IRC logs for #openttd on OFTC at 2024-09-23
⏴ go to previous day
02:06:07 *** Wormnest has quit IRC (Quit: Leaving)
02:14:59 *** gnu_jj_ has joined #openttd
02:18:29 *** gnu_jj has quit IRC (Ping timeout: 480 seconds)
02:34:58 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:47:34 *** D-HUND is now known as debdog
04:45:40 <DorpsGek> - Update: Translations from eints (by translators)
05:01:04 *** keikoz has quit IRC (Ping timeout: 480 seconds)
05:40:59 <andythenorth> but do vehicle name boxes support regexp?
05:44:24 <_pruple> jgrpp feature, clearly
05:45:40 <andythenorth> change for sake of change imho
05:45:56 <andythenorth> why isn't there a no-code spreadsheet in OpenTTD?
05:46:05 <andythenorth> so I can use it as a database?
05:46:35 <_pruple> you can probably create one using presignals
05:47:07 <_pruple> they must be useful for something, or they wouldn't be hidden behind a super-pro-advanced secret button
05:52:37 <andythenorth> conditional order to rename a train?
05:53:04 <andythenorth> imagine if there was a scripting language, with buttons and stuff
05:53:18 <andythenorth> then I could alias my trains by writing a script
05:53:25 <andythenorth> impossible dreams
05:55:03 <_pruple> someone* should write a better AI
05:56:21 <andythenorth> but should I weather these?
05:56:24 <andythenorth> they seem a bit clean
07:14:56 <DorpsGek> - Add: summary for week 38 of 2024 (by OpenTTD Survey)
08:16:49 <peter1138> 32bpp dirt overlays?
08:17:29 <peter1138> So this morning I woke up and my knee is back to hurting again... wtf.
08:19:20 <andythenorth> I've snagged my neck
08:19:24 <andythenorth> old man injuries
08:54:51 <reldred> peter1138: knees are horrible. Mine have started aching significantly more than usual. Plus my elbows. And my ankles.
08:54:59 <reldred> old people injuries indeed.
08:57:04 <peter1138> Annoying, in my mind, all of us still involved in this game are in our 20s still...
08:59:52 <peter1138> Also Belugas also does a bit of cycling these days
09:09:58 <peter1138> Trying to think how to unit test #12914.
09:29:07 <peter1138> Oh dear, a database transaction issue :/
09:29:24 <peter1138> > 40001: could not serialize access due to read/write dependencies among transactions
09:52:37 <mnhebi> reldred: don't need to be old. Just have joint psoriasis or Ehlers-Danlos...
09:52:58 <mnhebi> I'm only 35 and I'm almost retirement material with all the pain I have...
09:55:21 <reldred> Doc reckons I have rheumatoid arthritis, so, yeah
09:57:42 <mnhebi> Sometimes im fine walking, sometimes just walking 500 meters my legs feel like they are ready to be cut off..
10:37:25 <peter1138> Oh, forgot LogDebug() will show in production...
10:38:12 <peter1138> Only because production is running debug of course, but then without that error reporting is not so helpful.
10:42:13 <peter1138> It can be used for more than just enum classes, works with regular enum too, as long as they are tagged with the bitset stuff.
10:42:44 <peter1138> There are a few places where the test is not as simple as testing one bit 🙂
10:43:07 <xarick> I'm looking into whether I can use HasFlag for my AirportPlaneCompatibility PR
10:43:10 <peter1138> But I think this also makes switching more to enum class easier.
10:49:06 <peter1138> If y only has one bit set, then it's the same.
10:49:38 <peter1138> Reason I didn't do != 0 is because with 0 is not compatible with enum classes.
10:49:59 <peter1138> Although `to_underlying(x & y) != 0` would've works.
10:50:31 <peter1138> And `(to_underlying(x) & to_underlying(y) )!= 0` would work without the bitset stuff too.
