Config
Log for #openttd on 24th April 2020:
Times are UTC Toggle Colours
00:05:40  *** Wormnest_ has quit IRC
00:31:49  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfIF4
00:38:27  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfIF4
00:40:27  *** tokai|noir has joined #openttd
00:40:27  *** ChanServ sets mode: +v tokai|noir
00:47:20  *** tokai has quit IRC
00:57:33  *** Wormnest_ has joined #openttd
01:09:06  *** Flygon has joined #openttd
01:56:07  <DorpsGek_III> [OpenTTD/nml] FLHerne commented on pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfLfr
02:29:20  *** D-HUND has joined #openttd
02:32:41  *** debdog has quit IRC
03:09:14  *** k-man has quit IRC
03:16:07  *** glx has quit IRC
03:18:35  *** Wormnest_ has quit IRC
03:19:20  *** k-man has joined #openttd
03:56:15  *** k-man has quit IRC
04:12:18  *** k-man has joined #openttd
04:25:04  *** Wormnest_ has joined #openttd
05:07:04  *** k-man_ has joined #openttd
05:08:13  *** k-man has quit IRC
05:18:38  *** Wormnest_ has quit IRC
05:40:56  *** snail_UES_ has quit IRC
06:02:41  *** sla_ro|master has joined #openttd
06:18:30  *** andythenorth has joined #openttd
06:27:16  *** Wolf01 has joined #openttd
06:34:02  <TrueBrain> and _dp_ , you ended up being able to do some nasty shit?
06:40:06  <andythenorth> moin
06:52:17  <Wolf01> Hi
06:54:08  <andythenorth> what shall I do today?
06:54:09  <andythenorth> or
06:54:37  <andythenorth> shall I do what I'm obliged to do
06:54:49  * andythenorth knows the answer :|
06:57:12  <Xaroth> I wouldn't know... also, probably not good to ask TrueBrain as he'll likely mutter something about taking over the world.
06:57:51  <TrueBrain> its pinky, its pinky, its pinky and the brain brain brain brain brain brain brain brain
06:57:57  <Xaroth> :D
06:58:15  * andythenorth counters with log
06:58:35  <andythenorth> https://www.youtube.com/watch?v=-fQGPZTECYs
07:04:53  * andythenorth will mostly be doing grown up meetings today
07:04:57  <andythenorth> and SEO
07:09:54  <TrueBrain> my idea was the work all day on AWS, but it seems work requires a bit attention too
07:09:56  *** cHawk- has quit IRC
07:10:00  <TrueBrain> (took the day off, but .. what can you do)
07:10:36  <andythenorth> I want to take the day off, I have been productive this week
07:10:52  <andythenorth> but we have monthly information security meeting booked, not optional :|
07:11:14  <TrueBrain> I feel your pain
07:11:17  <andythenorth> the risk register won't review itself!
07:12:13  <andythenorth> I noticed that overnight, we have made nml much better
07:12:18  <andythenorth> GOOD
07:13:08  *** Samu has joined #openttd
07:13:11  <Samu> hi
07:18:12  <Samu> ahem, _dp_, do you have enough patience to improve this? https://pastebin.com/raw/d7vyCBiN
07:32:18  <DorpsGek_III> [OpenTTD/nml] andythenorth commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLtW
07:47:05  <_dp_> TrueBrain, not rly, unless it broke something that I didn't notice xD
07:47:17  <TrueBrain> :)
07:47:21  <TrueBrain> well, that is good I guess :P
07:48:49  <_dp_> csp seems to save it, non-csp browsers probably got xssed xD
07:49:30  *** supermop_Home has quit IRC
07:51:24  <_dp_> Samu, god no
08:07:07  *** Smedles has quit IRC
08:08:33  *** Smedles has joined #openttd
08:20:41  <TrueBrain> _dp_: that won't be good either; I assume there was some escaping going on tbh :)
08:21:12  <TrueBrain> owh, there is for most part
08:21:12  <TrueBrain> sweet
08:21:41  <_dp_> There is escaping but some xss doesn't rely on escaped stuff
08:21:56  <_dp_> eg it accepts javascript: and data: just fine
08:21:57  <TrueBrain> I like the data: urls :)
08:22:25  *** cHawk has joined #openttd
08:22:43  <TrueBrain> well, always feel free to (ab)use staging, and if you find a way to poke a hole, let us know :)
08:33:12  <_dp_> TrueBrain, considering I'm using same stuff for my websites it would be strange if I knew how to break it xD
08:33:57  <Eddi|zuHause> strange maybe, but not unheard of :p
08:41:44  <_dp_> modern frameworks are kinda boring
08:42:11  <_dp_> I remember in php days there were like 10 escaping methods with 9 of them being faulty xD
08:42:52  <TrueBrain> you have no idea how often this still happens ...
08:43:15  <TrueBrain> many companies don't know what OWASP top 10 is .. and they developer very large webapplications
08:43:22  <TrueBrain> s/developer/develop/
08:43:44  <TrueBrain> "we wrote our own ORM, it is faster"
08:43:47  <TrueBrain> etc
08:43:47  <TrueBrain> etc
08:45:40  <andythenorth> I used to write 'eval(user_input)'
08:45:46  <andythenorth> so that we could write code through the web
08:45:47  <andythenorth> oops
08:46:23  <_dp_> lol
08:46:37  <_dp_> and I was going full lex+yacc in private project xD
08:52:28  <TrueBrain> okay, I have a small Python script that writes the traefik configuration for the NLB .. now how do I run this when needed .. hmmm
08:55:05  *** Xetalim has joined #openttd
09:09:35  *** gelignite has joined #openttd
09:19:19  <DorpsGek_III> [OpenTTD/nml] planetmaker commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLOE
09:24:05  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLOM
09:48:43  *** cHawk has quit IRC
09:53:57  <Eddi|zuHause> <TrueBrain> "we wrote our own ORM, it is faster" <-- and each department of the company does that, so the final product ships 10 different versions of the same thing
09:54:09  <Samu> english question: "will be build" or "will be built"?
