01:38:30 Unstable branch on crawl.develz.org updated to: 0.25-a0-890-g62861fb195 (34) 02:10:20 Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-890-g62861fb195 03:24:16 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-890-g62861fb (34) 04:24:15 Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2651-g880b443feb 09:47:42 03Goratrix02 {Aidan Holm} 07* 0.25-a0-891-g4c921ad: Tweak phial of floods description (Flugkiller) 10(21 hours ago, 1 file, 5+ 4-) 13https://github.com/crawl/crawl/commit/4c921ade2430 09:59:08 New branch created: minor-refactors (6 commits) 13https://github.com/crawl/crawl/tree/minor-refactors 09:59:09 03Aidan Holm02 07[minor-refactors] * 0.25-a0-885-g7166e27: Move shop_struct to shopping.h 10(24 hours ago, 4 files, 27+ 23-) 13https://github.com/crawl/crawl/commit/7166e278e1eb 09:59:09 03Aidan Holm02 07[minor-refactors] * 0.25-a0-886-g9ab75b0: Move map_markers struct to mapmark.h 10(24 hours ago, 3 files, 47+ 48-) 13https://github.com/crawl/crawl/commit/9ab75b035933 09:59:09 03Aidan Holm02 07[minor-refactors] * 0.25-a0-887-gdf44acf: Further break externs.h apart 10(2 hours ago, 6 files, 541+ 514-) 13https://github.com/crawl/crawl/commit/df44acfd4894 09:59:09 03Aidan Holm02 07[minor-refactors] * 0.25-a0-888-gb0e87c4: Clean up describe_item() code 10(3 months ago, 1 file, 3+ 13-) 13https://github.com/crawl/crawl/commit/b0e87c482273 09:59:09 03Aidan Holm02 07[minor-refactors] * 0.25-a0-889-ga662fc3: Run timer callbacks in main thread 10(3 months ago, 3 files, 22+ 4-) 13https://github.com/crawl/crawl/commit/a662fc330b2d 09:59:09 03Aidan Holm02 07[minor-refactors] * 0.25-a0-890-g544e378: Remove raise_custom_event() 10(3 months ago, 3 files, 0+ 9-) 13https://github.com/crawl/crawl/commit/544e378b80f2 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-887-ga005f11: Split command processing out of show_map() 10(22 hours ago, 3 files, 650+ 609-) 13https://github.com/crawl/crawl/commit/a005f11e69e1 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-888-g1f792ce: Simplify arithmetic expressions 10(21 hours ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/1f792ce165bb 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-889-g680d8d9: Extract recentre_map_target() function 10(5 hours ago, 1 file, 17+ 14-) 13https://github.com/crawl/crawl/commit/680d8d9f77f5 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-890-gf30e781: Split viewport calculation out of show_map() 10(5 hours ago, 2 files, 63+ 84-) 13https://github.com/crawl/crawl/commit/f30e781499f2 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-891-g2ff48b6: Remove now-unused redundant map scrolling state 10(5 hours ago, 2 files, 2+ 8-) 13https://github.com/crawl/crawl/commit/2ff48b6cd075 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-892-g7ba4cb9: Don't exit map view on invalid commands 10(5 hours ago, 1 file, 2+ 7-) 13https://github.com/crawl/crawl/commit/7ba4cb98a868 10:00:12 03Aidan Holm02 07[refactor-show-map] * 0.25-a0-893-g23172d5: Extract coord_def clamp helper function() 10(4 hours ago, 2 files, 15+ 19-) 13https://github.com/crawl/crawl/commit/23172d59c050 10:05:17 aidanh: if you want a fun ui bug that I think is in the tiles ui 10:05:57 !bug 11833 10:06:02 https://crawl.develz.org/mantis/view.php?id=11833 10:06:17 I can replicate in local tiles, but the bug isn't in the beam code because the beam code is used for actually digging the squares 10:06:23 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-891-g4c921ade24 (34) 10:06:23 and that works correctly 10:06:43 also console uses the beam code and draws correctly 10:06:53 it seems to be a bug in grid2show! 10:07:04 and how grid2show interacts with barachi los? 10:07:41 ooo sounds interesting 10:07:54 it's not isotropic 10:08:12 aiming from up-right to down-left (inclusive, all the way around the circle of los) comes up one tile short 10:08:24 and around the rest of the angles it highlights correctly 10:09:26 btw i plan to merge support for minor tags > 255 and drop support for loading saves from 0.11 final (which is major v 33 so it's a bit weird anyway) 10:10:34 i.e. separately to the matter of breaking save compat and/or switching to a rolling system 10:11:24 oh yeah, this is definitely a bug in crawl_view 10:12:29 the window can be scaled down to smaller thann barachi los! 10:15:26 well this is far into parts of the codebase that I haven't looked at 10:27:55 so I'll give up for now 11:19:01 I'm having trouble rebasing my branch onto master 11:19:09 because I can't pull master 11:19:24 how so? 11:19:30 I think it might be because I added saves/ and morgue/ to gitignore 11:19:41 some diagnostics I ran: https://pastebin.com/K7n1CkuJ 11:20:07 yeah, you have commits on master 11:20:35 which unless you are working with a forked repository on your own github account is almost certainly not what you want 11:20:43 I think you also have uncommitted changes 11:20:49 which is causing the immediate problem 11:21:00 I can't commit those changes 11:21:03 specifically the .gitignore file as you said 11:21:06 you don't want to 11:21:26 then what do I do? 11:22:13 `git checkout -- crawl-ref/.gitignore` to reset that to master 11:22:38 saves and morgue are already in that gitignore fwiw 11:22:56 under the comment `# Crawl generated junk` 11:23:03 oh 11:23:16 I recall git complaining about saves/ and morgue/ 11:24:19 oh nm maybe you don't have commits on master, I think the line I'm seeing just means that upstream added 14 commits on top of wherever your HEAD is 11:25:17 petercordia: could you try pastebinning the output of this? git log --graph --oneline --abbrev-commit --date=relative --all 11:27:20 03Aidan Holm02 07* 0.25-a0-892-gd6d150f: Move shop_struct to shopping.h 10(25 hours ago, 4 files, 27+ 23-) 13https://github.com/crawl/crawl/commit/d6d150f7a2bd 11:27:20 03Aidan Holm02 07* 0.25-a0-893-g30cafcb: Move map_markers struct to mapmark.h 10(25 hours ago, 3 files, 47+ 48-) 13https://github.com/crawl/crawl/commit/30cafcbba1cf 11:27:20 03Aidan Holm02 07* 0.25-a0-894-g48f1b6f: Further break externs.h apart 10(3 hours ago, 6 files, 541+ 514-) 13https://github.com/crawl/crawl/commit/48f1b6f60fc1 11:27:20 03Aidan Holm02 07* 0.25-a0-895-gc1fa99a: Clean up describe_item() code 10(3 months ago, 1 file, 3+ 13-) 13https://github.com/crawl/crawl/commit/c1fa99a940d7 11:27:20 03Aidan Holm02 07* 0.25-a0-896-ga840853: Run timer callbacks in main thread 10(3 months ago, 3 files, 22+ 4-) 13https://github.com/crawl/crawl/commit/a8408532f3ae 11:27:20 03Aidan Holm02 07* 0.25-a0-897-gf770d96: Remove raise_custom_event() 10(3 months ago, 3 files, 0+ 9-) 13https://github.com/crawl/crawl/commit/f770d9629a5a 11:27:20 03Aidan Holm02 07* 0.25-a0-898-g4357695: Add support for minor tags > 255 10(7 days ago, 9 files, 147+ 43-) 13https://github.com/crawl/crawl/commit/435769570560 11:27:20 03Aidan Holm02 07* 0.25-a0-899-g1f02ff8: Improve save version mismatch reporting 10(5 days ago, 2 files, 31+ 23-) 13https://github.com/crawl/crawl/commit/1f02ff80f28f 