10:50:56 <peter1138> (But then you can't do other useful operations)
11:07:41 <xarick> actually, it's compatible
11:24:59 <peter1138> Yeah, AirportFTAClass::ALL would not make sense with it, but the others do.
11:25:38 <andythenorth> peter1138: 8bpp dirt recolours? 😛
11:26:51 <peter1138> Hmm, any idea what DepotCommand::LocateHangar is meant for...
11:27:11 <peter1138> Or... rather why it exists. Because it doesn't seem to be tested for.
11:27:15 <peter1138> Unless I broke that 😄
12:06:09 *** geertop has joined #openttd
12:06:09 <geertop> Ok how do you draw a locomotive?
12:06:23 <geertop> Like is there’s any rules to follow or something?
13:15:20 <xarick> const uint8_t GetNumTerminals() const
13:15:31 <xarick> what do dual const do here? I'm too noob
13:15:58 <LordAro> xarick: google: "const after function signature c++"
13:20:57 *** kuhnovic has joined #openttd
13:20:57 <kuhnovic> You can even get triple const when dealing with pointers: `const double* const doSomething() const.` . Just to make things extra confusing 😛
13:23:17 <LordAro> const per pointer, please
13:23:25 <LordAro> const double * const * const * const *
13:24:32 <peter1138> Passing pointers everywhere sure looks like C instead of C++ 😉
13:24:54 <LordAro> imagine being mutable by default
13:26:29 <peter1138> Imagine if it was untyped.
13:27:00 <LordAro> the amount of effort being put into making python typed is staggering
13:27:23 <xarick> well, it compiles, so it's probably ok
13:38:43 <kuhnovic> xarick: I'm biting my tongue so hard right now haha
13:42:45 <xarick> it does not like the first const, fine
13:43:28 <xarick> type qualifiers ignored on function return type
13:43:46 <xarick> 'const' type qualifier on return type has no effect
13:45:24 <xarick> I don't understand this type of english
13:48:05 <kuhnovic> I'm assuming you are looking at the Eror List in visual studio. Those one-liners generally explain part of the problem. You'll find more information in the Output window. Make sure you read carefully, the compiler might spit out more than you bargained for.
13:48:32 <peter1138> I means that 'const' does not make sense for the return type in that case.
13:49:09 <peter1138> (It only makes sense for pointer or reference types)
13:55:18 <kuhnovic> It doesn't make sense but you can still do it
13:55:40 <kuhnovic> But then we have warnings-as-errors, I guess that what Xarick is running into
13:56:48 <peter1138> Not seen this before.
13:56:48 <kuhnovic> github-advanced-secu: There's your answer xarick
13:57:36 <truebrain> sorry, I had to press a button in the GitHub interface
13:57:39 <truebrain> this might go terribly
13:57:51 <truebrain> owh, it doesn't auto-create a PR, so there is that
13:58:20 <truebrain> GitHub can now auto-fix some CodeQL issues 😛
14:03:35 <kuhnovic> Oh god, what could possible go wrong 😛
14:03:35 <xarick> if GetNumTerminals can't be in struct AirportFTAClass
14:05:32 <peter1138> Forgive me for hating it :p
14:05:56 <peter1138> It should go in a source file.
14:06:32 <_glx_> `const uint8_t ...` doesn't make sense 🙂
14:08:30 <_glx_> learn to read error messages
14:09:22 <kuhnovic> Well, CodeQL will fix that for him now 😛
14:10:04 <xarick> i removed the first const
14:12:10 <truebrain> kuhnovic: owh my, please no, let's not go that way! 😛
14:12:41 <kuhnovic> Hey I didn't enable this feature 😛
14:13:21 <_glx_> the syntax for the example is not confusing at all 🙂
14:13:25 <truebrain> owh, Discord has a new capability, "forwarding" of messages. At least, I never saw that. How is the bridge handling that ..... must .... test ......
14:14:11 <truebrain> I really wouldn't know how to convert that forward to IRC lingo ..
