00:22:12 create some secret lore by calling it Stafflebaf 00:22:40 sorr, Stafflbat 00:22:45 *sorry 00:44:47 alexjurkiewicz: thanks! mostly just "check the commits" unfortunately 00:45:29 things like extra species should be easy enough to integrate, but if there's any significant changes then integration might be more involved 01:21:17 mostly extra species. But also I've wanted to implement scrolling/second page/something similar for backgrounds, and an additional menu after the existing ones for "game modifiers" 01:24:50 advil and I were discussing some changes of that nature a few hours ago 01:25:42 game modifiers being things like seeds/pregen presumably? 01:34:02 I'm not entirely sure what you mean by scrolling/second page for backgrounds, since the backgrounds page does now scroll :) 01:37:47 Unstable branch on crawl.develz.org updated to: 0.24-a0-716-g2d0fce6 (34) 02:01:04 Windows builds of master branch on crawl.develz.org updated to: 0.24-a0-716-g2d0fce6 02:24:47 03Aidan Holm02 07* 0.24-a0-717-g5a4ac4e: Remove debug logging 10(74 minutes ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/5a4ac4e04deb 02:24:47 03Aidan Holm02 07* 0.24-a0-718-g8282249: Refactor memorise menu entry colour code 10(69 minutes ago, 1 file, 17+ 16-) 13https://github.com/crawl/crawl/commit/828224901cbf 02:24:47 03Aidan Holm02 07* 0.24-a0-719-g2d00395: Fix memorise menu entry letter colour 10(68 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/2d00395b2e96 02:24:47 03Aidan Holm02 07* 0.24-a0-720-gbbc45b0: Fix newgame menu highlight width on remote console 10(5 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/bbc45b03fdac 02:32:55 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-720-gbbc45b03fd (34) 02:53:30 fyi Battlestaff is an anagram of Flat Tab Fest 02:59:05 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-443-g80245de 03:31:58 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-2461-g77d781cab4 04:11:20 runewalsh (L27 OpWz) ASSERT(attacker) in 'attack.cc' at line 354 failed. (Zig:26) 07:19:19 aidanh: oh scrolling already built in!! nice! 07:19:48 aidanh: game modifiers are yes like seeds. for bloatcrawl it will also be things like "disable rune locks" and what have you 07:20:27 :) 07:21:23 right, that'd be doable, but bear in mind that the current startup / newgame code is pretty horrendous right now 07:22:56 don't worry, i'm well aware after adding a bunch of modifiers 07:23:12 heh 07:23:53 are the options mostly binary modifiers? I'm planning to add a checkbox widget at some point 07:24:33 half are binary, half are ternary, at least one I wish could take a user input number 07:25:21 if you want to see the full details, I can PM you an ascii 'screenshot' of the menu 07:25:45 that'd be helpful, yeah 07:26:22 my half-formed plan is to have field focus, and allow switching between fields with tab and shift-tab 07:30:11 hm, merged pretty easily. I guess I could rewrite the menu with the new style but it's optional. Really the only conflict was the jobs list and some minor alignment tweaks we had made 07:30:18 * alexjurkiewicz enables an extra 12 backgrounds 07:34:48 sweet 07:36:30 I take it that the hotkey-based approach for toggling options is born out of necessity 07:36:37 or is that a deliberate design decision? 07:36:51 my tab-based idea would admittedly require more keypresses 07:39:44 definitely the former, I think more keypresses is fine because it would make the UI much more conventional 07:42:59 took a while to figure out the reason this merge wasn't building was because it kept around both the old and new species_group::attach 07:51:49 aidanh: how do I dprf a shared_ptr? %s'ing it or .c_str()'ed doesn't work 08:14:47 aidanh: adding 20 items to a single column in the jobs view seems to cause a crash: ASSERT(y >= 0 && y < m_height) in 'outer-menu.cc' at line 226 failed. 08:15:07 to a group starting right at the top: coord_def(2, 0), 15, 08:19:02 aidanh: I'm also interested if the code supports using the numbers 0-9 after runing out of letters, which may soon be a concern. I can't quite find the code which allocates hotkeys to menu items 08:30:12 alexjurkiewicz: text->get_text() will return a formatted_string 08:31:28 the assertion is because the menu width and height are fixed when it's constructed 08:33:21 !source newgame.cc:1223 08:33:21 https://github.com/crawl/crawl/blob/master/crawl-ref/source/newgame.cc#L1223 08:35:34 you specify the hotkey, if any, so you can use whatever scheme you'd like 08:43:30 hm, so I could disable hotkeys for additional backgrounds beyond the 62nd 08:46:40 i'm not sure what you mean about the assertion. Is that a hint on how to fix it? 08:48:11 ah, sorry 08:48:20 !source newgame.cc:1089 08:48:21 https://github.com/crawl/crawl/blob/master/crawl-ref/source/newgame.cc#L1089 08:49:47 e.g. here the width and height for the species and job menus are set to 3 and 20 respectively 08:51:14 although I'm not quite sure why the height is 20 actually.. 