01:10:18 termcast.develz.org seems to be frozen on a request by Pacra 01:14:45 yeah I sent an email to snark about it 01:19:58 Unstable branch on crawl.develz.org updated to: 0.21-a0-364-gbd1fde4 (34) 02:00:19 Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-364-gbd1fde4 02:54:00 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-364-gbd1fde4 03:11:46 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-364-gbd1fde4 (34) 04:16:08 hey all, I'm working on getting the tiles viewport working with arbitrary pixel sizes (showing partial tiles at the edges), and I'm having issues with TileRegion's dx and dy being used for both resizing and render scaling 04:18:37 Currently I'm expanding the viewport region and shifting with a negative offset, but this approach is causing issues with inside(), so there are issues like viewport tooltips appearing spuriously 04:19:43 I'm wondering if it's better to make Region::inside() virtual? Or if it's better to set the viewport dx and dy = 1, and then re-implement the rendering code? 04:33:15 hrm 04:33:39 aidanh: would this have any implications for webtiles? 04:36:31 gammafunk: From the brief look I had at tileweb.cc, it doesn't use any of the tilereg-foo.cc classes 04:37:37 The changes I'm thinking of would be in tilesdl.cc and tilereg-dgn.cc, so webtiles probably wouldn't be affected 04:38:13 yeah, I suppose webtiles wouldn't know about those classes at all 04:44:07 aidanh: I can't give you a very good answer about that; It does seem odd to change dx and dy when all of the other regions use the TILE_X, TILE_Y values for their tile display 04:44:36 making it virtual seems cleaner to me but I don't know what the practical issues are for the tiles code 04:44:58 no one understands the Tiles code especially well 04:45:13 I recommend you ask |amethyst this question, as he might have a better idea 04:45:28 you can send him that question with !tell 04:45:43 e.g. !tell |amethyst MESSAGE 04:46:16 if you mention him he'll probably also see it 04:47:30 I suppose you sort of really are using dx = 1, dy = 1 conceptually, but it's probably that the class is not quite generalized enough to handle partial lines in the first place 04:47:38 s/lines/tiles/ 04:52:06 gammafunk: alright, thanks; I'll see what I can do 06:01:49 Skill target setting is buggy 13https://crawl.develz.org/mantis/view.php?id=11258 by NormalPerson7 07:24:22 -!- amalloy is now known as amalloy_ 07:52:54 -!- amalloy_ is now known as amalloy 08:33:10 03advil02 07* 0.21-a0-365-g814e052: Wipe out text input prefill more thoroughly (11258) 10(2 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/814e05216aec 08:33:39 surprised no one reported that yet 08:33:56 guess no one plays webtiles 08:49:01 aidanh: out of curiosity, what OS are you developing on? 08:49:17 advil: linux 08:49:31 ok 08:50:05 I can't really offer any advice for your question earlier, except to offer the worry that some of that stuff may be a bit OS sensitive 08:50:30 scaling, specifically 08:50:48 we don't really have anyone who even builds tiles on windows right now :-/ 08:52:39 Hm... I would've thought SDL would take care of most of that stuff 08:52:55 Or do you mean high-DPI scaling? 08:53:11 yeah, DPI scaling was tricky to get working 08:54:25 there are also some reported bugs that may still be related on windows that we don't have a good way of trying to reproduce 08:55:25 maybe driver related 08:56:00 anyways, the general line of your changes looks great, though I haven't looked at the details 08:56:25 Driver/platform details always suck :/ 08:56:40 Glad to hear 09:09:17 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-365-g814e052 (34) 11:07:24 advil: do you happen to know what ports the tiles SDL implementation is used on? 11:07:39 it is used on all 3 11:07:49 win/mac/linux 11:08:24 the only build target that isn't fully cross-platform is webtiles, which doesn't run on windows (but isn't designed to be run locally anyways) 11:08:41 but not android? 