Log for #openttd on 23rd January 2021:
Times are UTC Toggle Colours
00:41:13  *** Progman has quit IRC
00:42:29  *** HerzogDeXtEr has quit IRC
02:06:27  *** Flygon has joined #openttd
02:39:06  *** tokai|noir has joined #openttd
02:39:06  *** ChanServ sets mode: +v tokai|noir
02:45:58  *** tokai has quit IRC
03:20:09  *** WormnestAndroid has joined #openttd
03:56:22  *** didac has joined #openttd
03:57:00  *** debdog has joined #openttd
04:00:22  *** D-HUND has quit IRC
04:01:32  *** glx has quit IRC
04:02:39  *** supermop_Home_ has quit IRC
04:58:24  *** keith_ has joined #openttd
06:11:38  *** snail_UES_ has quit IRC
06:36:29  *** didac has quit IRC
08:15:45  *** sla_ro|master has joined #openttd
08:21:22  *** nielsm has joined #openttd
08:40:48  <DorpsGek> [OpenTTD/OpenTTD] perezdidac opened pull request #8603: Feature: Object class selection string filtering
08:40:54  *** qwebirc76033 has joined #openttd
09:00:37  <DorpsGek> [OpenTTD/OpenTTD] J0anJosep commented on pull request #8603: Feature: Object class selection string filtering
09:05:11  *** Progman has joined #openttd
09:43:22  *** DasPoseidon has joined #openttd
09:43:27  *** jottyfan has joined #openttd
09:46:10  *** HerzogDeXtEr has joined #openttd
10:04:58  *** Wolf01 has joined #openttd
10:38:54  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8550: Feature: "Remove all industries" button in scenario editor
10:42:04  <LordAro> TrueBrain: how does new-bananas handle this these days?
10:42:49  <TrueBrain> what made you dig that up? :D
10:43:03  <TrueBrain> given BaNaNaS now uses git repo, I do not think it is limited in any way to a max size
10:43:10  <TrueBrain> so the network protocol is in charge
10:43:21  <TrueBrain> so that problem is still happening, I am sure :)
10:43:26  <LordAro> the person who raised it originally mentioned it :p
10:43:27  <TrueBrain> test it out on staging!
10:44:48  *** jottyfan has quit IRC
10:45:02  *** nielsm has quit IRC
10:49:44  <TrueBrain> but yeah, I mentioned it a few weeks ago .. would love to make a new version of the network protocol, to fix all such quirks and things :)
10:49:56  <TrueBrain> it is just a lot of work, first to index all the things that are now so-so
10:51:10  <LordAro> is this the right place?
10:51:22  <TrueBrain> right place for what?
10:51:38  <LordAro> where the 'name length validation' occurs
10:51:48  <TrueBrain> no, that validates if everything fits in a packet
10:51:54  <TrueBrain> honestly, the issue is not with BaNaNaS
10:52:08  <TrueBrain> the network protocol sends strings
10:52:13  <TrueBrain> but the client receives them in fixed-size buffers
10:53:32  <TrueBrain>
10:53:52  <TrueBrain> as same thing happens with URL and with description, even with version
10:54:48  <TrueBrain> and sure we can limit it on the BaNaNaS side, but there are a few places that would need to be done: in the textbox on the frontend-web and somewhere in the validation. But I cannot remember we did any length-checks on strings
10:54:52  <TrueBrain> so it would need to be added
10:56:01  <LordAro> ah yes, got it now
10:56:12  <TrueBrain> seems the only way it "worked" in the old BaNaNaS, was because the MySQL was limited in chars :)
10:56:16  <TrueBrain> so that was more by chance :P
10:56:18  <LordAro> lol
10:56:21  <TrueBrain> or luck, what-ever
10:56:47  <TrueBrain> so yeah, up to you where you want to fix it :) But for BaNaNaS, add a new validation function to check (encoded) string length
10:57:06  <TrueBrain> <- should be hooked in somewhere there
10:57:20  <TrueBrain> I wouldn't do it in validate_packet_size, as that really checks for MTU
10:57:41  <LordAro> yeah
10:57:44  <TrueBrain> its a fail-safe, basically :P
10:58:22  <_dp_> is there a way to just make rev.cpp with cmake instead of full compile?
10:58:34  <_dp_> like, before i could do make $ttdir/src/rev.cpp
10:59:10  <TrueBrain> LordAro: I was wrong
10:59:11  <TrueBrain>
10:59:15  <TrueBrain> there is some length validation going on
10:59:21  <TrueBrain> but via the JSON schema
10:59:31  <TrueBrain> I forgot we validated the user input really really well :P
10:59:35  <TrueBrain> (shocker! :D)
10:59:53  <LordAro> :O
11:00:09  <TrueBrain> so a quick-fix would be to change that validate.Length into something that does it on encoded size, instead of decoded size
11:00:23  <TrueBrain> completely forgot I did add code for that, honestly :)
11:00:26  <TrueBrain> did not expect I would have ...