08:52:57 ah. So I'm a little confused about how the menu height works with display and footers and such. eg if I set the height to 100 and just add a couple more rows to what I have, does the empty space appear to the user? Does it push the footer off the page unless the user scrolls to the bottom? Or is this number more about internal memory management 08:56:21 internal memory management; it shouldn't appear, but it will use more memory than necessary 08:57:03 there's internally a vector of width*height pointers 09:01:51 thanks. it works!! in tiles mode too 09:02:51 is there any way to have header text for a column (or better: section) not scroll off the top of the screen, if the section has more items than fit into one window? 09:03:00 now you gotta go make some snazzy tiles for all those species and backgrounds 09:05:12 that'd be tricky, but probably doable... headers are on a grid, so you'd need to take them out of the grid and float them 09:07:59 that's probably a feature that the regular menus could use as well 09:10:02 i noticed the menu was much smaller in local tiles than terminal, since terminal can use for full terminal height. I haven't tested webtiles 09:12:04 could you upload a screenshot? 09:13:09 sure are you on discord? 09:13:09 er nope 09:14:08 https://cdn.discordapp.com/attachments/178440102454886400/628217989451218946/Screen_Shot_2019-09-30_at_11.13.20_pm.png 09:15:04 i wonder if this is dependent on resolution, actually 09:15:21 that... looks normal to me 09:16:43 is that full-screen? the popup has a margin of 50px above and below 09:17:37 yeah it is 09:17:44 the margin should probably be lerped from 10px to 50px based on screen-size 09:18:47 the limiting factor here is probably the tiles, but you could set a top and bottom margin of -1px on them 09:19:12 that'd be tile->set_margin_for_sdl({-1, 0, -1, 0}) 09:20:04 i might also set a wider layout for tiles 09:20:21 more columns. That will be annoying to code though, since I'd basically need two layouts 09:22:14 well whatever you do, don't follow in the footsteps of the abomination that does the small hud layout 09:23:03 never fear, bloatcrawl is full of the most maintinable and properly indented code 09:23:03 although I suppose in this case it'd be a direct coord_def -> coord_def conversion, with no state machine insanity 09:23:31 heh 09:23:59 actually there are a few things in bloatcrawl i'm very proud of the implementation of, like making undead state applicable to any species and a shapeshifter player species 09:36:00 hm shapeshifter player sounds fun 09:50:55 03Aidan Holm02 07* 0.24-a0-721-g5fed591: Reduce popup padding on small screens (alexjurkiewicz) 10(5 minutes ago, 2 files, 14+ 4-) 13https://github.com/crawl/crawl/commit/5fed5919c2cd 10:12:40 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-721-g5fed5919c2 (34) 10:31:38 aidanh: looks like maybe the main menu grid margin for console should be `grid->set_margin_for_crt({1, 0, 1, 0});` instead of `grid->set_margin_for_crt({1, 0, 2, 0});`? 10:32:08 with the current values at 80x24 I'm getting the game choice menu scrolling, and the save menu at the min size 10:32:24 not sure if there's something I'm missing though 10:34:09 I guess I'm also inclined to remove the "Options read from..." from that screen 10:34:09 you can see it with ctrl-p, though that's not very obvious to non-dev users 10:34:47 or maybe it should appear in the bottom status slot when you first start? 10:34:51 hm yes it looks like you're right 10:35:28 well, definitely about the menu scrolling; I haven't checked the margins yet 10:35:42 it looks right to me at 80x24 and 80x25 10:36:41 I wonder if "Read the instructions for legal details." is really necessary 10:37:35 I have some old mockups somewhere that I need to update 10:37:53 but I was basically planning to spirit all that away to an "About" screen 10:38:14 // Eventually, this should be something more grand. {dlb} 10:38:58 five years ago? ten? 10:39:56 wow, almost ten years 10:40:00 heh 10:40:01 the init file thing I want to change (a) because without a linebreak it's useless, and (b) because it's messing up window width on tiles with ling paths, which happens on windows 10:40:10 *long 10:41:41 right; I think it doesn't necessarily belong on that screen in the first place, but another approach is just to hard wrap it by inserting newlines every 80 chars or whatever 10:42:07 yeah, I think that's what happened before? 