IRC logs for #openttd on OFTC at 2024-04-20
#openttd / 2024 / 04 
            

00:05:07  *** Wormnest has quit IRC (Ping timeout: 480 seconds)



00:40:21  *** Wormnest has joined #openttd



01:13:18  *** Compu has quit IRC (Remote host closed the connection)



01:13:45  *** Compu has joined #openttd



01:16:01  *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)



01:34:37  *** Wormnest has quit IRC (Ping timeout: 480 seconds)



01:43:52  *** gelignite has quit IRC (Quit: Stay safe!)



01:56:57  <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #12538: Codechange: Emplace std::pair instead of creating then inserting. https://github.com/OpenTTD/OpenTTD/pull/12538



02:01:40  *** Compu has quit IRC (Quit: https://www.youtube.com/watch?v=5FjWe31S_0g)



02:16:02  *** Wormnest has joined #openttd



02:20:49  *** Wormnest has quit IRC ()



02:33:13  *** gnu_jj_ has joined #openttd



02:36:47  *** gnu_jj has quit IRC (Ping timeout: 480 seconds)



02:39:14  *** debdog has joined #openttd



02:42:46  *** D-HUND has quit IRC (Ping timeout: 480 seconds)



03:00:07  *** herms has quit IRC (Quit: bye)



03:03:55  *** herms has joined #openttd



03:59:28  *** Flygon has joined #openttd



04:08:33  *** keikoz has joined #openttd



04:23:40  <Rubidium> peter1138: looks like find_if is manually loop-unrolled



04:25:08  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12538: Codechange: Emplace std::pair instead of creating then inserting. https://github.com/OpenTTD/OpenTTD/pull/12538#pullrequestreview-2013019819



04:41:27  <DorpsGek> [OpenTTD/OpenTTD] eints-sync[bot] pushed 1 commits to master https://github.com/OpenTTD/OpenTTD/commit/08140fdca36b16a784e096df115b53c05ca8dbf3



04:41:28  <DorpsGek>   - Update: Translations from eints (by translators)



05:47:01  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12537: Codechange: Remove old group liveries savegame conversion. https://github.com/OpenTTD/OpenTTD/pull/12537#pullrequestreview-2013033364



05:47:41  *** HerzogDeXtEr has joined #openttd



05:53:29  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12435: Change: Disallow using Action A to load sprites above the baseset unless reserved. https://github.com/OpenTTD/OpenTTD/pull/12435#pullrequestreview-2013034040



05:54:13  *** Smedles has quit IRC (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)



05:55:36  *** Smedles has joined #openttd



05:57:49  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12536: Change: De-template BaseSetTextfileWindow. https://github.com/OpenTTD/OpenTTD/pull/12536#pullrequestreview-2013034501



06:02:08  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12508: Fix 3de8853: Industries accept/produce no cargo for pre-SLV_78 saves. https://github.com/OpenTTD/OpenTTD/pull/12508#pullrequestreview-2013034947



06:58:07  *** nielsm has joined #openttd



07:03:50  *** keikoz has quit IRC ()



07:10:37  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12442: Codechange: Use std::accumulate to determine infrastructure total rail/road pieces. https://github.com/OpenTTD/OpenTTD/pull/12442#pullrequestreview-2013044839



07:12:31  *** keikoz has joined #openttd



07:14:08  <DorpsGek> [OpenTTD/team] Porrumentzio opened issue #551: [eu_ES] Translator access request https://github.com/OpenTTD/team/issues/551



07:46:54  <peter1138> Rubidium: Yea, that's the conclusion I came to.



07:47:07  <andythenorth> what shall we do today?



07:47:27  <peter1138> Stop my neck from hurting



07:47:36  <andythenorth> osteo



07:50:39  *** Ox7C5 has joined #openttd



07:57:47  <peter1138> Also, I happened to look at the disassembly of SetupCargoForClimate.



07:58:55  <peter1138> It is a bit hilarious.



08:02:08  <peter1138> https://cdn.discordapp.com/attachments/1008473233844097104/1231152912256995379/image.png?ex=6624c7ff&is=6623767f&hm=ab8dbeb50d0dc8045cad84b0fc816331f4a6c55ba3fdd3716b90f6339588ec25&



08:02:08  <peter1138> 4 screenfuls of this...