14:14:20 <truebrain> (if anyone does have ideas, I am all ears)
14:15:05 <peter1138> _glx_: Except we don't use trailing return types much... yet.
14:15:17 <peter1138> (There's a couple of lambdas that do)
14:49:30 <xarick> * Get the number of terminals of a finite state machine of an airport.
14:52:15 <xarick> I'm removing GetNumTerminals from the struct AirportFTAClass
14:52:35 <xarick> putting it somewhere accessible by the callers
15:07:48 <xarick> is it still considered a refactor if i move a function from a file to another and have it a declaration?
15:08:25 <LordAro> what do you think a refactor is?
15:12:02 <xarick> making part of code from a big function into a smaller function then have the big function call the smaller function to avoid repetition
15:12:23 <LordAro> that's certainly one form of it, yes
15:28:17 <peter1138> Advanced Hidden Secret Setting: Random Vs Predictable town/city ratios.
15:30:40 *** talltyler has joined #openttd
15:30:40 <talltyler> Ctrl+Click something
15:31:25 <talltyler> Meanwhile, I have a PR to remove a setting 😉
15:31:43 <peter1138> `Change: City generation isn't random enough, make it a 1-in-X probability`
15:32:07 <talltyler> There are actually a lot of 100% or 99.9% unchanged settings in the survey that could be eliminated to reduce code complexity:)
15:33:07 <peter1138> Designed a schema for a new feature. Additional request "can we have it per account?"
15:33:07 <LordAro> left ctrl vs right ctrl
15:33:23 <peter1138> Left Ctrl AND right Ctrl.
15:34:45 <peter1138> I think it's time to start revoking repository permissions.
15:35:14 <peter1138> I'm getting out of hand.
15:39:45 *** Wormnest has joined #openttd
15:52:34 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
15:58:20 <xarick> wondering if j0anjosep is working on terminal groups, making it possible to order aircraft into the group the user specifies
16:05:19 *** XYZ has quit IRC (Ping timeout: 480 seconds)
16:09:34 <xarick> airportspec has no member nof_depots... whaat?
16:17:56 *** j0anjosep has joined #openttd
16:17:56 <j0anjosep> xarick: No, but it is possible to do so. It is not a priority for me.
16:26:32 <xarick> I imagine it's gonna take a lot of work, needs to know where the user clicks on the airport, needs to know where the aircraft is heading to, if it's busy to switch to the other group... hmm
16:35:57 *** HerzogDeXtEr has joined #openttd
16:49:02 <soylent_cow[m]> is there no way to see user passwords? i would like to restore them for players after server reboots, even if i have to manually type them up in server console
16:51:15 <peter1138> No. And in nightlies there are no passwords any more, replaced with a key-based system.)
16:56:29 *** gelignite has joined #openttd
17:20:47 <peter1138> I'll guess that nobody ever used that flag.
17:22:39 <_pruple> not many animated houses out there
17:23:03 <peter1138> Well, they're using animation but not in the intended way.
17:23:23 <peter1138> Gotta have the house pick itself up and rotate to match a road...
17:23:40 <_pruple> yes, I'm using that trick for my new houses, since people complained about them rotating 😛
17:25:30 <_pruple> I guess it probably makes them more performant, too, since it's not constantly checking the roads.
17:30:00 <peter1138> At long as the animation stops, yeah.
17:32:22 <_pruple> yep, it basically just uses the animation frame as a permanent storage for the orientation
17:33:48 <peter1138> Zephyris 'Full English' set only has 17 building types, yet...
17:34:34 <peter1138> Heavily randomised and automatic, which makes manual placement with the picker a bit, well, random.
17:47:55 <xarick> oh wow 10411 conflicts with 7 files
17:48:17 <LordAro> i thought that was the number of conflicts for a moment
17:50:10 <xarick> closest ship depot conflicts with the non existing npf
17:56:43 <andythenorth> npf is gone, no?