10:42:40 yeah 10:42:43 it'll mess up all the newly-regular spacing in console though 10:44:13 not sure what you mean 10:44:53 at 80x24 that'll typically force the save browser to be 2 lines I think 10:44:59 before, the linebreak spilled over into the line above the name box 10:45:09 oh, I see, yes 10:45:12 which I'm assuming your ui code will (very sensibly) keep from happening 10:45:43 https://www.dropbox.com/s/tn82biqbp3nliif/Screenshot%202019-09-30%2010.45.34.png?dl=0 10:46:09 yes, it'll get pushed down 10:46:32 we could set the text widget to ellipsize its contents and then give it a maximum height 10:46:53 but then it's not very useful to know that your rcfile is somewhere on C:\ 10:46:57 yeah 10:48:19 I think it might accomplish the same goals as the current design to show it in the bottom line on startup, until an action is taken 10:53:14 the instruction text could also be shrunk to one line instead of two 10:53:30 e.g: [arrows] choose option [enter] select [tab] repeat last choice 10:54:16 it's possible that of those only [tab] needs to be documented in 2019 10:57:06 i'd say it's a pretty safe bet, yeah 11:03:55 oh also in console I'm seeing highlighted buttons as white on lightgray, is that intentional? previous was black on lightgray, which was a lot easier to read 11:04:53 at the main menu 11:05:12 other menus look fine 11:07:37 i'm seeing black on lightgrey 11:09:10 your screenshot shows black on lightgrey? unless the colours for the main options are different 11:10:09 that screenshot was 0.23.1 11:11:46 doh 11:15:10 https://www.dropbox.com/s/i8qdc2umf3omqtd/Screenshot%202019-09-30%2011.15.06.png?dl=0 11:18:28 oh, do you have allow_extended_colours = true? 11:20:08 yes 11:20:23 TERM=xterm-256color 11:22:06 thanks, I'll take a look 11:23:55 Stable (0.23) branch on underhound.eu updated to: 0.23.1-89-g277498a001 11:24:17 and yeah, removing the header line does look nicer, although perhaps the legalese should be grey/darkgrey? 11:24:26 and/or the version number 11:25:32 are you tweaking it? I started trying various options, guess we should try to not conflict 11:27:11 no, i'm getting ready to immerse myself in the colour munging code in libunix.cc 11:27:22 it requires a special state of mind 11:27:50 so go for it 11:49:25 :q 11:49:28 whoops 11:50:18 aidanh: another thing, looks like there are some key shortcuts assigned to main menu buttons? e.g. `C` selects "Dungeon Crawl" and clears the input box 11:51:39 I can't reproduce that 11:51:56 definitely happens to me :-) 11:52:10 e.g. I have a save called qwC2 that I can't select by typing the name 11:53:09 there's a bug where if you start typing after a full match to a save, the textbox is cleared again 11:53:11 it sounds like that 11:54:07 ah yes I also have a save called qw 11:54:41 and qwC for that matter 12:35:39 03advil02 07* 0.24-a0-722-g261a0b2: Main menu improvements and streamlining 10(3 minutes ago, 7 files, 56+ 28-) 13https://github.com/crawl/crawl/commit/261a0b24175b 12:43:48 aidanh: not sure about this `full_name` logic, I wonder if it is needed at all? 12:46:40 it looks like it was there before but I'm having trouble getting it to do much in previous versions 12:47:53 e.g. in 0.23 from the code it looks like backspace is supposed to clear a full name match but it doesn't 12:48:45 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-722-g261a0b2417 (34) 12:49:13 hm, backspace clears a full name match here, at least on 0.24a; did you mean on 0.23 it doesn't? 12:49:26 yes 12:49:36 that one is the most sensible of these behaviors, but I think by making these all actually work you've exposed that they're a bit hard to predict from standard text entry behavior 12:50:07 if they didn't work before, I'd say rip them out now before anyone notices 12:50:24 I think the *intent* of the previous code has exactly the same partial name match bug we were just discussing, but it was broken 12:50:27 haha sounds good 12:50:55 it might be a case where it worked many versions ago, and then broke somehow 12:51:21 that sounds plausible 12:52:55 the only one I'm tempted to keep is the backspace one, but even that seems hard to predict 12:53:29 there's always ctrl-u 12:53:41 oh, that doesn't work 12:57:24 in testing old versions these features do sometimes work, but I can't figure out the pattern for when 13:01:45 03advil02 07* 0.24-a0-723-g3c9d5f2: Remove full_name logic from main menu 10(3 minutes ago, 1 file, 1+ 13-) 13https://github.com/crawl/crawl/commit/3c9d5f24f390 13:11:40 advil: btw do you have any opinion on what I mentioned earlier about using tab / shift-tab to switch focus? 13:12:09 between things like pregen options 13:14:02 I didn't see the discussion but that seems fine 13:14:15 Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-722-g261a0b2 (34) 13:14:29 I guess one thing to know is that the pregen yes/no tab thing only appears on debug builds (or if that's wrong, it will be true before release, can't remember) 13:14:39 so that window will have only one field really 13:15:52 you were planning on moving that screen into the species > background > name flow at some point, right? 