08:04:55  <peter1138> So not only is the loop unrolled, the array data is ... whatever you call "unrolling" of array data...



08:06:21  <peter1138> I copied my hand written find_if at that location and it got unrolled exactly the same, until I did #pragma nounroll.



08:06:33  <peter1138> Compilers please :S



08:21:23  *** matusguy has joined #openttd



08:24:50  *** virtualrandomnumber has joined #openttd



08:25:43  *** virtualrandomnumber has quit IRC ()



08:40:31  <peter1138> Okay, I can simplify it.



08:40:38  <peter1138> By undoing something from 20 years ago 😄



08:41:29  *** nielsm has quit IRC (Remote host closed the connection)



08:44:00  *** nielsm has joined #openttd



08:49:35  <locosage> Is there a any way to stop `make -j<n>` from hanging the entire system randomly?



08:51:25  <locosage> it used to work fine but broke after some update or smth :(



08:51:55  <FLHerne> depends why it's doing that, if it's consuming all memory then setting up oomd or earlyoom will stop it hanging the system



08:52:05  <FLHerne> (by killing it first if it tries to)



08:52:06  *** nielsm has quit IRC (Ping timeout: 480 seconds)



08:53:45  <FLHerne> I would recommend using ninja instead of make, it's a lot faster



08:54:02  <FLHerne> also ccache makes a huge difference on slower hardware



08:54:04  <andythenorth> I probably don't build enough to see it hang randomly 😛



08:54:22  <andythenorth> I always use -j<n>



08:54:38  *** Wolf01 has joined #openttd



08:55:36  <andythenorth> oof build has got slow



08:55:49  <locosage> can I use ninja with openttd?



08:55:55  <FLHerne> (to use ninja, use `cmake -Gninja [whatever]` instead of `cmake [whatever]`



08:55:56  <FLHerne> )



08:55:58  <FLHerne> yes



08:56:29  <FLHerne> it'll generate build.ninja instead of a makefile



08:57:25  <FLHerne> then just run `ninja` instead of `make`, it'll pick a sane number of jobs by default



08:58:11  <FLHerne> setting up ccache is a little bit more hassle but really is worthwhile IME



09:01:09  <andythenorth> 1m 40s to build



09:01:29  <andythenorth> previous run seemed slower



09:01:48  <peter1138> locosage: Use less n



09:03:13  <peter1138> It's either IO, or lack of memory (resulting in IO)



09:03:50  <locosage> I tried but it keeps hanging from time to time



09:04:03  <locosage> I can go really low but that kinda defeats the purpose



09:04:30  <Rubidium> locosage: ionice and nice maybe?



09:04:42  <andythenorth> hmm 1m 50s



09:04:45  <andythenorth> really unstable results



09:04:46  <locosage> nope, nice didn't help



09:04:55  <locosage> or rather it did until it didn't



09:05:32  <locosage> I guess I'll try using ninja for now, it worked first time at least



09:06:00  <peter1138> https://cdn.discordapp.com/attachments/1008473233844097104/1231168987627520040/image.png?ex=6635fa78&is=66238578&hm=e9322b76107eb1890bfde76f0a210dbd5b57910de3b702de813b28c062185df6&



09:06:00  <peter1138> Such load avg



09:07:56  <locosage> is clang less memory intensive? I should probably try it



09:09:05  <locosage> `load average: 18.89, 8.47, 4.37`



09:09:11  <locosage> probably hitting oom



09:10:18  <andythenorth> https://cdn.discordapp.com/attachments/1008473233844097104/1231170067572723735/image.png?ex=6624d7fa&is=6623867a&hm=5e5c934cfa3f9bd6bf0d51c908106b2e8b24a46b9c49640ed6fc88fc040f744d&



09:10:18  <andythenorth> should I make this 'w' for wharf on the minimap? 😛



09:10:48  <andythenorth> https://cdn.discordapp.com/attachments/1008473233844097104/1231170194798415932/image.png?ex=6635fb98&is=66238698&hm=baab3671664e9361964770199d33f27c75e8c83e079c07fa5d90156dd4f32487&