11:27:40 https://pastebin.com/LXAigUzB 11:28:27 that looks fine to me, I think it's just the uncommitted changes to .gitignore 11:28:50 there's a :...skipping... in the middle 11:28:53 I can't seem to add .gitignore 11:28:58 so if you reset that it should be able to pull I think 11:29:18 did you run `git checkout -- crawl-ref/.gitignore`? :-) 11:29:44 adding it is not what you want...in general when working with a forked repo like this you never want any changes on master 11:29:54 It told me 'error: pathspec 'crawl-ref/.gitignore' did not match any file(s) known to git.' 11:30:09 should be crawl-ref/source/.gitignore 11:30:14 well, adjust the path as needed...I used the relative path 11:30:24 er, not correct 11:31:03 not sure why there is even a .gitignore in source but that's not the one in the error 11:31:23 the main .gitignore is in crawl-ref 11:31:29 (relative to the repository root) 11:31:37 I've run git checkout -- crawl-ref/source/.gitignore without error message 11:32:03 but git status still complains about 'modified: crawl-ref/source/contrib/sdl2-mixer (modified content)' and nothing else 11:32:19 how about git pull now? 11:32:33 03Aidan Holm02 07[wide-minor-tag] * 0.25-a0-900-g91b7f2f: Allow phasing out support for old minor versions 10(5 days ago, 3 files, 39+ 3-) 13https://github.com/crawl/crawl/commit/91b7f2f49629 11:32:38 Branch pull/1375 updated to be equal with wide-minor-tag: 13https://github.com/crawl/crawl/pull/1375 11:32:49 and git pull upstream master still complains about crawl-ref/.gitignore 11:32:49 hm 11:34:44 something must be wrong with the checkout command to get an error when resetting the .gitignore file 11:34:54 even if it is unchanged that command will just complete silently 11:36:07 also if you run `git status` from a submodule it will tell you what the changes are 11:37:32 you mean I could cd to crawl-ref/source/contrib/sdl2-mixer to find out what's up there? 11:37:37 yeah 11:38:21 deleted: external/flac-1.2.1/doc/html/api/annotated.html 11:38:44 I deleted that thinking it had something to do with the annotated.cc file I created earlier 11:39:31 you should also be able to use `git checkout -- file` to restore that 11:39:35 you can run `git checkout -- external/flac-1.2.1/doc/html/api/annotated.html` to get it back 11:41:50 thanks. It's telling me HEAD detached at 16dbc95 when I'm in crawl-ref/source/contrib/sdl2-mixer, but beyond that there's no trouble 11:42:22 no complaints about 'modified: crawl-ref/source/contrib/sdl2-mixer (modified content)' anymore 11:42:44 that is what it should be showing, submodules are treated as detached HEAD things 11:44:04 Fixed it now 11:44:30 Looks like I put crawl-ref/.gitignore into skip-worktree 11:44:41 and I fixed it with 'git update-index --no-skip-worktree crawl-ref/.gitignore' 11:44:58 ahh 11:45:17 I guess that is why it gave the error as if the file simply wasn't there? 11:46:24 it's possible to set a global .gitignore if you really want to, that's probably a better strategy for that particular file: https://help.github.com/en/github/using-git/ignoring-files 11:47:20 or I guess there's the `.git/info/exclude` file 11:53:07 does rebase tell me if there are any conflicts I need to clear up? 11:53:21 if there's a merge conflict it'll stop on that file 11:53:28 great 11:56:17 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-899-g1f02ff80f2 (34) 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-900-g23cf7c4: Allow any floor to be annotated with ! 10(13 days ago, 5 files, 9+ 50-) 13https://github.com/crawl/crawl/commit/23cf7c4a6b18 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-901-g10306df: I forgot to include files in the previous commit 10(13 days ago, 2 files, 257+ 0-) 13https://github.com/crawl/crawl/commit/10306dfb8934 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-902-g2353aef: Finished implementing the new annotation function 10(7 days ago, 5 files, 49+ 44-) 13https://github.com/crawl/crawl/commit/2353aef62ee8 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-903-g8a299e2: Tidying up 10(7 days ago, 7 files, 181+ 271-) 13https://github.com/crawl/crawl/commit/8a299e25aca4 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-904-gacb45be: Tidying up whitespace 10(7 days ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/acb45bee25bf 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-905-g66ea916: Fix whitespace 10(7 days ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/66ea916af594 11:59:53 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-912-g43051db: Merge branch 'new_annotate' of https://github.com/petercordia/crawl into new_annotate 10(2 minutes ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/43051db5e7e9 12:00:54 I've messed up pushing the rebase 12:02:31 I ran 'git pull origin new_annotate' 12:02:42 then 'git push origin new_annotate' 12:04:46 I'm not 100% sure what happened but generally when you have rebased, you simply want to force push (`git push --force`) 12:05:52 it looks like you've just ended up with the master commits on your branch? 12:06:29 it looks like it put in the master commits after the branch commits? 12:06:46 and I can't figure out what happened in my local version 12:06:57 yeah, it'll do that, the ordering is just based on the commit timestamp 12:10:12 03advil02 07[headless] * 0.25-a0-857-g97782ae: Headless UI mode 10(11 months ago, 9 files, 360+ 5-) 13https://github.com/crawl/crawl/commit/97782aea74c8 12:10:12 03advil02 07[headless] * 0.25-a0-858-g3ab9431: Fix a bunch of unused parameter warnings 10(74 minutes ago, 1 file, 16+ 6-) 13https://github.com/crawl/crawl/commit/3ab9431d7620 12:10:12 03advil02 07[headless] * 0.25-a0-859-g7cca37c: Handle ctrl-c better on unix headless builds 10(2 minutes ago, 4 files, 16+ 1-) 13https://github.com/crawl/crawl/commit/7cca37cad636 12:11:15 in my local branch there are now a bunch of commits after HEAD, which are the one's I don't want 12:11:37 if I pretend they don't exist, are they likely to cause trouble? 12:12:15 probably :-) 12:12:19 let me look at this 12:12:40 git log --graph --oneline --abbrev-commit --date=relative --all* 43051db5e7 (origin/new_annotate) Merge branch 'new_annotate' of https://github.com/petercordia/crawl into new_annotate|\ | * 55e1b7d4e9 Fix whitespace| * e6dc51bfe1 Tidying up whitespace| * a1fc0308f1 Tidying up| * c9ce714574 Finished implementing the new annotation function| * 12:12:40 b89853c09d I forgot to include files in the previous commit| * 73ae0bab11 Allow any floor to be annotated with !* | 66ea916af5 (HEAD -> new_annotate) Fix whitespace* | acb45bee25 Tidying up whitespace* | 8a299e25ac Tidying up* | 2353aef62e Finished implementing the new annotation function* | 10306dfb89 I forgot to include files in the previous 12:12:41 commit* | 23cf7c4a6b Allow any floor to be annotated with !