00:45:31 Monster database of master branch on crawl.develz.org updated to: 0.33-a0-853-g30d10493ae 04:34:02 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 15:06:48 New branch created: reexperience (6 commits) 13https://github.com/crawl/crawl/tree/reexperience 15:06:52 03hellmonk02 07[reexperience] * 0.33-a0-815-gd4950f61e2: xp rework framework 10(6 days ago, 2 files, 79+ 153-) 13https://github.com/crawl/crawl/commit/d4950f61e2c6 15:06:52 03hellmonk02 07[reexperience] * 0.33-a0-816-gf3427fa0d7: Directly specify monster xp values 10(15 hours ago, 523 files, 540+ 342-) 13https://github.com/crawl/crawl/commit/f3427fa0d753 15:06:52 03hellmonk02 07[reexperience] * 0.33-a0-817-g3cf8a8abe1: handle derived undead 10(77 minutes ago, 1 file, 9+ 6-) 13https://github.com/crawl/crawl/commit/3cf8a8abe163 15:06:52 03hellmonk02 07[reexperience] * 0.33-a0-818-g88857c57c3: handle weird monsters 10(33 minutes ago, 7 files, 20+ 18-) 13https://github.com/crawl/crawl/commit/88857c57c3ed 15:06:52 03hellmonk02 07[reexperience] * 0.33-a0-819-gae9ef89b5c: handle gozag bribes 10(27 minutes ago, 3 files, 9+ 5-) 13https://github.com/crawl/crawl/commit/ae9ef89b5c42 15:06:52 03hellmonk02 07[reexperience] * 0.33-a0-820-gc07793a5a9: better legacy handling 10(67 seconds ago, 3 files, 17+ 16-) 13https://github.com/crawl/crawl/commit/c07793a5a989 15:09:00 New branch created: pull/4300 (6 commits) 13https://github.com/crawl/crawl/pull/4300 15:09:00 Branch pull/4300 updated to be equal with reexperience: 13https://github.com/crawl/crawl/pull/4300 16:10:53 <06mumra> @gammafunk gammafunk_temple_of_torment is detected as a teleport closet because it uses portals (and placement.lua doesn't recognise them as connectivity) looking at the vault, should the whole inside just be no_tele_into or is it expected someone might land in there and deal with it ? 16:14:07 <06pleasingfungus> https://herecomesthemoon.net/2024/11/two-factions-of-cpp/ trying to decide whether we count as "remotely modern" or "legacy cpp" per this article's terms 16:14:28 <06pleasingfungus> not sure, but i have a bad feeling... 16:16:19 <06mumra> https://cdn.discordapp.com/attachments/747522859361894521/1338649780737409076/image.png?ex=67abda42&is=67aa88c2&hm=e419295cf1bda8cf951e108a9beb06e44b912f52d6a55299a2a8993f6a03c3fb& 16:16:32 <06mumra> The capitalisation on "remotely" swings it completely i feel 16:19:53 <06mumra> And particularly the following paragraph describing "legacy" > Any C++ that relies on some utterly ancient chunk of compiled code, whose source has been lost, and whose original authors are unreachable. and > Any code where building any used binary from source requires more than a few button presses, or is straight-up impossible. crawl is actually pretty well-maintained and reliably buildable on multiple platforms without mystic 16:19:54 incantations 16:20:41 <09gammafunk> our general policy has been to not use no_tele_into for normal content except for ghost vaults (so players aren't ever forced to fight ghosts) and certain extreme cases. So no, it shouldn't have no_tele_into by our current policy. I think if placement.lua can't be made to understand transporters, it's probably best to just put such vaults in the list of exceptions 16:21:36 <09gammafunk> We are talking about revisiting the idea of no_tele_into for transporter and runed door vaults, but there doesn't seem to be concensus on that yet 16:21:42 <06mumra> yep, it has a list of exceptions, which i already added a harmless vault to 16:22:14 <06mumra> but on the way i've seen another vault that has no_tele_into already (just missed a glyph) but possibly shouldn't, according to this definition 16:22:45 <06mumra> hence why i checked anyway 16:22:49 <09gammafunk> yeah, like I said there are some exceptions for very extreme cases, like very early challenge vaults or vaults that place really weird terrain 16:22:51 <09gammafunk> what was the vault? 16:23:46 <09gammafunk> it's definitely not an ideal situation in that we can't programattically apply a ruleset we'd like nor show in the UI to the player when no_tele_into is relevant 16:24:05 <09gammafunk> but ghost vaults are all moving to a new portal (maybe next version, not sure if that's getting completed in 0.33) 16:27:15 <06mumra> actually, i was incorrect there ... i was thinking of grunt_qazlal_sundering but actually, the no_tele_into kprops make sense there, i just can't work out why it's getting disconnected zones approx 1/25 times 16:30:50 <06mumra> a couple of good improvements to placement.lua to go on the backlog - dump a screenshot of the map on an error, it's very hard to work out sometimes what could have actually gone wrong (and there are many exceptions already in the script that would be much easier to fix properly with a screenshot) - handle portals properly does anyone know if there's an existing lua method to dump the map in txt form? 16:31:24 <06mumra> screenshot() used by crash reports doesn't really work since it's only the player's view and there isn't a player in this instance 16:35:08 <09gammafunk> it'd be nice if the builder on c++ side could just correctly determine transporter connectivity 16:35:50 <09gammafunk> the travel algorithm can already use transporters with its floodfill approach and I think the builder is using a floodfill that could be adapted in the same way 19:35:28 03mumra02 07* 0.33-a0-854-g9c2e5f8391: Fix teleport closets and connectivity in numerous vaults 10(5 minutes ago, 11 files, 48+ 25-) 13https://github.com/crawl/crawl/commit/9c2e5f8391b8 20:57:25 I'd say crawl counts as remotely modern 20:57:38 C++11 is pushing it on that particular aspect, but you are on C++11 20:57:48 And you have checked collections etc by default 20:57:58 which even much of the "modern C++ world" continues to have head in