09:54:11  <TrueBrain> and they are all rubbish :P
09:54:32  <Samu> to be build, to be built?
09:54:45  <Eddi|zuHause> god forbid the departments actually talking to each other to coordinate
09:58:13  <doug16k> Samu, built
09:58:42  <Samu> openttd code comments have many of these "be build"
09:58:54  <doug16k> "built" when it tried to be the invalid word builded
09:59:36  <doug16k> past tense of build
10:00:01  <doug16k> ah it is also the future tense
10:00:51  <doug16k> I see why you ask
10:02:36  <Samu> 	/* If the two tiles are 1 tile apart and are sloped, or more than 1 tile apart, the pathfinder wants an aqueduct
10:02:36  <Samu> 	 *  to be built.
10:02:50  <Samu> or is it to be build
10:03:00  <doug16k> yes. definitely built
10:03:08  <Samu> ok, thx
10:04:50  <doug16k> that almost looks like a search and replace mistake, if I had to guess
10:05:05  <doug16k> comment bugs build up
10:05:28  <doug16k> very slowly
10:06:30  *** Smedles_ has joined #openttd
10:08:06  <doug16k> or someone knowing other languages and not perfect. I can't complain until I know two spoken languages really well :D
10:09:20  *** Smedles has quit IRC
10:11:10  <doug16k> I would love a conditional branch that branches "if the train carries mostly" and {passengers,mail,etc...} dropdown then goto
10:12:25  <doug16k> and look at the capacity of each type and see if the max one matches the criteria
10:16:11  *** andythenorth has left #openttd
10:16:17  *** syr has quit IRC
10:16:39  *** syr has joined #openttd
10:16:52  <LordAro> Samu: OTTD comments are about 70% written by Dutch or German people :p
10:25:41  <DorpsGek_III> [OpenTTD/nml] planetmaker requested changes for pull request #103: Blacken & Flake8 nml https://git.io/JfLGq
10:29:19  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLGG
10:32:12  *** Xetalim has quit IRC
10:39:25  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLG6
10:47:28  <Eddi|zuHause> doug16k: so if the train carries 40% passengers, 30% ore and 30% coal, can it justify going down the "passenger" branch even though it's 60% non-passengers?
10:49:18  <doug16k> put the priority order the same as the order it presents them when you buy. first one in that order with most wins?
10:49:46  <doug16k> ah
10:49:51  <Eddi|zuHause> that does not answer the question
10:50:19  <doug16k> it does carry mostly passengers
10:50:29  <Eddi|zuHause> but, does it really?
10:51:22  <Eddi|zuHause> can the party who got 40% of the vote claim they "won" the election, even though the two 30% parties can form a coalition against them?
10:51:55  <LordAro> Eddi|zuHause: as it turns out, yes
10:52:00  *** Tirili has joined #openttd
10:52:03  <LordAro> repeatedly.
10:52:56  <doug16k> count the cars, group by type, order descending by count,order_in_buy_screen, select top one
10:53:23  <doug16k> -order_in_buy_screen
10:53:54  <doug16k> will that not select passenger on your train?
10:54:10  <Eddi|zuHause> that's not the question
10:54:25  <Eddi|zuHause> the question is whether that decision is justified
10:54:47  <Eddi|zuHause> philosophy, not technique.
10:55:00  <Samu> @calc 303/333
10:55:00  <DorpsGek> Samu: 0.90990990991
10:55:07  <Xaroth> I think you're putting too much thought in it :P
10:55:08  <Samu> amazing :)
10:56:09  <Eddi|zuHause> i'm asking this because of two reasons: a) different players will have opposing opinions on this matter, and b) how to explain which of the methods is chosen, with very restricted space?
10:56:11  <doug16k> I want to send excessive passenger trains on an loop to a rule that says go to the sell-me maintenance building and stop
10:56:21  <Samu> yesterday it was 403 ticks, today it is down to 303
10:56:37  <Samu> making progress
10:57:42  <doug16k> what's more straightforward than conditional rule by trains primary purpose
10:58:15  <Eddi|zuHause> i'm not seeing how such an order would even help you with that
11:00:04  <Eddi|zuHause> also, if you zoom out one game design level, you should ask: how many other people could find this useful for other purposes
11:00:33  <Eddi|zuHause> and i'm very far on the "this doesn't help enough people to justify putting it in" side
11:01:29  <Eddi|zuHause> and that's completely ignoring the fact that conditional orders are very incompatible with cargo distribution
11:03:47  <doug16k> you want me to have all separate rule lists to micro on all the different trains
11:04:07  *** Xetalim has joined #openttd
11:04:08  <doug16k> I want one rule program to be able to make basic decision about skipping
11:05:15  <doug16k> or as tools to make it possible to make your cars go somewhere without too much trouble if you find you have too much of one type
11:05:53  <Eddi|zuHause> how is it too much to ask that you have separate groups/orders for passenger trains and mail trains?
11:05:54  <doug16k> sitting and micromanaging cars coming into a depot is not effective use of time
11:06:52  <doug16k> then you have to go micro each type of train a station can support?
11:07:04  <doug16k> figure out the insertion point
11:07:22  <Eddi|zuHause> i've no clue what you're talking about
11:07:58  <doug16k> if I have a bunch of different rules I will end up missing places unless I micro manage each type of cargo at each station
11:08:04  <Eddi|zuHause> there's certainly loads of sources for micromanagement in this game. i fail to see how "decide whether this is a passenger or mail train" is a major contributor to that
11:08:58  <Eddi|zuHause> if at all, that is a group management feature and not an order list feature
11:10:50  <Eddi|zuHause> in any case, adding a feature that only one player in the world knows how to use is not an effective use of UI-real-estate
11:15:58  <doug16k> ya why would people want to make a rule that helps them manage the amount of a type of train is in a loop. that would never be wanted
11:16:02  <Samu> can't get lower than 303 t.t
11:16:04  <doug16k> thanks for chatting
11:17:40  <Eddi|zuHause> how is it my fault that you can't bring your point across?
