00:01:28 -!- amalloy is now known as amalloy_ 00:18:09 Not many people fielding dev questions today. 01:57:27 Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-85-ga0cb861 02:13:34 -!- amalloy_ is now known as amalloy 02:53:13 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-85-ga0cb861 02:53:26 !messages 02:53:27 No messages for kenran. 04:07:04 New branch created: pull/560 (1 commit) 13https://github.com/crawl/crawl/pull/560 04:07:04 03kenranunderscore02 07https://github.com/crawl/crawl/pull/560 * 0.21-a0-82-g868b70c: Small Lua refactoring for bazaars 10(19 hours ago, 1 file, 31+ 36-) 13https://github.com/crawl/crawl/commit/868b70cf60ac 04:20:00 hm, I wonder if the -90px thing for fullscreen is intended exactly not for fullscreen, so as to have a reasonable window size 04:25:16 OS X is ignoring window size in fullscreen mode anyways 04:28:28 the fullscreen code is kind of weird in general, but I've stayed away from changing it because I can't easily test on non-mac. But on a mac the best bet for real fullscreen that behaves properly is to start windowed, and then use the sdl menu to set fullscreen. 05:07:00 03advil02 07* 0.21-a0-86-gb3e98cc: Ignore negative (relative) window sizes in fullscreen mode 10(4 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/b3e98cc55699 05:07:47 there could be use cases I haven't thought of but that change seems reasonable to me right now [in a slightly jet-lagged state] 05:11:23 I think I;d argue fullscreen and windowed should use different window sizes, since the meaning of "window size" in fullscreen is actually screen resolution 05:11:43 yeah that would be reasonable, PF suggested something like that in his original commit message 05:11:52 conflating them is just asking for badness when you toggle fullscreen 05:12:29 well, I suspect something to do with how fullscreen is set up is just fundamentally broken, w.r.t. toggling 05:13:17 at least based on fullscreen behavior on a mac 05:13:58 macs are just weird about fullscreen, I think. fullscreen likely behaves better on windows and linux 05:14:10 heh I think fullscreen behaves a lot better on a mac :D 05:14:21 since a few OS X versions ago 05:14:21 but you have to set it up right 05:14:40 I suspect that the way fullscreen is currently set up is using some SDL workarounds from like 10 years ago or something 05:16:30 it isn't properly in OS X fullscreen mode when you start it in fullscreen (no menu, no system fullscreen toggle, ignores C-cmd-f, ignores c-left/c-right (which switch between a fullscreen app and other desktops) 05:17:05 if you start in windowed mode, and then use standard OS X stuff to start fullscreen, all of that works 05:18:59 sounds like it's just set up for how Macs worked before a true fullscreen mode was introduced in Lion 05:52:31 -!- amalloy is now known as amalloy_ 06:10:30 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-86-gb3e98cc (34) 06:11:06 I'm currently pondering how to proceed with https://github.com/crawl/crawl/pull/555 06:12:00 I'd love some more feedback on that. should I comment on the PR, or just wait longer (I don't have a feeling for the dev response time, if there's such a thing, yet)? 06:12:58 amalloy e.g. said that he can't help but feel "why are there shops in a lighthouse?". I get that. would removing the ophan and making it into an "island hut" (similar to shoals) improve the vault maybe? 07:48:04 New branch created: pull/561 (23 commits) 13https://github.com/crawl/crawl/pull/561 07:48:05 03jungdai02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1110-g775d74f: 이거 되는건가? 10(7 weeks ago, 4 files, 97+ 52-) 13https://github.com/crawl/crawl/commit/775d74f7aa30 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1111-g0c2b500: ring of wild 10(7 weeks ago, 17 files, 86+ 29-) 13https://github.com/crawl/crawl/commit/0c2b50048c80 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1112-g2b76814: fix text 10(7 weeks ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/2b768144f295 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1113-ga582993: demigod portfolio 10(7 weeks ago, 7 files, 54+ 0-) 13https://github.com/crawl/crawl/commit/a58299375a58 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1114-g884925c: enable traning invocation for demigod 10(6 weeks ago, 2 files, 6+ 1-) 13https://github.com/crawl/crawl/commit/884925cf4f63 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1115-gebbc087: demigod-portfolio.cc 10(6 weeks ago, 7 files, 113+ 8-) 13https://github.com/crawl/crawl/commit/ebbc08732689 07:48:05 03sterilizedmilk02 {GitHub} 