IRC logs for #openttd on OFTC at 2019-01-09
⏴ go to previous day
00:00:03 <nielsm> general issue with UI input event handling
00:01:28 <Samu> * Check wether growing on a half-tile coast tile ends up blocking a water connection
00:01:38 <Samu> * @return true if building here blocks a water connection /
00:05:31 <Samu> now with imbuted CornerToDirection
00:06:12 <Samu> does it need more comments explaining stuff?
00:12:12 <Samu> looks like i can move opposite_track a bit further down
00:18:16 <Samu> gonna see what happens with 90 degrees forbidden, brb
00:24:59 <Samu> town constructors become too smart if they know a ship can't make 90 degrees they build on water
00:34:36 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
00:34:50 <Samu> how do I make it compile?
00:35:11 <Samu> ever since the azure switch thing, it doesn't start compiling my stuff
00:37:38 <LordAro> assuming you've rebased
00:38:01 <LordAro> just takes a moment to start
00:39:06 <Samu> command line scares me too much
00:39:10 <LordAro> azure-pipelines-ci.yml is in the repo
00:39:14 <LordAro> you only need to do it once
00:39:26 <LordAro> (and rebasing and pushing again will only serve to slow it down)
00:39:52 <Samu> Some checks haven’t completed yet
00:40:09 <Samu> waiting for status to be reported :(
00:42:01 <Samu> where is this azure guy?
00:42:52 <Samu> nevermind, i think i found it
01:05:23 <Samu> git rebase -i origin/master or upstream/master or something else
01:16:21 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
01:16:33 <Samu> no idea what i'm doing atm
01:25:30 <Samu> failed with weird errors
01:25:54 <Samu> doesn't seem caused by me
01:26:37 <glx> CI sometimes fails to work
01:28:11 <glx> 2019-01-09T00:18:08.5922077Z ##[error]*** b/src/town_cmd.cpp:910: Trailing whitespace: ' * '
01:28:40 <glx> but commit checker could be easier to understand
01:29:09 <glx> clic on the error/warning count
01:29:48 <glx> then commit-checker, then build and test
01:30:45 <glx> but it's not your fault if win64 failed
01:33:15 <Samu> is it git rebase -i origin/master
01:34:49 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
01:35:08 <Samu> trailing whitespace is fixed
01:36:46 <glx> should be upstream, you can check with "git remote -v"
01:37:17 <glx> the one without your username is the one you want
01:38:17 <glx> usually you just need to do "git fetch upstream" then "git rebase upstream/master"
01:38:48 <glx> and if there's no conflict you're fine, else you need to solve them
01:39:15 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
01:41:12 <Samu> it's not gonna build again?
01:42:17 <Eddi|zuHause> i set up my git to do "git pull" from original repo and "git push" to my repo
01:54:56 <Samu> oh snap, the first tile also need to be checked if it's valid t.t
01:55:37 <Eddi|zuHause> Samu: the error message is currently a bit hidden/buried.
01:55:41 <Eddi|zuHause> Step 2: find your build (the most recent one is at the top), click on the failed "commit-checker" task, and there on the log with all the errors
01:55:42 <Eddi|zuHause> Step 3: scroll to the very bottom of the log to get the actual description of what is wrong
01:57:35 <Samu> i fixed that already, glx told me where it was
01:58:26 <Samu> i have one different problem now, a simple !IsValidTile(tile) is missing
01:58:49 <Eddi|zuHause> did you roll back the rebase again?
01:59:20 <Samu> yes, but it didn't build (yet)
01:59:55 <Samu> dont know how that azure guy works
02:00:29 <Samu> shouldn't build now that i have another fix coming up, :o
02:01:20 <Eddi|zuHause> it won't build if you are based on a master commit that is too old
02:01:52 <Eddi|zuHause> you must rebase to current master, and then never go back to an older master again
02:02:53 <Samu> "git fetch upstream" then "git rebase upstream/master" ? glx told me to do like this
02:09:34 <peter1138> git pull upstream master
02:10:05 *** Thedarkb-X40 has joined #openttd
02:13:00 <Samu> git rebase -i upstream/master
02:14:46 <Samu> $ git rebase -i upstream/master
02:14:46 <Samu> Successfully rebased and updated refs/heads/prevent-town-growth-from-blocking-ships.
02:15:05 <Samu> had 2 commits, used fixup on the later
02:15:25 <Samu> now what? git pull upstream master?
02:16:54 <Samu> i have a 1 behind, 1 ahead
02:18:45 <Samu> Samu Committed Oct 2, 2018
02:21:23 <glx> no need for -i in this case, and --force will indeed work
02:23:35 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
02:24:11 <peter1138> I dunno, -i is useful to know if the rebase is going to do what you think it will be doing.
02:25:05 <glx> ah yes, and it's easy to stop it with -i
02:26:00 <Samu> just added the IsValidTile fix
02:27:28 <Samu> forgot that bridge heads could be tried to be built outside the map
02:29:56 <Samu> generating a map with 13312 towns
02:30:05 <Samu> ought to be enough for testing
02:30:27 <peter1138> Depending on how you're testing, 1 might be enough.
02:30:55 <Samu> erm dont remember if i got 90 deg on
02:31:01 <Samu> takes 10 mins to generate map
02:31:46 <Samu> it's debugging much faster
02:52:29 <Eddi|zuHause> if the town name generator runs out of names, it will take forever to try to generate the rest of the towns
02:53:02 <Eddi|zuHause> and even good town name generators will usually give up after about 3000 towns
02:53:19 <Eddi|zuHause> some don't even have 100
03:07:52 <Samu> somethnig really strange is happening bridges are still without problems
03:09:07 <Samu> during town generation, some bridges are initially built, then still during town generation, bridges are demolished later on
03:09:35 <Samu> as if a nearby town cleared them for some reason
03:13:38 <Samu> top image, there are bridges, bottom image, bridges are gone, the roads towards NW seems disconnected
03:24:35 <Samu> could towns be removed during world generation?
03:25:06 <Samu> like when it's trying to place a new town near another, but the other is so big?
