IRC logs for #openttd on OFTC at 2024-03-30
⏴ go to previous day
00:53:27 *** jinks has quit IRC (Remote host closed the connection)
01:23:24 *** Smedles has joined #openttd
02:08:00 *** Wormnest_ has quit IRC (Quit: Leaving)
02:25:42 *** Eddi|zuHause2 has joined #openttd
02:30:52 *** Eddi|zuHause has quit IRC (Ping timeout: 480 seconds)
03:20:20 *** D-HUND has quit IRC (Ping timeout: 480 seconds)
04:40:34 <DorpsGek> - Update: Translations from eints (by translators)
05:44:38 <merni> so, drop linux and mac support?
06:11:51 <peter1138> Source only on all platforms, build it yourself
06:22:37 <peter1138> Anyway, it's because access to lzma has been blocked by github.
07:37:57 <truebrain> That kinda makes sense .. so we will let all that settle and enjoy build failures till it does 🙂
07:40:03 <andythenorth> Might change my nick to BuildFailureEnjoyer
07:41:06 <andythenorth> I have been awake since 4am, but then again I noticed some of you have also slept improperly
07:50:37 <gebik4544> Bananas are broken... It tells me something has failed every time I try to do an autofill, validation or publish of an scenario that is attached here...
07:51:20 <gebik4544> I tried just before posting here, so checking logs might be good idea.
07:55:18 <truebrain> Yeah, we received the bug report from the backend already. I guess you made that scenario with a 14.0 RC or beta?
07:55:59 <gebik4544> truebrain: JGR 0.58.1
07:56:30 <truebrain> Yeah. Will be fixed.. "soon" (no eta)
07:58:10 <truebrain> peter1138: This btw also might impact our planned release 🙂
08:02:01 <andythenorth> Happy Easter weekend
08:10:31 <truebrain> I am guessing the CI will fail, but what-ever 😛
08:17:49 <truebrain> that is new, that sentry jumps in .. lol
08:23:52 <truebrain> well, although that PR fixes scenarios made with 14.0, it does not fix the above mentioned scenario .. but that might be JGRPP related
08:35:31 <truebrain> gebik4544: I fixed the brokeness, but it seems JGRPP has a change that makes the normal savegame reader fail. I need to investigate this further. But at least you can now upload vanilla 14.0 scenarios 🙂
08:37:27 <_jgr_> I've been migrating more chunks to use the new table format, it may be related to that
08:37:51 <truebrain> this reader knows how to read the common chunk formats; but it is overreading some blob
08:40:42 <truebrain> there is more than one NGRF chunk 😮
08:41:13 <truebrain> misreading my own code
08:41:21 <truebrain> but it is the NGRF chunk that is giving trouble
08:43:36 <truebrain> it can read 2 entries from the NGRF chunk fine, but it fails on the 3rd
08:44:49 <truebrain> very specific, it fails to read the md5sum
08:46:47 <truebrain> lol, yes, that is kinda broken 😄
08:47:03 <truebrain> it wrote in the savegame the md5sum is of length 108 bytes
08:47:13 <truebrain> or, and where it fails, it is of length 218072164 bytes
08:48:42 <truebrain> vanilla really writes 16 there
08:52:46 <truebrain> if I would be to guess, JGRPP doesn't write the length byte for fixed arrays (but it should, as the savegame has no information about the fact it is fixed)
08:54:07 <truebrain> yeah, if I don't read the length but force it to 16, it does work
08:54:18 <truebrain> _jgr_: did you copy the vanilla encoder, or did you make changes / optimizations yourself?