11:08:47 oh it might run on android 11:08:51 we don't maintain android 11:09:03 well that makes my life easier then :) 11:09:30 there is an android build that compiles but I'd be a little surprised if it really works; the best android app is a wrapper around webtiles over internet 11:10:07 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-362-gb87a002: Fix sidebar monster/inv background flicker on resize 10(2 days ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/b87a00276c2b 11:10:07 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-363-g5272efc: Fix tiles minimap viewport indicator leaving the sidebar 10(2 days ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/5272efcd9e28 11:10:07 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-364-g43569ea: Improve layout+spacing of tiles main dungeon view 10(2 days ago, 3 files, 26+ 6-) 13https://github.com/crawl/crawl/commit/43569ea1abe5 11:10:07 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-365-g91492a8: Stick message region to the bottom of the window 10(19 hours ago, 1 file, 5+ 0-) 13https://github.com/crawl/crawl/commit/91492a8beaf9 11:10:07 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-366-g6773b0b: Make DungeonRegion::inside() work correctly 10(3 hours ago, 3 files, 7+ 1-) 13https://github.com/crawl/crawl/commit/6773b0bfc06d 11:10:07 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-367-g141d5d0: Fix a few tiles rendering glitches 10(21 minutes ago, 1 file, 24+ 6-) 13https://github.com/crawl/crawl/commit/141d5d01bc9b 11:10:10 the small window mode is used outside android though 11:10:29 small window mode? 11:11:33 Ah, yeah 11:14:34 it was originally written for touchscreen devices but some of it impacts layout in tiles, I think the main thing that may interact with your stuff is the message overlay setting 11:14:56 !source tilesdl.cc:990 11:14:57 https://github.com/crawl/crawl/blob/master/crawl-ref/source/tilesdl.cc#L990 11:15:49 can't remember what other hud/UI things change, maybe some of the lower-right stuff 11:16:17 (most of that stuff in the lower right on local tiles is really neglected, because it doesn't exist on other build targets) 11:17:06 ah right, `git grep use_small_layout` will help 11:17:21 The overlay mode wasn't too bad actually; I mostly ignored that, and it all seems to work fine 11:18:39 the biggest problem I've had is the very ad-hoc way in which screen redraws are requested 11:19:26 The paradigm seems to be 'add set_need_redraw() wherever you need' 11:20:38 sounds about right 11:20:45 some of the redraw triggers are for both tiles and console, fyi 11:21:55 oh that sounds fun 11:23:15 yeah, doing ui stuff in crawl is a unique challenge given the diversity of the 3 build targets :) 11:23:36 luckily with the goals you seem to have, you can just ignore webtiles 11:23:57 Yupp 11:26:39 I think my PR is mostly complete; I was also interested in the possibility of sprucing up some of the other UI screens as well 11:27:11 for a tiles build, that is 11:28:37 but it seems that each ui layout has its own imperative implementation? 11:32:08 what do you mean by imperative implementation? Maybe: all the menus are laid out seperately; there's also a lot more shared code between console/local tiles for menus than for the main layout I think 11:32:25 well, the HUD is shared in the main layout 11:35:37 how do you trigger the bugs that 141d5d0 is fixing? 11:39:33 maybe https://crawl.develz.org/mantis/view.php?id=967 ? 11:40:09 the alt text bug? 11:41:06 get a monster in los, then move the mouse cursor over the monster's sidebar icon; its alt text should appear 11:41:41 move the mouse up off the sidebar monster row, about 50px into the minimap region, and the alt text will disappear 11:41:52 move the mouse back down over the monster, and the alt text *won't* reappear 11:42:31 that ancient bug report may be the cursor bug, not sure 11:43:18 I haven't seen that bug, I don't think 11:44:15 "Fixes cursors still being shown on the dungeon viewport even when the cursor has left the dungeon region." is what I was wondering about 11:44:25 Ah 11:45:09 Mouse over a monster in the sidebar, and a cursor will be shown over it in the dungeon viewport; move the cursor down and off the sidebar, and the cursor is still on the monster in the viewport 11:48:51 ah thanks 11:48:51 by 'down and off the sidebar' I meant 'move the mouse straight down and off the monster bar' 11:49:15 maybe be related to 967 11:49:51 I sometimes include replication instructions for stuff like that in commit messages if there's not a mantis issue 11:50:30 they can be really useful like a year later when someone's