13:16:04 or making it accessible from there somehow 13:16:51 I'm not sure how viable that's going to be though 13:17:20 bloatcrawl2 is doing some sort of game options screen; it'd be nice to consolidate the UI 13:18:07 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-723-g3c9d5f24f3 (34) 13:28:02 ah, also there's empty space at the bottom of the pregen menu on release 13:28:51 https://a.uguu.se/0t5LYnix4lnc_pregen.png 15:20:20 Inarius (L25 DgFE) ERROR in 'stairs.cc' at line 985: hell exit without return destination (Hell) 15:24:04 !crashlog 15:24:22 21209. Inarius, XL25 DgFE, T:125415 (milestone): https://underhound.eu/crawl/morgue/Inarius/crash-Inarius-20190930-192017.txt 15:25:03 now *that*s the kind of error I expected to be seeing like a month ago 15:27:09 03gammafunk02 07* 0.24-a0-724-gbe6272a: Use no_tele_into in a Bazaar map (wizard44) 10(6 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/be6272a8903a 15:32:53 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-724-gbe6272a890 (34) 15:39:07 03Aidan Holm02 07[ui-fixes] * 0.24-a0-722-gd725f74: Refactor curses colour and attr munging 10(2 hours ago, 1 file, 72+ 96-) 13https://github.com/crawl/crawl/commit/d725f7463c67 15:39:07 03Aidan Holm02 07[ui-fixes] * 0.24-a0-723-gf59dad7: Fix difficult-to-read main menu labels on console (advil) 10(42 minutes ago, 5 files, 54+ 41-) 13https://github.com/crawl/crawl/commit/f59dad7addd1 15:39:18 03advil02 07* 0.24-a0-725-gf1a930f: Don't pull abyss terrain from the vestibule 10(2 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/f1a930fc54b2 15:47:58 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-725-gf1a930fc54 (34) 16:03:40 aidanh: this weird menu border glitching is very replicable on windows in the vm but I just can't get it to happen on os x no matter what I do 16:04:07 there's very little platform specific code that I know of, but I can't find any other factors either 16:04:12 any ideas? 16:04:55 is this with the CDO trunk build? 16:05:11 I can get it with my own builds 16:05:27 example: https://www.dropbox.com/s/r806x6ukskmt0ao/Screenshot%202019-09-30%2016.04.59.png?dl=0 (though usually the right-hand line is clipped I think) 16:06:59 if you can think of anything to try, lmk 16:08:35 I take it you didn't see this in wine? 16:09:13 looks like horizontal lines are shifted down by 1px, and vertical lines are shifted left by 1px 16:10:11 no, and not in a win7 vm either I don't think 16:10:21 although I'll have to double check that last bit 16:21:20 what VM do you use? 16:28:08 virtualbox 16:39:32 !source LineBuffer::add_square 16:39:33 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/tilebuf.cc#L318 16:46:25 yes, I think it's something to do with that 16:47:11 and maybe an interaction with virtualbox's drivers? 16:47:15 which are not amazing 16:47:45 I'm trying rolling back the changes there in the two pixel-perfect commits, still building 16:50:30 those commits only affect the top and left edges, not all four, so there's something else going on 16:51:48 s/left/right/ 16:53:27 in that pic I posted if you shift the top edge right by 1 pixel and the right edge down by 1 pixel (+ maybe left?) you get a connected rectangle 16:56:28 like this: https://www.dropbox.com/s/xys1dcd1ubqenp4/Screenshot%202019-09-30%2016.55.38.png?dl=0 16:56:41 the right lines are still cut off in the main menu and there's a missing pixel 16:57:08 the current result is the same as the intermediate commit, so maybe whatever GL magic is happening now just doesn't work for the virtualbox drivers? 17:04:32 right, but i'm going off the background rectangle here 17:06:32 well, in a roundabout fashion; I'm basing my current understanding on that orc tile 17:08:39 its tile (orc_f.png I think, but I didn't verify) has clearences of 2px at the top and 0px at the bottom, so I'm reasonably sure the dark grey background is correctly drawn 17:08:45 vertically at least 17:09:18 yeah not to say that what it did before was right 17:09:35 just that it looked like a rectangle 17:11:54 right 17:13:07 well afaik local tiles never draws diagonal lines... 17:13:28 so one easy fix is just to simulate line drawing with rects 17:14:17 do you remember if the buggy display looks like what caused you to introduce the scale stuff? 17:15:00 I wonder if virtualbox is just not dealing with scaling correctly 17:15:02 iirc it had one corner missing 17:19:17 although i'm pretty sure it was the top-left corner 18:23:43 Unstable branch on underhound.eu updated to: 0.24-a0-725-gf1a930fc54 (34)