09:10:48  <andythenorth> I can make letters for these port industries



09:11:11  <andythenorth> bt for 'bulk terminal' would be possible



09:11:37  <andythenorth> will only work for SE coasts though 😛



09:22:23  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12442: Codechange: Use std::accumulate to determine infrastructure total rail/road pieces. https://github.com/OpenTTD/OpenTTD/pull/12442



09:23:39  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12435: Change: Disallow using Action A to load sprites above the baseset unless reserved. https://github.com/OpenTTD/OpenTTD/pull/12435



09:23:50  <peter1138> RIP asasignals



09:24:13  <andythenorth> hmm what about player-buildable industry tiles? 😛



09:24:24  <andythenorth> "I wish to customise this industry"



09:24:29  <andythenorth> what does Factorio do?



09:24:39  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12508: Fix 3de8853: Industries accept/produce no cargo for pre-SLV_78 saves. https://github.com/OpenTTD/OpenTTD/pull/12508



09:25:08  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12537: Codechange: Remove old group liveries savegame conversion. https://github.com/OpenTTD/OpenTTD/pull/12537



09:25:24  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12538: Codechange: Emplace std::pair instead of creating then inserting. https://github.com/OpenTTD/OpenTTD/pull/12538



09:28:11  <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #12481: Codechange: use std::array instead of C-style array for produced/accepts cargo https://github.com/OpenTTD/OpenTTD/pull/12481#pullrequestreview-2013060436



09:29:10  <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #12451: Codechange: replace last cases of endof with other (safer) constructs https://github.com/OpenTTD/OpenTTD/pull/12451#pullrequestreview-2013060548



09:29:58  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12481: Codechange: use std::array instead of C-style array for produced/accepts cargo https://github.com/OpenTTD/OpenTTD/pull/12481



09:30:32  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12536: Change: De-template BaseSetTextfileWindow. https://github.com/OpenTTD/OpenTTD/pull/12536



09:30:48  <peter1138> I better check master still compiles after all that ;D



09:30:50  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12451: Codechange: replace last cases of endof with other (safer) constructs https://github.com/OpenTTD/OpenTTD/pull/12451



09:31:19  <peter1138> Oops, that was a Codechange not a Change :/



09:35:56  <peter1138> locosage: clang also seems to produce slightly faster executables.



09:36:26  <peter1138> TB also noticed that I think.



09:38:04  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #12440: Codechange: replace many cases of lengthof with std::size https://github.com/OpenTTD/OpenTTD/pull/12440



09:38:58  <peter1138> I now only have 288 local branches...



09:40:22  <Rubidium> wow... double gross



09:44:26  <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #12539: Change: Use std::make_unique instead of passing new() https://github.com/OpenTTD/OpenTTD/pull/12539



09:50:42  <peter1138> Hmm, can I write a unit test for YAPF-internals...



09:51:08  <peter1138> Scared 😄



09:54:26  <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #12436: Codechange: use C++ containers for parsing the settings int lists https://github.com/OpenTTD/OpenTTD/pull/12436#pullrequestreview-2013063454



09:56:16  <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #12385: Codechange: move knowledge about 'packed' orders to the saveload code https://github.com/OpenTTD/OpenTTD/pull/12385#pullrequestreview-2013063688



09:57:22  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #12540: Codechange: replace cpp_lengthof with safe alternatives https://github.com/OpenTTD/OpenTTD/pull/12540



09:57:48  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12436: Codechange: use C++ containers for parsing the settings int lists https://github.com/OpenTTD/OpenTTD/pull/12436



09:58:11  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12385: Codechange: move knowledge about 'packed' orders to the saveload code https://github.com/OpenTTD/OpenTTD/pull/12385



10:00:47  <merni> locosage: are you still interested in https://github.com/OpenTTD/survey-web/pull/37 ?



