00:59:16 -!- indigaz23 is now known as indigaz2 01:57:59 03DracoOmega02 07* 0.34-a0-361-g77f475f057: Blame the player for fewer actions by confused enemies (Acrobat) 10(40 seconds ago, 2 files, 16+ 9-) 13https://github.com/crawl/crawl/commit/77f475f057e2 02:02:16 03dependabot[bot]02 {GitHub} 07[dependabot/pip/crawl-ref/source/webserver/requirements/tornado-6.5.1] * 0.34-a0-362-g9ecece3107: Bump tornado in /crawl-ref/source/webserver/requirements 10(51 seconds ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/9ecece310740 02:02:16 Branch pull/4566 updated to be equal with dependabot/pip/crawl-ref/source/webserver/requirements/tornado-6.5.1: 13https://github.com/crawl/crawl/pull/4566 02:04:52 <04dracoomega> I was thinking that somebody finally ought to do something about all those 'security alerts' that I keep being notified of when I pull, but this PR still seems to be failing checks in general. I don't suppose anyone more familiar with this end of things knows why? 03:31:34 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 05:17:56 Unstable branch on crawl.akrasiac.org updated to: 0.34-a0-361-g77f475f (34) 09:04:03 <09gammafunk> I'm guessing that tornado 6.5 doesn't support python 3.8 09:04:24 <09gammafunk> yep 09:04:28 <09gammafunk> @dracoomega > Prerequisites: Tornado 6.3 requires Python 3.9 or newer. The following optional packages may be useful: 09:08:59 <04dracoomega> Hmm... 3.9 is apparently already almost 5 years old. Is there some compatibility reason not to have upgraded (or just no compelling reason to have done so?)? 09:09:58 <09gammafunk> it would be something where we'd need to see what our older online server's python is 09:10:26 <09gammafunk> it's non-trivial to upgrade an installation of python on a live server 09:11:02 <09gammafunk> we seem to have 3.10 on CAO 09:12:48 <09gammafunk> I think switching to a newer python and tornado requirement would only affect new installations, provided that we don't put any e.g. 3.10+-isms into the webtiles 09:13:55 <09gammafunk> I'd have to remind myself on what the make file updates on the python side of things, if anything 09:17:00 <09gammafunk> looks like our makefile never messes with anything python-side as far as the webserver is concerned 09:19:57 <09gammafunk> I'm not sure if any of our existing servers use 3.9, but CAO is likely the oldest python install, and regardless this would only be relevant to said servers if they attempted a webserver reinstall or otherwise updated the server code. 09:20:16 <09gammafunk> all this to say that it's probably safe to bump the python requirement to 3.10, if that's enough, but we should update the appropriate places if we do that 09:21:16 <09gammafunk> er, I should say "if any of our existing servers use 3.8" 09:21:21 <09gammafunk> because apparently just 3.9 is sufficient 09:25:32 <04dracoomega> (Like, all of this is a bit out of my depth, I admit, but it seems like good principle to not ignore git warning about vulnerabilities on a long-term basis >.>) 09:45:47 <09gammafunk> well, the main thing about the vulnerabilities is that they'll always be in play until a server both has an updated python/tornado install (assuming we merge that PR) and restarts webtiles 09:46:27 <09gammafunk> so it's good to merge these PRs, but they don't actually directly improve security until admins perform server-side upgrades 09:48:14 <09gammafunk> the thing with these servers is that they tend to use python installs that come from the OS distribution, so we can't force a python/tornado upgrade upon them 09:54:46 <04dracoomega> ...does that mean that they also might already be using a version of tornado new enough to not have this vulnerability or something? I thought this was something bundled as a dependency of Crawl itself here 10:01:23 <09gammafunk> they might have a newer tornado, sure. You can think of it as a dependency that's only checked once, at the time the