IRC logs for #openttd on OFTC at 2024-04-28
⏴ go to previous day
00:02:36 <kamnet> yup: Hey I worked on doing an update to the Alaska USA scenario. All of the files should be available to download. Would you like to try it out?
00:11:41 <peter1138> Move all content that isn't in content_download away, then you can see what's missing easier.
00:12:00 <peter1138> That was for kamnet 🙂
00:17:21 <kamnet> I just picked the wrong file. Meh
00:26:25 <kamnet> reldred: Could you upload North American City Set so that it can replace Canadian Cities in scenario files I'm fixing?
00:38:03 *** Wormnest has joined #openttd
00:41:28 <reldred> It’s not called North American city set. It’s North American building set, it’s always been NABS, and it’s on bananas already.
00:42:02 <reldred> North American City Set is Oz’s version before it got@merged into Canadian cities.
00:42:17 <reldred> He added extra buildings that he drew, I can’t include them.
00:49:26 <peter1138> Looking a bit better in the rebase...
00:49:43 <peter1138> Yes, it took me this long to iron it out... and fix other things I noticed at the same time.
00:51:24 <peter1138> Oh, that's unify code, not unify into one window, heh.
00:56:33 <kamnet> reldred: Sorry, meant buildings... and like I thought I just added the wrong version that wasn't the latest. sorry to have bothered you
00:57:01 <reldred> re_nabs is the one you’re after. ae_nabs is the ancient version
00:57:23 <reldred> It only had small tweaks and bug fixes. Nothing earth shattering.
01:32:13 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
01:34:55 *** gelignite is now known as Guest2441
01:34:58 *** gelignite has joined #openttd
01:42:17 *** Guest2441 has quit IRC (Ping timeout: 480 seconds)
02:04:37 *** Wormnest has joined #openttd
02:15:56 *** gelignite has quit IRC (Quit: Stay safe!)
02:24:38 *** gnu_jj_ has joined #openttd
02:26:17 *** Flygon has quit IRC (Read error: Connection reset by peer)
02:27:48 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
02:33:50 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:08:06 *** Tirili has quit IRC (Remote host closed the connection)
04:41:17 <DorpsGek> - Update: Translations from eints (by translators)
07:02:12 <andythenorth> such international trading port
07:02:33 <andythenorth> much long distance trading by ship
07:13:39 <truebrain> dunno if ShipPF can handle 5 tiles 😛
07:15:18 <andythenorth> one day I'll write a circular tile search to prevent that placement ^
07:17:23 <truebrain> and is that day today?
07:21:13 <andythenorth> tracking an array of contiguous tiles in a grf callback
07:21:22 <andythenorth> is not a thing I wish to approach today 😛
07:24:01 <locosage> what's the array for? you only need like 1 or 2 vars for circular search
07:24:21 <truebrain> so what is on the agenda today andythenorth ?
07:24:48 <andythenorth> but also sprites
07:25:20 <andythenorth> this has to be transposed for 3 other coast directions
07:26:01 <andythenorth> AndythenorthIntelligence, yes
07:26:41 <andythenorth> I wrote a layout rotating algorithm, but rotating sprites requires drawing them
07:26:57 <truebrain> or you could make a 3D model of the sprite, and use that 😛 😛 😛
07:27:16 <andythenorth> I did test a generative game sprite maker
07:27:36 <andythenorth> couldn't be bothered to put in the training time needed
07:32:05 <truebrain> ghehe, I can imagine 😄
07:57:34 <truebrain> my idea was nice, the end-result not so much 😄
08:07:05 <truebrain> there, more visible 🙂
08:09:27 <LordAro> truebrain: will this be WT5?
08:09:38 <truebrain> or WT6? WT7? Where are we at?
