Atrik

A small annoyance is also the inability to close console with the associated hotkey if it's printable character. Comparing with a27 it did work there.

Just check the link https://releases.wildfiregames.com/rc/ , download for your os, now linux have a appimage so you don't need to do anything fancy. RIP you, market got fixed, you'll need to find another exploit.

Thanks! I will. Right now I'm noticing that formations got much worse in r28. They were already very buggy before, so these new bugs will really be frustrating. I'll try help on this first as it might still be time for bug fixs for r28.

No, same reason as @ffm2, and no entity get spawned. Just test it lol. Not sure how that happened, its not supposed to be like that. Random guess would be that the Carth stone tech get applied to all civs in p1. That is intentional, as the tech turns them into 5 pop houses. But you couldn't train females from them afterward iirc. It seemed a bit odd. Other proprieties of houses are likely to be applied to the wagon that might cause game-play weirdness.

I know nothing about this stuff but my first guess would be that you named something with a too long or invalid name.

I'd like to revisit the code a bit. It needs polishing in any cases but I might also make improvements. Since apparently it's too late to get anything into r28, I'll rather give it the time. I'd love to have your help and views on it! Especially since you seems to be very proefficient on performance implications, and UX in general (here is an opportunity to thank you for your work on 0ad too ). I'm still learning the review/git process so I don't understand what you have in mind here, as I also haven't went through a merged PR.

I think you might have to do something like this : Add a "submarine" section in there pathfinder.xml with a nonrestrictive "shore distance". Handle it in BuildRestrictions (maybe also some stuff here Templates.js). Then you should be able to add <PlacementType>submarine</PlacementType> in <BuildRestrictions> in templates. Don't forget to fix petra too!

Assuming the RC is same as release 0.28 git branch : Germans are now unplayable Stone gather rates is almost doubled compared to xml template for seemingly all units Last known bug for Germans i noticed and don't think got fixed (can't check cause can't play them): Fertility festival and loom become available in wagons after researching Migratory Resettlement. Don't think that is intentional.

My bad, the @manowar config tip for lfs works for me, I also had to run git remote remove origin

Ah! I missed that thread. I have a similar problem (still today). Can't update my fork. ssh -T git@gitea.wildfiregames.comSSH Says hi, but then it ask me for https authentication in a loop. No error, nothing, just keep asking credentials. I've tried a lot of different things as I though I was doing something wrong but no success here yet.

I ran into a disk failure, probably, the error was NOT related, however after restoring everything I've then been able to compile. Something was very probably outdated but I don't think it was the compiler since I tried both clang and gcc, clang had at least the most recent version and gcc was 12.x Now using gcc 13.3.0 It works seamlessly. Thanks.

Thanks, so what I've tried now is : ./update-workspaces.sh --cc=gcc-12 cd gcc make And ./update-workspaces.sh --cc=clang-18 cd gcc #Can't see any other folders make If i pass a compiler version i don't have to update-workspaces.sh, it tell me so. Therefor I'm guessing i do have to compiler I'm passing above. Yet compilation fails in both cases. With gcc /usr/include/fmt/core.h:2201:22: error: ‘constexpr void fmt::v8::basic_format_parse_context<Char, ErrorHandler>::on_error(const char*) [with Char = char; ErrorHandler = fmt::v8::detail::error_handler]’ called in a constant expression 2201 | context_.on_error(message); With clang In file included from ../../../source/ps/GameSetup/GameSetup.cpp:84: ../../../source/scriptinterface/FunctionWrapper.h:518:6: error: call to consteval function 'fmt::basic_format_string<char, std::basic_string<char> &>::basic_format_string<char[51], 0>' is not a constant expression 518 | "Failed defining function {:?} on the native scope.", classInfo->name)}; | ^

That do fix the error, but compilation fails later with another forward declaration and a bunch of new errors too. I have clang 18.1.8, rusc 1.91.0