03:45:34 <Samu> so it's not a bug caused by me, it's something of it's own
05:29:03 *** ToBeFree has joined #openttd
05:50:20 *** sla_ro|master has joined #openttd
06:14:43 *** HerzogDeXtEr has joined #openttd
08:13:29 *** andythenorth has joined #openttd
09:04:59 *** andythenorth has joined #openttd
10:11:44 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #4449: [Windows] Fullscreen as borderless window at desktop resolution https://git.io/fhZ3n
10:11:45 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #4449: [Windows] Fullscreen as borderless window at desktop resolution https://git.io/fhZ3c
10:14:28 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #5859: Calling VehicleEnterDepot for a train frees the reservation of the depot https://git.io/fhZ30
10:17:30 *** andythenorth has joined #openttd
11:02:41 *** Progman has joined #openttd
11:50:52 <DorpsGek_II> [OpenTTD/OpenTTD] lanurmi commented on issue #4449: [Windows] Fullscreen as borderless window at desktop resolution https://git.io/fhZGR
11:51:57 <LordAro> i'm detecting some sarcasm
12:06:37 <andythenorth> " as this PR is never going to get merged anyway. "
12:07:10 <andythenorth> I could argue, or I could accept I'll probably be dead 'soon'
12:08:49 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #4449: [Windows] Fullscreen as borderless window at desktop resolution https://git.io/fhZGK
12:12:51 <LordAro> see, i definitely disagree that the issue should've been closed
12:13:16 <LordAro> it's a valid issue, that's even had significant discussion on it since the gh move
12:13:28 <planetmaker> agreed... closing might be a bit over-zealous.
12:14:29 <planetmaker> it is an issue, but not yet resolved. And as it looks won't be by the person who contributed the patches since GH move
12:14:51 <LordAro> lock it to contributors by all means
12:14:53 <planetmaker> but hey, it's andy's activity on the issue which brought it to live again in the first place
12:17:47 <andythenorth> I closed it because it's not going anywhere, and the original contributor is whining
12:17:51 <andythenorth> even before today
12:17:59 <andythenorth> bearing in mind that issues can always be retrieved
12:18:34 <peter1138> Just hide them a little bit further and pretend they're not there.
12:20:05 <andythenorth> we close about 26 PRs a month
12:20:09 <andythenorth> and add about 31
12:20:19 <andythenorth> so we're not short of things we could already ship
12:20:54 <LordAro> i don't think things should be closed just because they're "not going anywhere"
12:21:02 <DorpsGek> andythenorth: 0.838709677419
12:21:15 <andythenorth> there's about an 83% chance of the PR going through currently
12:23:45 <andythenorth> LordAro: I'm not upset if they get re-opened
12:24:05 <andythenorth> I just see a pattern, might be false correlation
12:24:22 <andythenorth> in the last 2 years, the more issues I close, the more nice contributions we get :)
12:24:40 <andythenorth> you can send me the wiki link for 'cargo cult' if you want :P
13:10:15 <Eddi|zuHause> "the more drama i cause, the more activity happens" is maybe not the best correlation out there...
13:11:26 <Eddi|zuHause> it might have well-intended short term effects, but long-term is usually a lot more negative
13:23:03 *** chomwitt has joined #openttd
14:12:37 *** chomwitt has joined #openttd
14:18:51 *** sla_ro|master has joined #openttd
14:24:47 *** Flygon_ has joined #openttd
15:32:48 *** wodencafe has joined #openttd
15:36:25 *** Xaroth19 has joined #openttd
15:45:56 <andythenorth> after 4 days work
16:10:44 <peter1138> Has andythenorth closed #7031 yet?
16:11:15 <andythenorth> can't close any more
16:11:41 <andythenorth> wiki is all very wrong
16:11:45 <andythenorth> oh wait, wikis are always wrong
16:18:05 <nielsm> good it looks like the dumb "security vuln" non-vulnerability thread has died off
16:19:48 <andythenorth> progressive backoff of brute force attempts is fine
16:19:54 <andythenorth> that thread wasn't very helpful though
16:35:54 <Eddi|zuHause> did we ever even adress the fact that the passwords are stored in plain text in memory, and a server could just dump and collect them?
16:36:18 <andythenorth> I'll find the issue
16:36:59 <nielsm> and they are transmitted plain text over the network
16:42:20 <nielsm> if you want to store salted and hashed passwords on the server, you also need an authentication mechanism that protects against replay attacks
16:43:32 <peter1138> Fortunately it's already closed
16:43:47 <nielsm> IMO it would be more sensible to make it clear that OTTD company passwords are not secure, you should assume the server admin is able to read the password
16:43:47 <Eddi|zuHause> we need to ask people about their phone numbers, email addresses and credit card numbers and store those!! that'll never cause any problems in the future!
16:44:36 <Eddi|zuHause> at the bottom of that issue it's said the passwords aren't sent clear text
16:45:45 <Eddi|zuHause> dunno, i don't think i ever even looked at that part of the code
16:48:00 <Samu> gonna change variable names to make it more clear and easier to read code
16:48:13 <Samu> oh, and I have one question about where to make these checks
16:49:16 <Samu> I can check ealier, in conjunction with HasTileWaterGround or later, right before doing the build command
16:49:29 <nielsm> okay looks like the password is salted md5
16:49:29 <Samu> anyone familiar with town growth code?
16:50:07 <nielsm> the salt is calculated from server id and game seed
16:50:39 <Eddi|zuHause> not that md5 is a "safe" cryptographic algorithm or anything...
16:50:43 <andythenorth> automatic let's encrypt certs for all!
16:50:55 <andythenorth> start openttd server, generate a cert!
16:50:56 <nielsm> you may as well store it plaintext by now
16:51:07 <nielsm> instead of pretending it's secure
16:51:13 <andythenorth> nielsm: well it keeps out 7 year olds
16:51:16 <andythenorth> probably not more
16:51:30 <andythenorth> is all md5 cracked just on reverse lookups
16:51:43 <andythenorth> we used to use it to 'secure' high score tables in online games
16:51:48 <andythenorth> even then it was known vulnerable
16:52:25 <andythenorth> yeah reverse lookups are a thing
16:54:10 <peter1138> It's a compromise. For a secure server-side hash, the server doing authentication needs to be sent the cleartext password, which therefore requires transport level encryption (such as TLS)
16:54:33 <Samu> or if i can do the check later on
16:54:39 <Samu> right before the DoCommands
16:55:08 <peter1138> What we do 'protects' the password in transit as it is not sent as cleartext.