11:00:43  <LordAro> hehe
11:00:44  <TrueBrain> marshmallow is a very easy library, so adding that is not difficult
11:00:49  <TrueBrain> I just still think it is the wrong fix :P
11:01:01  <TrueBrain> (especially now with C++, those pre-sized buffers are not needed, I think :D)
11:01:08  <LordAro> indeed not
11:01:38  <LordAro> the real question, is that validate.Length(max=31) bytes or codepoints? :p
11:02:28  <LordAro>
11:02:31  <LordAro> i think codepoints :>
11:02:31  <TrueBrain> I already answered that for you: it currently does a decoded length check
11:02:46  <TrueBrain> no clue why you call it codepoints, it does the length of the unicode string :)
11:03:05  <TrueBrain> not of the utf-8 string
11:03:13  <TrueBrain> (which is, I guess, where OpenTTD limits it on)
11:03:39  <LordAro> sorry yes, i'm failing to remember proper terms for bytes vs <thing>
11:03:41  <TrueBrain> ValidateURL in that same file shows how you can overload it :)
11:03:52  <TrueBrain> in python, encode() vs decode() or unicode vs utf-8 :)
11:04:02  <TrueBrain> or bytes vs str
11:04:11  <TrueBrain> Python3 made this so much easier :P
11:04:11  <LordAro> yes
11:04:38  <TrueBrain> so yeah .. it should be somethinglike: len(value.encode())
11:04:38  <TrueBrain> :)
11:04:42  <TrueBrain> you figure it out, I am sure ;)
11:04:49  *** andythenorth has joined #openttd
11:04:53  <andythenorth> yo
11:04:55  <LordAro> :)
11:05:02  <LordAro> sup ands
11:05:23  <LordAro> _dp_: can't see anything obvious
11:06:48  <LordAro> grep the contents of build/ for rev.cpp :p
11:16:45  <andythenorth> where is my livery spec gist?
11:16:59  <andythenorth> oh here we go
11:17:00  <andythenorth>
11:18:53  <andythenorth> it's funny, all the authors who might use this are in discord, except me
11:19:03  <andythenorth> and the 1 or 3 people who might understand the spec are here
11:19:10  <andythenorth> newgrf lolz
11:22:57  *** Samu has joined #openttd
11:27:25  <DorpsGek> [OpenTTD/bananas-api] LordAro opened pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
11:27:28  <LordAro> TrueBrain: :)
11:29:04  *** frosch123 has joined #openttd
11:29:55  *** Progman has quit IRC
11:31:30  <TrueBrain> LordAro: version should be done the same, I think?
11:32:52  <DorpsGek> [OpenTTD/bananas-api] TrueBrain commented on pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
11:33:00  <TrueBrain> PRing untested stuff ....... I expected more of you LordAro  :P
11:33:26  <LordAro> shhh
11:33:27  <DorpsGek> [OpenTTD/bananas-api] LordAro updated pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
11:33:30  <TrueBrain> at the very least, please next time SAY you didn't test it :)
11:34:16  <TrueBrain> did you test it now? :D
11:35:57  <LordAro> nope :>
11:36:21  <TrueBrain> would you mind doing so? :)
11:37:15  <TrueBrain> my main issue with your current solution, is the way errors are presented back to the user .. not sure if that is ideal, but also not sure if it s worth fixing
11:37:16  <LordAro> you drive a hard bargin
11:37:26  <TrueBrain> basically, they get a byte-string in the error back
11:37:54  <TrueBrain> LordAro: well, more the issue, if we brick BaNaNaS, we have people with pitchforks telling us we did :P
11:39:06  <TrueBrain> maybe we should just copy/paste the whole call .. how ever ugly that sounds
11:39:09  <TrueBrain> not sure
11:39:16  <TrueBrain> depends on how ugly the error really is, I guess :D
11:39:21  <LordAro> mm, true
11:39:46  <TrueBrain> owh, no, you can overload _format_error again
11:39:50  <TrueBrain> and do a .decode() there
11:39:52  <TrueBrain> not sure that is pretty
11:39:57  <TrueBrain> but ... also solves the issue :D
11:40:07  <TrueBrain> and remember that call returns the value
11:40:13  <TrueBrain> so you need to decode that again
11:40:15  <TrueBrain> and return it
11:40:16  <LordAro> yeah
11:40:22  <TrueBrain> okay, you really need to test this :D
11:40:37  <TrueBrain> while reasoning in what it would do, I think this currently doesn't work :)
11:41:42  <TrueBrain> if you don't want to build your own frontend-web, you can use btw
11:41:44  <TrueBrain> prebuild for you!