11:18:46  <doug16k> I don't have to convince you, I have the source
11:19:32  <Eddi|zuHause> you don't have to convince me, but if you can't convince me, how would you convince any of the devs to include your suggestion?
11:19:36  <doug16k> I thought maybe you have a really good argument against it and everyone comes here with that rule
11:20:13  <Eddi|zuHause> as far as i'm aware you're the first person who ever wanted this specific rule
11:20:31  <doug16k> do you know assembly language?
11:20:41  <Eddi|zuHause> a bunch of people wanted other rules that would have helped their specific use case
11:20:44  <doug16k> they are conditional branches
11:20:58  <Eddi|zuHause> ... but they usually failed to make a case for how that would benefit a broader audience
11:21:05  <doug16k> I can't condition on train type
11:22:54  <Eddi|zuHause> yes, but making an order list shouldn't require of a player that they know assembly language
11:23:21  <doug16k> they already are using conditional branches if they ever use conditional order
11:23:22  <Eddi|zuHause> because the median openttd player would likely not have that kind of knowledge
11:23:48  <doug16k> that order window. that's assembly language. they don't know it, but it is
11:24:01  <doug16k> there are instructions, the operands are locations
11:25:30  <doug16k> there are a bunch of flag bits in the opcode for each instruction, for load all, etc
11:25:31  <Eddi|zuHause> and many many people already can't cope with that
11:26:16  <Eddi|zuHause> if an inexperienced player opens a window, which is overloaded with complicated stuff, they turn around and walk away
11:26:40  <doug16k> one more item in the conditional order drop down drives them away??
11:26:49  <Eddi|zuHause> yes
11:26:59  <doug16k> seems like you drive away any contributions too
11:27:38  <Eddi|zuHause> to be fair, most "contribuitions" anywhere in the world work like that
11:27:44  <doug16k> one chat with one person determines the fate of any idea?
11:27:50  <Eddi|zuHause> no
11:28:23  <Eddi|zuHause> i'm trying to break open your mind and reconsider your idea in a broader scope. if you can't do that, your idea is already dead on the spot.
11:28:52  <doug16k> you already have it there where you choose the criteria. age, etc
11:29:10  <doug16k> how open should my mind be? propose some new special location?
11:29:55  <doug16k> how about "carries only" and a car type. then no argument about what is the most
11:31:18  <Eddi|zuHause> well, there's many levels of scope that you could view this from. in the easiest case: how would a person that is not you use this feature for something that is completely different from your use case
11:31:21  <doug16k> I thought "mostly" made it usable in more scenarios
11:32:08  <Eddi|zuHause> and the most complicated scope is: the order UI is already a horrible mess, let's restruture it completely to make more sense of it, and actually adding stuff without making it more overloaded
11:32:31  <doug16k> you can do this: add a shortcut that skips stuff that many car types don't want. put a shortcut that skips them. then, put a way for me to say skip that if you are the wrong type
11:34:10  <doug16k> if the player micros that rule set, then cars will not all go that way for nothing
11:34:27  <doug16k> and sharing is easier
11:34:44  <Eddi|zuHause> and somewhere inbetween is: how would a player that doesn't know what he wants to do see whether this thing that you want to add is what he wants to do
11:34:53  <DorpsGek_III> [OpenTTD/nml] planetmaker commented on pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfLnb
11:36:35  <doug16k> exploring the amazing stuff you could do with rules might be a fun part of the gameplay
11:36:47  <doug16k> what restrict it to only simplistic use?
11:37:12  <milek7> so, squirrel controlled orders? ;d
11:37:15  <doug16k> for me that wrong an xhci driver in my kernel
11:37:19  <doug16k> wrote
11:39:05  <Eddi|zuHause> so you want to write a feature for the intersection of "players of openttd" with "people who once wrote a kernel driver", and expect that number to be larger than 1?
11:39:14  <doug16k> the redstone of openttd
11:39:54  <Eddi|zuHause> we have that playerbase covered. they built a CPU using trains and signals
11:40:32  <Eddi|zuHause> just look up "load balancer" and "priority signals"
11:41:23  <Eddi|zuHause> ... and these people have an endless supply of "if only we added <X> we could do more complex stuff" type of suggestions
11:42:34  <Eddi|zuHause> there are already more suggestions than anyone could ever code, so naturally we will be looking for reasons to turn them down
11:43:32  <Eddi|zuHause> and even if someone manages to come along with actual code to support their idea, there's still loads of reasons to turn them down
11:44:14  <Eddi|zuHause> if you build a jenga tower of code, any wrong step could topple over the tower and ruin the game for everyone
11:45:45  <Eddi|zuHause> just remember that scene in the meaning of life, where the guy eats just a tiny thing on top of all the other things he ate
11:46:27  <Eddi|zuHause> that's what its like to program for a long standing project like this
11:47:41  *** doug16k has left #openttd
11:48:16  <Eddi|zuHause> (i think i won? :p)
11:48:24  <TrueBrain> is this winning?
11:48:35  <TrueBrain> (I have not read ANYTHING you guys wrote for the last hour or so; I am just trolling :D)
11:49:17  <Eddi|zuHause> i don't think it's worth your time reading that :p
11:50:18  <Eddi|zuHause> TLDR: he requests a ridiculously niche feature, and then can't handle the fact that i'm telling him it has no chance of going in
11:51:07  <TrueBrain> that is not very polite :P
11:52:59  <DorpsGek_III> [OpenTTD/OpenTTD] koala12345 opened issue #8096: commands redirected to other stations  https://git.io/JfLcP
11:53:59  <Eddi|zuHause> it would have been less polite if i had used those exact words, yes :p
11:55:52  <TrueBrain> :D
11:57:52  <Eddi|zuHause> i was poking his idea from various sides, and he basically failed every test on how robust his idea is for anything...