16:56:03 <nielsm> but it doesn't protect against replay attacks, just makes it harder for the server admin to guess what the player typed
16:56:28 <Gabda> I found that town indexes are 2 bytes, so storing it for every tile in a cache can be painful
16:57:01 <Samu> Gabda are u familiar with town growth algorithm?
16:57:36 <Samu> the decisions that makes it follow a road, build a road or grow a house
16:57:47 <Gabda> my 11 days of experience does not cover that part yet :D
16:58:45 <Samu> ok, t.t I wanted to know exactly where I should check the GrowingOnWaterTile in the logic
16:59:05 <Samu> so far, I've been making it right before DoCommands or MakeHouse
16:59:22 <Samu> but I noticed there is a HasTileWaterGround check way before
16:59:47 <Samu> this check fails on coast tiles, so I wondered if I do the check next to those
17:00:00 <Samu> would be the correct way
17:01:04 <Eddi|zuHause> that's why a voronoi partition would help, the graph is O(#towns) size instead of O(#tiles)
17:01:15 <Samu> lines 1266, 1315 and 1340
17:01:28 <andythenorth> nielsm: replay attacks are probably relatively hard
17:01:44 <andythenorth> or at least, there are more interesting targets
17:01:45 <Eddi|zuHause> however, it only works with euclidean distance, not manhattan distance
17:01:58 <Samu> I would change them to if (HasTileWaterGround(whatever_tile) || GrowingonWateredTle(whatever_tile) return false;
17:02:10 <Eddi|zuHause> and doing that with integer maths might be a bit tricky
17:02:27 <Eddi|zuHause> (can't use floats, so no "sqrt" function
17:02:31 <nielsm> andythenorth yes, if your intention was to grief an ottd game you probably wouldn't invest in intercepting network traffic to/from the server
17:02:50 <nielsm> (if you could do that you could do much more interesting things anyway)
17:03:33 *** Wormnest has joined #openttd
17:04:15 <Gabda> and with manhattan distances, one domain can be really long as well, with a few tiles wide channel
17:05:23 <Eddi|zuHause> well, flood-filling the tile cache you could do with euclidean-squared distance
17:05:38 <Eddi|zuHause> that works fine for "which is bigger" comparisons
17:05:51 <Eddi|zuHause> but not the complex algebra needed for voronoi
17:06:37 <Eddi|zuHause> "complex" not in the "complex number" mathematical sense
17:07:33 <Gabda> but as the goal would be to have information on the tiles without calculation, what options do we have aside from storing it for each tile?
17:07:52 <Gabda> to have the answer in constant time instead of O(#towns)
17:12:23 <Eddi|zuHause> it's not constant, but i don't think it's O(#towns)
17:12:44 <Eddi|zuHause> you could do O(#towns) by a simple loop over all towns
17:13:52 <nielsm> my proposal for "security": 1) allow a TLS wrapped network protocol, 2) server has a secret key it uses to scramble company passwords with to store them in the savegame
17:14:15 <nielsm> with (2) the server will be able to reload the company passwords, but anyone else getting the savegame won't be able to read them
17:14:34 <nielsm> and (1) is just generic expectations
17:15:06 <Eddi|zuHause> actually, i think you can get away with just using the squared euclidean distance
17:16:41 <Gabda> i think i have to look into the graph representation deeper to underwtand what you mean
17:16:48 <Eddi|zuHause> when i scroll through my code, i see a few divisions that might be problematic, but no sqrt
17:18:15 <Borg> are u guys wasting time trying to encrypt company passwords?
17:18:39 <Eddi|zuHause> Borg: that's not the point
17:18:49 <Borg> except that admin can read them.. and u can sniff them.. is there an issue here?
17:18:54 <Borg> Eddi|zuHause: and the point is?
17:18:58 <nielsm> I still hold the best approach is to add a note to the password entry box that the server admin may be able to read anything you enter here
17:19:08 <Eddi|zuHause> Borg: the point is, people are stupid and reuse passwords that they're also using for "important" stuff
17:19:42 <Borg> stop protecting dumbs from natural selection..
17:19:55 <Gabda> limit the password in 5 a-z letters
17:19:56 <Eddi|zuHause> i would kickban you right now if i had op
17:20:05 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on issue #6967: BaNaNaS: Unhandled exception when uploading new https://git.io/fhZWA
17:20:07 <Borg> Eddi|zuHause: lucky.. u dont have it
17:20:46 <Gabda> noone has a pass that short, so they would need to think up a new one
17:20:54 *** ToBeFree has joined #openttd
17:21:15 <andythenorth> can borg handle the failure demand then?
17:21:16 <Gabda> and they will feel it is not that secure
17:21:19 <andythenorth> give him the support
17:21:20 <Borg> do NOT do anything.. best idea is nielsm ... a hint: password is NOT stored securly.
17:21:40 <andythenorth> it's a game with players
17:21:43 <Eddi|zuHause> Borg: people would go crazy
17:21:54 <Borg> then do NOT do anything..
17:22:01 <andythenorth> I am fine with doing nothing
17:22:02 <Borg> just do NOT waste time on such silly issue really
17:22:13 <Borg> but.. of course.. its your right to waste time :) do as you wish
17:22:20 <TrueBrain> security silly? Let me grap my popcorn, this gotta be good
17:22:30 <Eddi|zuHause> Borg: it's by far the easier apporach to do this properly than to spend the next few years turning down requests about "why does it try to scare me?"
17:22:34 <Borg> we talk about game.. not security..
17:22:52 <TrueBrain> Borg: please adjust your tone; it is rather annoying. the capitalization is unneeded. The constant references to "wasting time" is bullshit in an open source community. Be polite. Be a good citizen. Be less abusive. Tnx!
17:22:57 <andythenorth> Borg: are you aware of the context?