11:43:24  <LordAro> i was playing around with curl directly
11:43:28  <LordAro> just got a 500 :)
11:51:40  <Samu> hi
11:52:59  <Samu> look closely
11:53:08  <Samu> buses positions are moving
11:53:26  <Samu> which one has better alignment?
11:58:11  <andythenorth> left side (first image)
11:58:14  <andythenorth> why?
11:58:39  <Samu> oh :(
11:59:52  <Samu> i'm trying to match the number of frames a vehicle needs to traverse when going on the left lane vs right lane
12:00:26  <Samu> but for that, I have to create or remove an extra frame
12:00:35  <Samu> depending on the situation
12:01:18  <_dp_> Samu, keep in mind there is also frame doubling hack for short turns to make articulated vehicles follow properly
12:04:14  <Samu> i have 4 choices
12:04:58  <Samu> "17/11" vs "18/11" vs "18/10" vs "17/10"
12:05:46  <Samu> that image shows all the 4 as having "18/10"
12:06:53  <Samu> 18 frames needed to traverse the outer curve, 10 frames needed to traverse the inner curve
12:08:01  <Samu> that image on the right side, that is
12:20:58  <Samu> gonna test all "18/11" next, after lunch
12:21:41  <Timberwolf> This is interesting... running a game in 1.11 beta, my GS window claims "1.5 API compatibility in effect" but crashes on a call to GSCompany.ChangeBankBalance() because it's missing the new tile parameter.
12:25:59  <TrueBrain> _dp_ made a booboo? :D
12:29:21  <_dp_> hm... I did add a compate file....
12:29:38  <TrueBrain> and by the looks it should work
12:30:21  <TrueBrain> Timberwolf: you didn't happen to only copy the openttd executable and lang folder to an existing one, did you? :D
12:30:24  <TrueBrain> (just to be sure :P)
12:30:37  <Timberwolf> Oh yes, that would be exactly the stupid think I did :)
12:30:48  <TrueBrain> please also copy the bin folder ;)
12:32:18  <Timberwolf> There we go, problem solved :)
12:32:26  <TrueBrain> PEBKAC! :D
12:32:53  *** keith_ has quit IRC
12:34:23  <LordAro> TrueBrain: i'm trying, but i have no idea how to use docker
12:34:35  <TrueBrain> then run it in a venv
12:34:37  <TrueBrain> but what is failing?
12:35:00  <LordAro> i have no idea
12:35:04  <LordAro> i think i'm running the container
12:35:13  <LordAro> but i'm really not sure
12:35:17  <LordAro> there's no ports
12:35:27  <TrueBrain> you executed the commands from the README?
12:37:59  <LordAro> no, don't be silly
12:38:41  <TrueBrain> if only there was this file that helps people get this setup and running .........
12:38:51  <TrueBrain> :P
12:38:54  <LordAro> if only
12:39:04  <LordAro> ok, the web-client appears to be running now
12:39:55  <LordAro> though it's trying to connect to staging api, not my api? how do i change the args that the container runs with
12:40:26  <TrueBrain> <- it does by default
12:40:43  <TrueBrain> so unless you are doing your own thing, and not copying pasting from the README, it should "just work" :)
12:41:22  <TrueBrain> in case you are not using Docker
12:41:24  <LordAro> exactly, i am doing my own thing
12:41:26  <LordAro> aren't i?
12:41:37  <TrueBrain> I cannot look in your machine, and you are not being descriptive :P
12:41:42  <TrueBrain> so ... yeah .... what can I say :)
12:42:40  <LordAro> well i'm changing bananas-api
12:43:02  <LordAro> so i want the bananas-frontend-web to connect to my local instance
12:43:07  <LordAro> not the staging instance
12:43:37  <TrueBrain> but you are not following the README?
12:44:00  <DorpsGek> [OpenTTD/bananas-api] LordAro updated pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
12:44:30  <TrueBrain> as copy/paste from the README should work; if not, I would love to know
12:44:33  <TrueBrain> as that should setup a development setup
12:44:42  <LordAro> it works fine
12:44:42  <TrueBrain> if you are not: look at what the commands do, and replicate that :)
12:44:47  <LordAro> but the frontend does not connect there
12:44:59  <TrueBrain> the frontend, if following the README, should connect to the localhost API
12:45:08  <LordAro> the docker image does not
12:45:08  <TrueBrain> that is why I ask .. but you are a bit confusing in your answer :P
12:45:13  <TrueBrain> it should :)
12:45:15  <LordAro> you are confusing in your answers!