08:10:10 <LordAro> nono, it has to be WT<N>
08:10:20 <LordAro> or new new new new new WT
08:11:13 <truebrain> frosch already has spoken
08:22:57 <Rubidium> though... if you're going to do a massive rewrite afterwards, maybe call it second edition nile is less expensive? ;)
08:23:22 <truebrain> someone is awake this morning 😛
08:26:36 <truebrain> or you needed 12 minutes to come up with that .. I can't tell 😛
08:45:50 <Rubidium> truebrain: lets just say that I'm not watching the channel like a hawk, just sporadically check whether something interesting happened
08:46:11 <truebrain> so you are just awake on this fine Sunday morning 😄
08:52:06 <truebrain> couldn't resist going with "main stream"
09:38:46 *** gelignite has joined #openttd
09:39:51 <michi_cc> We have a PR department now? 😁
09:59:21 <peter1138> Hmm, I need a `SZSP_VERTICAL` or `SZSP_HORZIONTAL` depending on whether the stack is in a horizontal or vertical container.
10:12:38 <truebrain> I just spend 40-ish minutes building something that is very niche and not actually all that useful 😛 But it was fun!
10:22:38 <peter1138> Sounds like my whole career.
10:33:16 *** ahyangyi has joined #openttd
10:33:17 <ahyangyi> A sunday that's also a workday for me -- I wish it doesn't sound like my whole career 😛
10:36:00 *** Tirili has quit IRC (Remote host closed the connection)
10:44:56 <peter1138> Presumably you are on Linux.
10:46:12 <peter1138> For MIDI on Linux we can Fluidsynth as long as its installed.
10:47:29 <peter1138> Or if you have timidity installed then that can be used.
10:47:44 <peter1138> Our Linux MIDI driver is not very good.
10:48:11 <peter1138> It's either fluidsynth, or pass off to an external command (which defaults to timidity) and if that does exist it just does what you are seeing.
10:49:01 <peter1138> It's surprising really because I'm a Linux user and I'm also into my MIDI, so I should've got around to making it better.
11:04:59 <peter1138> The prebuilt OpenTTD should support fluidsynth, so installing that (& its data) should work.
11:16:40 *** telk5093 has joined #openttd
11:16:40 <telk5093> nile? does it come from the Nile river in Egypt?
11:18:35 <peter1138> Bah, I used a std::set, but you can't update items in a set 😮
11:24:42 <ahyangyi> You can remove an item then add the new version though
11:28:37 <peter1138> But doing that for all items...
11:29:35 <ahyangyi> I guess making a new std::set would be faster
11:40:49 <peter1138> Well, at least I templated this so only need to write it once.
11:42:28 <peter1138> And in theory I only need to do this if NewGRFs are changed (either from starting/loading a game, or in game)
11:43:03 <peter1138> Yay, no longer crashes 😄
11:49:50 *** asasnat has quit IRC (Quit: User went offline on Discord a while ago)
12:12:48 <peter1138> Okay, happy with this refactor (of a branch that isn't even a PR let alone merged...)
12:13:13 <peter1138> It would now be possible to add a grfid filter without having to rewrite everything.
12:14:55 <peter1138> Now to catch up with the other stashes 😄
12:24:50 *** jenkings has quit IRC (Quit: User went offline on Discord a while ago)
12:41:58 <peter1138> I might've broken this...
12:46:24 <peter1138> Correction, I just haven't written it yet 😄
12:51:25 <peter1138> Hmm, I guess I need to iterate all specs to find grfids.
12:51:44 <peter1138> Wait. I was going to leave this as "possible to do" rather than "done"
12:52:28 <jfs> peter1138: std::map though? if the lookup key is constant at least
12:53:18 <peter1138> Given what it needs to do, and how often it needs to happen (rarely) I went with iterating the old set and creating a new set.
13:30:00 <truebrain> frosch123: after a bunch of infra, finally: https://nile-preview.openttd.org/ 🙂 It is far from done, ofc. The left button doesn't work, the right button does. All pages work, except the actual string editor page. That is next on the list 🙂
13:30:21 <truebrain> but the navigation and stuff should be done now
13:38:27 <peter1138> "Get started" does not.