* | 1f02ff80f2 (upstream/master, origin/master, origin/HEAD, master) Improve save version mismatch reporting* | 4357695705 Add support for minor tags > 255* | f770d9629a Remove raise_custom_event()* | a8408532f3 Run timer callbacks in main thread 12:13:10 it sort of looks like you just did a merge commit instead of a rebase? 12:13:47 so I did a rebase, and then a pull 12:14:00 and the pull implemented a merge 12:14:03 oh hm 12:14:11 I'm surprised that even did a ff merge 12:14:21 usually once you rebase the remote branch is fairly incompatible 12:14:22 aidanh: I just tested the latest commits on windows, and it's causing problems, I think it's probably a840853 12:14:30 %git a840853 12:14:30 07Aidan Holm02 * 0.25-a0-896-ga840853: Run timer callbacks in main thread 10(3 months ago, 3 files, 22+ 4-) 13https://github.com/crawl/crawl/commit/a8408532f3ae 12:14:35 CPU uasge is through the roof 12:14:48 where normally I don't see any cpu load whatsoever 12:15:04 how are you building? 12:15:08 also it looks like closing crawl didn't actually kill the process? I need to test that again 12:15:21 windows 10 64 bit mingw-w64 12:15:29 I notices you mentioned i686 12:15:59 that's 32 bit 12:15:59 so may be related to 32 vs 64 somehow 12:16:04 I'm using the exact instructions from INSTALL.md 12:16:09 and those are 64 bit 12:16:18 mingw-w64-x86_64 12:17:13 petercordia yeah it looks like there are multiple copies of your commits somehow so this probably isn't ignorable 12:17:15 i686-w64-mingw32 should support both 32 and 64; it's used for official builds after all 12:17:48 Goratrix: what do you have tile_tooltip_ms set to? 12:17:57 looks like I fixed the online version at least 12:18:36 default? 12:18:41 have not changed that setting 12:18:50 doing some more tests 12:19:01 it may be related to fullscreen? 12:19:19 petercordia ah looks like it to me too 12:20:33 yeah, that's what the last person complained of 12:21:25 played for a bit in windowed mode, cpu was ok, but when I closed the app window with the windows x button (in top right corner) it closed, but task is still running in task manager eating 20% cpu 12:21:45 advil, is there some slightly more thorough check we could do before I reset my local version to the online version? 12:22:32 if you're feeling really paranoid you could just clone a new local repository from petercordia/crawl and mv your current one somewhere else, then you still have it 12:22:48 but otherwise eyeball https://github.com/crawl/crawl/pull/1369/files and see if it looks right to you 12:22:50 https://imgur.com/Q3Z5Vrz 12:23:18 Goratrix: any chance you can profile where it's spending its time? 12:23:52 btw, are you using MSYS2? 12:24:11 ugh, this win 10 evaluation vm is absurdly slow 12:25:28 yes, MSYS2 as per the official instructions 12:26:58 would this help? https://imgur.com/QY77SHn 12:28:08 unfortunately not 12:29:01 it happens only when the window is force-closed 12:29:08 not when properly saving 12:29:20 looks like 12:30:42 uhh i think i can reproduce this on linux 12:31:08 it's just it uses much less cpu, so I never noticed 12:31:42 `return SDL_PushEvent(&event);` seems odd 12:32:19 according to the docs the callback should return the time until the next check 12:32:22 or 0 on error 12:32:42 that value is in ms 12:32:55 so if SDL_PushEvent returns 1, that's a pretty tight loop 12:33:20 e.g. see the example around here https://wiki.libsdl.org/SDL_AddTimer#Remarks 12:33:29 so it looks like everyting is ok when playing, both windowed and fullscreen, and the problem only happens when the window is force-closed. the crawl.exe gets stuck running at very high cpu and needs to be killed 12:33:47 oh 12:34:15 it also happens when exiting via ESC from main menu, stuck again 12:34:27 yeah looks like it 12:35:11 interesting that it still works pretty much ok on linux, but falls apart on windows 12:37:57 if my diagnosis is right it's probably fair to call it a bug in sdl, but still probably easier to work around than actually get fixed 12:38:26 well the cpu load may not really be a bug, but the fact that the thread seems to get stuck is 12:38:45 well, still my bug :) you can see in the original commit that the original timer callback returned zero 12:39:09 the cpu load only happens with the stuck process, not when the game is actually running 12:39:36 I thought initially that it was while running, but it was the previous stuck one doing it 12:40:27 SDL_PushEvent() can return 1 on success, which causes high CPU usage, or 0 on failure, which cancels the timeout like it should 12:40:36 lol 12:40:42 and one more thing, you actually need to start a game for it to happen 12:40:53 so it looks like it depends on the queue implementation, and i wouldn't be surprised if the queue hooks straight into whatever windows has 12:40:55 good bug 12:40:55 aidanh also is it intentional that there's now a static _timer_callback in both tilesdl.cc and windowmanager-sdl.cc? 12:40:56 not just open the main menu and ESC, then it closes properly 12:41:15 Goratrix: that's because this is only used to show tooltips after a delay 12:41:20 aha 12:42:02 advil: yeah that's intentional 12:42:10 well, harmless anyway 12:44:13 oh, I see that the one in tilesdl.cc is still used by something 12:44:39 03Aidan Holm02 07* 0.25-a0-900-g364e023: Fix high CPU usage on Windows builds (Goratrix, advil) 10(7 minutes ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/364e023d3122 12:45:19 it's a no-op callback used just to break out of the event loop 12:46:20 oh btw re the annotations pr, I only just discovered that you can annotate from interlevel travel 12:47:48 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-900-g364e023d31 (34) 12:52:02 ebering: fun bug; it does seem like env.tile_bg gets the right flag set, but it's still not rendering properly 12:55:53 I've also noticed that closing crawl 'improperly' now has a different effect in linux 12:55:53 after I press the little x the program doesn't stop automatically 12:56:03 it used to 12:56:24 Now I can't even kill it with control-c 12:57:13 as of which commit? 12:57:28 I just downloaded master 12:58:02 the most recent time before that was last week, I think 12:59:05 do you have the fix that was done 15 minutes ago? I'm just compiling windows, will test it in a minute 13:01:19 yeah, working fine here after 364e023 13:01:23 thanks! :) 13:02:12 great, thanks for the help! 13:31:13 the wizard help menu allows the user to input commands whilst looking at the help 13:31:20 does anyone know how it's done? 13:34:25 iirc the menu exits immediately and returns the key pressed 13:39:25 looking at command.cc, help menus like the wizard one use int show_keyhelp_menu(const vector &lines) 13:39:36 whereas help messages use void show_specific_help(const string &key) 13:40:16 !source show_keyhelp_menu 13:40:35 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/command.cc#L499 13:40:39 the former have an input which is a vector of strings, and the latter have an input which is a header in help.txt 13:41:01 the FS_EASY_EXIT tells the menu code to exit on the first invalid keypress, and that function returns it (return cmd_help.get_lastch()) 13:41:20 it seems a bit odd that they are so different. Which is prefered? 