17:22:58 <Borg> security for me is: OpenTTD packet handling can be exploited to get a shell... that IS problem
17:23:25 <Borg> TrueBrain: yeah.. and allow idiots to take over the world.. no thank you
17:23:59 <Eddi|zuHause> Borg: you know, these kinds of "get a shell" attacks are usually chains of vulnerabilities that start with "user password is transmitted in plain text"
17:24:23 <Borg> Eddi|zuHause: uh.. did you ever done any security stuff? shell codes? injections.. whatever..
17:24:41 <Borg> belive me.. that vectors have nothing to do w/ password storage
17:24:46 <Borg> this is completly different vector
17:24:51 <Eddi|zuHause> right, we have a winner for secon... third ignore list entry
17:25:07 <Borg> andythenorth: not fully.. but still
17:25:19 <Eddi|zuHause> ah, the quietness
17:26:56 <TrueBrain> hmm ... my own bot refuses to accept my credentials .. this is annoying
17:27:27 <nielsm> grr windows updates again busting my debug symbols cache
17:27:29 <Borg> andythenorth: before I will read futher.. does openttd rcon can exec anything?
17:27:49 <Borg> and.. is it by default disabled?
17:28:23 <andythenorth> I have literally zero knowledge of ottdsec
17:28:34 <Borg> okey.. time to do RTFS then
17:29:29 <DorpsGek> TrueBrain: TrueBrain
17:29:38 <TrueBrain> finally, DorpsGek, took you long enough
17:30:44 <TrueBrain> @calc 60 * 60 * 3600
17:31:05 <TrueBrain> @kban Borg 90000 I asked nicely. Sadly, I got a not-so-nice-answer. So .. try again in 24h
17:31:06 *** DorpsGek sets mode: +b *!~borg@87-99-43-52.internetia.net.pl
17:31:07 *** Borg was kicked by DorpsGek (I asked nicely. Sadly, I got a not-so-nice-answer. So .. try again in 24h)
17:31:10 <TrueBrain> right, that took WAY too much time
17:31:42 <TrueBrain> we might not have a CoC, but no need to be a dick
17:31:51 <Eddi|zuHause> doesn't happen often enough to be a reflex? :p
17:32:01 <TrueBrain> yeah .. I lost my touch there :P
17:32:19 <peter1138> So, er, what was the issue there?
17:32:28 <LordAro> i'm inclined to agree with borg's general point, it has to be said
17:33:05 <Eddi|zuHause> i'm inclined to oppose his "general point"
17:33:19 <TrueBrain> I don't think his point was relevant
17:33:25 <TrueBrain> his way of delivering the message was not ... nice
17:33:41 <TrueBrain> and if someone asks you to be less .... hostile .. the wrong answer is saying "no" to that :)
17:33:43 <Eddi|zuHause> well, that needs no discussion
17:36:13 <TrueBrain> Eddi|zuHause: ^^ :)
17:36:33 <Eddi|zuHause> yeah, i threatened him with kickban for no reason
17:37:13 <Eddi|zuHause> then TrueBrain appeared and kickbanned him
17:37:47 <TrueBrain> Eddi|zuHause: owh, I didnt even read that part
17:37:50 <Eddi|zuHause> LordAro: my point is, "it's a just game" is never an excuse to not do it properly
17:37:54 <TrueBrain> that alone would indeed have been reason enough :)
17:38:02 <nielsm> (needs some line breaking)
17:38:24 <TrueBrain> nielsm: I like the pragmatic solution there :)
17:38:51 <nielsm> "don't use a password you care about" basically
17:39:01 <TrueBrain> we had talks about that months ago already
17:39:08 <Eddi|zuHause> do we also need a GDPR blob?
17:39:16 <TrueBrain> guess we tried to solve it a bit too complicated :D
17:39:29 <nielsm> that's up the the individual server admin how they manage personal information :)
17:39:48 <TrueBrain> OpenTTD's services are mostly GDPR compliant
17:39:55 <TrueBrain> I say mostly, as there is one more thing I would like to fix
17:39:58 <nielsm> include a default MOTD system via the story book? :D
17:40:22 *** HeyCitizen has joined #openttd
17:40:41 <TrueBrain> okay, I only now caught up with what the discussion was about :P
17:40:57 <Eddi|zuHause> i mean more like, when you open the multiplayer window, a popup saying "the servers are not operated by the OpenTTD team, we are not responsible with what the servers do with your data"
17:41:09 <TrueBrain> LordAro: I work for a security company; trust me if I tell you OpenTTD really needs to help people not being stupid :D
17:41:17 <TrueBrain> (IT security company)
17:41:35 <TrueBrain> Eddi|zuHause: feels a bit over the top at first, but I see your point
17:41:45 <TrueBrain> Eddi|zuHause: possibly only the first time you go there or something?
17:42:16 <Eddi|zuHause> like a "don't bother me again" checkbox?
17:42:27 <Eddi|zuHause> probably overkill
17:42:30 <TrueBrain> nielsm: btw, last time we talked about passwords, we figured out the best solution was an Authentication Server :D
17:42:33 <nielsm> so, how do I set a maximum width on a WWT_LABEL widget and make it automatically line break?
17:42:34 <TrueBrain> (talking about overengineering)
17:44:21 <TrueBrain> nielsm: another suggestion that floated by months ago, was instead of setting a password, let the server return a phrase when you want to "protect" your company
17:44:25 <TrueBrain> removing the human factor completely
17:44:55 <nielsm> TrueBrain, i.e. server makes a password for you?
17:45:05 <TrueBrain> or something like that
17:45:07 <Eddi|zuHause> nielsm: there might be a "write multiline string" function?
17:45:24 <Samu> maybe dont call it password
17:45:25 <TrueBrain> you can save this easily in savegames, etc
17:45:39 <Samu> call it ... hmm pin, or so
17:45:45 <Eddi|zuHause> TrueBrain: how are people supposed to remember random strings thrown at them?
17:45:47 <TrueBrain> passphrase, pin, ...