11:58:06  <Eddi|zuHause> and then i got more and more direct, and apparently that pissed him off
12:04:09  <DorpsGek_III> [OpenTTD/OpenTTD] James103 commented on issue #8096: commands redirected to other stations  https://git.io/JfLcP
12:04:19  <milek7> he had a point though
12:04:21  <milek7> projects with more haphazard development are more fun :P
12:06:18  <DorpsGek_III> [OpenTTD/OpenTTD] koala12345 commented on issue #8096: commands redirected to other stations  https://git.io/JfLcP
12:06:18  <DorpsGek_III> [OpenTTD/OpenTTD] koala12345 closed issue #8096: commands redirected to other stations  https://git.io/JfLcP
12:12:49  <Eddi|zuHause> you need the right balance between conservatism (keep the project running) and experimentalism (add new features)
12:13:13  *** WormnestAndroid has quit IRC
12:13:26  *** WormnestAndroid has joined #openttd
12:42:57  *** nielsm has joined #openttd
12:49:04  <TrueBrain> ugh .. so much time has gone into this already, and only now I notice that you cannot enable proxy-protocol for outgoing connections from Traefik
12:49:05  <TrueBrain> lame
12:53:59  <Eddi|zuHause> clearly you should write your own system
12:55:01  <TrueBrain> seriously, it turns out there is no software that does both TCP and UDP with proxy support on the backend connection
12:55:03  <TrueBrain> super annoying
12:55:59  <TrueBrain> well, proxy support v2, that is
13:09:15  <Eddi|zuHause> dunno what that even means
13:13:13  *** KenjiE20 has quit IRC
13:17:14  *** KenjiE20 has joined #openttd
13:17:39  <Samu> down to 260 ticks!
13:18:20  *** KenjiE20 has quit IRC
13:19:02  *** KenjiE20 has joined #openttd
13:19:07  *** andythenorth has joined #openttd
13:19:13  <andythenorth> newgrf orders
13:19:17  <andythenorth> newgrf UI also
13:19:20  * andythenorth leaves again
13:19:24  *** andythenorth has left #openttd
13:21:27  <DorpsGek_III> [OpenTTD/nml] glx22 commented on pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfL4g
13:27:21  <DorpsGek_III> [OpenTTD/nml] TrueBrain commented on pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfL4F
13:28:38  <TrueBrain> I hope that made sense ..
13:38:46  *** Xetalim has quit IRC
13:40:32  *** snail_UES_ has joined #openttd
13:47:04  *** mcbanhas has joined #openttd
13:47:08  <mcbanhas> hello
13:47:46  <mcbanhas> LordAro: you got a minute?
13:49:12  <milek7> doesn't nginx support udp proxying?
13:58:43  <LordAro> @topic get 3
13:58:44  <DorpsGek> LordAro: Don't ask to ask, just ask
13:58:47  <LordAro> mcbanhas: (no)
14:01:07  <mcbanhas> I assumed asking for time did not fall into that category :p
14:01:58  <LordAro> perhaps, but i am here all the time, and i will get to whatever you ask when i can
14:02:08  <LordAro> alternatively, there might be someone else here who can help
14:02:24  <mcbanhas> Ok, sure
14:06:07  <mcbanhas> Just in case you remember: a month or so ago I submitted a PR for fixing a lot of stuff with in-game text and tooltips. Because reviewing it would be a tricky process and submitting multiple PRs for each category of changes was out of the question, we agreed the best way to handle it would be for me to create a manual of style for general text, much like the one for coding style.
14:07:15  <mcbanhas> I'm almost done writing the manual of style, and I've even made a thread about it for people to discuss and suggest changes.
14:08:27  <mcbanhas> Once I'm done I'd like you to go over it, given you're an EN-UK native speaker, so you can give your final approval or suggest any ammendments necessary.
14:09:35  <mcbanhas> Any other native EN-UK speaking core dev would also do, of course, but I'm not aware of any other atm.
14:12:35  <LordAro> depends how active peter1138 is feeling
14:15:20  <mcbanhas> The first version of the document would be in markdown format. I would upload it in the forum thread or something like that. Once approved, I would transcribe the whole thing to the wiki.
14:16:32  <mcbanhas> I could also do a wiki draft instead, if you'd prefer that, since I would upload example pics, and it would be more legible I guess.
14:26:27  <Samu> back to 309 ticks :( im sad, I regressed
14:28:07  <Samu> it could be worse
14:28:20  <Samu> 6 more ticks than what I used to have
14:28:34  *** Xetalim has joined #openttd
14:28:38  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfIF4
14:43:12  *** glx has joined #openttd
14:43:12  *** ChanServ sets mode: +v glx
14:45:44  <Samu> looks like my conflict tile checker wasn't 100% accurate
14:45:51  <Samu> it was overchecking
14:46:24  <Samu> excessive conflicts when in truth there weren't any
14:46:40  <Samu> less conflicts = more nodes to check = more ticks :( oh well
14:46:50  <Samu> can't have it all
14:47:28  <Samu> I'm still unsure about Array vs AIList
14:47:38  <Samu> which one would give me more speed
14:56:12  *** Xetalim has quit IRC
15:02:01  <DorpsGek_III> [OpenTTD/OpenTTD] sheepo99 commented on issue #8095: Feature suggestion: option to automatically space out vehicles in a shared order https://git.io/JfIUi
15:08:20  *** supermop_Home_ has joined #openttd
15:08:25  *** gelignite has quit IRC
15:08:37  <supermop_Home_> hello
15:08:38  *** Tirili has quit IRC
15:12:06  <Samu> confirmed, AITileList is faster
15:12:31  <Samu> hi
15:12:43  <supermop_Home_> so tired
15:13:31  <Samu> i miss a feature from squirrel 3
15:13:40  <Samu> array.find(item)
15:13:44  <Samu> or value
15:14:35  <Samu> but it's ok, we have AITileList.KeepList, i can just smash a list into another and check for conflicts this way
15:15:16  *** Xetalim has joined #openttd
15:21:29  *** Progman has joined #openttd
15:26:56  *** Xetalim has quit IRC
15:28:42  <TrueBrain> 
15:28:42  <TrueBrain> 2020-04-24 15:28:19 INFO Dropping invalid packet from ('10.0.89.85', 38856): PacketInvalidData('packet too short')
15:28:46  <TrueBrain> WE HAVE COMMUNICATION
15:28:49  <TrueBrain> I REPEAT
15:30:43  <TrueBrain> hmm .. and I don't have native IPv6 .. tha tis annoying
15:39:03  <planetmaker> huston, we have communication? :)
15:39:19  <planetmaker> +o
15:44:05  *** Wormnest_ has joined #openttd
15:51:54  <SpComb> TrueBrain: implementing your own TCP/UDP proxy?