08:57:50 <_jgr_> truebrain: That chunk is using my own, not the vanilla backend
08:58:20 <truebrain> Sadly, I cannot fix it from my end, as I literally don't have the information to do so 😦
08:58:52 <truebrain> which was a bit the trouble with fixed-length arrays .. I had to repeat the length every time, as there was no place to store it was fixed-length 😦
09:07:20 <truebrain> ah, you have `saveload` for the vanilla loading, and `sl` for your own 🙂 I see 🙂 (sorry, I am a curious person, and kinda hoping to find a way I can just load this on the backend 😄 )
09:08:50 <_jgr_> Where possible, chunks in sl just forward to the vanilla code in saveload
09:09:36 <truebrain> and nice that you are switching to the header structure too btw 🙂 Must not be easy .. (I remember how much effort it took me 😛 )
09:10:23 <truebrain> okay, so the SlArray in your save/load doesn't do a `SlWriteArrayLength(length);`. Which makes sense. And which works fine when you save/load with that code. But yeah, it does mean it cannot be read as a table-format, as that length information isn't stored anywhere in the savegame
09:10:31 <truebrain> meh .. means I also can't hack around it 😦
09:11:14 <truebrain> that was btw one of the more subtle changes in my changeset .. that could have been written down with a bit more care on my end 🙂
09:11:19 <_jgr_> I'll have it fixed before I next do a release
09:20:32 <truebrain> meh; still trying to find a way to know that this is a savegame without the length bytes there, and fix it. But it is tricky 🙂 (sorry, I just like to support on the backend what-ever is thrown at it 😛 )
09:23:23 <truebrain> _jgr_: do I have it correct that you don't store a savegame versions as much, but more flags to what things are in the savegame?
09:24:43 <_jgr_> Yes, the main savegame version is basically frozen now, the first chunk in the file contains the versions of the various features
09:25:13 <truebrain> okay; and would I be wrong to assume that whatever you cook up to add this byte, it would be either a feature bump or a new feature?
09:26:33 <_jgr_> Yes, I think a feature version bump should be enough
09:28:36 <_jgr_> The version chunk has its own format though, I don't want to make you have to write a parser for that as well
09:29:42 <truebrain> a nice way to get a bit of insight what you have been up to 🙂
09:31:36 <truebrain> yeah; found that part, and it is also accurate 🙂 Which is nice 😉
09:37:42 <truebrain> `version_label 1 7 b'jgrpp-0.58.1'` w00p
09:55:42 <xarick> AroAI stalling at times
09:56:15 <andythenorth> such downgrading
09:57:46 <truebrain> _jgr_: I would appreciate your opinion on #456, to see if I did something stupid. Either way, it also means there is no pressure for you to fix this in JGRPP anymore 🙂 Also, feel free to implement the addition of the length indicator in any other form; I just made a wild assumption you would bump `table_newgrf_sl` (and the other `table_XXX_sl`, but by no means that is actually required 🙂 Any
09:57:46 <truebrain> indication which savegame does or does not have the length indicator will work 🙂
09:59:45 <_jgr_> Yes, I was planning on bumping that version
10:00:18 <_jgr_> I don't think that I'll need to bump others as I was testing using your standalone savegame readers for earlier work on this
10:00:42 <truebrain> I did not check if any other chunks were impacted; but if you did test, you indeed wouldn't have to 😄
10:00:50 <truebrain> we don't use `SL_ARR` that much
10:01:10 <truebrain> but that is even better news, means my loader works fine like this 🙂
10:04:05 <truebrain> and I am happy I found a relative simple solution 😄 Took a bit of creativity 😄
10:16:16 *** Eddi|zuHause2 is now known as Eddi|zuHause
10:28:08 <_jgr_> #456 looks all OK to me
10:29:41 <truebrain> Teamwork makes the dreamwork! Sweet!
10:39:50 <truebrain> gebik4544: should be all fixed now!