webserver is installed. Nothing in our make targets that are built as part of updating crawl would update anything on the python side of webtiles server (the make file does do updates to the webtiles client, however) 10:01:59 <09gammafunk> The webserver install is just this separate step that's done when the server is setup (and then whenever an admin feels like updating their webtiles server) 10:02:30 <09gammafunk> of course we do have CI that go through the webserver installation, but those are just CI that we use for informational purposes 10:03:27 <04dracoomega> So git is basically just warning that we aren't requiring the use of a safe version, not specifically that we are obligatorily using a vulnerable version? 10:04:29 <09gammafunk> yeah, it's warning that our formal requirements aren't safe 10:07:01 <09gammafunk> and these are only requirements in terms of someone actually deciding to use them (e.g. with pip to do an install) 10:07:43 <09gammafunk> although for requirements on just python, we actually do have a number of build components that do get updated via our makefile (e.g. the monster/species yaml generation stuff) 10:10:04 <09gammafunk> but there's no requirments.txt file involved for those, it's just whatever functionality we put in those python scripts might depend on some base version of python 10:23:16 <04dracoomega> Well, that's I guess a less concerning situation than I had assumed it might be, at least 11:00:01 hello 11:03:28 During gameplay, a crash made further progress completely impossible. After obtaining two runes, I was unable to enter areas such as Elf:1, Vault:2, Slime:1, and Depths:1. I tried using a distortion weapon to access the Abyss, and it was accessible. Could someone please help me? I have a dump URL available 11:05:16 https://archive.nemelex.cards/saves/BlackTorch-crawl-git-77f475f057-250712-1754.tar.bz2 11:06:21 <04dracoomega> I will take a look 11:07:25 ty 11:08:39 https://archive.nemelex.cards/morgue/BlackTorch/crash-BlackTorch-20250712-134132.txt 11:08:48 <04dracoomega> Yeah, I already pulled up the crashlog 11:08:48 Here's the link to the crash file as well 11:08:57 <04dracoomega> (The save is very helpful, though) 11:11:45 <04dracoomega> The issue seems to be an invalid dummy unrand generating in Vaults:2, though unclear why/how this is happening 11:13:26 If I hadn't visited Vaults:2 at all, would I have been able to access other areas normally? 11:14:19 <04dracoomega> No, no. Floors in the game are pre-generated in a specific order. You haven't been to Vaults:2 at all yet, but entering many other places (such as Elf or Depths) will pregenerate the rest of Vaults automatically 11:14:36 <04dracoomega> Where it is somehow generating an invalid item and then crashing as a result 11:15:31 <04dracoomega> I am still trying to figure out how this occurred, but I'm afraid that may take a while 11:15:49 I understand. If the bug gets fixed later, will I be able to continue playing with this data as is, or will I just have to ctrl+q? 11:17:23 <04dracoomega> I think it is very possible that the save will be recoverable once I figure out the cause of the bad generation here 11:17:35 <04dracoomega> Since the floor has not been generated 'yet', so the save file does not itself contain invalid information 11:17:52 <04dracoomega> And once the problem is fixed, Vaults:2 should generate in a valid state 11:19:19 Ge0ff (L7 TeSt) (D:5) 11:19:28 Ge0ff (L7 TeSt) (D:5) 11:21:00 i see. ty 11:21:38 <04dracoomega> @ge0ff ...you know, at first I was like "What in the world are you doing to get a crash in describe_god, but on second thought, I don't think this is actually my problem to fix for some reason 😛 11:22:06 <04dracoomega> And I can trust to the passage of 15 years to have fixed it for me 11:24:31 <12ge0ff> That's just old Ash crashing when i switch to the detailed description on the religion screen 11:28:21 <09gammafunk> nice test 11:29:31 <04dracoomega> In any case, somehow Hat of the Alchemist is being generated (seemingly normally) and then somehow being edited later on to have the wrong unrand ID. It even still has the proper inscription! 