15:52:15  <TrueBrain> no, I tried very very VERY hard to avoid that
15:52:20  <SpComb> that would be a fun project in Go. I should have volunteered earlier :P
15:57:44  <SpComb> if this is still about the AWS IPv4/IPv6 TCP/UDP thing we discussed earlier, and it requires some custom implementation of that, I might be interested in the implementation of something like that
15:58:04  <TrueBrain> I do not want to implement my own proxy; that would be VERY difficult to pull off
15:58:17  <TrueBrain> but atm I do have nginx running, on proxy-protocol v1, doing the same
15:58:28  <SpComb> I can't remember all the details of this
15:58:31  <TrueBrain> but if you enjoy rebuilding an NLB on AWS, there might be some bits and pieces I need help with :P
15:59:04  <TrueBrain> hmm ... CDK can be a real PITA to deploy mutations
16:00:52  <SpComb> this was about the lack of IPv6 support for AWS NLB?
16:00:55  <TrueBrain> yes
16:01:32  <TrueBrain> so what I currently have is an ASG with user-data that spins up EC2 instances with nginx + python script that provisions the correct configuration
16:01:49  <TrueBrain> next task is to make some lambdas and connect them to hooks to automate the coming and going of services
16:02:13  <SpComb> but if nginx ngx_stream_proxy_module can implement it, then that's probably the least in terms of maintenance overhead
16:02:22  <TrueBrain> that is what I did now
16:02:26  <TrueBrain> sadly, it is proxy-protocol v1 only
16:02:37  <TrueBrain> which is a bit annoying to pick up on the server-side of things
16:02:41  <TrueBrain> but .. I guess it will have to do
16:02:45  <SpComb> yeah, so it just needs some config automation for nginx?
16:02:52  <TrueBrain> that part is done now
16:02:58  <TrueBrain> there was not much "just" about it
16:03:03  <TrueBrain> took .. 6 hours to build? :D
16:03:13  <TrueBrain> 120 lines of Python code :P
16:04:02  *** Flygon has quit IRC
16:05:58  <TrueBrain> okay .. lets see if this redeployment doesn't destroy the ECS cluster :D
16:06:18  <SpComb> I haven't dealt with AWS ASG:s yet, but I will need to soon... however, my plan was to use kube as a layer over all of the service orchestration
16:06:42  <TrueBrain> I have run away from k8s as fast and far as I could
16:07:13  <SpComb> so just ASG the kube nodes and use EKS or whatever, and keep all the actual services as kube stuff, not AWS-specific
16:07:21  <SpComb> *or use
16:07:44  <TrueBrain> anyway, EKS / ECS, it won't make a real difference in this case
16:08:01  <TrueBrain> if you cannot use the ALB and NLB
16:08:07  <TrueBrain> you have some work to do :D
16:08:25  <TrueBrain> so if you use EKS, but can I suggest ECS, I hope you only have HTTP traffic to route :D
16:09:00  <SpComb> yeah my work usecase is pure HTTP, like all sensible usecases are :P
16:09:23  <TrueBrain> I provisioned that with CDK, ECS and ALB, very easy to deal with
16:09:30  <TrueBrain> even stuff like: on this domain, that path has to go to this pod
16:09:35  <TrueBrain> and on that path to the other pod
16:09:37  <TrueBrain> very nice :)
16:09:51  <TrueBrain> okay, this didn't kill the ECS, so that is good ..
16:10:48  *** Xetalim has joined #openttd
16:13:02  <SpComb> was this the master-server/browser or bananas or what service?
16:13:29  *** andythenorth has joined #openttd
16:13:30  <andythenorth> YO
16:13:52  <TrueBrain> it is now bananas
16:13:57  <TrueBrain> but it will also be used for master-server
16:14:44  <SpComb> why do I remember UDP being involved
16:15:45  <TrueBrain> Masterserver
16:15:57  <TrueBrain> YES! Security Group change was accepted, and now traffic is flowing again :)
16:16:05  <SpComb> bananas was some custom TCP protocol?
16:16:58  <glx> except for actual downloading yes
16:17:56  <DorpsGek_III> [OpenTTD/nml] andythenorth commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLws
16:19:40  <TrueBrain> oh-oh, my ASG keeps on rebuilding instances
16:19:42  <TrueBrain> that is not good :D
16:19:48  <andythenorth> LordAro do you want to investigate Blacken further, or should I read the rest of that PR and then approve it?
16:20:10  <LordAro> (black)
16:20:19  <LordAro> i do want to investigate further
16:20:36  <LordAro> there are definitely a few things in there that would benefit from disabling formatting
16:20:43  <LordAro> not all that pm pointed out imo, but some
16:22:18  <DorpsGek_III> [OpenTTD/OpenTTD] sheepo99 commented on issue #8092: Swedish translation issue in the performance rating window https://git.io/JfTUv
16:22:18  <andythenorth> I used to do the alignment thing, it was common on coop projects
16:22:27  <andythenorth> I have a few cases of it remaining, e.g. https://github.com/andythenorth/iron-horse/blob/master/src/global_constants.py#L114
16:22:47  <andythenorth> but generally I dropped it, the readability tradeoff isn't high versus the messing around with spaces
16:26:53  <DorpsGek_III> [OpenTTD/nml] FLHerne commented on pull request #103: Blacken & Flake8 nml https://git.io/JfLwX
16:32:15  <andythenorth> FLHerne like nml/global_constants/py?
16:32:23  <andythenorth> .py *
16:34:04  <FLHerne> andythenorth: And action0properties, action2var_variables, ...
