00:20:40 ??is cao back 00:20:40 I don't have a page labeled is_cao_back in my learndb. 00:24:00 ??is cao down 00:24:03 3 hours, 11 minutes, 24 seconds since last activity (cao) 00:33:51 i guess there's no eta on that 00:34:36 no 00:34:51 we're not sure when rax will have time to get to the server 01:09:31 * geekosaur wonders if he (and/or |amethyst) should be holding a sysadmin boot camp or something 01:19:59 will i have to recompile crawl to test vaults i am making? 01:21:07 no 01:21:45 well, depends. some oddnesses can only be diagnosed with a debug version of crawl --- but try some other way first as it's a lot of message spam 01:24:49 thanks! 01:24:54 ooh kb_arrival_jail ought to have Ba and Gn representatives added 01:25:19 although i guess it only has 26 slots 01:37:46 nikheizen: des cache is rebuilt when files have changed each time you run crawl, just have to remember save and restart crawl if you make a change to des while crawl is running 01:37:56 which sounds a bit obvious, but it's easy to forget 01:38:21 yeah the arrival jail might need to have more space added 01:39:37 i see 01:39:50 the things you get from a debug build of crawl aren't usually necessary for vault work; pretty much only if you're doing some higher-level stuff like layouts 01:39:53 that's good because my computer is not fast to compile 01:40:11 yeah i ain't doing anything more than some babby lua yet likely 01:46:09 -!- Menche is now known as Menche\lettuce 02:04:49 dunno, I had to use it to figure out why a vault I made was screwing up when it got rotated 02:09:32 good morning! my pull request got approved, but am has had some very constructive ideas on how to improve it. is it okay to just make another commit to address these things, or should I reply first? 02:09:33 kenran: You have 1 message. Use !messages to read it. 02:09:37 !messages 02:09:38 (1/1) gammafunk said (4h 39m ago): I made commit 04744e2 (0.21-a0-79-g04744e2). Now you can disable the shop halos in your vault by having its setup call be setup_bazaar(_G, false) instead of the usual setup_bazaar(_G) 02:12:14 !tell gammafunk thanks a lot! I might incorporate it in my forest vault or not, but it'll surely make it into the next idea :) 02:12:15 kenran: OK, I'll let gammafunk know. 02:17:04 kenran: you can just make another commit and then reply; it's not so much order of communication as it is just doing the communication 02:17:04 gammafunk: You have 2 messages. Use !messages to read them. 02:17:39 and if the person you want to talk to is already in channel, you can just tell them instead of sending a !tell :) 02:17:47 tell is usually for when people aren't in channel 02:18:32 if you start the message with "gammafunk:" like I did when talking to you, most irc clients have a special notification so it gets the intended recipient's attention 02:18:55 gammafunk: yeah, I was just stupid; tried to avoid the pinging but !tell gammafunk pings you as well... 02:19:08 it's still early. gotcha! 02:19:32 yeah, pinging is just fine 02:20:01 there are a few admins we ask people not to ping unless it's important, but in general for anyone actually in channel, ping away 02:20:19 I would prefer to just read the message in channel rather than have to check messages 02:20:22 !messages 02:20:23 (1/2) MarvinPA_ said (12h 32m 11s ago): re gnolls, i can't say i'm particularly enthused about it but some form of the stat lock thing does definitely sound less fundamentally broken than the current implementation 02:20:26 !messages 02:20:26 (1/1) kenran said (8m 11s ago): thanks a lot! I might incorporate it in my forest vault or not, but it'll surely make it into the next idea :) 02:20:27 !messages 02:20:28 No messages for gammafunk. 02:20:46 wow amalloy with the fancy github replies 02:21:00 he's making the rest of us look bad 02:21:10 the code-review feature was such a big improvement 02:21:15 they're really fancy, but also really useful 02:21:36 will another commit make one of you have to review again btw? 02:21:46 none of us have to review ever, really 02:21:49 or does it depend on the severity of the changes in some way? 02:21:51 ahh ok 02:21:53 I forget how notifications work there 02:21:58 but if you comment, at least 02:22:01 the comment count is updated 02:22:08 but don't worry, at least one of us will look at it 02:22:13 ...eventually 02:22:27 man, the good old days 02:22:44 I would put a pastebin link of my first vaults in chat 02:23:00 5 minutes later a grunt commit with those vaults, his edits and no discussion 02:23:04 I wasn't