11:29:41 <04dracoomega> the +0 DUMMY UNRANDART 1 {rElec rPois rF+ rC+ rN+ Will+ rMut rCorr} 11:53:53 <04dracoomega> Okay, I've found the problem 11:53:56 <06regret-⸸nde※> https://cdn.discordapp.com/attachments/747522859361894521/1393666703765934200/image.png?ex=687400c3&is=6872af43&hm=2bd7537e2a5721596a0118d75d0352404c52f2267749e307c220ad39bc8864e3& 11:54:28 <04dracoomega> The mundane specified tries to squash the plusses and such after the item generates, and is... incorrect in two different ways if the item happened to become an artefact. 11:54:51 <04dracoomega> For randarts, it just sets their plusses to 0 (but they keep everything else). For unrands, they become flatly buggy. 11:55:05 <04dracoomega> It's just that mundane isn't used a lot and unrands are very rare 11:56:48 <04dracoomega> Probably ISPEC_MUNDANE just shouldn't be allowed to make an artefact (only to later on fail to make it mundane again properly.) This is probably better than trying to un-artefact-ify it after the fact. 12:08:30 03DracoOmega02 07* 0.34-a0-362-gf73b091665: Don't generate buggy mundane unrands (BlackTorch) 10(63 seconds ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/f73b0916658d 12:08:31 <04dracoomega> BlackTorch: Once your server catches up to the newest update, your game should be playable again 13:56:17 Hi, I'm setting up mingw64 to compile some test versions and following the INSTALL.md instructions 13:57:08 when i put in the ` python3 -c "import yaml" ` command it doesn't give me anything - no error, no message, just opens the console again. 13:57:25 Does that mean it's working? 14:15:10 ??install 14:15:10 install[1/1]: See the following for installation and compilation: https://github.com/crawl/crawl/blob/master/crawl-ref/INSTALL.md 14:15:34 nobody_serenity: if you don't get an error, you're good 14:16:01 ty ! 14:16:12 nobody_serenity: I'd recommend trying WSL over msys2 by the way 14:16:30 we should probably list that as the recommended way to set up compilation 14:16:56 what is it exactly? 14:17:24 the instruction list looks dramatically shorter that's for sure 14:17:36 "Windows Subsystem for Linux" 14:17:52 (which is a Linux subsystem for Windows) 14:18:06 yeah that's the part that got me ngl 14:18:24 "but i'm not using linux. how does it work for me" 14:18:59 it's generally going to give you better compilation times than msys 14:19:52 i see 14:20:10 Also IME it's a lot less painful but then I'm already a Unix beard 14:25:11 i'll try it next time i compile, for sure 14:25:20 also can windows not have the ccache perchance? 14:26:25 WSL is Linux - once you've got it you can install ccache 14:27:25 oki 15:39:27 Unstable branch on underhound.eu updated to: 0.34-a0-362-gf73b091665 (34) 20:05:15 03hellmonk02 07[greathelmet] * 0.34-a0-175-ga376bf2ac6: Orbs of conjuring 10(32 minutes ago, 11 files, 28+ 6-) 13https://github.com/crawl/crawl/commit/a376bf2ac672 20:05:15 03hellmonk02 07[greathelmet] * 0.34-a0-176-g4b17f6b7a2: Orbs of glass 10(3 minutes ago, 9 files, 20+ 3-) 13https://github.com/crawl/crawl/commit/4b17f6b7a246 20:05:15 03hellmonk02 07[greathelmet] * 0.34-a0-177-gbdec800ac2: beautiful tile recolors 10(77 seconds ago, 2 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/bdec800ac2b6 20:05:15 Branch pull/4642 updated to be equal with greathelmet: 13https://github.com/crawl/crawl/pull/4642 22:35:18 Unstable branch on crawl.develz.org updated to: 0.34-a0-362-gf73b091665 (34) 22:57:51 Windows builds of master branch on crawl.develz.org updated to: 0.34-a0-362-gf73b091665 23:21:50 Unstable branch on cbro.berotato.org updated to: 0.34-a0-362-gf73b091665 (34) 23:55:15 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-362-gf73b091665