12:45:30  <TrueBrain> <- the docker image, if started as mentioned in the README, uses the localhost API
12:45:40  <TrueBrain> if you start it in any other way, it will default to the built-in API, which is staging
12:45:43  <LordAro> "--api-url", ""
12:45:46  <LordAro> no it doesn't?
12:45:56  <TrueBrain> owh, that is new
12:45:56  <TrueBrain> haha
12:46:20  <TrueBrain> I just blame the maintainer at this point ;)
12:46:37  <TrueBrain> anyway, that is why I asked first: did you follow the README :) You said no, which makes it really difficult for me :P
12:46:46  <TrueBrain> if you did, and it fails, it means our README is wrong :D
12:47:05  <LordAro> i *wasn't* following it, no
12:47:17  <TrueBrain> but you understand you are not helping me helping you at that point :P
12:47:20  <LordAro> :p
12:47:27  <TrueBrain> seems only the Makefile connects to the localhost
12:48:11  <TrueBrain> docker run --rm -p openttd/bananas-frontend-web:local --api-url http://localhost:800 --frontend-url http://localhost:5000 run -p 80 -h
12:48:16  <TrueBrain> with the correct port
12:48:19  <TrueBrain> lol, 800 is wrong ofc
12:48:36  <TrueBrain> and you can use the image instead of the local one if you like
12:53:10  <LordAro> aha, now i'm getting somewhere
12:57:32  <LordAro> so apparently the message does not actually include the value
12:59:32  <DorpsGek> [OpenTTD/bananas-api] LordAro updated pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:00:01  <DorpsGek> [OpenTTD/bananas-api] LordAro commented on pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:00:44  <DorpsGek> [OpenTTD/bananas-api] TrueBrain commented on pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:01:07  <TrueBrain> I cannot believe it works without return, honestly :o
13:02:20  <TrueBrain> possibly it makes the name None now, after validation :D
13:03:00  <LordAro> ...i didn't actually get as far as testing it actually uploads, i'll admit
13:03:05  <LordAro> only tested the validation
13:03:32  <TrueBrain> I doubt I would have tested it myself :P
13:03:38  <TrueBrain> I was just comparing what the validate URL thingy did
13:05:44  <DorpsGek> [OpenTTD/bananas-api] LordAro updated pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:05:53  <LordAro> does it actually have any return value at all, perhaps?
13:06:10  <LordAro> any used*
13:06:39  <TrueBrain> if I remember correctly, you can also modify the value in validate functions
13:06:45  <TrueBrain> I might be completely wrong
13:07:07  <TrueBrain> okay, and sorry LordAro , I thought it was possible to have the frontend-web via Docker connect to a locally run API, but .. I remember now: Docker made that REALLY difficult
13:07:12  <TrueBrain> as there is not an IP that is "host"
13:07:19  <TrueBrain> it depends on the Docker configuration
13:09:13  <TrueBrain> regression is ratting you out :D Hihi, I love regression sets ..
13:09:24  <LordAro> that was a dumb mistake
13:09:26  <DorpsGek> [OpenTTD/bananas-api] LordAro updated pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:09:48  <TrueBrain> you should see my PR fixes sometimes :P
13:11:39  <DorpsGek> [OpenTTD/bananas-api] TrueBrain approved pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:11:39  <TrueBrain> tnx LordAro :)
13:11:42  <LordAro> :)
13:12:03  <TrueBrain> especially for fixing a bit more than needed, just because you ran into it; that is appreciated :)
13:12:22  <DorpsGek> [OpenTTD/bananas-api] TrueBrain merged pull request #80: Fix: Validate against the byte length of strings, rather than their unicode length
13:12:23  <LordAro> trying to replicate oauth with curl was great fun
13:12:37  <TrueBrain> your fun is different from my fun :P
13:15:18  <TrueBrain> now the question .. should we validate existing packages for violations? :)
13:15:43  <TrueBrain> hmm .. in fact ... I am pretty sure if we deploy this on production, that is exactly what happens
13:15:53  <TrueBrain> and it will fail to load the content that violates the schema :D
13:16:01  <TrueBrain> LordAro: for shit and giggles, checkout the BaNaNaS repo in your API folder
13:16:03  <TrueBrain> and start the API
13:17:22  <TrueBrain> that validates all the metadata, and shows any content we might have that violates the new constraint
13:17:38  <LordAro> just the one
13:17:40  <LordAro> BaNaNaS/newgrf/5a425801
13:17:51  <TrueBrain> we might want to fix that before deploying to production
13:17:59  <LordAro> (assuming it doesn't stop after failing the first one)
13:18:04  <TrueBrain> it doesnt :)
13:18:33  <TrueBrain> or maybe better: really shouldn't :P
13:18:51  <TrueBrain> but that is not bad, just 1
13:20:52  <LordAro> hmm, i can't shorten it enough just by removing the obviously "duplicated" (websites, attribution)
13:22:35  <TrueBrain> remove the "my website" part?