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1113-g20bb73f: 작은 오류수정 10(6 weeks ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/20bb73fdeafe 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1116-ga6e1990: poison demigod 10(6 weeks ago, 9 files, 81+ 35-) 13https://github.com/crawl/crawl/commit/a6e19904d86b 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1117-g9837d0c: buff poison got a bit 10(6 weeks ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/9837d0ca4589 07:48:05 03gguo02 07https://github.com/crawl/crawl/pull/561 * 0.20-a0-1118-g5b4be13: poison god 10(6 weeks ago, 2 files, 146+ 0-) 13https://github.com/crawl/crawl/commit/5b4be132c113 07:48:05 ... and 13 more commits 07:55:53 https://crawl.develz.org/trunk/ page has incorrect (Last updated ... ) dates under windows download links 08:19:27 currently "dgn.set_branch_epilogue" is only used for bazaars afaict. if I needed to clear the branch epilogue that was set earlier, what would you think was cleaner: 08:20:33 (1) create dgn.clear_branch_epilogue (2) change set_branch_epilogue such that when setting nil, the epilogue for that branch is then set to nullptr? 08:21:40 well, there's also (3) I could create a lua function "do_nothing" and set that as branch epilogue, but I don't like that 13:11:51 03advil02 07* 0.21-a0-87-g268a196: Fix behavior of tile_full_screen with some associated refactoring 10(3 minutes ago, 4 files, 34+ 25-) 13https://github.com/crawl/crawl/commit/268a196dd2b0 13:56:12 I got an email about restoring a password, and went to change it, but the passwd.db3 file in /chroot/dgldir is... empty? Are passwords handled in another way now? 13:56:15 (on CAO) 14:00:46 -!- amalloy_ is now known as amalloy 14:07:28 kenran: yes, i think removing the ophan is a fine solution. monsters trapped in glass is a gimmick that shouldn't be used often 14:09:37 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-87-g268a196 (34) 14:15:21 thanks amalloy! I'll probably prepare another pr first that refactors some stuff and fixed older vaults, and then update the lighthouse vault. its "markup" should then be a little simpler 14:31:25 03Nikhil Addleman02 {amalloy} 07* 0.21-a0-88-g13b9641: Fix alignment of Items table in crawl_manual.rst 10(2 days ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/13b96418aaa0 14:55:32 thanks amalloy 14:57:12 thanks amalloy 15:09:30 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-88-g13b9641 (34) 15:33:57 -!- Fixer_ is now known as Fixer 17:31:09 the branch epilogue in bazaar.des is actually only needed for halos. the randomisation of the tiles doesn't need to be in there, and when checking older vaults I found some that actually don't want to use it, so I want to make this controllable as well. 17:31:50 that's why I separated colour randomisation and floor halos (though the latter make no sense without the former as it relies on the predefined tilesets) 17:32:10 well I don't see what that has to do with trying to clear a branch epilogue 17:32:16 that comes now :) 17:32:27 you can simply add an argument to control the behaviour of what the epilogue does 17:34:02 phew, it's hard to express myself here... 17:34:21 wait a sec, I'll pastebin the state of the file (everything working as intended now) 17:36:17 https://pastebin.com/1W0aqbih 17:36:36 you want to pastepin a patch 17:36:56 or a diff 17:37:32 git format-patch --stdout origin/master > mybranch.patch 17:37:35 from your branch 17:38:18 does this work for you as well? https://github.com/kenranunderscore/crawl/tree/fix_older_bazaars 17:39:06 no, in that it's just a link to the tree 17:39:10 you need some kind of diff thing 17:39:26 hm yeah I didn't know you didn't have that button, I'm sorry 17:39:32 still new to those features 17:39:52 https://github.com/crawl/crawl/compare/master...kenranunderscore:fix_older_bazaars this maybe? if that doesn't work, I'm going to paste a diff 17:40:17 yeah I found the link 17:40:33 without looking at it too closely, it seems fine, but you don't need to set a do nothing function 17:40:47 if the bazaar epilogue is never set in the first place 17:40:56 I gues it might be relevant for multiple bazaars 17:40:58 there's a problem when you have two bazaars 17:41:00 yeah 17:41:08 but I'd say that it's a bad hack 17:41:10 that's why I was asking, but it was hard for me to get to that point :D 17:41:15 yeah it is 17:41:34 it would be better to simply have the conditioning in the epilogue 17:41:52 or set one of two epilogues like I did 17:42:00 I'd prefer that 17:42:18 actually what you could do I suppose 17:42:22 is call it with an empty string 17:42:29 that should also erase any epilogue 17:42:31 that's checked 17:42:42 what's checked? 