18:00:22 <LordAro> andythenorth: it's mildly backwards, but an old npf-touching PR will conflict
18:02:11 <_jgr_> As merge conflicts go it's pretty easy to resolve though
18:11:18 *** tokai|noir has joined #openttd
18:11:18 *** ChanServ sets mode: +v tokai|noir
18:18:14 *** tokai has quit IRC (Ping timeout: 480 seconds)
18:46:32 <peter1138> Loading FIRS in a debug build...
18:47:26 <peter1138> Loading Iron Horse in a debug build...
18:49:10 <_jgr_> Sounds like you'd be better off with RelWithDebInfo
18:53:36 <peter1138> Not sure about this. Can't really do columns.
18:55:09 <peter1138> (Not that country flags are useful here, just using them as an example of any sort of badge)
18:56:50 <andythenorth> peter1138: compiling Iron Horse on this old CPU 😦
18:56:59 <andythenorth> it's back to 1 minute 😦
19:51:06 <andythenorth> hmm new CPU time?
19:56:50 <xarick> 6 of them were because of "victims"
19:59:43 *** gelignite has quit IRC (Quit: Stay safe!)
20:09:05 *** abboh_132 has joined #openttd
20:09:05 <abboh_132> Is a dot "." missing here at the end or is it intended?
20:12:00 <_glx_> I think most tooltips don't end with a dot
20:13:16 <abboh_132> I see, so I won't add it. Thank you!
20:13:18 <peter1138> Yes, it's intended.
20:20:17 <peter1138> Ooh, touched a nerve.
20:39:14 *** nielsm has quit IRC (Ping timeout: 480 seconds)
20:42:36 <xarick> the remote is different than the local
20:47:00 <peter1138> If you knew what you'd written you'd be able to redo it 🙂
20:49:40 <talltyler> peter1138: “I was rude and somebody didn’t appreciate it” 🤦
21:00:03 <talltyler> (Not criticizing you, but the other person)
21:07:05 <xarick> need to refresh my memory
21:08:03 <xarick> all my changes are before CheckShipReverse unified behaviour
21:14:27 <xarick> okay, I kinda remember something
21:15:00 <xarick> trying to solve a edgy case
21:15:36 <xarick> return UpdateOrderDest(v, order, may_reverse) && may_reverse;
21:16:07 <xarick> FindClosestDepot(may_reverse)
21:17:13 <xarick> when departing from a dock while the next order is find closest depot, it is allowed to reverse, so the pathfinder needs to know whether it can pf bothways
21:26:15 <xarick> bed, cyas goodnight. I'll look at this tomorrow
21:36:29 *** keikoz has quit IRC (Ping timeout: 480 seconds)
21:51:00 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
21:53:07 <peter1138> Such patriotism... whoops
21:56:24 <andythenorth> is there a zzzz badge?
21:56:46 <peter1138> Turns out upper_bound is not lower_bound 😄
21:56:56 <peter1138> There can be. You just need to define it.
21:58:19 <andythenorth> think there's a savegame cursor we could use 😛
21:58:37 <peter1138> Might be a bit large.
22:17:48 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
23:13:46 <peter1138> Okay, what sets were broken by FIRS?
23:14:50 <peter1138> Weird, I did lots of testing but never put pictures in the PR. Maybe that would've helped :p
23:14:58 *** Ox7C5 has quit IRC (Ping timeout: 480 seconds)
23:16:31 <peter1138> I did them here, didn't I...
23:20:33 <peter1138> Christ no wonder, it didn't mention fixing anything in particular at all.
23:23:21 <belajalilija> peter1138: Do you mean those that struggle with non vanilla cargoes?
23:23:31 <belajalilija> Swedish trains is one i think
23:26:08 <belajalilija> Are you making it so old grfs work again?
23:26:33 <peter1138> Already done. But not all of them.
23:27:25 <belajalilija> The US train set is old, it might be another that doesn’t work
23:45:29 <peter1138> It has a translation table.
23:46:12 <peter1138> This stuff has been available for decades, some sets just didn't bother.
continue to next day ⏵