10:41:45 <xarick> ah, AroAI is constantly trying to build a road non-stop without sleep
10:49:40 <frosch123> wake up, read a sentry mail, notice it's already fixed, live's good 🙂
10:55:14 <truebrain> and think of it like this .. that Sentry issue was (relatively) easy to solve. But now imagine you get the report for xz 😛
10:56:44 <frosch123> i am sure it was disclosed by java people just before easter, just to take revenge from log4j
10:57:03 <truebrain> it is one big conspiracy? 😄
10:58:23 <truebrain> I do like HackerNews. The whole xz thread, has a thread inside about whether 2FA protects against anything or not ... 😛
10:58:25 <frosch123> to me the most funny part is: xz affects ssh, even though ssh has no intended call tree that ends up in xz
10:58:33 <frosch123> so it must all be static initialisation of shared libs
10:58:39 <frosch123> blame c++, c never had such issues :p
10:59:04 <truebrain> yeah .. I did not know you could do these things via a library before yesterday 😛
10:59:08 <truebrain> never heard of ifunc before
10:59:12 <truebrain> maybe that is not a bad thing 😄
11:00:16 *** keikoz has quit IRC (Read error: Connection reset by peer)
11:02:23 <frosch123> static initialisation is fancy 🙂 you can do many cool things with it, like declaring test-cases without having to explicitly call them. and then the great static-initialisation-fiasco footgun, though maybe it is rather a leggun
11:03:03 <truebrain> test-infrastructure in C(++) has always been awkward
11:03:16 <truebrain> using weak functions to mock stuff .. it is just blegh
11:04:12 <truebrain> lol, so the latest vcpkg shows very odd behaviour on Linux; but although they have .exe and .pdb on their release page, they don't seem to have Linux debug symbols 😛
11:04:16 <truebrain> and the binary is stripped of them
11:06:45 <truebrain> haha, so vcpkg can't build xz atm, as GitHub suspended the repo (for good reason) .. the suggestion of people? "just use a mirror of xz for now" ... euh ..
11:06:54 <truebrain> that sounds like a terrible choice 😛
11:15:55 <frosch123> worse than naming a key-value store "valkey"?
11:16:49 <truebrain> reminds me that Redict is the new Redis 😛
11:19:57 <truebrain> grrr @ performance benchmarking on GitHub Runners ... sometimes you just end up on a node that is too busy to do an accurate reading
11:20:29 <truebrain> `High deviation detected: 5.71% after 13708.01 ms` .. between 5 runs, there was a 5% standard deviation of runtime
11:20:36 <truebrain> that is just too much
11:20:36 <frosch123> yeah, how could you ever detect a slowdown if 0.2s in ottd startup?
11:20:49 <truebrain> I am not interested in a 0.2s slowdown 🙂
11:21:01 <truebrain> I am interested in those +10% CPU time over a 10s run 😄
11:25:26 <truebrain> that said, I guess I should also publish flamegraphs, so we can more easily see our hotspots 🙂
11:58:21 <truebrain> GitHub sometimes does have a sense of humor 😄
12:48:50 <truebrain> missing a header, I guess 😛
13:05:37 <peter1138> Or, in miles, 69. Nice
13:06:02 <truebrain> ... how ever old we are, we are still little children 😛
13:07:38 <_glx_> the vcpkg crash you reported seems to be the hard to trace kind
13:07:52 <truebrain> yeah, when I manually compile vcpkg-tool, the problem does not exists
13:07:57 <truebrain> even if I compile with their commands
13:08:00 <truebrain> so it is something on their buildfarm
13:08:07 <truebrain> I was afraid to find another xz, so I stopped looking
13:08:08 <_glx_> my guess would be memory corruption somewhere
13:08:19 <peter1138> Fixed length arrays without lengths eh
13:08:42 <truebrain> `### unhandled dwarf2 abbrev form code 0x25`
13:10:21 <truebrain> _glx_: valgrind reports no errors when running with my own build binary ..
13:10:27 <truebrain> and their binary has no symbols, so nothing to tell about that
13:11:17 <_glx_> maybe some undefined behaviour
13:14:18 <peter1138> Constinit exists but I don't it does much for static initialisation fiasco
13:16:04 *** Eddi|zuHause has quit IRC ()
13:17:37 <peter1138> Oof 186 bpm this time, and that's with a new battery.