13:39:01 <peter1138> Am I looking at it wrong?
13:41:49 <peter1138> Also... "from the comfort of your own browser"
13:42:09 <peter1138> Updating translations in-game would be neat 😄
13:42:59 <truebrain> peter1138: ` The left button doesn't work, the right button does.` <- I did mention that, yes 🙂
13:43:18 <truebrain> peter1138: what browser is that? As the browser is doing it wrong 🙂
13:43:20 <peter1138> Okay, I was thinking left MOUSE button, lol
13:43:51 <truebrain> peter1138: and the fact that the left button didn't work, didn't change that perception? 😄
13:44:15 <peter1138> Well the button is labelled "Get Started" not "Left"
13:45:28 <peter1138> Clicking on a string is a blank page (just the header)
13:45:52 <truebrain> peter1138: `All pages work, except the actual string editor page`
13:46:00 <truebrain> reading is not your finest skill this Sunday, is it? 🙂
13:46:09 <truebrain> or you never used eints
13:46:11 <truebrain> that is also possible 😛
13:46:15 <peter1138> I never used eints.
13:46:33 <peter1138> Only logged in one time to check once after an account was made.
13:47:05 <peter1138> But I clicked on a string. I am not clairvoyant enough to know that is the string editor page.
13:47:13 <truebrain> peter1138: ah, your window is too small. This component library should make that pretty, but it seemly doesn't
13:47:15 <peter1138> Though fair, it is likely.
13:47:50 <truebrain> the browser window too? 🙂
13:48:07 <peter1138> What other window would I mean?
13:48:26 <truebrain> it needs 1584px atm before rendering correctly 🙂
13:48:28 <truebrain> but that is fixable 😉
13:55:33 <peter1138> Interesting serialisation of data.
13:56:06 <peter1138> More optimised, but also looks magic.
13:56:24 <peter1138> Hmm, some json in it though.
13:59:53 <peter1138> Responses from the API.
14:00:09 <peter1138> I was looking at developer view (f12)
14:00:25 <truebrain> there are 2 APIs here, one for the webapp itself, and one for the actual data
14:00:37 <truebrain> the first is NextJS, which is mainly Javascript
14:02:46 <LordAro> truebrain: "push-xosmuplqvwmm"
14:04:03 <truebrain> luckily for you, they disappear as soon as things are merged 🙂
14:05:37 <truebrain> peter1138: fixed! I misunderstood the documentation 😄
14:06:45 <truebrain> I see I also need to fix mobile design a bit on a few places, but that is not for today 🙂
14:19:11 <frosch123> truebrain: did chatgpt write the frontpage text? :p
14:19:49 <truebrain> frosch123: No 😦 I wrote that 😢
14:20:05 <frosch123> when did you learn to write like that?
14:21:42 <truebrain> frosch123: not sure that is a stab or not 😛
14:21:51 <frosch123> it looks cool though 🙂
14:22:24 <truebrain> tnx (I think .. really unsure at this point 😛 )
14:25:35 <frosch123> maybe it needs to be that project dropdown
14:25:54 <truebrain> no, it is the current page
14:26:00 <truebrain> I guess it is mostly weird as it is not at the end
14:26:13 <truebrain> so either it shouldn't be a link, or I need to break consistency between pages
14:26:51 <truebrain> by clicking the project again?
14:27:06 <truebrain> I am not sure what you mean with "I can't get back"
14:27:17 <frosch123> hmm, true, that is also how the all-languages works
14:27:37 <truebrain> I guess I could keep the dropdown
14:29:45 <truebrain> I btw kept the eints links, for when people have them in their bookmarks; but the links are weird 😛
14:38:48 <peter1138> My player's been appreciating the better ship pathfinding.