10:00:47  <merni> The relevant key seems to be `"revision"`



10:00:47  <merni> https://github.com/OpenTTD/survey-web/blob/4104b6b154c0f3a159dfa7122b7b53fc6a16e296/analysis/__main__.py#L197



10:00:52  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12539: Change: Use std::make_unique instead of passing new() https://github.com/OpenTTD/OpenTTD/pull/12539#pullrequestreview-2013064163



10:01:52  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #12440: Codechange: replace many cases of lengthof with std::size https://github.com/OpenTTD/OpenTTD/pull/12440



10:02:50  <DorpsGek> [OpenTTD/OpenTTD] PeterN dismissed a review for pull request #12297: Change: Add per-company group numbers. https://github.com/OpenTTD/OpenTTD/pull/12297#pullrequestreview-1944399347



10:02:53  <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #12297: Change: Add per-company group numbers. https://github.com/OpenTTD/OpenTTD/pull/12297



10:04:35  <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #12297: Change: Add per-company group numbers. https://github.com/OpenTTD/OpenTTD/pull/12297#issuecomment-2067625383



10:06:47  <merni> This bit in the description is missing something:



10:06:47  <merni> > When creating groups, the default group name is "Group ", where index is a global number starting at zero, and shared by all companies.



10:06:58  <merni> "Group <index>"?



10:09:35  <peter1138> https://cdn.discordapp.com/attachments/1008473233844097104/1231184984488939561/image.png?ex=6636095e&is=6623945e&hm=60c0426f8c5ca04bba0114c12f26ff839418598263bcf256fcf56dced8328e7e&



10:09:36  <peter1138> 😮



10:11:19  <merni> It probably thinks it's HTML



10:11:27  <merni> may need to be somehow escapedf



10:16:38  <kuhnovic> peter1138: There probably is some ugly way using all that CRTP magic 😆



10:19:14  <peter1138> Hmm, still seems weird seeing process IDs in the millions. They were 16 bit for so long...



10:19:56  <peter1138> Back when you'd get exploits that worked by reusing process IDs as they wrapped easily...



10:20:53  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12539: Change: Use std::make_unique instead of passing new() https://github.com/OpenTTD/OpenTTD/pull/12539



10:22:08  <peter1138> > 288 branches on the wall, 288 branches. Take one down and pass it around, 287 branches on the wall.



10:22:17  <merni> What about patches



10:23:00  <peter1138> Too many to count, but if it's a patch file it will be ancient and not worth bothering with.



10:24:07  <merni> I love how many mentions issue #1 has god



10:24:10  <merni> *got



10:26:58  <peter1138> Heh. I wonder which are intentional. But I'm not going to look.



10:31:27  <merni> I doubt even one is



11:06:31  <locosage> merni: ty, I'll consider



11:06:48  <locosage> already have custom survey server and stuff in survey that vanilla doesn't support though



11:14:17  *** _Artea is now known as Artea



11:30:40  <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #12540: Codechange: replace cpp_lengthof with safe alternatives https://github.com/OpenTTD/OpenTTD/pull/12540#pullrequestreview-2013073855



11:36:23  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12540: Codechange: replace cpp_lengthof with safe alternatives https://github.com/OpenTTD/OpenTTD/pull/12540



11:41:32  <peter1138> Hmm, how many places do we reinterpret an array...



11:55:05  <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541



12:03:34  <DorpsGek> [OpenTTD/team] glx22 commented on issue #551: [eu_ES] Translator access request https://github.com/OpenTTD/team/issues/551



12:20:39  <andythenorth> hmm



12:20:44  <andythenorth> should we give ships reflections?



12:21:16  <andythenorth> or in fact, just reflecting water feature?



12:21:44  <peter1138> Huh.



12:21:56  <peter1138> Well funnily enough, I'm not sure what the operator precedence should be :p



12:23:44  <peter1138> (8 * sizeof(Tindex)) I think.



12:24:20  <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541



12:26:55  <Rubidium> isn't that like static_assert(Tmax_size == std::numeric_limits<Tindex>::max()); ?



12:27:46  <Rubidium> well <= instead of ==



12:27:49  <peter1138> Almost.



12:27:50  <peter1138> Yeah



12:28:49  <peter1138> Thanks,



12:28:51  <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541



12:29:08  <peter1138> One reason why I'm haven't just going for every single cast in every file, makes these things easier to see.



12:29:22  <peter1138> Words :S



12:30:54  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013080269



12:32:29  <Rubidium> oh... that's breaking quite hard :(



12:34:45  <peter1138> Hmm.