13:17:47 <_glx_> hmm the x86_64 macos failure is bad luck in timing (arm64 is not affected because it was able to cache lzma in a previous run)
13:18:34 *** Eddi|zuHause has joined #openttd
13:21:44 <_glx_> that will be a problem soon
13:27:48 <frosch123> hmm, is there a standard procedure to report people who pushed their pypi upload token to their gh repository?
13:28:07 <truebrain> if they are a secret partner, the token will already be blocked
13:28:34 <truebrain> but that is the only "standard" I know in this field
13:36:50 <frosch123> ok, in 2021 pypi was a secret scanner partner. let's assume they still are
13:37:06 <truebrain> then at least they got a call from GitHub telling the token was added to a repo 🙂
13:38:55 <_glx_> and token should be invalidated
13:39:25 <frosch123> a coworker of me uses a tiny python package, and wanted to contribute back some minor fixes. they regretted that so hard... turned out the "maintainer" is 60+, has no clue whatsoever, and I suspect their nephew helped them upload their thing to github/pypi in the first place
13:41:08 <frosch123> though, maybe normal for a belgian :p
13:57:41 <peter1138> Hmm, I guess we lose gpg signing information of commits sense we always squash/rebase.
13:57:59 <truebrain> squashing via the interface adds the gpg signature (that GitHub has) 🙂
14:04:25 <peter1138> In the context of xz, that isn't that useful.
14:24:58 <truebrain> no; which is the biggest issue in general with GPG signing 🙂
15:28:06 <peter1138> Of course, if the commits are all signed by the now bad-actor, it isn't that helpful either.
15:29:20 *** Wormnest_ has joined #openttd
15:50:36 *** keikoz has quit IRC (Ping timeout: 480 seconds)
15:56:54 <truebrain> lol, modifying xkcd with the wrong font, how bad 😛
16:06:20 <Rubidium> oh noes... has the bad actor changed the past of xkcd?!?
16:09:44 *** CornsMcGowan[m] has quit IRC (Quit: Client limit exceeded: 20000)
16:36:48 <Eddi|zuHause> probably misrepresenting the number of projects being targeted
16:37:06 <peter1138> Who's our bad actor?
16:42:29 <peter1138> (Well, only something wasm related)
16:45:19 <truebrain> Who was pushing for unittests in OpenTTD? 😄
16:45:37 <peter1138> Those test saves look like binary data to me.
16:46:03 <truebrain> And who pushes those title games?
16:46:31 <truebrain> Also uses xz for compression!!!!!
16:46:50 <peter1138> From the comments on the voting discussion you'd think it was rigged...
16:47:46 <truebrain> (Mandatory disclaimer: For those not picking up, I am trolling and not serious)
16:51:58 <frosch123> they are not wrong. what kind of "community" would meet up i brussels? must be fake
16:52:37 <frosch123> raise a distraction balloon by removing block signals, and commit other stuff behind it
16:53:14 <Eddi|zuHause> you mean, like, rip out perfectly functional systems before the replacement system is ready?
16:54:38 <truebrain> And the removal of shares is also suss
16:55:11 * Rubidium wonders how many admin-protocol libraries are (still) broken for 14
17:02:03 <Rubidium> it seems that: not broken, possibly broken but I'm not sure and definitely broken are roughly equally spread
17:05:18 <_glx_> how many use admin port ?
17:05:29 <truebrain> I was thinking if we should replace admin protocol with a http server and json .. would be far easier to use for anyone
17:08:33 <_glx_> "don't open terminal if you don't know what you're doing"
17:11:42 *** keikoz has quit IRC (Ping timeout: 480 seconds)
17:28:29 <andythenorth> imagine typing `brew install zx` instead of `brew install xz`
17:29:35 <truebrain> and one less compromised system!
17:32:09 <peter1138> Imagine playing Doom instead of creating PRs.