puzzling over the commit 11:50:51 Right; I'll amend the commit 11:51:00 well, not a big deal 11:51:14 if you need to do some rebasing later it could be something to add though 11:51:27 Ok 11:52:12 now it's in the dev logs so someone could find it 11:57:29 in general we do tend to put a bit more detail in the commit messages (with PR comments not usually being that easy to find after a PR is merged), e.g. 43569ea1abe5 could have some of the information in the PR comment 11:57:32 here's an example 11:57:35 %%git a058ae749a5284471e7911be9b99eafb9b5a80ec 11:57:41 %git a058ae749a5284471e7911be9b99eafb9b5a80ec 11:57:41 07advil02 * 0.21-a0-32-ga058ae7: Don't display local tiles tooltip on out of view cell 10(5 months ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/a058ae749a52 11:57:50 (you may have actually fixed the thing I couldn't figure out there) 11:59:36 or 11:59:38 %git 8880629c5b67f052fbbdc78c1e18eb5524268f44 11:59:38 07advil02 * 0.20-a0-867-g8880629: Check map bounds for local tiles right-click (10600) 10(7 months ago, 1 file, 7+ 2-) 13https://github.com/crawl/crawl/commit/8880629c5b67 12:00:31 Sure, makes sense; I'll add some notes 12:01:35 I'll try to take a look at the actual code at some point today :) 12:09:32 Sounds good :) btw, after b87a002 there's nothing using m_flavour anymore; should I just delete that code? 12:23:18 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-362-gf2c7069: Fix sidebar monster/inv background flicker on resize 10(2 days ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/f2c7069f9c16 12:23:18 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-363-g3d44c76: Fix tiles minimap viewport indicator leaving the sidebar 10(2 days ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/3d44c7646a51 12:23:18 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-364-g86c7c90: Improve layout+spacing of tiles main dungeon view 10(2 days ago, 4 files, 38+ 7-) 13https://github.com/crawl/crawl/commit/86c7c90f975f 12:23:18 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-365-g77dd2fa: Fix a few tiles rendering glitches 10(2 hours ago, 1 file, 24+ 6-) 13https://github.com/crawl/crawl/commit/77dd2fa73bcd 12:23:37 hmm if I understand that code right it was randomly changing the flavor of wall/floor tiles on resize? with that change does it just use the default wall/floor tile all the time? 12:24:26 I'm not sure why it would change them on resize, but randomizing the wall/floor tiles somehow still seems like a good thing 12:27:12 I guess it had to change them on resize because the grid size could change 12:28:14 but I wonder if there's a way of doing that that keeps the randomness and doesn't flicker 12:31:45 it uses the current index for flavour instead now 12:32:24 So its not just the same 32x32 image for each tile (although it is the same across restarts) 12:33:48 If you'd like it to vary randomly, the obvious solution is to have m_flavour_seed be a random number used to seed a sequence of flavour offsets 12:33:56 ah I see 12:34:06 it's even in the new commit message :) 12:34:23 ;) 12:34:54 I guess if it looks fine then that works 12:35:15 then the answer to your earlier question is, yes, eliminate m_flavour 12:36:23 Ok 12:37:22 separate question: should blowgun+poison-needle targeting ignore phantoms? seems like it should, as they're poison-immune 12:59:53 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-362-g5ceeddb: Fix sidebar monster/inv background flicker on resize 10(2 days ago, 4 files, 4+ 23-) 13https://github.com/crawl/crawl/commit/5ceeddb97a8c 12:59:53 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-363-gb2181da: Fix tiles minimap viewport indicator leaving the sidebar 10(2 days ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/b2181da973cf 12:59:53 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-364-g8c76618: Improve layout+spacing of tiles main dungeon view 10(2 days ago, 4 files, 38+ 7-) 13https://github.com/crawl/crawl/commit/8c76618138a2 12:59:53 03Aidan Holm02 07https://github.com/crawl/crawl/pull/620 * 0.21-a0-365-g8a0fa41: Fix a few tiles rendering glitches 10(2 hours ago, 1 file, 24+ 6-) 13https://github.com/crawl/crawl/commit/8a0fa41a65e4 13:04:15 03advil02 07* 0.21-a0-366-g9170151: Prevent cursor from disappearing on right arrow in skill menu (10760) 10(5 minutes ago, 2 files, 2+ 0-) 13https://github.com/crawl/crawl/commit/917015170dac 13:05:09 Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-365-g814e052 (34) 13:46:42 are any of you going to the RL thing in SF next month? 