sand about 20:58:44 You have a linter set up 20:58:47 and CI 20:59:20 The make files are very old school and not idea 20:59:22 ideal 20:59:27 but they are reliable enough overall 23:00:40 <06pleasingfungus> yeah, the make situation was what had me leeriest 23:00:50 <06pleasingfungus> but i agree we are Probably Technically Basically Modernish 23:00:55 <06pleasingfungus> 🎊 23:07:32 There's a PR that tries to get us into the cmake era 23:07:47 I need to see if I can drag that over the line sometime maybe, idk 23:11:38 FWIW, fixing builds and refactoring code is kinda of my day job, just a little bit :D 23:11:43 And something I enjoy 23:12:08 But how much I'm likely to do is inversely proportional to how hard it is to get PRs merged :P 23:14:20 *cough* at least 3 ci and build PRs are ready for merge https://github.com/crawl/crawl/pulls/Cgettys *cough* :P 23:21:57 <04dracoomega> I do apologize for how little I've been able to get to PRs in recent days (though honestly, a few of those are outside my own comfort zone anyway. Though I'm not sure who's they're in that's active) 23:23:00 <04dracoomega> I am still hammering away at undead species stuff over here 23:24:08 <04dracoomega> I did plan to go through a handful of backlog of things at some point before 0.33 is over (though I'm also behind schedule on talismans, aaaaaa ^^; ) 23:24:51 Wasn't intended as a critique of any particular dev or anyone. I know we all have lives (ok, maybe I don't :P), just lightheartedly following up 23:25:43 <04dracoomega> No, I mean, it's a fair point. It's useful work, and it never feels good to just let it sit there. There's some pretty big things that have sat for a while that I really do need to bite the bullet on in some form or another 23:26:10 New branch created: pull/4301 (1 commit) 13https://github.com/crawl/crawl/pull/4301 23:26:11 03Cgettys02 07https://github.com/crawl/crawl/pull/4301 * 0.33-a0-854-gebea520e8e: fix: unistd.h inclusion for macOS 10.14 10(2 minutes ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/ebea520e8eed 23:27:39 I'm also happy to go thru PRs and try to triage (I've already left comments on some I've seen go by). But that's only useful insofar as one trusts my judgment in some sense :D 23:35:16 Unstable branch on crawl.develz.org updated to: 0.33-a0-854-g9c2e5f8391 (34) 23:35:45 <06pleasingfungus> i'm going to one-up the 62 item slots PR with a 63 item slots commit 23:37:18 I'll 1 up that with a 64 item slot PR :P 23:38:18 more seriously though @pleasingfungus, won't that break q1 / r1/etc type bindings? 23:39:11 right, it does from wha tit says 23:39:23 don't think it's end of world but will miss that personally 23:39:50 I have r1 bound to identify, then Home and end bound to identify options and scrolls respectively via macros :D 23:48:37 <04dracoomega> That is part of why it wasn't done. (Though a few weeks ago, there was some extended discussion on the subject in #dcss that ended up with an alternate proposal that I am personally fond of, which was separating 'consumable inventory' and 'equipment inventory' altogether, which I think has a number of advantages: -There are literally not enough consumable types in the game to use up all letters, so the player could literally never 23:48:37 run out of room. -Not competing with equipment letters means that you could have consumables auto-bound to specific inventory letters every game, if one wanted (without needing inscriptions or anything.) -There are already virtually no actions for which a consumable or a piece of equipment are a valid action anymore (since breadwielding was removed). Dropping is nearly the only one (and there's no need to drop consumables if you can never overflow the 23:48:38 limit anyway) -By making this much space, that would pretty much be the final word on fighting against the inventory limit forever, I think. (Some people have argued that 62 wouldn't actually help a lot over 52, though other people think it would. But this definitely would.) 23:49:10 <04dracoomega> (Non-trivial implementation refactoring that I'm in no rush to do right now, but I could totally see trying for 0.34) 23:51:15 Right, I'm a big fan of the consumable inventories idea 23:52:41 <04dracoomega> Oh, you might have been there for that, I guess 23:52:49 <04dracoomega> (It's hard to remember these things ^^; ) 23:53:47 Right, been a busy few weeks, think I was 23:54:03 dunno if I was first but I proposed some version of that :D 23:54:04 https://discord.com/channels/205277826788622337/205316046230388737/1328299094237515827 23:54:26 Perhaps another interesting orthogonal question is, do all slots have to be the same / do all binds have to be one letter. let's say you made all potions pa-z, scrolls sa-z, so you'd do say dpa to drop your first potion.Ta-da, 3 inventories (scroll, potion, everything else) and it might be cleaner?idk, just somethign that popped into my brain at 2amBut feels like it'd get rid of inventory problems a lot, even if you cut down the normal inventory to 26 it' 23:55:53 <04dracoomega> Not entirely clear to me that there's meaningful advantages of this over a merged consumable inventory or not. I guess you could use the same letter for 'quaff x' and 'read x' for different items? 23:57:50 Windows builds of master branch on crawl.develz.org updated to: 0.33-a0-854-g9c2e5f8391 23:57:52 Oh, merged inventory tottaly makes more sense 23:57:59 I was just quoting my contribution to that discussion 23:58:28 <04dracoomega> Oh, okay ^^; 23:58:28 *merged consumable inventory I mean 23:59:07 I don't know if I was first with that basic idea 23:59:15 but I dropped that idea at like 2am then went to bed :D 23:59:46 Then missed most of the follow-up conversation over the next few days that seem to have refined that idea to your above summary