17:32:24 <truebrain> 100 hours in a Factorio map .... I say NOTHING
17:37:40 <Eddi|zuHause> dunno, sounds like a low number for such a game :p
17:56:45 <silent_tempest> So while I've been developing on the font picker I've just been using these names:
17:57:09 <silent_tempest> Detail to FS_SMALL
17:57:09 <silent_tempest> Primaty to FS_NORMAL
17:57:09 <silent_tempest> News to FS_LARGE
17:57:31 <silent_tempest> I don't like it.
17:57:40 <andythenorth> peter1138: imagine playing blitz instead of recoding FIRS ports
17:58:19 <peter1138> Normal, small, large, are the names everyone is used to.
17:58:30 <silent_tempest> I think I should use one set of names for everything. And my default would be to remove the detail/primary/large and move to small/medium/large
17:58:55 <silent_tempest> Does anyone hate the small/medium/large scheme?
17:59:20 <peter1138> And normal should be first, as it's the most important.
17:59:35 <silent_tempest> The only advantage to using the defail/primary/news scheme is that it give some hint to those fonts are actually used
18:00:15 <silent_tempest> peter1138: I can't tell is sarcasm or serious...
18:02:35 <locosage> I'd seriously agree with that
18:02:41 <locosage> there is also monospace font btw
18:04:38 <silent_tempest> There is a monospace font. But I haven't convinced myself some one will actually ever want to change it other to than to maybe enlarge it?
18:05:34 <silent_tempest> It only gets used to display the text files in the like the Readme window
18:05:45 <silent_tempest> You would never see it if you just open the game and play
18:05:51 <locosage> GS can also use mono font
18:05:58 <locosage> so if you change normal it's best to change mono too
18:06:46 <peter1138> Not including monospace would be a strange decision.
18:07:47 <peter1138> E.g. I hardly ever see news, so why change the large font?
18:08:02 <silent_tempest> News it part of the game UI.
18:08:13 <silent_tempest> Mono isn't unless a script is using it
18:08:21 <silent_tempest> Not something I was aware was possible
18:09:10 <silent_tempest> I've always found the Mono name to be confusing
18:09:14 <silent_tempest> I know what a mono space font
18:10:04 <silent_tempest> But seems weird to have a monospace font with ever making it obvious where it would be used
18:10:17 <silent_tempest> If the UI had:
18:10:30 <silent_tempest> Mono seems out of place in that list
18:11:09 <locosage> I guess it can be called Monospace Normal
18:11:15 <silent_tempest> I think I should rename it at least in the UI text string
18:11:38 <silent_tempest> "Tertiary Font"?
18:12:22 <frosch123> sans-serif, condensed, serif, mono :p
18:24:20 *** test has quit IRC (Remote host closed the connection)
18:34:57 <Eddi|zuHause> silent_tempest: well, normal settings have a description. there you could explain where each font might commonly be used
18:35:25 <Eddi|zuHause> but i don't see how "tertiary font" is any more descriptive
18:35:43 <silent_tempest> Good idea. They've just been generically Do X or adjust X but yeah they can be way more useful
18:37:30 <silent_tempest> Eddi|zuHause: It's tertiary in the sense it's a medium sized font but used over the medium font in limited special cases
18:38:18 <Eddi|zuHause> dunno, seems like a complete misnomer
18:38:34 <Eddi|zuHause> it's not even the 3rd font
18:38:52 <Eddi|zuHause> at least "mono" is a description of what the font looks like
18:39:02 <Eddi|zuHause> but "tertiary" is a description of nothing
18:39:30 <silent_tempest> it's a monospave by default
18:39:32 <merni> Technically quaternary I guess?
18:40:12 <Eddi|zuHause> silent_tempest: it's the intended shape.
18:40:13 <merni> Maybe "text file font"?
18:40:49 <Eddi|zuHause> it's the only reason for the setting to exist at all
18:41:02 <merni> what's the reason for having it as a monospace by default? Alignment for some kind of ascii tables/columns?