13:41:32 show_keyhelp_menu has worse code 13:41:38 but better functionalit 13:41:47 functionality 13:42:29 for annotations help, i'd say use show_specific_help 13:46:37 03Aidan Holm02 07* 0.25-a0-901-g968d922: Fix targeters sometimes being too short for Barachi (#11833) 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/968d922a6d55 13:47:40 ebering: ^ those pesky coordinates.. 13:47:48 might be worth having separate types for those 13:47:53 may I change show_specific_help, from void show_specific_help(...) { .... show_keyhelp_menu(formatted_lines);} to int show_specific_help(...) { .... return show_keyhelp_menu(formatted_lines);} ? 13:49:06 why? so that you can input commands from the annotations help? i don't think that's worth it, overall 13:49:37 it'd be weirdly inconsistent if we allowed that from that menu, but not others, and I don't want to do all the work of adding keybinds for the other menus 13:50:05 I'd like to have that feature in as many menus as possible 13:50:21 I've never found it weird that I could do it in wiz help 13:50:55 I'd suggest it should be the default behaviour for any new help menu which gets added 13:51:41 Bug reported on my fork that I think could happen on mainline, though it'd be rarer. An octopode with Antennae 3 reads acquirement. It tries to give armour, crashes. (Not sure if it'd happen on mainline, but it seems plausible) 13:55:34 I was worried it could cause bugs, but it sounds like it wouldn't 13:56:10 Actually I suppose it'd have to be Antennae 3 and sacrificed an arm to Ru on mainline (since shields and armours are one acquirement type)...but still rare circumstance where it'd try to acquire armour and there wouldn't be any valid armour. 13:56:30 Bcadren: while we appreciate bug reports, as you probably ought to know by now, a commen in IRC that says, "this thing happens on my fork and I haven't tried to otherwise replicate it but it may happen" is not a great bug report 13:56:34 *comment 13:57:02 and it's also just going to go down the drain in this channel unless someone is paying attention and gets interested in that moment 13:57:54 Fair point. I don't have a current copy of mainline to wizmode test if it happens there...I'm just going to fix the bug in my fork and move on. 13:59:39 petercordia: another reason i prefer not to support this is that some commands have varying effects and you probably want visual context 14:00:24 e.g. on interlevel travel, a bunch of commands depend on the currently focused tile 14:00:34 Doesn't happen in vanilla often enough to get it with one stack of a dozen ?acq 14:03:20 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-901-g968d922a6d (34) 14:10:23 oh haha good catch with the barachi thing aidanh 14:11:19 probably affects the passwall targeter too 14:11:23 %git cae90e3dc 14:11:23 07advil02 * 0.22-a0-426-gcae90e3: Refactor passwall and add a targeter 10(1 year, 11 months ago, 13 files, 356+ 112-) 13https://github.com/crawl/crawl/commit/cae90e3dcecc 14:11:26 it's from that commit 14:11:50 heh, yeah i'm really considering adding gc_coord, vp_coord, and ep_coord types 14:11:52 I'm surprised that didn't crash under some circumstances 14:12:14 although... `rg -o coord_def | wc -l` gives me 4246 instances, so maybe not :) 14:13:43 does not help that ep and gc are terrible variable names 14:14:57 -!- Guest67469 is now known as bh 14:16:14 yeah I can never remember which one is which 14:19:55 aidanh: dang good find 14:21:00 this mysterious Guest67469 is unmasked 14:25:48 03Nikolai Lavsky02 {ebering} 07* 0.25-a0-902-gf09614b: Clean up some preferred food-related code 10(8 days ago, 5 files, 3+ 18-) 13https://github.com/crawl/crawl/commit/f09614b0b8e6 14:25:48 03Nikolai Lavsky02 {ebering} 07* 0.25-a0-903-g529d17c: Remove a couple of obsolete menu_colour rules 10(8 days ago, 1 file, 0+ 4-) 13https://github.com/crawl/crawl/commit/529d17c01339 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-904-g30a2e19: Allow any floor to be annotated with ! 10(13 days ago, 5 files, 9+ 50-) 13https://github.com/crawl/crawl/commit/30a2e1919b2e 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-905-g65e0260: I forgot to include files in the previous commit 10(13 days ago, 2 files, 257+ 0-) 13https://github.com/crawl/crawl/commit/65e026076eb3 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-906-gd2f3356: Finished implementing the new annotation function 10(7 days ago, 5 files, 49+ 44-) 13https://github.com/crawl/crawl/commit/d2f3356826ae 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-907-gf090a0a: Tidying up 10(7 days ago, 7 files, 181+ 271-) 13https://github.com/crawl/crawl/commit/f090a0a4e821 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-908-gb6a941e: Tidying up whitespace 10(7 days ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/b6a941e8d46f 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-909-g3441200: Fix whitespace 10(7 days ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/3441200ddc55 14:31:24 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-910-g2f9ab96: Encorperated comments from reviewers 10(5 minutes ago, 5 files, 58+ 50-) 13https://github.com/crawl/crawl/commit/2f9ab96f4ca9 14:33:04 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-903-g529d17c013 (34) 14:35:02 I think I've followed all the recommendations, except to move the creation of a list of branches to branch.cc 14:35:40 which I didn't do because I still find the way keys are used a little daunting 14:35:41 oh i wouldn't worry about that one, i had a look and the branch_iterator is fine 14:42:08 03Aidan Holm02 07* 0.25-a0-904-ge98a5be: Split command processing out of show_map() 10(27 hours ago, 3 files, 648+ 611-) 13https://github.com/crawl/crawl/commit/e98a5be7b875 14:42:08 03Aidan Holm02 07* 0.25-a0-905-ga6a3aef: Simplify arithmetic expressions 10(26 hours ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/a6a3aeff3227 14:42:08 03Aidan Holm02 07* 0.25-a0-906-g235c7b5: Extract recentre_map_target() function 10(10 hours ago, 1 file, 17+ 14-) 13https://github.com/crawl/crawl/commit/235c7b5e9340 14:42:08 03Aidan Holm02 07* 0.25-a0-907-g2fe5a39: Split viewport calculation out of show_map() 10(10 hours ago, 2 files, 63+ 84-) 13https://github.com/crawl/crawl/commit/2fe5a39f5e4a 14:42:08 03Aidan Holm02 07* 0.25-a0-908-g0b7a428: Remove now-unused redundant map scrolling state 10(10 hours ago, 2 files, 2+ 8-) 13https://github.com/crawl/crawl/commit/0b7a42810195 14:42:08 03Aidan Holm02 07* 0.25-a0-909-gf80317f: Don't exit map view on invalid commands 10(9 hours ago, 1 file, 2+ 7-) 13https://github.com/crawl/crawl/commit/f80317f37c15 14:42:08 03Aidan Holm02 07* 0.25-a0-910-gca36f61: Extract coord_def clamp helper function() 10(9 hours ago, 2 files, 15+ 19-) 13https://github.com/crawl/crawl/commit/ca36f614e438 14:49:38 I've now merged all commits, so it might be ready for pulling 14:50:44 I've reproduced this acquirement bug in vanilla (octopode, antennae 3, sac hand). Looks like a fairly obvious error in acquire.cc. Not always on the first scroll. Mantis issue, or still reading this? 14:51:58 maybe mantis issue; i for one am about to hit the hay 14:52:23 03advil02 07* 0.25-a0-911-g237ada3: Add explicit header for iswalnum (#1372) 10(2 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/237ada30ef69 14:52:32 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-904-g63b2e0f: Allow any floor to be annotated with ! 