16:34:52  <FLHerne> Some of those have 20+ lines in a table
16:35:08  <FLHerne> Which is just an unreadable mess without formatting
16:35:21  <SpComb> TrueBrain: but from what I remember of the previous discussion, my suggested long-term approach was to implement a new HTTP server for bananas/masterserver, and then implement a custom UDP/TCP -> HTTP proxy for backwards-compatbility... assuming that some new version of the OpenTTD client itself would just use HTTP as well
16:35:56  <SpComb> which is probably also way more work than scripting some nginx configs... although I wonder what the actual proxy backend is going to be?
16:36:02  <FLHerne> andythenorth: If anything, global_constants isn't as bad, because it's mostly just one value per line
16:36:36  <FLHerne> So the alignment makes it harder to scan the values, but at least they're not jumbled together
16:36:46  <TrueBrain> SpComb: we have a deadline of .. 6 more days to get this fixed. So all solutions involving long-term are already out of the window ;)
16:36:51  <TrueBrain> but I am off for dinner now; talk after
16:37:37  <andythenorth> FLHerne I am +/-0, I used to care much more about formatting like this for readability
16:38:05  <andythenorth> and I used to argue with people who wanted to pep-8 everything regardless of readability
16:38:32  <SpComb> TBH just dropping IPv6 support is probably also an acceptable short-term solution
16:38:55  <andythenorth> now I am more [shrug emoji]
16:39:11  <TrueBrain> no, it is not. That means 15+% of our users won't be able to use the service. That is completely unacceptable even, as it goes about solutions :)
16:41:57  <supermop_Home_> can't we just put the site on gophernet instead?
16:56:06  *** syr has quit IRC
16:56:29  *** syr has joined #openttd
16:59:59  <SpComb> replace both with editable wiki pages: List of Multiplayer Servers
17:00:01  *** Xetalim has quit IRC
17:01:20  <nielsm> TrueBrain do you know how many of those 15% of users have ipv6-only connections and how many have both ipv4 and ipv6 but just prefer ipv6 for services that support it?
17:04:44  <LordAro> a much much smaller number, i'd bet
17:05:35  *** gelignite has joined #openttd
17:15:14  *** gnu_jj_ has quit IRC
17:35:26  *** frosch123 has joined #openttd
17:42:59  *** gnu_jj has joined #openttd
17:44:34  <TrueBrain> nielsm: lot of that 15% is from Asia, which are known for their IPv6 only
17:44:43  <TrueBrain> but it is difficult to say without disabling IPv6
17:44:50  <TrueBrain> nevertheless, we are not going to do that :)
17:46:32  <TrueBrain> so .. network paths are made .. now I need to implement Proxy Protocol v1 I guess .. TO THE BATMOBILE
17:49:34  *** Progman has quit IRC
17:50:20  *** Progman has joined #openttd
17:56:53  <andythenorth> quak
17:57:16  * andythenorth was hoping overnight fairy would have styled this https://dev.openttdcoop.org/attachments/download/9642/upload_new_package.png
17:57:27  <andythenorth> turns out...no fairies, imps, pixies or goblins
17:58:38  *** Xetalim has joined #openttd
18:00:12  <frosch123> if the table would have align-top, it's not so bad
18:01:24  <DorpsGek_III> [OpenTTD/bananas-frontend-web] andythenorth commented on pull request #16: Format the elements in the page header https://git.io/JfL6c
18:01:31  <andythenorth> frosch123 I'll do that now
18:01:45  <frosch123> hmm, i think we should delete the subheaders
18:01:57  <andythenorth> I tried to write a nice reply to Auge, but my brain is fail, end of the week
18:02:09  <andythenorth> subheaders?
18:02:16  <frosch123> they had a purpose when i added them initially, but then other stuff changed, and now they are pointless
18:02:36  <frosch123> https://bananas.staging.openttd.org/package/base-graphics/4f474658/2020-04-12T16:26:12+00:00 <- the first line after "OpenGFX 0.5.5"
18:03:26  <andythenorth> oh yeah I've been deleting those as I go :)
18:03:31  <andythenorth> I assumed you'd agree :)
18:08:48  <DorpsGek_III> [OpenTTD/bananas-server] TrueBrain opened pull request #5: Add: support Proxy Protocol v1 (if requested) https://git.io/JfL6K
18:08:52  <TrueBrain> there we go; that was relative painless
18:09:00  <andythenorth> ok I needed to rebase, there was a conflict
18:09:05  <andythenorth> wish me luck :|
18:09:11  <TrueBrain> luck!
18:09:15  <DorpsGek_III> [OpenTTD/bananas-server] TrueBrain updated pull request #5: Add: support Proxy Protocol v1 (if requested) https://git.io/JfL6K
18:13:00  <andythenorth> hmm, total manual picking of lines needed :)
18:13:02  <andythenorth> oof
18:14:10  <frosch123> you indened <tbody> :)
18:14:25  <frosch123> don't change indenting, it breaks everything
18:16:19  *** y2kboy23 has quit IRC
18:18:34  *** y2kboy23 has joined #openttd
18:23:48  <andythenorth> git hates me :)
18:24:22  <andythenorth> you don't fancy merging it yet? :P
18:24:32  <frosch123> shall i rebase your stuff?
18:24:40  <andythenorth> if you don't mind
18:24:45  <andythenorth> otherwise I can actually fix it
18:24:59  <andythenorth> but I am eating a burger
18:25:40  <frosch123> abbde5f06d67b3681e29e4d25c27d9fb49935b74  is your HEAD?
18:25:59  <andythenorth> 84baed588bb0ebfb0cf4a2906dae1b097c3f6ffc
18:26:00  <TrueBrain> hmmm .. burgers ....
18:28:03  <mcbanhas> Finished the first draft of my manual of style
18:28:12  *** supermop_Home_ has quit IRC
18:28:38  <andythenorth> TrueBrain one is usually enough? o_O
18:28:59  <mcbanhas> I'd never imagine that in my life I would ever end up doing one of the most boring things ever conceived by men.