17:45:53 <TrueBrain> Eddi|zuHause: write them down ? :D
17:46:02 <TrueBrain> it is not a security thingy
17:46:14 <Eddi|zuHause> sounds like something unlikely to happen
17:46:50 <TrueBrain> that is indeed the biggest drawback of that method
17:47:06 <TrueBrain> (third time I agree with Eddi|zuHause today; shit is happening!)
17:47:40 <TrueBrain> so I like the pragmatic solution nielsm is proposing :)
17:48:01 <TrueBrain> and indeed what Samu suggests .. don't call it password .. and possibly don't even mask it
17:48:06 <TrueBrain> that should give so many hints :P
17:48:38 <nielsm> it's not being masked during entry
17:48:55 <TrueBrain> Set the name of your animal
17:49:19 <TrueBrain> public secret handshake :P
17:49:24 <Eddi|zuHause> your mom's maiden name
17:49:27 <TrueBrain> Sacro: I await your patch :D
17:49:47 <TrueBrain> but yeah, some Authentication Server moes the trust to a single entity
17:49:52 <TrueBrain> instead of any server :P
17:49:53 <nielsm> yeah you should submit your public key to the server and authenticate your company with your private key
17:50:13 <TrueBrain> (with client-side certificate)
17:50:39 <Eddi|zuHause> introduce people to letsencrypt?
17:50:42 <TrueBrain> we can make an index of the public keys :)
17:50:57 <TrueBrain> and we are back to overengineering the shit out of this :P
17:51:18 <andythenorth> just don't run a server?
17:51:25 <TrueBrain> remove multiplayer!!
17:51:26 <andythenorth> do clients have any ports open etc?
17:51:31 <andythenorth> if not running server?
17:51:38 <TrueBrain> I remember the time I added code to track players ... for some reason this was not a like addition to the game :P
17:51:42 <TrueBrain> I was so naive back then
17:51:54 <andythenorth> I don't get any warnings from my OS about OpenTTD trying to use network services, but eh
17:51:56 <TrueBrain> andythenorth: clients never have ports open; servers do :)
17:52:01 <Eddi|zuHause> now for something completely differences, i have two annoyances with the CI: 1) in the log of the failed task, the whole log is repeated with [error] instead of the actual error message, and 2) the error message is not propagated to the github PR
17:52:23 <TrueBrain> Eddi|zuHause: 1) can be solved, but someone has to do that :)
17:52:32 <TrueBrain> 2) they do, but only for Windows atm
17:52:41 <TrueBrain> I am not sure we did something wrong, or Azure Pipelines are not there yet
17:52:44 <TrueBrain> but yes, it is annoying as fuck
17:53:19 <andythenorth> so we started trying to understand how to 1
17:53:26 <Eddi|zuHause> and possibly we should run the commit checker first, and only run the other tasks after that one came out ok? for earlier failure abort
17:53:38 <Sacro> Disable port fowarding except via SSH tunnel
17:53:46 <Sacro> require pub key to ssh tunnel, bingol
17:53:53 <TrueBrain> Eddi|zuHause: good point, hmm .. will look into that
17:54:29 <Eddi|zuHause> Sacro: we'll have about 3 players after that change.
17:54:37 <TrueBrain> Eddi|zuHause: I removed that part, as I thought: if we run it all at once, it is done quicker. But that is only for when it succeeds :D
17:54:39 <Sacro> how many more does a server need?
17:54:45 <glx> and of course a better error message TrueBrain ;)
17:54:57 * Sacro wonders if he could pay OpenTTD at work
17:54:59 <Eddi|zuHause> TrueBrain: i assume the commit checker is the one failing the most often
17:55:15 <Eddi|zuHause> TrueBrain: also i noticed some random task failing because it couldn't get opengfx from mirror
17:55:17 <TrueBrain> Eddi|zuHause: yeah .. I think we should make it two CIs
17:55:23 <TrueBrain> one for the commit checker, the other for the rest
17:55:34 *** synchris has joined #openttd
17:55:36 <TrueBrain> that ... should never happen
17:55:40 <TrueBrain> link me the logs in PM please
17:55:47 <TrueBrain> sadly, I have to go now, so I will look into that this weekend
17:55:52 <TrueBrain> but please tell me about these things :)
17:55:56 <TrueBrain> cannot fix what I cannot see :)
17:56:07 <Eddi|zuHause> uhm, have to dig for that
17:56:11 <andythenorth> we have room now
17:56:39 <TrueBrain> owh, I was going; ciao!
17:57:27 <andythenorth> but there's a patch, it's long-standing, it's in JGRPP
17:57:32 <glx> TrueBrain: builds 153 and 173 for example
17:58:24 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #3000: Allow train to find depot along reserved path https://git.io/fhZlw
17:58:25 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #3000: Allow train to find depot along reserved path https://git.io/fhZlr
17:59:55 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #5753: Game Script: GSText in info.nut's GetSettings() method https://git.io/fhZlS
17:59:56 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #5753: Game Script: GSText in info.nut's GetSettings() method https://git.io/fhZl9
18:01:08 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #5875: Bananas: online content doesn't stop downloading when canceled https://git.io/fhZl7
18:01:09 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #5875: Bananas: online content doesn't stop downloading when canceled https://git.io/fhZl5
18:02:03 *** HerzogDeXtEr has joined #openttd
18:02:12 <Eddi|zuHause> i need an IRC client that could filter the DorpsGek_II messages and display them in a different font/colour/size
18:02:50 <glx> I modified my event handler to not ding on DorpsGek_II :)
18:03:51 <glx> all the ding ding spam was annoying
18:04:22 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #6095: File failed to decompress - Error window not on top https://git.io/fhZlx
18:04:23 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #6095: File failed to decompress - Error window not on top https://git.io/fhZlp
18:04:46 <glx> because each issue closing means 4 dings ;)
18:05:14 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #6226: Crash when no baseset en invalid font in configuration https://git.io/fhZlh
18:05:39 <glx> and andythenorth I suggest you /nick ;)
18:05:53 *** andythenorth is now known as andythenorth_stop_ping
18:07:36 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #6383: Bananas: "Select upgrade" for content tries to download old versions https://git.io/fhZ8f
18:07:37 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #6383: Bananas: "Select upgrade" for content tries to download old versions https://git.io/fhZ8J
18:07:48 *** andythenorth_stop_ping is now known as andythenorth
18:09:02 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #6452: Locked GS parameter value can be reseted to default value https://git.io/fhZ8T
18:09:03 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #6452: Locked GS parameter value can be reseted to default value https://git.io/fhZ8k
18:11:15 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on issue #6539: "Load scenario" dialog remains focusable while downloading missing online content for a scenario https://git.io/fhZ8Y
18:11:16 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed issue #6539: "Load scenario" dialog remains focusable while downloading missing online content for a scenario https://git.io/fhZ8O
18:11:49 <andythenorth> just 120 left to go
18:12:00 <andythenorth> apply now, limited stocks!