12:35:01  <peter1138> Should've actually tried compiling it, right? 😄



12:35:48  <peter1138> Ah, of course, the company pool...



12:37:15  <peter1138> Although actually that suggests the original check was wrong.



12:37:18  <Rubidium> oh, apparantly std::numeric_limits<enum>::max() == 0; maybe replacing Tindex with std::to_underlying_type_t<Tindex> works... when every index is an enum



12:37:29  <merni> I want to work in a company with a pool



12:37:39  <peter1138> Ahhh!



12:37:55  <Rubidium> merni: maybe Homer's job? :D



12:38:14  <merni> ah, yes, swim in the nuclear waste



12:39:15  <Rubidium> it's unlikely that you'll die from nuclear exposure when swimming in those pools



12:39:16  <truebrain> merni: Lifeguard?



12:39:29  <merni> huh?



12:40:03  <truebrain> Those people work for a company with a pool!



12:40:16  <merni> oh



12:40:16  <merni> true



12:40:24  <merni> but I don't want to *work* in the pool!



12:40:35  <truebrain> Be their manager



12:41:14  <Eddi|zuHause> Rubidium: unless you dive and actually touch the nuclear waste



12:42:24  <peter1138> Hmm, how to switch between <underlype_type<type>> and <type> depends on is_enum_v...



12:44:52  <Rubidium> Eddi|zuHause: https://youtu.be/EFRUL7vKdU8?t=210 ...



12:45:13  <Eddi|zuHause> i'm pretty sure i saw that video (without clicking on it)



12:45:33  <Rubidium> peter1138: yeah, that's the tricky bit



12:46:49  <Rubidium> peter1138: maybe with std::conditional_t ?



12:56:01  <peter1138> `static_assert(Tmax_size <= std::numeric_limits<std::conditional_t<std::is_enum_v<Tindex>, std::underlying_type_t<Tindex>, Tindex>>::max());`



12:56:02  <peter1138> ...



12:56:07  <peter1138> does not work 😦



12:58:11  <peter1138> <https://stackoverflow.com/questions/69284927/c-stdconditional-t-wont-compile-with-stdis-enum-v-and-stdunderlying-type>



12:58:14  <peter1138> Oh, snap.



12:59:34  <peter1138> The answer doesn't seem relevant there.



13:07:54  <peter1138> There's always `Tmax_size <= MAX_UVALUE(Tindex)` but that's still C-style really...



13:13:47  <peter1138> On the other hand, the intention is cleared than the bit-shifting...



13:14:03  <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #12533: Backport master into release/14 https://github.com/OpenTTD/OpenTTD/pull/12533



13:15:59  <_glx_> oh no, there was another requested backport



13:16:32  <_glx_> I should have check before merge



13:16:35  <Rubidium> peter1138: the MAX_UVALUE is definitely clearer than what it's now. And static_assert(static_cast<Tindex>(Tmax_size) == Tmax_size) is arguably not better thatn the MAX_UVALUE



13:34:29  <DorpsGek> [OpenTTD/OpenTTD] JGRennison opened pull request #12542: Change: [Linkgraph] Improve distance scaling algorithm in demand scaler https://github.com/OpenTTD/OpenTTD/pull/12542



13:43:23  <peter1138> It's also wrong 😄



13:43:50  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #12543: Codechange: simplify access to the current screenshot format https://github.com/OpenTTD/OpenTTD/pull/12543



13:44:57  <DorpsGek> [OpenTTD/OpenTTD] PeterN dismissed a review for pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013080269



13:45:00  <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541



13:57:42  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013091084



13:59:29  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 updated pull request #12543: Codechange: simplify access to the current screenshot format https://github.com/OpenTTD/OpenTTD/pull/12543



14:07:44  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013092326



14:11:21  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013092859



14:11:24  <DorpsGek> [OpenTTD/OpenTTD] PeterN dismissed a review for pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013091084



14:11:27  <DorpsGek> [OpenTTD/OpenTTD] PeterN updated pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541



14:15:05  <DorpsGek> [OpenTTD/OpenTTD] glx22 opened pull request #12544: Codechange: [Script] Use std::unique_ptr for Company::ai_instance https://github.com/OpenTTD/OpenTTD/pull/12544



14:15:24  <peter1138> Hah, I have a patch for that 🙂



14:16:04  <_glx_> it's in my head since I rework ai_config



14:16:43  <peter1138> I had to mark `Company::~Company()` as virtual though, to avoid the compiler trying to optimise it anyway and then complaining that it doesn't know the sizeof AIInstance.