worried, just curious :D github is kinda different from the stuff I work with, and I haven't had the opportunity to try out advanced features 02:23:12 s/chat/channel/ 02:23:28 well he might of !firestormed me after making the commit 02:23:50 s/of/have/ 02:24:08 heh 02:24:15 might of/have is the English equivalent to the German das/dass 02:24:22 it seems to me 02:24:58 truly grunt has the might of firestorm 02:25:03 the problem is the contraction "would've" sounds exactly like "would of" 02:25:12 might've might of 02:25:18 or whatever you're putting before "have" 02:26:55 now that I've used it a bit, I quite like the branch system as it's used in git(hub) 02:27:44 there's nothing worse than being forced to work with shelvesets in tfs for instance 02:35:08 hrm 02:35:12 i don't know anything about TFS, but the worst thing in source-control that i know about is merging branches in CVS 02:35:13 !source ng-init.cc:238 02:35:18 https://github.com/crawl/crawl/blob/master/crawl-ref/source/ng-init.cc#L238 02:35:29 I need to do a check for if this is set 02:35:37 but I'm actually not sure of the best way to check 02:35:59 get_vector() looks like it does initialization if the thing is unset 02:36:03 you mean whether the property exists? 02:36:16 yeah 02:36:34 well, what I need to do there is not create a new vector if it already exists 02:36:34 for mapstat/objstat 02:36:50 otherwise it crashes after the first iteration 02:37:13 looks like I could do those flag checks but I'm not sure if there's a better way 02:37:47 gammafunk: https://github.com/crawl/crawl/blob/d0efa662383971e981fb52f84055cffa5108b718/crawl-ref/source/store.h#L257 02:38:05 ok 02:38:25 hrm, must have just missed that when looking through that class, thanks 02:44:17 it's so funny using this FOO_BAR_KEY pattern for hash-like things in C++ after switching from python 02:45:02 I bet there are at least some stubborn programmers who do things like that in their python code as well 02:46:53 New branch created: pull/552 (2 commits) 13https://github.com/crawl/crawl/pull/552 02:46:53 03kenranunderscore02 07https://github.com/crawl/crawl/pull/552 * 0.21-a0-79-gb88d852: Tree form now fully suppresses the blink and teleportitis mutations 10(6 hours ago, 1 file, 6+ 0-) 13https://github.com/crawl/crawl/commit/b88d8529697b 02:46:53 03kenranunderscore02 07https://github.com/crawl/crawl/pull/552 * 0.21-a0-80-gd68ad71: Trying to blink when that is not possible now prints 'You cannot blink...' instead of 'teleport' 10(6 hours ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/d68ad7107abb 02:47:08 oh 02:47:39 kenran: for commit message, the first line should be 72 char max (it's the commit title), the second line a blank line, then the commit body 02:48:00 which we like a max line length of I think 75 chars? I use 72 02:48:24 so your second commit there needs a reformatted commit message 02:48:56 and I would try just rephrasing the title as something like 02:49:21 "Print a better message when an attempted blink is not possible" 02:49:55 it's not quite necessary to list what the exact message you change it to is, since that's in the commit itself, which is easy to read 02:50:15 phew, thanks :D so much to learn 02:50:16 but if you did want to put that, definitely don't try to fit all of that in the commit title. Longer descriptions like that go in the commit message body 02:50:19 np 02:52:55 https://chris.beams.io/posts/git-commit/ is a good article about git commit messages, though of course don't feel you need to learn everything all at once 02:55:19 is the body necessary at all for such an easy commit, or do you sometimes just leave it? 02:55:45 you can definitely omit the body for simple commits. the article mentions that 02:56:29 yeah, your second commit would be just fine with a title something like what I mentioned (exact wording is not important, just communicate the change you're making), and no commit message body 02:57:06 gammafunk: ok, I amended it and shamelessly stole your title word by word 02:57:16 03kenranunderscore02 07https://github.com/crawl/crawl/pull/552 * 0.21-a0-80-g0fe7644: Print a better message when an attempted blink is not possible 10(6 hours ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/0fe764482265 02:57:18 that's alright 02:57:31 oh nice, Cheibriados even updates it in here 03:00:22 is it okay to just slap my vaults at the bottom of the relevant .des file? arrival/simple.des mentions that they are in order 03:01:14 that's for the existing vaults 03:01:22 for your new ones, yeah I would