18:14:38 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
18:16:47 <Samu> changed some variable names
18:17:12 <Samu> and with that the code seems more readable :), also made the checks to happen earlier rather than later
18:17:54 <Samu> I know I said this before, but now I believe the code is ready for review *cross fingers*
18:20:43 <Samu> well, the pipeline azure dude isn't building...
18:21:09 * andythenorth should patch to enable flipping articulated vehicles
18:21:27 <andythenorth> but newgrf author has to handle *all* of it except the bit flip
18:22:06 <Samu> why doesn't it build? I did exactly as told
18:22:24 <Samu> git rebase -i upstream/master
18:23:47 <nielsm> because the pipeline is busy building other things
18:25:17 <Samu> the last build was yesterday night
18:25:25 <Samu> i just pushed 10 mins ago
18:26:05 <Samu> OpenTTD CI In progress — This check has started...
18:26:56 <andythenorth> after 4 days work on my pixel generator
18:27:01 <andythenorth> I am ready to draw 3 trains with it
18:27:17 <glx> ah you had time to do other stuff ?
18:27:43 <andythenorth> 24 / 7 slaving over python
18:29:52 <planetmaker> what about... "good first issue". Such issues are relatively simple
18:30:03 <planetmaker> at least I'd think so :D
18:30:08 <planetmaker> I might completely err
18:30:43 <andythenorth> I might go re-open some later
18:30:54 <andythenorth> I think to fairly encourage 'first issue' PRs
18:31:01 <andythenorth> we have to be reviewing PRs
18:31:10 <andythenorth> otherwise it's a bit misleading
18:31:45 *** frosch123 has joined #openttd
18:31:53 <planetmaker> that's way more important
18:32:09 <andythenorth> let's review a frosch123 PR!
18:32:21 <andythenorth> it's good to encourage new contributors!
18:32:24 <planetmaker> it might simply leap into the water :)
18:32:48 <andythenorth> I can't comment on the code, or the savegame implications
18:32:51 <andythenorth> or the MP implications
18:32:57 <andythenorth> or the pathfinder risks
18:33:22 <andythenorth> but I can do maths :P
18:35:51 <planetmaker> I don't understand how that fixes it. But I don't see how fixing that has any implications on anything relevant
18:36:02 <planetmaker> except that the game becomes more visually appealing
18:36:22 <andythenorth> the vehicle has no forward motion
18:36:26 <planetmaker> (the last 'e' in the sentence is relevant) :D
18:36:27 <andythenorth> in the fixed version
18:36:44 <andythenorth> in the non-fixed version, there is some acceleration, but the vehicle can't actually move
18:36:48 <andythenorth> so the motion counter increases
18:37:31 <andythenorth> I can't approve my own PR :P
18:37:42 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker approved pull request #7018: Fix: Don't increase motion counter while train is waiting at non-path… https://git.io/fhZ8V
18:38:01 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker merged pull request #7018: Fix: Don't increase motion counter while train is waiting at non-path… https://git.io/fh3Nb
18:38:22 <planetmaker> oh :P Was it reserved for him?
18:38:41 <planetmaker> sorry, should that have been the case
18:38:57 <andythenorth> oh you have merge rights too :D
18:39:07 * andythenorth didn't get the memo sorry
18:39:46 <planetmaker> I may not know much about github. But I could do much :P
18:40:12 <andythenorth> I might even use it if it's added
18:40:27 <andythenorth> new contributor also :)
18:43:19 <planetmaker> I actually would like it to read "Default company colour" as well. But feels a bit like bike shedding
18:44:03 <andythenorth> Could go on for days :)
18:44:06 <Eddi|zuHause> <planetmaker> I don't understand how that fixes it <- imagine you are blind and run into a wall, that wall bounces you back 100 distance units and disorients you so you don't remember there was a wall. now you have 100 distance units to cover some acceleration, and inevitably hit the wall again, at a non-zero speed
18:44:23 <andythenorth> nice analogy Eddi|zuHause :P
18:44:38 <Eddi|zuHause> now with the "fix" you don't bounce 100 units back, and every attempt at accelerating will now just push against the wall
18:44:50 <andythenorth> I think it makes nicer desert rivers
18:45:36 <planetmaker> you made PR out of all of those? :)
18:45:57 <andythenorth> I am not *just* closing issues :)
18:46:06 <andythenorth> I am trying to get some patches through
18:46:12 <andythenorth> but I can't review if I open them
18:46:23 <andythenorth> there is a little bit more complaining about ticket closing here
18:46:32 <andythenorth> than there is making and reviewing PRs :P
18:46:36 <andythenorth> I guess irc is cheaper
18:46:47 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker approved pull request #7015: Change: Desert tiles are now half-desert if a neighboured tile is non… https://git.io/fhZ8i
18:48:11 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker merged pull request #7015: Change: Desert tiles are now half-desert if a neighboured tile is non… https://git.io/fh3Ho
18:48:27 <planetmaker> cleaning issue list like this is... lazy and convenient :P
18:48:33 <planetmaker> for me at least ;)
18:50:42 <andythenorth> much easier for small things
18:50:48 <andythenorth> than curl patch, blah blah
18:52:05 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth approved pull request #6998: Feature #4115: default company colour setting https://git.io/fhZ8A
18:52:41 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker commented on pull request #6998: Feature #4115: default company colour setting https://git.io/fhZ8h
18:53:18 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on pull request #6848: Feature : New Trees Placement Algorithm "Forest". https://git.io/fhZ8j
18:53:19 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed pull request #6848: Feature : New Trees Placement Algorithm "Forest". https://git.io/fbx93
18:54:48 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on pull request #6846: Feature: Allow flip of train vehicles in depot independently of NewGRF property https://git.io/fhZ4J
18:54:49 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth closed pull request #6846: Feature: Allow flip of train vehicles in depot independently of NewGRF property https://git.io/f9kZm
18:55:34 <andythenorth> Samu: you going to rebase this one? o_O
18:55:41 <andythenorth> personally I think these checks are boiling the ocean
18:55:47 <andythenorth> but nielsm approved it ages ago
18:55:51 <andythenorth> it just didn't get merged
18:56:18 <andythenorth> are we also going to check road construction for blocking vehicle routes?