18:41:04 <Eddi|zuHause> could have just used the normal/medium font
18:41:24 <Eddi|zuHause> yes, for readme.txt files that have ascii table formatting
18:42:03 <Eddi|zuHause> (the readme.txt files existed first)
18:42:34 <merni> I don't think there are any such files now, but I guess newgrf readmes could use them
18:43:01 <merni> "any such files" in the base game, I mean
18:44:31 <peter1138> It always was for text files in NewGRFs.
18:45:51 <merni> "text file viewer" or "file viewer" font sounds okay to me
18:46:11 <merni> after all there are some people who even use proportional fonts for code
18:46:22 <peter1138> Also, I think the sliders should just go with something like "smaller" and "larger" rather than numbers that don't mean anything.
18:47:07 <merni> peter1138: Where fonts are involved people probably are used to numbers. Of course they will be scaled by interface scale here
18:47:47 <silent_tempest> peter1138: Interesting thought. I mean they do mean something as I use the actual number. Smaller -> Larger might better for the user though? Not sure. I'll think about it
18:47:50 <silent_tempest> It's a simple change
18:52:10 <_glx_> The number is scaled by GUI zoom
18:54:54 <Rubidium> merni: ever seen the "Help & Manuals" button in the intro menu? That links to our readme, changelog, known bugs and license.
18:56:13 <peter1138> I might have asked this before, can basesets have the same version limitations on Bananas as NewGRFs?
18:57:43 <peter1138> Bah, £73 for one bike tyre :S
18:58:39 <locosage> the only important quality of mono font is it being monospace, calling it anything else would be misleading
18:59:58 <silent_tempest> locosage: If I give the user a button change it that is no longer guranteed to be the case
19:00:20 <peter1138> Same with the font command.
19:00:23 <merni> Rubidium: Well, I've definitely seen it, but I haven't seen any ascii alignment stuff in it
19:00:27 <peter1138> Or the settings file.
19:00:38 <locosage> it was never guaranteed, user is supposed to chose monospace font for, well, monospace font xD
19:00:46 <silent_tempest> peter1138: Seems expensive.
19:00:54 <peter1138> (But also, if you are listing fonts using the appropriate calls it is possible to restrict to monospace, but I'd advise against it)
19:01:46 <silent_tempest> peter1138: I wouldn't bother trying to force them
19:01:54 <locosage> but if you call it "tetriary" or smth how would user know it's supposed to be monospace?
19:02:22 <silent_tempest> Monospace then?
19:02:54 <merni> ladysadieviaGitHub: Lol, sorry silent_tempest :)
19:02:54 <merni> Now that you explained how the `name` confused you, it makes more sense to me
19:03:47 <merni> Not that I would have ever thought of giving a font a nickname... not seen any program that allows that
19:04:00 <silent_tempest> silent_tempest: I like "Monospaced" I'm going to use that
19:04:22 <silent_tempest> merni: No but I have seem programs let you name your settings
19:04:24 <peter1138> merni: Most games don't have a font picker 😉
19:04:46 <merni> peter1138: I said program, not game :)
19:04:59 <peter1138> I know, OpenTTD is more than a game.
19:05:02 <peter1138> It's... a way of life.
19:05:04 <silent_tempest> I mean the easiest example is when games let you name a Keyboard profile
19:11:25 <Eddi|zuHause> silent_tempest: we have a hotkeys.cfg as well, so... :p
19:12:11 <Eddi|zuHause> (but that may be a project for another day)
19:13:13 <silent_tempest> The next thing I want to see is a grf file. One that I bet will have an audience of myself and literally no one else
19:14:11 <Eddi|zuHause> wouldn't be the first grf with that premise :p
19:15:32 <silent_tempest> I want my crazy ass OG graphics and red water back... 😛
19:17:47 <silent_tempest> Though on a related note.... And something that would be a far larger project that I have a passing interest in would be to colorize the game GUI. Like the OG game didn't have any gray windows. They were all sorts of colors.
19:18:50 <peter1138> There's a Toyland to Mars conversion, but I don't know how far it goes.