18:29:59  <andythenorth> you have been on Zoom calls all day too? :o
18:31:13  <mcbanhas> No, I use Jitsi.
18:33:26  <peter1138> I have been biked... and now I have been beered.
18:33:42  <milek7> surely nobody is using ipv6-only connection
18:35:59  <peter1138> Why not?
18:36:09  <peter1138> It is the future.
18:36:57  <andythenorth> beer
18:37:14  * andythenorth wonders when beer will go out of fashion
18:38:15  <peter1138> It is the future.
18:41:22  *** supermop_Home has joined #openttd
18:41:35  <supermop_Home> i am pro-burger
18:42:06  <nielsm> I am having fish and wine
18:44:07  <mcbanhas> White or red?
18:45:05  <supermop_Home> lockdown = no access to orange wine
18:46:23  <andythenorth> supermop_Home chemicaltown has gone quite deep into chemicals now
18:46:29  <andythenorth> not so much New Extreme :P
18:47:43  <TrueBrain> red fish .. what kind of fish is that?
18:48:22  <nielsm> salmon?
18:48:23  <nielsm> kind of
18:48:26  <mcbanhas> TrueBrain: red snappers are otherwise known as redfish if you go to a fishmonger.
18:48:56  <nielsm> but what I'm having is white fish and white wine
18:48:57  <mcbanhas> They're great as an oven roast.
18:49:03  <TrueBrain> hmm, even in dutch is it "red"
18:49:04  <TrueBrain> funny
18:49:15  <TrueBrain> learn something new every day :)
18:50:32  <supermop_Home> andythenorth "you're not from brighton" economy?
18:51:24  <supermop_Home> andythenorth should i buy this? https://www.discogs.com/The-Chemical-Brothers-Surrender/release/140845
18:52:13  <andythenorth> unknown :)
18:52:44  <supermop_Home> MD Walkman might still work
18:55:34  <mcbanhas> How to make a great snapper roast: scale the snapper, don't skin it or make it into fillet. Slice the Snapper into 4 slices including head & tail. Skin and quarter some Russet potatoes. In a wide pan with olive oil sweat one large onion and some tinned tomatoes. Put the snapper slices in the center of a wide ceramic oven tray with the potatoes spreaded around and season with salt + pepper. Pout the onion mix on top of it all,
18:57:08  <mcbanhas> Final result should be something like this: https://media-cdn.tripadvisor.com/media/photo-s/03/8d/d0/74/marques-da-varjota.jpg
18:58:04  <mcbanhas> You can use other types of fish too, of course, but snapper is cheap and easy to find.
19:04:50  *** tokai has joined #openttd
19:04:50  *** ChanServ sets mode: +v tokai
19:11:43  *** tokai|noir has quit IRC
19:21:43  <frosch123> andythenorth: https://github.com/frosch123/bananas-frontend-web/commits/style-classic-openttd
19:22:19  <frosch123> andythenorth: you wrote somewhere that "html validation" is last. please extend that list with "fix indentation" is also last :)
19:22:41  <andythenorth> +1
19:22:45  <andythenorth> it also produces big diffs :P
19:23:22  <andythenorth> sometimes I do it first, with an automated tool ;)
19:25:23  <frosch123> aw, the footer i c&p from main website does not work
19:25:51  <frosch123> i thought the css would have been magically present
19:26:36  <andythenorth> might just need a class or id
19:28:41  <andythenorth> hmm my local branch won't rebase to yours, I'll just delete it and re-branch
19:43:26  <TrueBrain> @calc 0.0000002083 * 10 * 60 * 5
19:43:26  <DorpsGek> TrueBrain: 0
19:43:32  <TrueBrain> well, that is not helpful
19:44:15  <peter1138> Almost Lego... https://www.youtube.com/watch?v=Ae7XLg3RFWY
19:46:02  <TrueBrain> it does look awesome :)
20:00:58  <Samu> what does lhs and rhs stand for
20:02:12  <Samu> return lhs._type == rhs._type && lhs._unVal.raw == rhs._unVal.raw;
20:02:13  <andythenorth> left hand side
20:02:14  <Samu> like in this
20:02:15  <andythenorth> right hand side
20:02:24  <Samu> oh ok
20:04:15  <andythenorth> frosch123 I keep finding new package edit forms :D
20:08:23  <andythenorth> can't remember how to get to version info page though
20:09:37  <andythenorth> maybe I need an archived package
20:15:30  <andythenorth> found it :D
20:16:06  <TrueBrain> anyone up for a quick review of https://github.com/OpenTTD/bananas-server/pull/5 ?
20:16:16  <TrueBrain> that would fix most of my infra issues :D
20:19:32  <frosch123> http://devs.openttd.org/~frosch/multirowheader.png <- i hated the overflowing nav bar, so i stackoverflowed that
20:20:30  <andythenorth> it's quite a challenge that nav
20:20:35  <andythenorth> stacked navs are hard to use
20:20:43  <andythenorth> but eh, we just need something
20:21:08  <andythenorth> I can see 4 things that we can move into a different row
20:21:29  <andythenorth> frosch123 also I've done all the pages I could find
20:21:41  <andythenorth> 'done' isn't done
20:21:52  <andythenorth> I don't know about form validation style, user messages etc
20:22:20  <frosch123> as long as all text is readable :)
20:22:57  <andythenorth> my automated checker just found a few things to clean up
20:22:59  <andythenorth> I'll do that
20:23:50  <nielsm> TrueBrain: I can make an approving review but it won't be reviewing the contents of the PR
20:24:27  <andythenorth> "openttd-64-ipv6.gif" o_O
20:24:37  <TrueBrain> nielsm: you don't know Python? :D
20:24:50  <nielsm> I don't have any experience with the modules involved
20:25:12  <TrueBrain> does anyone, really? :D
20:26:27  <Samu> I am bored
20:26:35  <Xaroth> lgtm
20:26:59  <DorpsGek_III> [OpenTTD/bananas-server] Xaroth approved pull request #5: Add: support Proxy Protocol v1 (if requested) https://git.io/JfLDV
20:27:12  <andythenorth> 15 contrast errors
20:27:22  <andythenorth> no html validation errors
20:27:35  <TrueBrain> sadly, that approval doesn't allow me to merge :P
20:27:39  <TrueBrain> but tnx Xaroth  :)
20:27:52  <Xaroth> <3
20:27:53  <frosch123> also, isn't xaroth your brother or something?