19:02:28 <planetmaker> someone on the internet is always wrong. Anotherone likely right
19:02:46 <andythenorth> Samu: JJ isn't a reviewer currently
19:02:51 <andythenorth> I happen to agree with JJ though
19:02:58 <andythenorth> these checks are not winning anything
19:04:11 <andythenorth> IMO: *either* fix the upstream issue (ships have no routes of any kind)
19:04:20 <andythenorth> *or* fix the downstream issue (AIs can't handle failure)
19:04:56 <Samu> that change is more geared towards AIs indeed
19:05:13 <planetmaker> it's somewhat an anti-griefing change
19:06:02 <Samu> a dock can still be blocked by other means, in that regard it's not complete
19:06:22 <andythenorth> there are 3 or 4 of these PRs
19:06:38 <andythenorth> I don't want to just reject them, because I don't want to disappoint
19:07:15 <Samu> well, i was told to make separate patches for every issue. this is one of the cases where everything makes sense together and little sense alone
19:08:24 <andythenorth> I guess I just go up a level and ask 'why fix this at all?'
19:09:32 <Samu> yeah... seems that no matter what I explain, it will seem lacking
19:10:00 <andythenorth> well you learnt a lot of code :)
19:10:37 <Samu> EnsureNoDocking would need to be checked on a global level if it is to have any real meaning
19:10:53 <Samu> currently, it's only doing that check when building docks only
19:11:12 <Samu> because that's where I saw AI's getting most problems
19:11:25 <andythenorth> I can only think that we do something like allowing players to buy sea tiles
19:11:31 <Samu> they would block each other's docking tiles
19:11:33 <andythenorth> which negates the problem, via ownnership
19:11:45 <andythenorth> but owning tiles just allows more ways to block
19:12:17 <andythenorth> why do we need to protect the AIs from each other? o_O
19:12:34 <Samu> but yeah, AI authors could do better :o
19:13:16 <andythenorth> it's probably very hard for an AI to understand why a ship is lost
19:17:08 <DorpsGek> Xaroth19: I don't recognize you.
19:17:34 *** Xaroth19 is now known as Xaroth
19:22:22 <Xaroth> quick, run, before TrueBrain is back.
19:22:35 *** Eddi|zuHause has left #openttd
19:22:45 *** Eddi|zuHause has joined #openttd
19:22:53 <Eddi|zuHause> i don't think that was right
19:24:36 <Samu> my next step would be making it global
19:25:19 <Samu> so, im unsure what to say atm
19:25:23 <andythenorth> I can think of literally no viable solution
19:25:34 <andythenorth> that reduces AI accidents or griefing
19:25:45 <andythenorth> without also increasing AI accidents or griefing
19:26:12 <andythenorth> we could store 'a ship has used this tile' in every water tile
19:26:18 <andythenorth> but when would we clear it?
19:26:34 <andythenorth> I have savegames with ships that take months or nearly a year to run a route
19:26:55 <andythenorth> then I can 'own' islands by running ships around them
19:27:15 <andythenorth> we could require players to buy water tiles
19:27:40 <andythenorth> but then I can trap competitor ships by buying the water
19:27:50 <Samu> i mean global, but not in that manner
19:28:10 <Samu> have all other building methods to check for DockingTiles
19:28:38 <Samu> don't think it's required to store whether a ship used a tile
19:29:12 <andythenorth> what are DockingTiles?
19:29:34 <Samu> tiles that "belong" to a dock where the ship stops to load/unload
19:30:28 <andythenorth> how does that stop ships getting trapped or blocked?
19:30:33 <andythenorth> is it 2 different issues?
19:30:48 <Samu> sec, let me create a situation
19:32:49 <Samu> most AIs like to build like that
19:33:02 <Samu> the EnsureNoDockingTile would prevent it
19:34:18 <andythenorth> ok so that's a specific issue
19:34:27 *** Progman has joined #openttd
19:35:00 <Samu> it's what the function does, but atm, it's only checked when placing docks, and not when doing any other thing
19:35:17 <Samu> like terraforming, etc...
19:35:20 <andythenorth> AIs should try harder
19:36:14 <Samu> there could be the case where the dockting tile could belong to two different owners
19:36:58 <Samu> that's what stopped me from pursuing further checks
19:37:24 <Samu> i wouldn't know how to make terraform work
19:38:08 <Samu> there's also docks being build on land, the 3rd tile may not necessarily contain water (yet), until I build a canal there
19:38:35 <Samu> who would own the docking tile?
19:40:48 <Samu> would i automatically enforce the 3rd tile to become a "docking tile" even if it doesn't have water (yet)?
19:41:06 <Samu> and would i prevent someone else from building a road there, for instance?
19:42:09 <Samu> so, yeah, i'm not entirely confident to pursue a global feature
19:46:18 *** gelignite has joined #openttd
19:51:15 <andythenorth> recognise ocean boiling
19:53:28 <DorpsGek_II> [OpenTTD/OpenTTD] Ufiby commented on issue #6967: BaNaNaS: Unhandled exception when uploading new https://git.io/fhZ49
19:58:56 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick commented on pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhZ4x
20:24:18 <Samu> and i said ready for review, i hate myself
20:24:54 <Samu> i need to change a x into an y
20:24:59 <Samu> does it need rebase again?