10(13 days ago, 7 files, 192+ 53-) 13https://github.com/crawl/crawl/commit/63b2e0fe48dd 14:52:43 yeah, mantis or github issue 14:52:43 thanks for actually checking it Pinkbeast 14:57:27 advil: To be fair bcadren did produce the reproduction steps after a bit. 14:57:44 Possible acquirement crash when no armour slots available. 13https://crawl.develz.org/mantis/view.php?id=12247 by damerell 14:58:10 Submitted #12247 15:13:22 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-910-gca36f614e4 (34) 15:18:04 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-911-g237ada30ef (34) 15:32:22 03advil02 07* 0.25-a0-912-g4d83e08: Let `easy_confirm=all` allow lowercase stat selection (#1349) 10(7 minutes ago, 2 files, 16+ 6-) 13https://github.com/crawl/crawl/commit/4d83e081ddb9 15:32:56 advil: I'm also flipping through mantis and gh issues 15:33:02 and was aabout to look at that one! 15:33:07 heh 15:33:16 I did a tagging pass 15:33:25 that one was low-hanging fruit so I just went and did it 15:33:41 I'm currently going through the PR backlog, prepping 628, 978, and now 1232 15:34:10 If you've got a list of "low but not that low" fruit I'm thinking of doing a devblog post like the one for 0.23 about bugs we'd like to fix 15:35:59 probably some of the gh ones could be good; I have to say that that new bug template I added at some point has definitely made these reports better on average 15:36:24 wonder if we can do a mass purging of old mantis reports 15:36:34 maybe that doesn't matter that much 15:36:45 it's kind of hard to do actually because mantis is so slow these days 15:36:54 right 15:37:34 i'm curious as to why it's so slow.. it also seems like it's entirely single-threaded 15:38:11 i.e. if i have a page load on one tab and a bug update on the other, the load will be blocked by the update 15:38:25 literally no one has done any maintenance on it in like 10 years 15:38:31 well, that's not quite fair 15:39:13 maybe its database just needs a VACUUM? random thought 15:39:29 I'm also curious why it's so slow but I think only napkin has the permissions to dig into that 15:39:34 yeah, it hasn't received a lot of attention certainly 15:39:35 I'm sure its database needs *something* 15:41:01 personally I'd be happy to move entirely over to gh for new bug reports, but there's just a ton of information buried in mantis that I wouldn't want to drop entirely 15:42:04 yeah, I know you had some concerns about the issue interface not being as rigorous information-gathering-wise compared to mantis 15:42:14 or maybe that wasn't you, but neil? it was brought up a long time ago 15:42:33 maybe neil...but I think the issue templates help alot, and now github allows multiple issue templates as well 15:42:56 probably searching for some of the structured info is still better in mantis? to the extent data entry was done in the first place, at least 15:44:26 if there's any info that isn't being succesfully collected, anyone should feel free to either edit the current (only) issue template or add a new one 15:44:41 e.g. for scoring I have two templates that people have to choose from when filing a new bug: https://github.com/crawl/scoring/issues/new/choose 15:46:21 "Please do not use this to request features, unless you are one of Crawl's 15:46:23 developers." 15:46:30 wow, a power I never knew I had 15:46:35 haha 15:46:39 I wonder if we really need that 15:47:05 e.g. a literal reading of that might have discouraged that easy_confirm thing I just addressed 15:47:48 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-912-g4d83e081dd (34) 15:48:52 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-912-g0fa6652: Allow any floor to be annotated with ! 10(13 days ago, 7 files, 194+ 53-) 13https://github.com/crawl/crawl/commit/0fa66527d970 15:48:52 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-913-g849e6dd: Fix whitespace after merge conflict 10(15 minutes ago, 2 files, 1+ 3-) 13https://github.com/crawl/crawl/commit/849e6ddb6efd 15:51:57 why does the gh editor nag you about long commit titles but provide no means whatsoever for linebreaking the commit message 15:53:25 03advil02 {GitHub} 07* 0.25-a0-913-gf370ab6: Issue template tweaks 10(9 seconds ago, 1 file, 1+ 6-) 13https://github.com/crawl/crawl/commit/f370ab680416 15:53:45 huh why is yaml suddenly failing in gh actions 15:57:20 possibly the package should now be python3-yaml? 15:59:01 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-914-gd432caa: Unbrace. 10(5 minutes ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/d432caa921f2 16:02:39 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-913-gf370ab6804 (34) 16:03:40 yeah alexjurkiewicz and ebering were discussing that yesterday 16:12:00 oh yeah in fact that is what I needed for dgamelaunch-config recently on ubuntu 16:12:52 really it would be safer to install this through pip 16:15:22 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-915-g1f74b9f: simplify code, and add space to prompt 10(3 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/1f74b9faf214 16:16:29 I bet what changed is https://wiki.ubuntu.com/FocalFossa/ReleaseNotes 16:17:30 they moved to 20.04 already? 16:18:07 our tests aren't pinned to a particular release? 16:18:33 they are 16:18:39 they're pinned to 18.04 16:18:55 when I dug into it its a version difference in github images 16:19:17 well, they all say "ubuntu-latest" which will change, but yeah these still seem to bionic 16:19:44 in github virtual environment version 20200406.2 our build works and in 20200426.1 it doesn't 16:20:01 both 18.08 for the ubuntu version 16:20:03 er 16:20:04 18.04 16:21:04 re using pip, i was thinking about having repo-wide requirements.txt files 16:21:15 03advil02 07* 0.25-a0-914-g17231af: Try a different yaml package name for actions 10(7 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/17231afb6c68 16:21:36 that would probably be better, the webtiles tests already handle yaml more cleanly 16:22:26 should be possible for repo-wide lists to include webtiles-specific sublists, and build pinned lists for both repo-wide and just webtiles 16:22:40 ubuntu's python package naming is generally kind of annoying, I've noticed 16:22:55 also it's super annoying that recent ubuntu doesn't give you a `python` bin of any sort 16:23:18 how do you mean? 16:23:24 /usr/bin/python exists on my system 16:23:28 I realize