19:19:24 <silent_tempest> peter1138: I checked out one mars theme but it did not change the water
19:19:35 <silent_tempest> Red dirt and blue water was.... Bad Lol
19:20:18 <silent_tempest> I dunno why but I love those teal vehicle windows...
19:30:14 <peter1138> Yeah, Deluxe switched most of them to boring grey.
19:31:31 <Eddi|zuHause> i don't know if you even can change the water
19:31:47 <Eddi|zuHause> that seems to be something you'd have to do internally
19:32:31 <Eddi|zuHause> i remember in the original mars conversion, you could disable animations, and then have martian water on earth, or earth water on mars
19:32:51 <silent_tempest> If I can't to it in a grf than you'll see PR for it eventually 😛
19:33:09 <Eddi|zuHause> because the water cycle changes would only take effect on the next animation frame
19:41:51 *** salut3585 has joined #openttd
19:41:51 <salut3585> how can I set which version of OpenGRX OpenTTD can load?
19:42:44 <peter1138> Baseset can be selected under Game Options > Graphics
19:43:38 <salut3585> oh, I did not know it I will check
20:00:44 <salut3585> There I only have OpenGFX.
20:00:44 <salut3585> I tried changing openttd.grf and orig_extra.grf but it doesn't make any sense since they don't have many images and OpenTTD checks for OpenGFX every time and loads it if it doesn't. And can't load OpenTTD without OpenGFX
20:00:44 <salut3585> I've made some changes to OpenGFX and OpenTTD and they work well locally, but I need some way to ensure that the new version of OpenTTD can only load the new version of OpenGFX images. And for the old version of OpenTTD, the old version of OpenGFX is used.
20:00:44 <salut3585> These are system changes and I would like to use a base grf file for these images.
20:03:32 <salut3585> Another way is to get the maximum number of index files and place them in that location.
20:06:58 <peter1138> Okay, this is sounding like a serious X/Y problem.
20:12:15 <salut3585> is there an expert who can give advice on how to add new sprites without collisions between different versions?
20:16:21 <_glx_> if you add them via action5 and after existing ones (if reusing an existing action5 type, quite usual for GUI additions) it should still work in older versions
20:16:55 <peter1138> media/baseset/openttd/openttdui.nfo
20:17:23 <peter1138> Increase the 191 // OPENTTD_SPRITE_COUNT
20:17:28 <peter1138> Add your sprite at the end.
20:17:41 <peter1138> Increase OPENTTD_SPRITE_COUNT in src/table/sprites.h
20:17:47 *** Wormnest_ has quit IRC (Ping timeout: 480 seconds)
20:18:15 <silent_tempest> Adding the Monospaced font to the settings UI has made it bigger. I assume that is okay? Anyone object?
20:18:59 *** Wormnest_ has joined #openttd
20:19:01 <_glx_> not showing font sizes under the slider could reduce the used space
20:19:01 <peter1138> You could put the slider next to the Size: label.
20:19:39 <peter1138> And the slider labels should be in the small font
20:20:51 <_zephyris> I do worry that it's an extra tab and a big screen, when 99% of users will never touch these settings...
20:20:51 <silent_tempest> peter1138: Huh the rest of them are like. I didn't notice. I can change it.
20:21:34 <silent_tempest> Until I added the monospace at the bottom it didn't make the options menu noticably bigger
20:21:44 <peter1138> Also, only the interface scale slider is full width, the music/sound volume sliders are next to their labels, so use them as a guide 🙂
20:22:32 <silent_tempest> I had done that initially but thought it made the labels looke squished so I moved them up at some point
20:22:46 <silent_tempest> There's been so many minor changes to this design as I've worked on it
20:23:17 <silent_tempest> On a positive note I'm coming quite proficent how the NWidger code works and how to use it to get what I want. Lol
20:23:22 <_glx_> (that's why most of us use a large stick)
20:23:47 <peter1138> Well at least I understand it as well 😉
20:25:10 <silent_tempest> _glx_: As long as a designer doesn't summon me to a meeting to talk about what we could do to change the dialog layout so he can get his string to have the perfect alignment on this dialog this is a step up. Lol 🙂
20:25:30 <peter1138> There's always andy suggesting css but we ignore that.