13:22:39  <TrueBrain> in both languages
13:22:42  <LordAro> yeah
13:22:57  <LordAro> still 653 bytes :>
13:23:10  <TrueBrain> wow ....
13:23:25  <TrueBrain> did not expect that the one violating it, would be violating it that badly :P
13:23:26  <LordAro> most CJK chars are 3 bytes per char, iirc
13:23:54  <TrueBrain> wipe description, contact author? Dunno ..
13:24:32  <LordAro> oh wait, no
13:24:40  <DorpsGek> [OpenTTD/OpenTTD] Kuhnovic updated pull request #8550: Feature: "Remove all industries" button in scenario editor
13:24:42  <TrueBrain> you have to fix it twice :P
13:24:47  <LordAro> it's enough for the global, i neglected to fix it for the version
13:25:14  <TrueBrain> people are not getting the "global" thing, I have noticed :)
13:25:35  <LordAro> the error messages are decidedly unclear :p
13:26:06  <LordAro> oh, still 543/569 chars
13:26:22  <TrueBrain> yeah, but this error should never appear :P
13:26:44  <LordAro> this error, but the error on the website isn't the nicest formatted thing in the world :p
13:27:44  <LordAro> so, blank the description? i've no idea how we can contact the author though...
13:28:21  <TrueBrain> owh, for the frontend-web we have an issue saying: this should be done better
13:28:32  <TrueBrain> people said they were going to work on it, also no CSS etc
13:28:40  <TrueBrain> but ... as you might have noticed over the years ... that is always just talk
13:28:45  <TrueBrain> :P
13:28:57  <TrueBrain> the Author, we can contact them via GitHub I think? Not sure honestly :)
13:29:12  <LordAro> no private messaging on GH
13:29:19  <TrueBrain> owh, yeah, DOH
13:29:21  <TrueBrain> forgot :D
13:30:37  <TrueBrain> wipe it, make an issue with @username?
13:31:21  <frosch123> maybe they have a forum pm
13:35:33  <frosch123> but since they are chinese town names, you can probably also delete the english description
13:35:59  <frosch123> if you can't read the description, you probably also can't read the townnames
13:36:06  <LordAro> heh, true
13:36:07  <TrueBrain> works for me
13:36:11  *** DasPoseidon has quit IRC
13:36:14  <LordAro> :+1:
13:37:20  <LordAro> PR, or just commit straight to master?
13:37:43  <TrueBrain> PR please
13:38:03  <TrueBrain> just so we can leave text WHY it happened :)
13:38:30  <LordAro> yeah
13:38:46  <TrueBrain> and I think I added a linter to PRs :P
13:41:27  <LordAro> maybe we should have dorpsgek notifying on PRs
13:41:59  <TrueBrain> add a .dorpsgek.yml, yeah
13:42:48  <Samu> wow, 18/11 looks good!
13:42:56  *** glx has joined #openttd
13:42:56  *** ChanServ sets mode: +v glx
13:44:59  <LordAro> TrueBrain: done :)
13:47:55  <DorpsGek> [OpenTTD/BaNaNaS] TrueBrain merged pull request #77: Add: DorpsGek config file
13:48:21  <LordAro> i see you, force merging PRs
13:48:22  <LordAro> :p
13:48:29  <TrueBrain> yeah ... rebase is for the weak
13:48:49  <TrueBrain> oops, I forgot you have to wait ~3 minutes before merging another PR in that repo
13:48:50  <TrueBrain> owh well
13:49:17  <TrueBrain> nothing a rerun can't fix
13:49:26  <Samu> andythenorth, - check it out
13:49:47  <TrueBrain> LordAro: well, now to tag bananas-api :)
13:50:06  <LordAro> TrueBrain: there's a few repos like that, i suppose there's no way of specifying "don't cancel this job if a new one comes in" ?
13:50:43  <TrueBrain> for this repo the problem is that I restricted the Lambda to only have 1 concurrent job
13:50:48  <TrueBrain> so if 2 actions request a reload at the same time
13:50:50  <TrueBrain> one bounces
13:50:59  <TrueBrain> I should add a queue in front of it
13:51:00  <Samu> 18/11 does look better
13:51:02  <TrueBrain> so they are both accepted
13:51:10  <TrueBrain> but ... lot of work for very rare cases it going wrong
13:51:13  <LordAro> mm
13:52:05  <DorpsGek> [OpenTTD/BaNaNaS] zbx1425 commented on pull request #76: Fix: Remove English part of description for Chinese townnames GRF
13:52:14  *** DasPoseidon has joined #openttd
13:52:43  <TrueBrain> LordAro: up to you to tag btw :)
13:53:55  <DorpsGek> [OpenTTD/bananas-api] LordAro created new tag: 1.0.8
13:55:45  <TrueBrain> w00p
13:58:37  <Samu> time to test 17/10
13:59:17  <andythenorth> Samu 'why'
13:59:24  <andythenorth> will I like the answer?