it's compatible with the binary naming PEP 16:23:39 either your system is old or you created a symbolic link at some point 16:23:56 no, didn't create a symlink 16:24:04 if you install current packages on a clean bionic system you do not get a file at that location 16:24:05 well it seems we have a quorum 16:24:14 my ubuntu is 19.10 16:24:24 in fairness, I have not tried on 19 16:24:40 but I promise you, it does not create that on 18 16:25:00 well on my 18 system I had default install of python2 and python3, and python was python2 16:25:22 I didn't get any feedback on the release timeline I floated a bit ago 16:25:56 well, I've said that the general time frame works for me; the outstanding issues are CAO maintenance stuff, Sequell being not-really-working 16:26:04 and then whether we wanted to try a tournament revamp 16:26:10 possibly other things 16:26:16 yeah, the revamp is more the question 16:27:02 now is the time to start on one if we want to finish it in time 16:27:02 the safest thing is to just run the save tournament we have been, maybe we could tweak the god banners and think of a good Wu idea 16:27:04 you may have had that on your 18 system but you would not get that if you installed python packages on a clean bionic system right now, I promise you 16:27:36 I'm not sure why you're arguing with me, I spent like 20 hours over the last week dealing with issues like this for dgamelaunch-config :-) 16:28:15 well, I'm not arguing really 16:30:24 in terms of a more ambitious revamp, the old scripts are pretty opaque to me 16:30:40 doesn't look like that commit worked 16:30:56 the hacked up scoreboard library taken from zxc, kramin, and chequers scoreboard project I use for csdc works pretty well 16:31:02 after initial ingest it runs v fast 16:31:38 I don't think we want to use an entirely new set of tournament scripts; the issue isn't even the set of scripts we use, it's the tournament format 16:31:56 or rather, we have to settle on a format first 16:32:30 -!- aidanh_ is now known as aidanh 16:32:48 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-914-g17231afb6c (34) 16:33:34 the existing scripts have run fine for last two tournaments on CDO, and that server is still available and our best bet for running this tournament 16:35:50 well, if we have a significant change in new rules we hvae to implement them 16:37:11 we do, but I think the rules are the most difficult/contentious part 16:39:11 and fwiw: https://packages.ubuntu.com/bionic/amd64/python-minimal/filelist does show /usr/bin/python for python-minimal, a dependency of https://packages.ubuntu.com/bionic/python (obviously python2) 16:40:34 the main issues with rules changes were: 1) we wanted to sort of remove the idea of single-player and team first place but 2) we'd need an almost entirely new set of team cheevos, or at least some way for those teams to feel sufficiently meaningful 16:41:06 ah, ok. well, python3 does not install that binary 16:41:22 yeah, python3 indeed refuses to call itself just 'python' 16:41:54 good news though! in 20.04, python2 isn't even installed, and is part of universe 16:42:51 people have kicked around ways to maybe make single player/team competitions not just involve spending more hours, but it was elliptic's opinion that this is sort of just not possible 16:42:54 oh god. yeah i dealt with that on a server i was messing with a few days ago at work 16:43:22 which remains annoying for all of our stuff, which assumes that the binary can be called as `python` 16:43:22 what a PITA 16:43:27 I was wondering if it would make sense to cap (or soft-cap) streaks this time? I know that suggestion has been floating around 16:43:50 yeah we could try some soft limits like that to help a bit 16:44:34 my sort of general concern about the non-competition thing was that it would make clans feel sort of pointless unless we had a nice set of achievements thought out for them to try 16:44:52 and it's sort of more "taking away" in general to remove the competitions, even if those two comptetitions are kind of inherently bad 16:46:23 Berder (L27 FeAr) ASSERT(attacker) in 'attack.cc' at line 351 failed. (Pan) 16:47:36 so we could try just the soft-cap thing and maybe come up with a decent Wu banner for sake of completeness, both of which would be easilly done in the timeframe and with existing scripts 16:48:00 something I've thought about 16:48:03 I'm all for new tournament format proposals that more systematically eliminate grinding 16:48:12 oh wow I started typing and then I just saw Berder's crash 16:48:18 name recognition 16:48:18 ok 16:48:21 back onn topic 16:50:42 hm this is too long for irc 16:51:43 the basic sketch is we rate players in a variety of categories; in place of "points earned" mushed together, a players "tournament score" is the sum of their ranking in each category 16:51:55 for clans you sum the min ranking of the clan's players in each category 16:52:29 lower score is better 16:52:37 plus you have individual category top 3's 16:53:02 so a clan could have a player focused on streaks, a player focused on nemelexes, etc. 16:54:16 it needs fleshing out and working correctly depends on the choice of categories 16:55:22 03advil02 07* 0.25-a0-915-g5fcb2d6: Attempt to install python more carefully for actions 10(31 seconds ago, 2 files, 18+ 1-) 13https://github.com/crawl/crawl/commit/5fcb2d6f646f 16:56:10 full disclosure that was a cargo cult commit 16:58:01 this is crawlcode, we hold things to a higher standard 16:58:50 ebering: so the idea is it doesn't matter if someone wants to streak 25 wins, the best they can do is 1st in streaks? 16:59:03 03advil02 07* 0.25-a0-916-gadb52c9: Fix a directory name 10(25 seconds ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/adb52c904e68 16:59:24 one issue that comes to mind is that some of our categories are pretty silly, aren't they? 17:00:33 so I guess most of the current "general scoring" section would effectively be gone 17:01:15 "most high scores" could remain 17:01:48 the species/background/god points would be effectively gone 17:02:19 nemelex choic isn't exactly a category unless we make it one, and there's probably be a lot of ties 17:02:38 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-916-gadb52c904e (34) 17:03:15 I suppose earliest win is an ok category, just a ranking based on order of your first win 17:03:27 ok, good, looks like that sorted out the main issue, now to see if the change breaks any of the other targets 17:03:43 then turncount/game score/realtime are all fine in that system 17:04:50 like I said, needs some fleshing out 17:04:57 not sure how we'd handle the special achievement points in this system 17:05:05 but the basic premise is that you can't do better than first in any category, so a diversity of playstyles is rewarded 17:05:13 I was thinking of a "banners" category 17:05:32 oh, so "most banners"? 