13:55:32 I think amalloy might be? 13:56:24 or was thinking about it, at least 13:56:49 oh, yeah he's on there: https://roguelike.club/ 14:01:55 I'll be there *if* I'm in town for work 14:09:18 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-366-g9170151 (34) 15:20:19 Butchering behaviour for multiple corpses is unfair 13https://crawl.develz.org/mantis/view.php?id=11259 by NormalPerson7 15:30:47 ^FR remove butchering 16:09:50 https://crawl.develz.org/tavern/viewtopic.php?f=8&t=24402 16:31:32 I imagine that at this point butchering is just going to wait for food removal 17:44:42 03elliptic02 07* 0.21-a0-367-gfeb02eb: Fix title weirdnesses when two skills are tied. 10(27 minutes ago, 2 files, 21+ 17-) 13https://github.com/crawl/crawl/commit/feb02eb07801 17:44:42 03elliptic02 07* 0.21-a0-368-g45f8505: Prevent random switching of gnoll titles when their top skills are tied. 10(24 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/45f8505ef0cd 17:48:37 I think the only thing remaining on my gnoll cleanup to-do list is making the skill training messages less spammy by combining them ("10 of your skills increase to level 7!") but doing that seems sort of tricky 17:49:13 so if anyone else is interested in doing that, please go ahead! 17:58:23 is there a way to build crawl to get the tiled AND the terminal version? So one can start it with `crawl --tiles` for example? 18:01:48 <|amethyst> fgerling: there is not 18:02:04 <|amethyst> fgerling: you can do webtiles and terminal in one binary, but not local tiles 18:02:43 Oh, okay. Thanks for the info. :) 18:07:12 Unstable branch on underhound.eu updated to: 0.21-a0-368-g45f8505 (34) 18:09:29 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-368-g45f8505 (34) 18:31:32 tiles and console binaries can coexist though 18:31:57 "make && cp crawl crawl-console && make TILES=yes" would work 21:00:56 -!- Rast- is now known as rast 21:32:02 -!- mikee__ is now known as mikee_ 22:44:54 Cannot train Polearms: Manual of Polearms + Okawaru's Heroism likely culprit 13https://crawl.develz.org/mantis/view.php?id=11260 by djt190ps4 22:47:13 relieved to see that is a 0.20 game 22:47:22 but that save is definitely bugged 22:50:08 no blog posts since 20.1 making me read changelogs 22:50:26 yeah, I'll do one eventually 22:50:40 isn't it almost tournament and release time 22:51:02 I would say not 22:51:22 I guess not then 22:51:25 at least I have a couple things I want to get in, definitely no tourney/release in october 22:51:39 well maybe I have time to write a patch if I have time in my life to do such a thing 22:51:43 advil: it just can't train polearms because it isn't carrying one 22:52:00 oh duh 22:52:20 ah, having a manual enables training 22:52:28 IMO that's the bug 22:52:30 I uh hope they realize that a lajatang is not a polearm 22:52:49 (that manuals enable training) 22:52:58 yeah, plausible 22:58:13 bug report: blog post said "See you all in the Hall of Zot!" but i have not yet encountered a gammafunk unique please fix thanks 22:59:20 at least I can't think of a good reason why manuals let you train skills, I vaguely remember this being the case since the start of the training restrictions though since "get manuals of invocations to get an invocations title on a non-invo god" was a theoretical thing 23:02:14 I guess if you find a manual of shields early on the right char you might start using it 23:08:04 I mean, if that's true then it is probably also true that some people might train X skill without the manual 23:09:18 the only reason I can think to allow this is so that people won't be unhappy/confused if they find a manual and can't figure out to use it 23:09:57 at the moment we don't even let people train necromancy skill just for a pain weapon 23:10:11 or fire just for a staff of fire 23:12:41 actually I suspect this is just a legacy of the time when manuals required actively turning them on 23:22:33 03advil02 07* 0.21-a0-369-g397cdb2: Make the manual bonus match the help (11139) and simplify 10(4 minutes ago, 1 file, 2+ 13-) 13https://github.com/crawl/crawl/commit/397cdb2507d8