20:26:37 <_glx_> I still remember how the GUI was before using objects
20:26:53 <peter1138> Fixed pixel positions.
20:27:17 <_glx_> I think the GUI rewrite killed miniIN
20:27:37 <_glx_> or maybe it was the c->cpp
20:28:29 <_glx_> hmm no, I remember doing some syncs related to GUI, so post cpp
20:29:16 <_glx_> but any huge rewrite was a patch killer
20:38:08 <silent_tempest> Not sure why moving the labels inline with the slider made the slider taller... But it is a welcome change
20:40:24 <silent_tempest> Also without the font tab and with:
20:40:37 <silent_tempest> Full window screen shot for reference.
20:42:31 <peter1138> As long as it fits in 640x480 at 1x 😉
20:45:16 <silent_tempest> peter1138: Issue -> Closed Reason: Cannot Reproduce. 😛
20:51:58 <peter1138> I think it was suggested 😄
20:51:58 <silent_tempest> how was that fixed?
20:52:40 <peter1138> That whole window is gone.
20:52:50 <peter1138> It's the a scrollable tree now.
20:53:10 <silent_tempest> I see though it appears broken?
20:53:51 <peter1138> You need to select an option.
20:54:02 <silent_tempest> Me doing dumb things ?
20:54:21 <silent_tempest> No no the panel at the bottom
20:54:28 <silent_tempest> It shouldn't be there right
20:54:40 <peter1138> Yes, but probably not that big.
20:54:42 <frosch123> the panel shows a description, if you select a setting
20:54:59 <frosch123> it's sized to fit the larget description
20:55:10 <frosch123> so it never resizes when you select something else
20:55:23 <peter1138> Do we have a very long description added somewhere... 😄
20:56:39 <peter1138> Although as you are using a non-standard taller font, that will affect things too.
20:56:50 <frosch123> wallclock text is rather long
20:58:53 <silent_tempest> Hmn I see like that on every build on my system
20:59:00 <frosch123> but i prefer it using blank lines, than the wall-of-text in some cdist settings
20:59:12 <silent_tempest> Not something I did directly. It could be the custom font though
20:59:25 <peter1138> We could limit the number of lines and add... another scrollbar.
21:00:40 <peter1138> Although with 2x interface scale at 1920x1080 and the proper fonts, it looks fine.
21:00:48 <frosch123> hmm the panel is also too large here
21:00:58 <frosch123> it was at least 4 lines more than needed
21:03:51 <frosch123> ah, it assumed the smallest width
21:04:21 <frosch123> the panel does not change height, when resizing the window horizontally
21:04:45 <frosch123> it's height is choosen, so that walltext help fits with minimal width
21:05:52 <peter1138> silent_tempest: Easy 😄
21:58:26 *** Flygon has quit IRC (Remote host closed the connection)
22:19:11 *** keikoz has quit IRC (Ping timeout: 480 seconds)
22:41:11 *** Wormnest_ has quit IRC (Ping timeout: 480 seconds)
22:44:14 *** argoneus has joined #openttd
22:44:51 *** argoneus has joined #openttd
22:58:41 <silent_tempest> Thanks for the tip
22:58:54 <silent_tempest> A lot of this is right. And a lot of it is wrong lol
23:16:09 *** Wormnest_ has joined #openttd
23:27:07 *** ChanServ sets mode: +v tokai
23:32:30 *** bryjen has quit IRC (Quit: Leaving)
23:33:52 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
23:38:11 *** gelignite has joined #openttd
23:52:23 <truebrain> pfff, images are for people that are getting odly old 😛
23:52:43 <truebrain> most likely the Cloudflare Worker is not forwarding the quest properly; just spotted something else odd too .. something to dive into tomorrow 🙂
continue to next day ⏵