17:05:33 where all banners get points and the category is ranked on banner points 17:05:36 yeah 17:05:51 right, was going to say how do you deal with different banner difficulty, but if you do points for banners, that works 17:05:52 this would allow the special bonus banners to count for a lot in this category 17:06:18 oh I guess all our banners are worth either 50 points or 0 points 17:07:15 so I guess one thing to point out is that this format does encourage "grinding" to improve your rankings in the competitive categories 17:08:28 I suppose it doesn't fundamentally stop anyone from grinding their less-than-first-place ranking by churning out the relevant type of game but due to all the different categories being essentially equally rated overall 17:08:59 hrm, someone with lots and lots of free time does still have an advantage 17:09:20 maybe it's enough of an improvement by spreading out the points over categories evenly 17:10:17 well I think without a hard limited play format more play will always mean better results 17:10:17 yeah 17:10:49 proposed format sounds kind of compelling 17:11:01 it would at least mix up the tournament competition and we could see how it plays out 17:11:07 I'll try to flesh things out abit more and put it on the dev wiki 17:12:22 heh, that dir fix 17:12:32 maybe that directory should be called webtiles and not webserver! 17:13:21 haha 17:13:44 it's an quite the error for me to make, given how much time I've spent in that directory 17:14:03 I think all banners getting some points is a great idea, I've actually always wondered why so few get points 17:15:44 people already do plan the "trying to get a lot of banners" even without that, but I bet it would be more fun with a leaderboard of some kind 17:15:44 *do play 17:43:10 03advil02 07* 0.25-a0-917-gfd9b2de: Explicitly install python for catch2 tests 10(76 seconds ago, 1 file, 9+ 0-) 13https://github.com/crawl/crawl/commit/fd9b2de1c411 17:45:16 ebering: hm (re pyyaml on github actions). I didn't read all of scrollback, did you compare the changes 4 days ago and find anything relevant? 17:47:45 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-917-gfd9b2de1c4 (34) 17:49:35 alexjurkiewicz I've fixed it but I basically didn't try to figure out what was going wrong, I just used the python install action so that the builds are a lot more consistent, and used one of the webtiles requirement.txt files to get pyyaml installed 17:50:28 ah yes, i see it. sensible! 17:50:49 aidanh was suggesting we have a repository-wide requirement.txt file, which can be subbed in for the webtiles one I used at some point 17:51:59 ebering: basing overall tourney rank solely on ranks in a bunch of specific categories is a cool idea (though I suspect you don't want to add up the rankings, but instead take a harmonic mean or something so that people aren't penalized as much for one category that they really hate and are rank 1000 in) 17:52:48 (it's also a short step from there to the suggestion I've made in the past, which is more or less the same thing but without overall rank :P) 17:56:59 advil: the main argument against giving out points for more banners is that a lot of the banners are only really interesting to do once or twice ever, and we wanted to be careful not to make people feel like they were a chore to be done once per tournament 17:57:44 03zuurr02 {ebering} 07* 0.25-a0-918-g7c84a75: Add auto_butcher_max_chunks option 10(2 years, 6 months ago, 4 files, 23+ 4-) 13https://github.com/crawl/crawl/commit/7c84a75a335d 17:57:44 03Alex Jurkiewicz02 {ebering} 07* 0.25-a0-919-g6329fa5: Add -branch-json 10(1 year, 3 months ago, 7 files, 53+ 2-) 13https://github.com/crawl/crawl/commit/6329fa547ddb 17:57:44 03Alex Jurkiewicz02 {ebering} 07* 0.25-a0-920-g0acc9a3: Add a version of json_mkstring that accepts string 10(1 year, 3 months ago, 2 files, 6+ 0-) 13https://github.com/crawl/crawl/commit/0acc9a31795f 17:57:44 03Alex Jurkiewicz02 {ebering} 07* 0.25-a0-921-g73108e5: Use the new json_mkstring(string s) function 10(1 year, 3 months ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/73108e5cc250 17:57:44 03Alex Jurkiewicz02 {ebering} 07* 0.25-a0-922-g9fd4d98: Add recommended jobs and species to -playable-json 10(1 year, 3 months ago, 3 files, 26+ 0-) 13https://github.com/crawl/crawl/commit/9fd4d98c1605 17:57:44 03sdynet02 {ebering} 07* 0.25-a0-923-gd0b4d5d: Remove summons from the Pain card 10(5 months ago, 3 files, 7+ 23-) 13https://github.com/crawl/crawl/commit/d0b4d5de9998 17:57:44 03sdynet02 {ebering} 07* 0.25-a0-924-g3194668: Improve the Cloud card 10(4 months ago, 2 files, 5+ 22-) 13https://github.com/crawl/crawl/commit/31946683ed5e 17:57:44 03ebering02 07* 0.25-a0-925-gf7cfe1f: Allow monsters to place on traps (#1283) 10(46 minutes ago, 2 files, 0+ 19-) 13https://github.com/crawl/crawl/commit/f7cfe1fbb0d6 17:57:44 03ebering02 07* 0.25-a0-926-g3419481: Don't pain bond friendly and neutral monsters (#1204) 10(30 minutes ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/3419481e9726 17:57:44 03ebering02 07* 0.25-a0-927-g39f137c: Prevent inadvertent opening of ebering_ghost_funeral_grove (#1077) 10(13 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/39f137c8b766 17:58:04 banners were also originally intended as something fun for people who didn't care about the points competition as much 17:58:06 (I think it's fine to have a "most banners" ranking though) 18:05:54 elliptic: yeah, I was inspired by your old category suggestion 18:06:10 I like the suggestion of the harmonic mean 18:06:25 probably a weighted harmonic mean would be the best, depending on the category type 18:09:15 ^status 18:09:15 14 Crawlers. CBRO disk usage=100% (135GB) | RAM usage=28% (4GB)| uptime/CPU= 18:09:16 up 157 days, 15:40, 0 users, load average: 0.01, 0.10, 0.20 (4 Cores) http://status.berotato.org 18:17:40 cbro disk is full, and I've messaged johnstein about it, in case anyone else gets reports and is wondering about it 18:24:15 Unstable branch on underhound.eu updated to: 0.25-a0-927-g39f137c8b7 (34) 18:28:30 wow big ups to this PR storm ebering & advil 18:28:46 I mostly did issues, ebering did all the PRs 18:29:04 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-927-g39f137c8b7 (34) 18:29:28 auto_butcher_max_chunks seems like a great thing to have on by default. I'll add it to my PR. Does 10 seem reasonable? 