put them at the end 03:02:02 I'm not quite sure where the informal tutorial is supposed to end there 03:02:16 but in other des files we generally put the new vaults in order of addition, as well 03:02:29 just making sure to put them in the right section of the file, for files that have that 03:02:43 but simple.des doesn't have any sections 03:04:00 the ######## blocking seems to be inconsistent as well 03:04:49 that's not a thing we do in many files 03:04:58 it seems to be based on batches of vaults added 03:06:04 we don't usually do blocking like that in other files; the blocking we do is often by type of vault (where type maybe be somewhat loosely defined) 03:06:26 I would say that you don't need to add any blocking 03:07:12 I guess I could add it for the newer vaults for consistency, or something, but it doesn't really add much; it's neither documenting nor organizing in any meaningful way 03:10:23 cool 03:12:22 Unstable branch on crawl.beRotato.org updated to: 0.21-a0-79-g04744e2 (34) 03:45:19 Uh 03:45:39 I think kicks are getting 5 less damage than they're supposed to 03:45:41 and have been for ages 03:46:39 !source melee-attack.cc:995 03:46:40 https://github.com/crawl/crawl/blob/master/crawl-ref/source/melee-attack.cc#L995 03:49:10 hrm 03:49:35 why does it say 03:49:47 I've done thousands of trials with fsim and talons 3 certainly are not doing 9 base damage 03:49:51 // Max talon damage: 9. 03:50:00 Because it's supposed to be 5 + 1 + talons level 03:50:06 looks like those comments are all just out of date, yeah 03:50:11 But compare it to headbutt 03:50:13 was there a commit nerfing their damage? 03:50:45 If I give a vault an ORIENT: of southwest, but not no_rotate, can it be placed in any corner? 03:51:02 ah 03:51:07 or should i give it both 03:51:44 minmay: yeah that does look like a disparity 03:52:36 I'm not sure if there was a commit nerfing the damage or not, I can't look further right now 03:53:14 I'm trying to find when this happened 03:54:01 nikheizen: would it be find with simply orient north? 03:54:43 hoof kicks or talon kicks or both? 03:55:25 they idea of those other ones is that you need to be in a corner for some reason 03:55:33 I'm actually not sure how rotation plays with those vaults 03:56:22 ProzacElf: all kicks, including tentacle spike slaps 03:56:32 huh 03:56:43 i wasn't aware that tentacle spikes counted as a "kick" 03:57:55 hmmm, i'll just stick to one side orient for now, since explicitly placing in a corner is kind of lame 04:04:40 03kenranunderscore02 07https://github.com/crawl/crawl/pull/551 * 0.21-a0-81-g421f4c0: Some improvements to a bazaar vault 10(in the future, 1 file, 14+ 12-) 13https://github.com/crawl/crawl/commit/421f4c050a8d 04:05:51 %git 3093310fbde4d5918d3f4a179848b93244a7d743 04:05:51 07PleasingFungus02 * 0.16-a0-420-g3093310: Arguably refactor aux attacks 10(2 years, 10 months ago, 2 files, 221+ 116-) 13https://github.com/crawl/crawl/commit/3093310fbde4 04:05:57 This is the commit that introduced the bug 04:06:14 Kicks have been doing far less damage than than they should for almost 3 years 04:06:25 arguably 04:07:12 ...I realize that link is a 404 04:07:44 but you can always "git diff 3093310fbde4d5918d3f4a179848b93244a7d743^ 3093310fbde4d5918d3f4a179848b93244a7d743" if you're curious 04:08:21 No wonder tengu felt bad 04:09:27 heh 04:09:39 well, there are arguably other reasons. but that is probably a considerable contributor 04:20:53 minmay: fwiw that's the same as `git show 3093310f` 04:24:27 ah 04:24:32 yes that is easier 04:35:13 dang 04:35:22 well 04:35:29 PF did say "arguably refactor" 04:35:58 hehehe 420 commit bet he was high when he made it blaze it 04:46:19 do you really want to hear the duvessa ceiling fan story again 04:59:44 .gfkmap map~~wizard_prison 04:59:45 6. billohreally the Warrior (L19 DsFi of Okawaru), splashed by a deep elf sorcerer's acid on Depths:2 (evilmike_wizard_prison) on 2017-06-22 01:05:14, with 292277 points after 53550 turns and 2:11:54. 07:41:31 is it possible to get shoals-like tide behaviour elsewhere? 