02:45:47 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-404-g291f19a 03:02:36 Crawl manual still has a paragraph about OOD monsters. 04:25:33 New branch created: pull/641 (1 commit) 13https://github.com/crawl/crawl/pull/641 04:25:33 03Yermak02 07https://github.com/crawl/crawl/pull/641 * 0.21-a0-405-gec4c2c2: Remove "Out of the depths" paragraph from crawl manual 10(7 minutes ago, 1 file, 0+ 13-) 13https://github.com/crawl/crawl/commit/ec4c2c26e4af 06:35:56 -!- amalloy is now known as amalloy_ 06:40:56 for (int idir = 1; idir < 8; (idir += 2) == 9 && (idir = 0)) 06:41:00 Wow, what a dirty hack 07:14:00 Is there any information about which platforms/targets crawl supports? 07:17:25 specifically, i'm trying to wrestle ncurses into returning KEY_RESIZE from getch(), but for some reason crawl has its own SIGWINCH handler 07:29:54 The quote from the code, monster.cc#L281: "Floundering reduces movement speed". But the converse is not true, at least for the player! Large species don't fumble in shallow water, but the player still gets penalty for moving in there. How about we allow large player species move in shallow water without penalty too, for consistency? 07:29:56 "some reason" is crawl needs to know the current screen size so it can redraw properly or abort if it won;t fit the new size. why do you need KEY_RESIZE? 07:30:42 floundering is not fumbling 07:32:29 yes, except that crawl currently only responds to screen size changes when you press a key, not as resize events happen 07:32:31 floundering *is* the speed penalty; water is more viscous than air, fast moves get slowed abruptly. fumbling is you slip because you're not securely planted, and is in some sense the opposite. 07:32:42 yes, it's true. I use bad reasoning. But idea still stands. 07:33:15 oh, is this ther same 'reasoning' as curare in the real world causes asphyxiation but somehow in crawl that is unanticipayed? 07:33:20 no 07:33:53 it's unfair that monster ogres move freely in shallow water and player ogre don't 07:34:14 I think this part was simply missed when coding. 07:34:25 monsters have lots of advantages over players... iood comes to mind 07:35:15 extra power? maybe 07:35:24 but I need KEY_RESIZE to be returned through getch() so I can block on incoming events instead of polling 07:35:54 and that's the behaviour you get by default with ncurses, _unless_ you set a signal handler for some reason 07:36:45 It would be simpler for players to understand if shallow water didn't do any harm when you go from normal species to large, instead of just losing one aspect out of two - what we have now, 07:39:06 aidanh, my recollection is crawl crashed on resize until that handler was added 07:39:21 so if you think it needs to be removed in order for things to work, you get to debug that crash 07:40:05 Ah, well that should be fun 11:19:32 Okay, Hetzner are being annoying about data recovery here. 11:19:56 They want to charge me extra money for adding a third hard drive (which would end up being permanent anyway) 11:20:08 So I'd like to know if anyone has any good alternatives to Hetzner 11:20:16 Because I'm simply going to pull the data out and set up somewhere else. 12:32:05 I need someone to pin this on the forums to the top of the tech support subforum: https://crawl.develz.org/tavern/viewtopic.php?f=9&t=24450 12:46:32 -!- amalloy_ is now known as amalloy 13:43:46 Okay, I've set up a redirect on my server that shows a page with information about what's going on, pinning is less critical. 14:44:34 geekosaur: Hm, not sure what crash you're referring to; I removed the signal handler and there's zero difference 14:45:24 well, crawl shits the bed if you make the window height too small, but it always did that 15:17:35 -!- jefus- is now known as jefus 15:30:32 <|amethyst> %git 502efc8e3 15:30:32 07|amethyst02 * 0.11-a0-1287-g502efc8: Ignore ncurses KEY_RESIZE. 10(6 years ago, 1 file, 3+ 2-) 13https://github.com/crawl/crawl/commit/502efc8e3f68 15:32:16 <|amethyst> just KEY_RESIZE isn't enough, if you change the window size while crawl isn't waiting for a key 15:32:42 <|amethyst> AIUI 15:33:05 <|amethyst> but we were definitely seeing KEY_RESIZE, even with SIGWINCH, before that commit of mine 15:37:51 you were seeing them even with the sigwinch handler? 16:00:36 oh, the reason it gets crashy if you don't use a signal handler is because WIDTH and HEIGHT are updated immediately before get_wch() returns with your KEY_RESIZE 16:01:36 but if you're not on the main view, e.g. in a menu, the resize handler won't call crawl_view.init_geometry(), which updates termsz to match 16:05:44 lines & cols, not width & height