18:31:03 Drawing this card enshrouds the user's foes in rings of sight-blocking clouds. <-- s/sight-blocking/opaque/ ? A little less awkward 18:58:01 hmm yes words 19:00:22 I wonder what my odds are of being able to upgrade py27 on cao 19:00:26 they seem slim 19:03:34 from 2.7.x to 2.7.latest? is it worth it? 19:03:58 it wasn't before, but I can't get the b2 command line utility to work on the version it has 19:04:30 which is thwarting my plans to get data off of there 19:04:53 I guess I could just wget stuff and then upload it from my computer 19:05:14 was hoping to just start a script running on cao though 19:08:42 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1378 * 0.25-a0-928-g7bb1e67: Change explore_auto_rest default to true 10(3 days ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/7bb1e67ea103 19:08:42 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1378 * 0.25-a0-929-g8f8fa73: Change auto_butcher default to always 10(3 days ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/8f8fa7308758 19:08:42 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1378 * 0.25-a0-930-g2f50b46: Set auto_butcher_max_chunks default to 10 10(4 minutes ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/2f50b461b5a7 19:08:58 til there are 10 different internal torment types 19:09:07 I mean, I knew that since I added one, but geeze 19:09:35 ebering: ^ updated as requested (plus auto_butcher_max_chunks) 19:14:08 New branch created: pull/1383 (1 commit) 13https://github.com/crawl/crawl/pull/1383 19:14:08 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1383 * 0.25-a0-928-gbaf1198: Change show_more default to false 10(6 minutes ago, 2 files, 1+ 5-) 13https://github.com/crawl/crawl/commit/baf11984204c 19:14:19 The build passed. (master - 39f137c #13192 : Edgar A. Bering IV): https://travis-ci.org/crawl/crawl/builds/682424804 19:23:43 Ok, I typed in aa bit more about these ideas but I'm done with crawl stuff for today and haven't had time to flesh out the categories 19:23:46 https://github.com/crawl/crawl/wiki/New-Tournament-Scoring 19:28:10 aidanh: for your ubuntu chroot script, I needed to do three extra things on CAO: bind-mount /dev/urandom (also did /dev/random, but apt seems to only use the former), fix the permissions on tmp (to 1777), and run `apt-get clean` once or twice before the apt commands. I do seem to have a working chroot now! 19:28:47 I ran through it manually so I haven't really tested it in automated form 19:31:54 dang 19:46:13 well, that was extremely convoluted but I got the b2 utility working in an ubuntu 19 chroot and did a proof of concept upload 19:48:26 wow, so is the chroot you're running on cao actually ubuntu? 19:48:41 I guess it's standardized that way isn't it 19:49:00 seems kind of funny how we're running an ubuntu chroot on a (regular) debian system 19:57:56 i love that even after years of playing I can still run across vaults i've never seen before. a giant half level hive 20:00:08 yeah, it's ubuntu 19 constructed via aidanh's script 20:11:54 hm. is it a known bug that sixfurhys can move through rock? 20:12:10 one tile of glass, i mean 20:13:32 it definitely is a bug if they're moving through glass, but their movement is pretty weird, so you'd want to be sure it actually happened 20:13:43 watching the ttyrec would probably reveal it 21:22:07 advil: woah, cool! is that just the crawl chroot, or a build chroot as well? 21:22:21 it's just the base chroot really 21:23:30 okay 21:23:36 for the build chroot, i was thinking it'd be good to make it build everything from vendored deps 21:25:03 although I guess if the chroot is reproducible it's less important 21:33:08 advil: oh, hm, can the ubuntu chroot run binaries produced by the debian system outside it? 21:33:36 I mean, I'm sure you checked before declaring success but I remember running into issues there 22:23:16 03Alex Jurkiewicz02 {Aidan Holm} 07* 0.25-a0-928-g2618fe8: Change explore_auto_rest default to true 10(3 days ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/2618fe8269ee 22:23:16 03Alex Jurkiewicz02 {Aidan Holm} 07* 0.25-a0-929-gd22f3de: Change auto_butcher default to always 10(3 days ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/d22f3de0d23f 22:23:16 03Alex Jurkiewicz02 {Aidan Holm} 07* 0.25-a0-930-g9f5dfa9: Set auto_butcher_max_chunks default to 10 10(3 hours ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/9f5dfa981578 22:30:11 ebering: re bugs we'd like to fix: https://crawl.develz.org/mantis/view.php?id=12056 22:32:53 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-930-g9f5dfa9815 (34) 22:53:45 ty 23:00:21 I proposed an alternative show_more mode to be used by default. Would appreciate feedback: https://github.com/crawl/crawl/pull/1383#issuecomment-623046433 23:01:44 tl;dr, there's no --MORE-- prompt, but instead an extra message like "7 new messages above (press ctrl-p to view)". This would be a new show_more mode, the existing true/false modes remain for advanced players. 23:09:16 that might work, but I do question the idea that we want to by default hide or at least obfuscate the fact that the player did something to generate a lot of text 23:09:31 the issue being that probably something happened they should be aware of 23:10:36 I'm not really sure if there's a good general approach to an issue like this; if the aim is getting players to learn, either messages are helpful to that end, and slowing them down to read it, or messages are bad to that end and we shouldn't be generating them, it feels like 23:10:54 s/down to read it/down to read it is good/ 23:11:34 i think the right way is just to flash the more prompt if the player hits keys while it's active 23:11:35 it is possible that something like message coloration and priority (sufficiently important ones get a color and either do trigger more or would if they would be missed) 23:12:06 and possibly switch --more-- to --more (press space to continue)-- 23:12:37 hrm, how is that different from the current default? 23:12:47 what would you see if you generted a more but hit nothing? 23:12:52 *generated 23:15:39 currently there's no visual feedback whatsoever 23:15:57 hit nothing? 23:17:34 -!- Tux[Qyou] is now known as aruor 23:18:04 -!- aruor is now known as Tux[Qyou] 23:18:45 I guess I don't know what you mean by flash the more prompt; flash it onscreen, out of the message area? 23:18:58 so, an additional indicator essentially? 23:21:37 i mean literally flash the piece of text on the screen saying --more-- :) 23:21:52 e.g. flip the background and foreground colours for 0.5 seconds 23:24:05 ok, I guess that's the same effect as what I said, in essense; you're just saying we'd make the prompt more visually distinct, so the player even gets that it's there 23:24:23 to address the "I don't know why this is pausing, but I figured out I can hit space to carry on" issue 23:24:33 yeah, exactly 23:25:46 seems reasonable to me 23:50:18 Game crash after using grand finale 13https://crawl.develz.org/mantis/view.php?id=12248 by Flugkiller 23:50:44 re this more stuff, I think it's definitely a bad idea to make any changes to it this close to a release 23:50:44 all of the alternatives I'm hearing would need testing IMO 23:51:05 and removing default more seems like not the way to go 23:51:15 if this is all such a concern, why not add something to the tutorial about it? 23:51:35 *concern for new players 23:55:45 Game crashed after going back a page in the browser when you have to select a new skill to train 13https://crawl.develz.org/mantis/view.php?id=12249 by Flugkiller 23:59:26 yeah, I'm not in any particular rush to change the --more-- prompt either