20:27:58  <Xaroth> hahahahahahahahahahahahahaha
20:28:22  <andythenorth> oh dear https://dev.openttdcoop.org/attachments/download/9644/1-Errors.png
20:28:24  <TrueBrain> even if he was, are you implying I can bribe him?
20:28:30  <TrueBrain> as that means you really don't know him :P
20:28:31  <Xaroth> damnit, now there's  lemonade all over my desk, thanks frosch123.
20:28:52  <andythenorth> really, who would do that? ^^^
20:29:39  <Xaroth> To be fair, I don't see TrueBrain as one who would bribe somebody just for a PR approval :P
20:29:40  <DorpsGek_III> [OpenTTD/bananas-server] frosch123 approved pull request #5: Add: support Proxy Protocol v1 (if requested) https://git.io/JfLDr
20:29:59  <TrueBrain> ty kind
20:30:00  <DorpsGek_III> [OpenTTD/bananas-server] TrueBrain merged pull request #5: Add: support Proxy Protocol v1 (if requested) https://git.io/JfL6K
20:30:33  <frosch123> andythenorth: yay, top-score on that analysis
20:30:41  <andythenorth> today's winner
20:30:52  <andythenorth> the lack of form labels really angers the validator
20:31:08  * andythenorth looks in current bananas
20:32:46  <Samu> I am Xarick
20:32:58  <andythenorth> ok current bananas provides some labels, inconsistently
20:33:01  <Xaroth> I am not.
20:33:16  *** Xetalim has quit IRC
20:34:11  <andythenorth> frosch123 what's the next step?  I'm not sure where I'll be next few days
20:34:17  <andythenorth> well I know I'll be in my house :P
20:34:26  <andythenorth> but OpenTTD time might be very variable
20:34:38  <andythenorth> PR the current state?
20:35:15  <frosch123> if you push to your gh, i can pull, fix the navigator selection, and pr it
20:35:45  <andythenorth> ok everything is pushed to style-classic-openttd
20:35:56  <andythenorth> which is recently based on your branch
20:36:20  <frosch123> i'll PR it tomorrow then
20:36:24  <andythenorth> we can use the PR for remaining issues, like the lack of form labels?
20:36:32  <andythenorth> I don't have a GH issue for this iirc
20:55:00  <TrueBrain> this AWS ... it is such a beast :)
20:55:12  <TrueBrain> my Lambda was not put in a VPC, so it was not able to validate the server was up ..
20:55:17  <TrueBrain> took me only ... 30 minutes to figure out :P
20:57:01  <TrueBrain> I think I created 200+ EC2 instances today :)
21:04:47  *** frosch123 has quit IRC
21:05:41  *** Samu has quit IRC
21:07:06  *** nielsm has quit IRC
21:07:59  <andythenorth> everyone is leaving!
21:08:05  * andythenorth should go to sleep
21:10:54  <TrueBrain> enjoy your night sleep andythenorth  :)
21:10:57  <TrueBrain> you deserve it :D
21:11:06  <andythenorth> I do!
21:11:09  <andythenorth> all the meetings!
21:11:12  <andythenorth> bye
21:11:15  *** andythenorth has left #openttd
21:12:22  <supermop_Home> i'm still in a meeting!
21:19:32  <TrueBrain> :(
21:19:35  <TrueBrain> is it remote?
21:21:49  *** cHawk has joined #openttd
21:23:26  <supermop_Home> video call via slack
21:24:04  <supermop_Home> so much of the day is slack or zoom meetings that i end up having to work late at night to actually accomplish anything
21:24:42  <supermop_Home> also my remote connection to my workstation at the office seems to work better late at night
21:29:12  <TrueBrain> I like when I can disable my webcam or put something silly on it, as that means I can walk around the house freely
21:29:16  <TrueBrain> which makes calls a lot better
21:39:28  <supermop_Home> unfortunately i am stuck in the one room as my wife is often in meetings in the other!
21:48:52  <TrueBrain> welcome to 2020; you get a new lifestyle, and you get a new lifestyle, and you get a new ...
21:49:33  *** jinks has quit IRC
21:51:17  *** Progman has quit IRC
21:52:59  *** cHawk- has joined #openttd
21:56:20  <TrueBrain> okay, over IPv4 the content service is working now
21:56:24  <TrueBrain> how am I going to test IPv6 .. hmm
21:59:05  *** cHawk has quit IRC
22:01:34  *** jinks has joined #openttd
22:04:07  *** cHawk- has quit IRC
22:04:34  <TrueBrain> yeah, IPv6 also works :D
22:04:35  <TrueBrain> w00p
22:05:38  <DorpsGek_III> [OpenTTD/nml] LordAro commented on pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfL9s
22:08:18  <TrueBrain> that took a lot more time than I expected, but .. the result is pretty clean and nice
22:08:42  <TrueBrain> tomorrow I hope to attach a DNS entry to it, as that alllows people to test it out :D
22:33:48  *** cHawk has joined #openttd
22:51:03  *** cHawk has quit IRC
22:55:36  *** mcbanhas has quit IRC
22:56:23  *** cHawk has joined #openttd
22:57:19  *** Wolf01 has quit IRC
23:04:57  *** sla_ro|master has quit IRC
23:07:47  *** cHawk has quit IRC
23:13:26  *** cHawk has joined #openttd
23:21:07  <DorpsGek_III> [OpenTTD/nml] glx22 updated pull request #102: Fix #100: follow PEP440 when generating version https://git.io/JfIF4
23:41:50  *** tokai|noir has joined #openttd
23:41:50  *** ChanServ sets mode: +v tokai|noir
23:48:40  *** tokai has quit IRC
23:56:03  *** gelignite has quit IRC

Powered by YARRSTE version: svn-trunk