14:52:33 <truebrain> frosch123: deployed a change to the navigation; I think this makes more sense now. The only one that is not weird, is if both go to All. As depending on which is first, you go to another page (Project vs Language). Not sure that can be solved. Not sure both are needed. But I copied eints for now 😄 (maybe 1 page showing all projects and their languages is better, than having one showing all
14:52:33 <truebrain> projects, and one showing all languages)
14:55:38 *** Wormnest has joined #openttd
14:57:06 <frosch123> not sure whether the all-projects-all-languages page has a use-case
14:57:19 <frosch123> the one-project-all-languages show the translation progress, which is useful
14:57:46 <frosch123> the all-projects-one-language is for translators, and probably should also show the progress?
14:57:59 <truebrain> yeah, I do have that planned
14:58:02 <truebrain> just not implemented 🙂
14:58:09 <truebrain> but the all/all is just a bit weird
14:58:21 <yup> And you somehow push new versions into Arch AUR. Or is it other people's business?
14:58:45 <frosch123> i am unsure what the progress bars show
14:58:51 <frosch123> they look not-to-scale
14:58:59 <truebrain> they aren't! Hover over for actual information
14:59:01 <frosch123> but i guess that is a good thing
14:59:05 <truebrain> but showing a 1px wide something is not useful 😉
14:59:10 <frosch123> otherwise you do not see, if 1 is missing
15:02:01 <peter1138> yup: Well that's up to whoever maintains Arch.
15:03:20 <truebrain> frosch123: next step is to think about how to show editing of strings .. I get why eints does what it does, but I also think it doesn't work. So I am thinking of working from the `/translation/...` pages, but not sure how yet 🙂
15:03:38 <truebrain> (navigation-wise, that is)
15:04:09 <truebrain> currently in eints it is really hard to get back to any overview when you are editing a string
15:04:21 <frosch123> editing translations is all nice, until you get 7 cases
15:05:13 <frosch123> eints uses a static page, and repeats the base language and the weird-but-not-history for every string, so the page becomes huge
15:05:28 <truebrain> I am not even talking about the editing itself, just the navigation 😄
15:06:10 <frosch123> when i used eints, i mostly searched the overview page for some word, and the opened all the strings in new tabs
15:06:50 <frosch123> last time it was when i touched all strings related to the years/periods thingie
15:07:28 <truebrain> lolz @ cases (I had to assign myself to a language that had cases 😛 )
15:07:54 <truebrain> anyway, the flow in eints is just a bit weird, to first show all missing and invalid strings, but once you click any, you lose that overview totally
15:08:18 <truebrain> so yeah .. your solution works for that, but is silly that it is needed 🙂
15:08:24 <truebrain> need to toy around a bit to see what works
15:08:56 <frosch123> can we have a two column layout with a filterable list of all strings on the left, and selecting one allows editing it on the right
15:09:14 <frosch123> so i can search for "period" on the left, see all strings with that word, and then click through them and change them
15:09:15 <truebrain> the translation page already started that a bit, with a ToC on th eleft
15:10:11 <frosch123> eints tried to figure out related string automatically, i am not sure whether that was ever useful, or whether manual filtering is just better
15:10:20 <frosch123> probably an AI should do it :p
15:10:34 <frosch123> "people who translated this string, also translated this one"
15:10:42 <_glx_> when using eints I just have many tabs
15:11:00 <truebrain> I wonder how useful the categorizing is with "Outdated", "Missing", etc
15:11:38 <_glx_> outdated means the translation might be still fine, but needs confirmation
15:11:48 <truebrain> I know what it means; that was not the question 🙂
15:12:11 <truebrain> on that page strings are categorized based on it
15:12:12 <_glx_> urgent vs less urgent 🙂