14:17:19  <_glx_> I just added an include to allow compilation



14:17:39  <_glx_> else it complained about destructor for incomplete type



14:17:53  <_glx_> had to do the same for ai_config



14:17:59  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #12545: Fix: allow only 255 league tables, as 255 is the invalid id sentinel https://github.com/OpenTTD/OpenTTD/pull/12545



14:18:51  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541#pullrequestreview-2013094105



14:20:02  <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #12545: Fix: allow only 255 league tables, as 255 is the invalid id sentinel https://github.com/OpenTTD/OpenTTD/pull/12545#issuecomment-2067688446



14:20:15  <_glx_> and VS tells me the 2 headers are not used in company_cmd.cpp 😉



14:28:45  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on pull request #12545: Fix: allow only 255 league tables, as 255 is the invalid id sentinel https://github.com/OpenTTD/OpenTTD/pull/12545#issuecomment-2067691284



14:29:52  <DorpsGek> [OpenTTD/OpenTTD] PeterN approved pull request #12545: Fix: allow only 255 league tables, as 255 is the invalid id sentinel https://github.com/OpenTTD/OpenTTD/pull/12545#pullrequestreview-2013098038



14:39:43  <peter1138> https://cdn.discordapp.com/attachments/1008473233844097104/1231252967752077343/image.png?ex=663648af&is=6623d3af&hm=932d7888a081b1182a7aed27fae2c6d95cb5edcf175c5583e3ecb1047aab4466&



14:39:43  <peter1138> cppcheck seems confused...



14:40:07  <peter1138> first_unused is size_t...



14:40:39  <_glx_> yeah size_t size is platform dependant



14:41:14  <_glx_> but I don't see why it would complain here



14:41:14  <peter1138> Sure, But it's not an address value.



14:43:17  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler approved pull request #12297: Change: Add per-company group numbers. https://github.com/OpenTTD/OpenTTD/pull/12297#pullrequestreview-2013099649



14:50:16  *** Wormnest has joined #openttd



14:52:53  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 approved pull request #12544: Codechange: [Script] Use std::unique_ptr for Company::ai_instance https://github.com/OpenTTD/OpenTTD/pull/12544#pullrequestreview-2013100866



14:53:07  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12545: Fix: allow only 255 league tables, as 255 is the invalid id sentinel https://github.com/OpenTTD/OpenTTD/pull/12545



14:53:18  <DorpsGek> [OpenTTD/OpenTTD] JGRennison opened pull request #12546: Fix #12509: Timer period modifications may violate invariants of TimerManager::base_timer_sorter std::set sorting https://github.com/OpenTTD/OpenTTD/pull/12546



14:55:56  <andythenorth> It would make no sense to build docks into FIRS port industries?



14:58:49  <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #12544: Codechange: [Script] Use std::unique_ptr for Company::ai_instance https://github.com/OpenTTD/OpenTTD/pull/12544



15:37:56  <FLHerne> andythenorth: gameplay-wise, not



15:38:29  <FLHerne> it does feel a bit odd though



15:50:16  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12541: Codechange: Replace C-casts in pool functions. https://github.com/OpenTTD/OpenTTD/pull/12541



16:18:24  <peter1138> _glx_: Got another backport PR incoming.



16:19:37  <_glx_> we already have 2 🙂



16:29:14  <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #12547: Fix 952d111: Houses and industry tiles could accept incorrect cargo. https://github.com/OpenTTD/OpenTTD/pull/12547



16:29:48  <peter1138> Might have to make a contrived NewGRF set up to test this...