14:00:06  <andythenorth> your goal is to equalise the time taken to traverse different radius curves?
14:00:12  <andythenorth> or I misunderstand?
14:00:26  <Samu> yes, that's it exactly
14:02:22  <andythenorth> why?
14:02:40  <andythenorth> that's kinematically unsound IMHO
14:03:24  <andythenorth> does it improve the (odd) appearance of RVs as they turn in curves?
14:03:45  <Samu> going diagonally from north to south has a smaller distance than going diagonally from south to north
14:04:33  <andythenorth> because there's a bug?
14:04:53  <andythenorth> or you just think the behaviour is designed wrong?
14:05:16  <Samu> I believe it's more about the design being wrong, than a bug
14:05:34  <andythenorth> road vehicle animation through curves does suck
14:05:40  <andythenorth> but I never figured out why
14:05:51  <andythenorth> I assumed it was just bad newgrf offsets, but you are using base set
14:06:51  <Samu> going diagonally from east to west is equal to west to east
14:07:36  <Samu> not bugged
14:07:54  <Samu> it's only when going south to north vs north to south that there's a discrepancy
14:09:34  *** supermop_Home has joined #openttd
14:09:41  <supermop_Home> good morning
14:09:52  <_dp_> rv movement should be just separated from the visuals imo
14:11:02  <Samu> using the example in that screenshot, going from the left depot to the right depot, the bus traverses an inner curve first, with 11 frames, then the second tile with an outer curve, with 17 frames
14:11:10  <Samu> it totals 28 frames
14:11:56  <Samu> going the opposite direction, from right depot to left depot, it is 10 frames + 18 = 28
14:12:55  <Samu> doing it for north depot to south depot, it is: 10 + 17 = 27
14:13:21  <Samu> and from south to north, it is: 11 + 18 = 29
14:13:40  <Samu> 27 vs 29, this is the discrepancy i'm talking about
14:13:56  <Samu> it's faster going from north to south, than it is going from south to north
14:14:40  <andythenorth> and if you change the drive side?
14:15:19  <Samu> it's also bugged, but i haven't tested much, i think it becomes 29/27 instead, it's faster going south to north
14:17:20  *** sla_ro|master has quit IRC
14:22:46  <Samu> also haven't tested trams
14:23:07  <Samu> i'm working with these cases first before I make a decision
14:32:47  *** iSoSyS has joined #openttd
14:33:16  *** jottyfan has joined #openttd
14:33:49  *** iSoSyS has quit IRC
14:41:46  <Samu> 17/10 tested, added a picture
14:42:21  <Samu> gonna test last one, 17/11, brb
14:45:27  *** andythenorth_ has joined #openttd
14:51:03  *** andythenorth has quit IRC
14:57:32  <Samu> 17/11 tested, new picture
14:59:49  <Samu> so, it looks like those with 10 inner frames look bad
15:00:04  <Samu> looking at the 4 pictures
15:00:23  <Samu> have to decide between 17/11 or 18/11
15:04:38  <_dp_> oh, did we switch to c++17 already?
15:05:51  *** roadt_ has quit IRC
15:09:29  <LordAro> up
15:09:31  <LordAro> yup*
15:11:50  <Samu> i think 18/11 looks best, personally, need a 2nd opinion
15:13:23  <Samu> i also believe 18/11 will be easier to do savegame conversion, because we're only adding frames, not removing any
15:13:48  *** roadt_ has joined #openttd
15:21:56  *** roadt_ has quit IRC
15:23:55  <Samu> I can't decide
15:31:12  *** roadt_ has joined #openttd
15:32:23  *** nielsm has joined #openttd
15:48:52  <Samu> is there a way to add a newgrf to a savegame, need to test trams :p
15:48:57  <Samu> iguess not
15:55:37  <FLHerne> Samu: There is, but your save might blow up
15:55:54  <FLHerne> (generally it doesn't)
15:56:57  *** Flygon has quit IRC
16:00:26  *** Wormnest has joined #openttd
16:10:30  *** jottyfan has joined #openttd
16:47:16  *** sla_ro|master has joined #openttd
16:49:05  *** snail_UES_ has joined #openttd
17:06:31  *** gelignite has joined #openttd
17:08:04  <andythenorth_> hmm
17:08:20  <andythenorth_> snow on industries above the snowline....but
17:08:24  <andythenorth_> what if the industry is hot?
