  *** andythenorth has quit IRC
21:27:05  <Tirili> Could someone send a copy of the text-article behind this links to me via pm?:
21:27:06  <Tirili> https://www.ln-online.de/Lokales/Ostholstein/Sierksdorf-Vodafone-Kunden-ohne-Internet-Telefon-und-Fernsehen
21:28:01  <glx> there's a paywall :)
21:28:06  <Tirili> oh well
21:28:11  <Tirili> thx though
21:38:03  <TrueBrain> in case you missed the news of the day: https://github.blog/2021-06-29-introducing-github-copilot-ai-pair-programmer/
21:38:11  <TrueBrain> no clue what to think about this ... really curious if it really works
21:38:21  <TrueBrain> sounds like stackoverflow integrated in your editor :P
21:39:06  *** spnda has joined #openttd
21:40:05  <spnda> btw to the irc: ive edited my comment regarding the unordered_map thing. turns out that std::hash makes it a heck of a lot slower so yeah, stay with std::map.
21:40:22  <glx> hehe
21:42:16  *** spnda has quit IRC
21:58:32  *** iSoSyS has joined #openttd
22:00:48  *** WormnestAndroid has quit IRC
22:06:12  *** WormnestAndroid has joined #openttd
22:09:06  *** WormnestAndroid has quit IRC
22:09:21  *** WormnestAndroid has joined #openttd
22:11:16  <Tirili> ha actually some websites work, if you give them a little time.
22:11:26  <Tirili> The link with the paywall is an example.
22:11:45  <Tirili> And it's a paywall where you can read the text, when you open the sites's sourcecode.
22:13:15  *** iSoSyS has quit IRC
22:22:19  *** sla_ro|master has quit IRC
22:33:03  *** Progman has quit IRC
22:33:20  *** Tirili has quit IRC
22:49:53  *** snail_UES_ has quit IRC
22:57:13  *** snail_UES_ has joined #openttd
23:15:42  *** WormnestAndroid has quit IRC
23:16:45  *** WormnestAndroid has joined #openttd
23:28:53  *** tokai has joined #openttd
23:28:53  *** ChanServ sets mode: +v tokai
23:35:18  *** tokai|noir has quit IRC

Powered by YARRSTE version: svn-trunk