16:30:04  <peter1138> Found it why looking at code, rather than while looking at cargos in game 🙂



16:35:33  <peter1138> Dear AI authors, please fix your AIs 😄



16:41:49  <Rubidium> Dear AIs, please fix yourself... or is that too soon :(



16:44:10  <DorpsGek> [OpenTTD/OpenTTD] PeterN commented on pull request #12547: Fix 952d111: Houses and industry tiles could accept incorrect cargo. https://github.com/OpenTTD/OpenTTD/pull/12547#issuecomment-2067725971



17:15:38  *** matusguy has quit IRC (Remote host closed the connection)



17:35:22  <peter1138> When you refactor something and then wonder if it's any better...



18:22:27  <andythenorth> standard for me 😛



18:49:09  <DorpsGek> [OpenTTD/OpenTTD] PeterN opened pull request #12548: Codechange: Add constants for original input/output cargo counts. https://github.com/OpenTTD/OpenTTD/pull/12548



18:59:53  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 opened pull request #12549: Fix 5008706: improved scenario editor tooltips in map generation stages are out of place https://github.com/OpenTTD/OpenTTD/pull/12549



19:03:09  <DorpsGek> [OpenTTD/nml] Quantom2 opened issue #325: Increase D0xx string limit please https://github.com/OpenTTD/nml/issues/325



19:04:19  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 commented on issue #10474: [Crash]: Crash when changing settings when config file is synced using Dropbox https://github.com/OpenTTD/OpenTTD/issues/10474



19:21:01  <peter1138> Heh by definition the limit of D0XX is... FF?



19:23:15  <peter1138> Judging from the stack trace, that is the same issue.



19:24:56  <DorpsGek> [OpenTTD/nml] glx22 commented on issue #325: Increase D0xx string limit please https://github.com/OpenTTD/nml/issues/325



19:25:23  <_glx_> limit is 3FF, but still we can't increase it



19:26:01  <_glx_> there's ways to work around



19:26:41  <_glx_> my guess is they use a lot of engine name callbacks



19:33:57  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler approved pull request #12549: Fix 5008706: improved scenario editor tooltips in map generation stages are out of place https://github.com/OpenTTD/OpenTTD/pull/12549#pullrequestreview-2013138400



19:36:21  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler approved pull request #12548: Codechange: Add constants for original input/output cargo counts. https://github.com/OpenTTD/OpenTTD/pull/12548#pullrequestreview-2013138606



19:45:58  <peter1138> > Yes, exactly. The slow path calls a non-inline function, which the compiler should be less likely to inline. And that worked pretty well in the past. But since C++20 every member function of std::vector is constexpr and that means they're implicitly inline. So now the compiler thinks we've asked it to try to inline everything, even the large, cold functions that are not marked inline in C++17



19:45:58  <peter1138> and earlier. This is the subject of https://gcc.gnu.org/PR93008



19:45:59  <peter1138> Hmm



19:50:52  <DorpsGek> [OpenTTD/OpenTTD] PeterN merged pull request #12548: Codechange: Add constants for original input/output cargo counts. https://github.com/OpenTTD/OpenTTD/pull/12548



19:56:31  <DorpsGek> [OpenTTD/OpenTTD] rubidium42 merged pull request #12549: Fix 5008706: improved scenario editor tooltips in map generation stages are out of place https://github.com/OpenTTD/OpenTTD/pull/12549



20:22:33  *** Ox7C5 has quit IRC ()



20:51:33  *** keikoz has quit IRC (Remote host closed the connection)



20:51:43  *** keikoz has joined #openttd



20:53:59  *** Flygon has quit IRC (Remote host closed the connection)



21:01:54  *** keikoz has quit IRC (Remote host closed the connection)



21:02:04  *** keikoz has joined #openttd



21:45:07  *** keikoz has quit IRC (Ping timeout: 480 seconds)



22:04:48  *** tokai has quit IRC (Quit: c('~' )o)



22:44:26  *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)



23:10:42  <DorpsGek> [OpenTTD/nml] glx22 commented on issue #325: Increase D0xx string limit please https://github.com/OpenTTD/nml/issues/325



23:29:59  *** Smedles has quit IRC (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)



23:59:22  <DorpsGek> [OpenTTD/OpenTTD] telk5093 opened issue #12550: [Bug]: ``screenshot minimap`` does not generate files in dedicated server again https://github.com/OpenTTD/OpenTTD/issues/12550