17:42:44 empty string, I mean 17:42:51 the one function checks that case 17:43:05 set_branch_epilogue in l-dgn.cc I think 17:44:03 I also thought about maybe just adapting that check 17:44:25 yeah, patching that function to allow setting an empty string specifically 17:44:29 would be a good compromise 17:44:40 to get the behaviour you described (empty string -> erase epilogue; actually I think the epilogues themselves are empty strings to start with when not set) 17:44:57 that's what I told you earlier 17:45:01 it's just an array of strings 17:45:05 thanks! sorry it took me so long to get to the point 17:45:25 so since it's an array of strings 17:45:38 if you set the array entry to a string, the level builder will not attempt to use it 17:47:05 yeah that was my alternative (2) in my initial question; I never planned to keep the do_nothing in case you thought that :D 17:47:16 I just wasn't sure if that was OK 17:47:30 it doesn't affect other things anyway, since nothing else uses the epilogues iirc 17:48:53 I'm not sure why calling floor_halo outside of an epilogue leads to strange things, but I don't want to bother with that as they're working fine otherwise 17:57:28 it's not so much as "outside of an epilogue" as it is likely to be "before the level is completely generated" 17:59:15 that sounds likely 17:59:23 various level postprocessing things happen, so if you instead do it in map placement, those can be overwritten 17:59:57 suspicious functions like _dgn_map_colour_fixup(); 18:00:14 or something happening before the epilogue call 18:00:36 yeah I found all sorts of strange behaviour 18:01:16 if I didn't use bazaar_setup at all but just put the TAGS and KFEATs there myself, then the game crashes whenever one vault sets LFLOORCOL for instance 18:02:09 if that weren't the case, I could have just adapted some (2 to be precise) older vaults to not use it and set their desired colors themselves 18:02:31 Unstable branch on underhound.eu updated to: 0.21-a0-88-g13b9641 (34) 18:06:13 putting tags, kfeats, and lfloorcol in a vault won't cause a crash 18:06:26 those are extensively used in many vaults 18:06:40 it's possible that setting the halo during map placement causes a crash 18:07:14 what the setup function is doing wrt those DES commands is the same as what would be happening if you put the commands directly in the map 18:07:31 only the halo is something that would potentially need running as an epilogue 18:07:32 you're right, but in the setting I had it did. I think I know now what it was: 18:08:30 I hadn't used bazaar_setup at all to find out how things worked. and when I created the vault in wizmode, I had to have entered another vault before being able to use &P, so there was already an epilogue set 18:08:54 and then the setting of the halos clashes with the LFLOORCOL that doesn't fit 18:09:12 why do manticores do almost as much melee damage as emperor scorpions 18:11:44 I would look at the actual backtrace in the morgue (or directly in a debugger) before making too many theories 18:11:57 but the crashes are more related to whatever halos are doing 18:12:06 which should probably be done differently 18:12:43 there's probably no good reason why setting these halos requires the code to be run after map placement 18:13:08 I had a look at the method in question in (I think) tileview.cc, and it's ... not nice to look at :D 18:13:16 s/after map placement/at the end of level generation/ 18:13:49 that would be a crash when the game is running, well after level generation 18:14:01 sounds like tile/color data is simply being set incorrectly 18:14:23 but yeah, if you can't debug further, you can't really know specifically why it's happening at all 18:14:47 @?manticore 18:14:47 manticore (05H) | Spd: 10 | HD: 9 | HP: 40-58 | AC/EV: 5/7 | Dam: 26, 14, 14 | fly | Res: 06magic(40) | XP: 529 | Sp: throw barbs (2d13) [11!AM, 06!sil] | Sz: Large | Int: animal. 18:14:51 @?emperor scorpion 18:14:51 emperor scorpion (15s) | Spd: 10 | HD: 14 | HP: 101-139 | AC/EV: 20/12 | Dam: 3008(poison:28-56), 1509(claw), 1509(claw) | Res: 06magic(60), 03poison, 12drown | Chunks: 14noxious | XP: 1632 | Sz: Giant | Int: animal. 18:16:32 I forget the history of how they were buffed to their current state 18:16:54 they're supposed to be somewhat nasty shoals monsters 18:17:01 but they can show up in lair quite a bit 18:17:30 they're OOD in lair 18:17:45 in dungeon from D:11- 18:22:51 for some reason, default_manual_training = true doesn't seem to get respected when I include it from another RC 18:24:58 -!