15:12:12 <frosch123> missing is a lot more effort to fix
15:12:15 <truebrain> I wonder if anyone actually cares
15:12:46 <truebrain> as on eints too, once you start editing, you don't see this at all anymore
15:12:50 <frosch123> outdated is also a different workflow, so i would hate alternating between missing and outdated strings, it's a context-switch for me
15:12:51 <truebrain> (which creates the need to open more tabs, I guess)
15:13:21 <frosch123> otoh, i think errors and warnings can be the same category: strings with issues
15:13:28 <truebrain> so we kinda need 3 columns 🙂 1 for "what to do", second for doing, third for context
15:13:43 <truebrain> I split up the "invalid" category, as here too: errors are more important to fix than warnings
15:13:58 <_glx_> often I have to open other strings in tabs for the context
15:14:19 <_glx_> because the "similar" strings are not always useful
15:15:01 <truebrain> yeah, and searching through the categorized strings is also a bit meh; you want to search through everything, and see what to do next too
15:15:09 <truebrain> I have some ideas, but they require testing 😄
15:15:20 <truebrain> there is a lot of horizontal space these days, making it a bit easier
15:15:42 <frosch123> let's hope we have no translators on mobile :p
15:15:51 <truebrain> I do plan to make it mobile friendly 🙂
15:15:56 <truebrain> just gathering context is impractical
15:16:20 <truebrain> just also no real need to break on mobile
15:17:26 <truebrain> well, dinner first! 😄
15:31:13 *** yup has quit IRC (Quit: Page closed)
15:56:42 <peter1138> Hmm, perhaps the "All" button should be part of the class list.
16:20:03 <peter1138> Nice, ObjectSpec::GetByGrf() does not work because there are specs with grffile as nullptr 🙂
16:21:10 <_glx_> lighthouse, transmitter, headquarter ?
16:28:33 <peter1138> Well, fixed it for now.
16:29:22 <peter1138> Not quite what I want 🙂
16:38:39 <peter1138> StringID widest_title...
16:38:53 <peter1138> That was the solution I did, but clearly never got around to PRing it.
16:39:37 <peter1138> And for the object my current PR wipes that out..
16:47:06 <Rubidium> peter1138: what number does that current PR have?
16:47:22 <peter1138> Oops, current branch I meant 🙂
17:06:47 *** HerzogDeXtEr has joined #openttd
17:32:07 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
17:40:08 *** Wormnest has joined #openttd
18:15:10 <peter1138> The existing auto-sizing feature of the object picker leads to some ... awkward window layouts.
18:16:18 <peter1138> So I have removed it 🙂
18:16:32 <andythenorth> The clipped version looks actually designed
18:16:49 <peter1138> The only "grace" is that the box for a 1-view object takes up the same as as a 4-view object.
18:28:22 <peter1138> Although I'm contemplating doing what the road stop view does.
18:52:44 *** petike873 has joined #openttd
18:52:44 <petike873> 18+ Free Content and onlyfans leaks :peach: :underage:
18:52:44 <petike873> discord.gg/nudesporn
18:53:52 <merni> Discord Moderator: spam ^
19:00:47 <peter1138> > 34 files changed, 1683 insertions(+), 1675 deletions(-)
20:07:05 *** gelignite has quit IRC (Quit: Stay safe!)
20:39:08 *** nielsm has quit IRC (Ping timeout: 480 seconds)
21:09:15 *** kale91 has quit IRC (Quit: User went offline on Discord a while ago)
21:33:06 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:53:24 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
22:03:46 <peter1138> If a window has multiple filter boxes, what should a hotkey layout setup like? 😄
22:16:51 <peter1138> Only other window that has two editboxes is the network server list window, with no hotkeys.
22:21:27 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:40:18 *** Wormnest has joined #openttd
23:32:32 *** Wormnest has quit IRC (Ping timeout: 480 seconds)
23:42:14 *** tokai|noir has joined #openttd
23:42:14 *** ChanServ sets mode: +v tokai|noir
23:48:58 *** tokai has quit IRC (Ping timeout: 480 seconds)
continue to next day ⏵