20:26:35 <andythenorth> when do we get concerned about action 2 / varaction 2 performance? o_O
20:26:45 * andythenorth is making switches as though they have zero cost
20:28:05 <Samu> oh god I feel so bad when I fail
20:28:19 <Samu> hope no one was reviewing it yet
20:29:44 <Samu> tile_2 wasn't using diffc.y for the y coordinate
20:29:58 <Samu> was using diffc.x because i'm blind
20:30:17 <Eddi|zuHause> andythenorth: main question there is how often the callback is called
20:30:50 <andythenorth> I suspect the impact is worse for me, compiling, than for OpenTTD
20:31:14 <andythenorth> fundamentally I'm adding a lot of branches, but the depth probably isn't that great
20:31:19 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
20:31:35 <Eddi|zuHause> andythenorth: if you ever are at a situation where you can optimize the run time at a cost of increased compile time, do that.
20:32:05 <andythenorth> it's kind of utilitarian though
20:32:41 <Samu> do I have to rebase force push again?
20:34:51 <andythenorth> when there are complex rules for which vehicle sprite to show
20:34:54 <andythenorth> then there are layers
20:35:00 <andythenorth> multiplication kicks in :P
20:35:35 <Samu> I guess I know the answer already
20:35:55 <Samu> even for a typo, i need a rebase
20:37:38 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6931: Change: Prevent town growth from blocking ships https://git.io/fhsRY
20:38:01 <Eddi|zuHause> the callback depth is probably most important, also the complexity of the variable calculations. the number of choices in each switch is probably irrelevant for performance
20:38:37 <andythenorth> if only we had an FPS meter....?
20:39:49 <planetmaker> we have an FF speed-o-meter
20:41:10 <andythenorth> what I could do with Eddi|zuHause is a var to compare current railtype and powerered-ness on ELRL :P
20:41:21 <andythenorth> thart would be very helpful in 1.9.0
20:41:54 <Eddi|zuHause> i guess i should start to implement the suggestions from the ticket
20:43:32 <andythenorth> I should finish pantographs for EMUs
20:43:36 <andythenorth> I blame Eddi|zuHause and nielsm
20:43:49 <Eddi|zuHause> andythenorth: we have a performance dislpay now
20:43:56 <andythenorth> I know, I was being silly :)
20:44:09 <andythenorth> what's the near-equivalent to sunk cost fallacy?
20:44:16 <andythenorth> when the effort will return
20:44:24 <andythenorth> but if you'd known in advance
20:44:29 <andythenorth> you might not have bothered :P
20:45:05 <Eddi|zuHause> but sometimes the effort to find out how much effort it would be is the same as just doing it
20:46:11 <andythenorth> making automatically generated, railtype responsive pantographs in 3 variants
20:46:17 <andythenorth> was a lot of work
20:48:24 <andythenorth> OTOH it's a solid use case for sprite layers
20:50:45 <Eddi|zuHause> "we put all this effort into sprite layers, now we must use them for something" is almost definitely a fallacy
20:53:32 <andythenorth> if the pantographs didn't have up / down states, I'd have rendered them into the vehicle sprite
20:54:20 <andythenorth> so what about user bits for train vehicles then? o_O
20:54:46 <andythenorth> need to be get / set on cb
20:55:38 <andythenorth> oh and I want a cb for 'vehicle was flipped by player' :P
20:58:46 <planetmaker> hahaha. That would solve the problem: you could then visually flip-back the vehicle and make it like it was never flipped :P
21:02:30 <andythenorth> I would count the number of flips, in a store
21:02:38 <andythenorth> mod(2) would give the actual flip state :P
21:05:47 <andythenorth> oh we didn't patch nml for 64 railtypes? o_O
21:06:08 <andythenorth> looks fine to me
21:06:35 <andythenorth> drive-by-contributions :D
21:06:50 <planetmaker> CI has not been setup?
21:07:23 <planetmaker> oh, nml, not OpenTTD
21:07:52 <andythenorth> should maybe have CI
21:13:27 <planetmaker> it would be no issue to run it on the existing CI I have for NewGRF (that's the plan - it also has tests for nml there)
21:14:11 <planetmaker> but curse V. I need to complete a factory :P
21:22:54 *** andythenorth is now known as Guest289
21:22:54 *** andythenorth has joined #openttd
21:26:43 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #6988: Change: AI/GS ScriptBridge::GetName takes one extra parameter to refer the vehicle type https://git.io/fhfaC
21:56:22 <frosch123> LordAro: function overloading and default parameters do not work in the squirrel api
21:56:46 <frosch123> the api wrapper can't handle that
22:28:02 *** m3henry has joined #openttd
22:48:44 *** Progman has joined #openttd
22:53:51 <andythenorth> 'over-engineered much'
22:54:22 <planetmaker> I guess NML is just really missing the appropriate trigger at the repo... building on DevZone works...
22:56:10 * andythenorth wonders if electric steam engines are now one line change
22:58:44 <andythenorth> apparently they are
23:04:35 *** andythenorth has left #openttd
23:08:21 <planetmaker> is that another name for fuel cell trains? ;)
23:09:28 <peter1138> " Today, I can still eat: 2707 out of about 5000 cals"
23:09:35 <peter1138> fitbit, why you so mad
23:30:01 <Eddi|zuHause> oh, i totally forgot we have the user bits on vehicles...
23:30:26 <Eddi|zuHause> that's one of those too-specialized-to-be-useful features of NewGRFs
23:33:13 <Samu> I'm looking at nielsm suggestion
23:33:32 <Samu> it doesn't handle service at nearest depot
23:34:13 <Samu> what if there are 500 ship depots ? :(
23:34:59 <Samu> it will only check next depot every tile it walks into a new tile
23:36:12 <Samu> i'm unsure how's this list kept refreshed, seems magic
23:38:24 <Samu> or maybe im seeing this wrong, brb
23:41:01 <Samu> ah, it pathfinds for every depot
23:41:17 <Samu> doesn't look like a good solution
23:42:00 <Samu> i dont know about my solution stallness either, must check
continue to next day ⏵