17:08:25  <andythenorth_> like a furnace?
17:09:17  <Eddi|zuHause> it's well-isolated, for energy efficience
17:10:14  <supermop_Home> andythenorth_ draw melted patches if cargo delivered
17:11:01  <Eddi|zuHause> it might also just be snowing faster than it can melt
17:13:46  <andythenorth_>,_Russia.jpg
17:15:32  <Eddi|zuHause> well, yeah, even a "hot" industry isn't just consisting of the "hot" parts
17:31:47  <Samu> road vehicles can overtake on curves? I forgot if they can
17:35:18  <Samu> next on testing is...
17:35:46  <Samu> tram, and articulated
17:55:21  <Samu> i need non articulated trams first
17:55:30  <Samu> is there a newgrf with them?
18:00:27  <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8603: Feature: Object class selection string filtering
18:01:11  <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8603: Feature: Object class selection string filtering
18:01:16  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8603: Feature: Object class selection string filtering
18:02:00  <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8603: Feature: Object class selection string filtering
18:30:07  <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8603: Feature: Object class selection string filtering
18:37:02  <Samu> just tested tram, for curves they use road data
18:37:29  <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8603: Feature: Object class selection string filtering
18:38:10  <Samu> and tested articulated, basically it's just another vehicle, no disconnections occurred
18:38:40  <Samu> next task is... the savegame conversion, if it's actually needed
18:39:41  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master
18:39:42  <DorpsGek>   - Update: Translations from eints (by translators)
18:44:28  <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8603: Feature: Object class selection string filtering
18:44:49  <Samu> i need a bigger sample
18:45:34  <Samu> do you guys have a handy savegame with tons of road vehicles, articulated or not, and tram vehicles, articulated or not?
18:46:09  <Samu> I kind of expected issues during load
18:46:18  <Samu> and I'm not getting any :(
18:49:29  <DorpsGek> [OpenTTD/OpenTTD] J0anJosep commented on pull request #8603: Feature: Object class selection string filtering
18:50:22  <Samu> wow... jesus,  i joined a random polish server, was kicked because of my name
18:50:29  <Samu> lack of name
18:50:51  <Samu> didn't even have time to savegame
18:51:35  <Samu> anyways, i need a savegame with tons of road/tram vehicles, articulated/non-articulated, plz
18:55:11  <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8603: Feature: Object class selection string filtering
18:56:43  <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8603: Feature: Object class selection string filtering
19:02:19  <_dp_> Samu,
19:02:35  <_dp_> also
19:03:01  <_dp_> dunno about articulated though
19:05:23  <supermop_Home> man people suck
19:05:32  <TrueBrain> pull your trousers up
19:07:50  <supermop_Home> idk why i ever read the openttd problems page
19:10:32  <TrueBrain> what page? :)
19:11:57  <andythenorth_> I read it a few times a year
19:12:04  <andythenorth_> it's approximately the same as reddit
19:12:18  <andythenorth_> except that 50% of reddit is "JGRPP Mac build when?"
19:13:50  <supermop_Home> one from November complaining about wiki
19:14:11  <TrueBrain> the one I called him out on his bullshit?
19:14:13  <TrueBrain> that was fun :D
19:14:15  <supermop_Home> yah
19:15:57  <supermop_Home> the forum is much less " here is a bug, this is broken" or "please help figure out X" and instead is mostly "do Y for me for free" and "you did z in a way different than how i want it, I hate you"
19:16:59  <andythenorth_> scrollbar demanded
19:17:33  <TrueBrain> they call that toxicity
19:17:38  <TrueBrain> and sadly, the forums is a bit full of it ..
19:17:42  <TrueBrain> reason I stopped reading it in 2006
19:17:57  <TrueBrain> sometimes someone from the forums joins Discord
19:18:03  <TrueBrain> he learns REALLY quick that is not appreciated there
19:18:05  <TrueBrain> it is fun
19:18:08  <supermop_Home> andythenorth_ i demand removing 64 railtypes in game, based on that image
19:18:17  <andythenorth_> I have no opinion
19:18:32  <supermop_Home> scroll bar would be fine
19:18:33  <andythenorth_> I do wonder if removing NRT might be good that
19:18:38  <andythenorth_> thought *
19:18:41  <andythenorth_> though *
19:18:43  * andythenorth_ needs to sleep
19:18:57  <supermop_Home> just the visual messiness of that list makes me dispair
19:19:03  <andythenorth_> so much ongoing debate about NRT
19:19:11  * andythenorth_ much regrets
19:22:05  *** DasPoseidon has quit IRC
19:22:08  *** DasPoseidon has joined #openttd
19:30:32  <TrueBrain> funny ... on reddit the beta release post received upvotes, and even a comment of praise
19:30:38  <TrueBrain> looks at forums
19:30:40  <TrueBrain> shrugs
19:30:44  <TrueBrain> always away
19:30:56  <TrueBrain> always? walks
19:30:57  <TrueBrain> typing, hard
19:33:44  <TrueBrain> orudge: that are a lot of voicemail messages we have now :D lol :)
19:33:57  <TrueBrain> at least, I assume those are linked to the VoIP number :)
19:34:07  <frosch123> are they legit, or just spam calls?