I'm not able to compile a28 or main, the compilation fail with this error: ../../../source/scriptinterface/ModuleLoader.h:86:15: required from here /usr/include/c++/11/bits/stl_pair.h:218:11: error: ‘std::pair<_T1, _T2>::second’ has incomplete type 218 | _T2 second; ///< The second member | ^~~~~~ In file included from ../../../source/dapinterface/DapInterface.cpp:30: ../../../source/scriptinterface/ModuleLoader.h:50:15: note: forward declaration of ‘class Script::ModuleLoader::CompiledModule’ 50 | class CompiledModule; | ^~~~~~~~~~~~~~ make[1]: *** [atlas.make:233: obj/atlas_Release/MiscHandlers.o] Error 1 make[1]: *** Waiting for unfinished jobs.... JSInterface_GameView.cpp JSInterface_Renderer.cpp make[1]: *** [engine.make:390: obj/engine_Release/CConsole.o] Error 1 make: *** [Makefile:122: engine] Error 2 AtlasObjectJS.cpp make[1]: *** [graphics.make:417: obj/graphics_Release/FontManager.o] Error 1 make[1]: *** Waiting for unfinished jobs.... CCmpMotionBall.cpp make[1]: *** [simulation2.make:371: obj/simulation2_Release/CCmpAIManager.o] Error 1 make[1]: *** Waiting for unfinished jobs.... AtlasObjectText.cpp CommonConvert.cpp make[1]: *** [dapinterface.make:142: obj/dapinterface_Release/DapInterface.o] Error 1 make: *** [Makefile:92: dapinterface] Error 2 Some time ago I already had this error when compiling main, and though it was because of some WIP, but now I'm guessing the problem is on my end since it's still here. What am I doing wrong? Debian build Thanks

@AlexHerbert those are the cav you get after Marian reform (Roman upgrade in fort). Once researched, this tech change almost all units of the Romans; It's mostly a tech to make infantry stronger, for cav, stats aren't affected but icon still are. It's new from a27.

Hi @Thomas Stewart, you likely joined a old lobby, make sure you reinstall the new version of the game (currently a27). Downloads are available there: https://play0ad.com/

I've tested all cases I could think of but we can always give it a couple days to see if something broken is discovered. If not, I don't have a gitea account or repo set up, and I might be lazy to set one up (a bit of other reasons too that makes me not want to do it maybe...). I've made a version of the code SnappingWallsPr.zip that use function overwrites and comments cleanup to make it easier if someone want create a PR for main game.

Good catch for the cached entities. Alright I think I've finished this up so we might have it now. Cool enough by replacing the method to retrieve snappable entites, a couple vanilla bugs for fixed just like that; notably the inability to continue wall placement in fog of war, but also inability to build from captured walls. I've listed all features achieved in opening comment. I've tested everything I could thing of to see if any exceptions are left unhandled but I think everything seems so much better now then it ever was.

0AD clearly lacks campaigns, and Hyrule is one of the rare mod to bring one. That's what i know. But beside this I'm not a huge single player, as I'm mostly caught in multiplayer already, so I don't feel best positioned to say more. If ever since it was my case at some point : building the game on linux is super easy, but reading through the build instructions is confusing. Someone should make something better with tabs or something for each os, instead of the current giant monolith... I made a summary here for building a27 on debian.

On the campaign, I'm not sure where to find the enemy after destroying the first base (got all the map explored), so I'm unable to finish first mission. Beside this very good job on the mod

It seems to have been fixed, the newest upload says 5 downloads. Wonder what the bug was. You probably got confused by the counter resetting because I'm reuploading shamefully the mod every time I find some bugs coming from the mod.. 4rd reupload due to poorly handled exceptions ... Also on @Vantha's tip I've tried to see if I could get some of the logic in guiinterface. I couldn't import much logic there but in SetWallPlacementPreview we could control additional visual effects of the previews. I also rewrote some of it to be able to reason from it... Anyways... If the mod/code is working as I think it is now (finely), my guess is that it is still not possible to PR it as long as it is a constant storing tower-like structures, usable as start/end points for wall segments.

Thanks, I've just realized I've slept an stupid, potentially error-generating mistake so I've just re-uploaded again with tiny fix... There are also small compromises I made as said in the opening comment... And the buildings templates that can be used as end wall sections (towers) are defined arbitrarily in a constant etc... So there are a few things that needs extra work.

Unlikely to be accepted as a PR. It lengthen the code by a bit; and is probably not how one would go about to implement all of the different features I've added, without the constraint of trying to make it into a compatible mod..

Wish I could, however this would either be very complicated, or needs to be as a incompatible mod setting some references point in fort templates or something. I wanted this mod to be compatible.

Updated : Significant performance improvement (First version was bad, this one fix it but also improve wall placement performance impacts in general) Now other towers can be used as a starting or end point for wall placement instead of wall towers. This doesn't prevent you from walling easily around them : you can also snap walls and palisades to their edges! (The threshold distance makes it so easy to choose between the two you won't even be bothered).