- amalloy is now known as amalloy_ 18:46:58 -!- xnavy_ is now known as xnavy 19:01:51 <|amethyst> gammafunk: is this a new game? 19:01:59 yes 19:02:01 <|amethyst> hm 19:02:37 works fine when set directly in the RC 19:03:55 it works for me when I include from another RC 19:27:11 New branch created: pull/562 (2 commits) 13https://github.com/crawl/crawl/pull/562 19:27:11 03kenranunderscore02 07https://github.com/crawl/crawl/pull/562 * 0.21-a0-89-g9e22181: Small Lua refactoring for bazaars 10(34 hours ago, 1 file, 31+ 36-) 13https://github.com/crawl/crawl/commit/9e221812da44 19:27:11 03kenranunderscore02 07https://github.com/crawl/crawl/pull/562 * 0.21-a0-90-ga05c852: Fix some old bazaar vaults 10(in the future, 3 files, 84+ 102-) 13https://github.com/crawl/crawl/commit/a05c852071da 19:58:22 Poison and Bribe Branch 13https://crawl.develz.org/mantis/view.php?id=11123 by Lavandula 20:19:53 -!- lobf_ is now known as lobf 20:46:13 -!- GiantOwl is now known as Kalir 21:14:07 -!- lobf is now known as Slim-Charles 21:14:07 -!- Slim-Charles is now known as Slim_Charles 21:15:18 -!- The topic of ##crawl-dev is: Crawl Development | Logs: http://s-z.org/crawl-dev/ | People with +v have commit access. | Please keep general Crawl-related chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to a pastebin service, please. 21:15:20 -!- The topic of ##crawl is: Play Dungeon Crawl Stone Soup online now! Type ??servers for instructions. | http://crawl.develz.org | FooTV: http://termcast.develz.org - ??footv for instructions | See also ##crawl-offtopic 21:48:25 !vault nicolae_ossuary_entry_false_entrances 21:48:26 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/portals/ossuary.des#L183 21:48:46 this ossuary entrance has a high chance of making the portal inaccessible without digging or teleportation 21:49:03 (and when it does it also means you could get stuck in the chamber with the portal, if you teleport there and can't teleport/dig back out) 22:21:37 good tier lists minmay 22:22:00 meh 22:22:02 between this and your spell guide I think I have a much better understanding of how you approach the game difficulty now 22:22:16 he'll never win my approval 22:22:22 nice try though 22:22:24 i wonder if gozag gold distraction could be nerfed at low "piety" somehow 22:22:44 or if it's even necessary. potion pettion comes online p. quickly 22:22:54 I think I'm going to merge Floodkiller's gnoll patch without the stat choice commit 22:23:00 for those who have been following that 22:23:36 Lasty: btw, I'm not sure if I heard from you wrt Gnolls and the current proposal (locked stats at 7/7/7 with no way to change) 22:24:47 MPA was mostly against them having a stat choice allowed 22:25:51 you could give them slowly increasing stats as the game goes on 22:26:01 1/1/1 + 1/1/1 every 3 levels 22:26:55 well 1/1/1 is too drastic, but yeah it's possible to lower their starting levels some 22:27:32 I will probably just merge the stat lock commit (with any necessary fixes) and then we can see about further changes 22:27:43 maybe some devs will have time to play test it a bit 22:35:29 Sounds interesting 22:35:36 I wonder if players will find them fun 22:38:33 is players having fun even in the design philosophy? 22:38:55 explicitly against iirc 22:38:56 _players_? No. 22:38:57 but yeah, FK did mention how he felt pushed especially much into something like summons 22:39:00 Just devs! 22:40:40 I do wonder how much more true that is on gnolls than other species 22:40:49 It's not like summons are normally bad 22:41:42 I also wonder if the ideal expression of this concept -- easy early game, difficult late game, diversified skill -- is actually interesting. 22:41:54 As in, maybe if we pulled this off perfectly 22:42:00 no one would enjoy it 22:58:34 Yes please merge it so i can win this garbage race and get my cheevo 22:58:54 gnsu boys 22:59:35 if stat increase maybe 1/1/1 every 5 or 6 rather than 3 23:03:41 6/6/6 plus 1/1/1 every 6 levels for total of 10/10/10 plus bonus devil cred 23:04:03 the latter being the most important element 23:14:35 -!- pi____ is now known as quechido 23:26:48 checkers, johnstein: can I get a hellcrawl update whenever you have time 23:27:33 havent pushed anything but some bugfixes but I am thinking about setting up some sort of hellcrawl mini tourney or weekly challenges so I can continue to procrastinate on implementing permabuffs 23:32:28 -!- amalloy_ is now known as amalloy 23:59:01 hellmonk: :o updating 23:59:15 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-88-g13b9641 (34)