19:34:14  <TrueBrain> haven't listened to them :)
19:34:28  <frosch123> neither me :p
19:35:44  <frosch123> does github have polls? other than thumb up/down? we could put the titlegame competition on gh :p
19:36:18  <TrueBrain> a Discussion is closest you get, I think .. with a reply per entry
19:36:39  <TrueBrain> if you make it into a QA, you can upvote them
19:37:09  <frosch123> i don't remember the last one, but the two before that had drama on forums, so maybe a different platform is better
19:37:28  <TrueBrain> <- you can see people can upvote answers
19:37:29  <frosch123> (can you imagine drama in a titlegame competition :p )
19:38:09  <TrueBrain> so we can make a category Titlegame for that :)
19:38:18  <TrueBrain> and yes, sadly, I can imagine ..... which is really sad, really :P
19:45:19  *** DasPoseidon has quit IRC
19:45:23  *** DasPoseidon has joined #openttd
19:45:41  *** tokai has joined #openttd
19:45:42  *** ChanServ sets mode: +v tokai
19:50:00  *** DasPoseidon has quit IRC
19:50:05  *** DasPoseidon has joined #openttd
19:52:37  *** tokai|noir has quit IRC
19:53:01  <DorpsGek> [OpenTTD/OpenTTD] giordy commented on issue #8038: Fullscreen mode in MacOSX is almost unusable
19:54:13  <DorpsGek> [OpenTTD/OpenTTD] LordAro reopened issue #8038: Fullscreen mode in MacOSX is almost unusable
19:56:59  *** DasPoseidon has quit IRC
19:57:06  *** DasPoseidon has joined #openttd
19:59:43  *** DasPoseidon has quit IRC
19:59:49  *** DasPoseidon has joined #openttd
20:02:48  *** DasPoseidon has quit IRC
20:02:55  *** DasPoseidon has joined #openttd
20:03:18  <TrueBrain> stupid OSX :P
20:03:18  *** DasPoseidon has quit IRC
20:03:24  *** DasPoseidon has joined #openttd
20:04:22  *** DasPoseidon has joined #openttd
20:05:02  *** jottyfan has joined #openttd
20:05:18  *** DasPoseidon has quit IRC
20:13:40  <Samu> got a crash with wentbourne, now that's something
20:59:51  <andythenorth_> yeah that macOS stuff is the same I recorded
21:00:34  <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on issue #8038: Fullscreen mode in MacOSX is almost unusable
21:07:07  <Samu> have to update x->pos and y->pos of all vehicles during afterload, hmm this is getting complicated
21:19:06  *** sla_ro|master has quit IRC
21:21:11  *** Progman has joined #openttd
21:22:13  <Samu>
21:22:33  <Samu> roadveh_movement
21:22:53  <Samu> 4 for drive on left, 4 for drive on right
21:24:24  <Samu> 2, 10, 12, 13 are for drive on left, 18, 19, 21, 26 are for drive on right
21:27:51  <Samu> i need something resembling a conversion table
21:27:56  <Samu> hmm how am i gonna do this
21:28:00  <Samu> think
21:32:39  <Samu> also gonna have to convert v->direction in a few occasions
21:51:06  *** Samu has quit IRC
22:01:23  <DorpsGek> [OpenTTD/team] lanurmi opened issue #132: [fi_FI] Translator access request
22:07:59  <DorpsGek> [OpenTTD/team] glx22 commented on issue #132: [fi_FI] Translator access request
22:12:08  *** frosch123 has quit IRC
22:17:26  <orudge> TrueBrain: think it's the online business directory. Will phone back on Monday...
22:43:26  *** jottyfan has quit IRC
23:29:18  *** andythenorth_ has quit IRC
23:30:47  *** andythenorth has joined #openttd
23:33:00  *** Wolf01 has quit IRC
23:35:27  <DorpsGek> [OpenTTD/OpenTTD] mattkimber opened pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites
23:46:07  *** andythenorth has quit IRC

Powered by YARRSTE version: svn-trunk