08:11:26 just emailed john but i restarted CAO 08:11:35 i'm on a business trip so very unresponsive but will keep an eye out 08:33:00 rax: thanks a lot 08:36:11 %git 6d440e639429f90010e7edeb4b5ab9c0f799f624 08:36:11 07PleasingFungus02 * 0.16-a0-884-g6d440e6: Arguably refactor aux attacks 10(2 years, 10 months ago, 2 files, 221+ 116-) 13https://github.com/crawl/crawl/commit/6d440e639429 08:36:22 %git 9187861b2f459159172a40db871ef680a8348ed8 08:36:22 07PleasingFungus02 * 0.16-a0-883-g9187861: Refactor punch aux-attack code 10(2 years, 10 months ago, 2 files, 54+ 67-) 13https://github.com/crawl/crawl/commit/9187861b2f45 08:37:10 so cao let me get to the dgl menu via ssh, but webtiles won't connect 08:40:42 <|amethyst> ok, webtiles should be working too 08:40:43 |amethyst: You have 1 message. Use !messages to read it. 09:00:35 that kick bug is weird because it's hard for me to see how PF could have done that accidentally given the nature of the refactor, but it can't possibly be on purpose as written 09:00:54 maybe it's just the order in which he refactored different attacks, and he did that one first or something 09:03:05 03advil02 07* 0.21-a0-80-g1f6c8c0: Restore base value of 5 max kick damage (minmay) 10(16 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/1f6c8c026d15 09:10:16 advil: how long does it usually take for this kind of patch to be deployed in trunk? 09:10:33 or does that differ substantially between the server 09:10:35 s 09:10:40 <|amethyst> ??rebuild 09:10:41 rebuild[1/2]: http://crawl.akrasiac.org/rebuild/ http://underhound.eu:81/rebuild/ http://crawl.berotato.org/crawl/rebuild/ http://crawl.xtahua.com/rebuild/ https://crawl.jorgrun.rocks/rebuild/ Bug |amethyst or Nap.Kin for CDO. Use your powers wisely. 09:10:42 <|amethyst> ??rebuild[2 09:10:42 rebuild[2/2]: Trunk update start times in UTC (second number during summer time): CAO: 1700; CDO: 0600/0500; CUE: 2300/2200; CBRO: 0800/0700; CXC: 0500/0400; CPO: every 15 minutes; CJR: every hour 09:10:42 the servers update on different cycles, but everything updates within a day 09:10:52 and we can trigger manual rebuilds 09:11:06 oh nice I didn't actually know about rebuild[2 09:11:08 ah cool, so I can play a Te tomorrow :D 09:19:10 !lg * te !alpha s=cv o=-cv / won -graph 09:19:13 556/96272 games for * (te !alpha): https://shalott.org/graphs/7069139ec9200a2b610908e88e68689e41790b62.html 09:51:32 any oddball reports or server oddities on cao? 10:09:42 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-80-g1f6c8c0 (34) 10:40:40 how often does crawl save to cs? 10:40:47 every turn? 10:42:22 <|amethyst> when you enter a level 10:42:37 <|amethyst> but that also includes things like using X[] to look at a different level 10:42:46 <|amethyst> or even Jiyva's passive offlevel item eating 10:48:58 ok. so my server-crashing oopsie may have caused a partial rollback of one level? 10:51:21 <|amethyst> or corruption of a save, though I think we at least try to be resilient against killing the crawl process 10:54:24 I suppose this will be a test of that 10:55:15 |amethyst: how would recommend killing persistent zombie crawl processes at 99%? 10:55:24 when kill PID doesn't work 10:58:30 <|amethyst> kill -9 is fine, just don't give -1 as the PID :) 10:58:54 <|amethyst> if it's using 99% CPU for a long time, it's almost certainly not in the process of saving 11:00:28 yea. I wasn't worried about that particular process. ty 12:54:44 03Johannes M02 {amalloy} 07* 0.21-a0-81-gc5150db: Add an autumn forest themed bazaar vault (#551) 10(9 seconds ago, 1 file, 61+ 0-) 13https://github.com/crawl/crawl/commit/c5150db817a2 12:55:44 aw dangit, github's squash+merge didn't make it clear to me i was doubling the commit message. oh well 13:09:34 Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-81-gc5150db (34) 13:21:09 -!- Mindiell_ is now known as Mindiell 13:53:53 amalloy: thanks for merging 13:57:03 New branch created: pull/553 (1 commit) 13https://github.com/crawl/crawl/pull/553 13:57:03 03cut1less02 07https://github.com/crawl/crawl/pull/553 * 0.21-a0-81-g4a5dd93: (Console) Use magic mapping colors 10(14 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/4a5dd939d268 14:35:00 -!- lobf_ is now known as lobf 15:37:46 I use FTILE and TILE extensively in a vault I'm working on (another bazaar), but after saving and coming back into the game, the colours have changed. I assume they changed to what the vault would look like without (F)TILE. Is it the branch epilogue that is doing this to me? 15:44:47 i'd guess not: it certainly looks like the branch epilogue is only supposed to be run during map generation 15:46:17 then that's rather strange 16:40:32 kenran: you use tile settings in your forest clearing bazaar, right? does it behave the same? if not, then i'd say try to think of things that are different between the two that might cause this 16:42:52 amalloy: I think it actually might, though it's not as obvious. yesterday I was thinking that after loading, I had way more green trees than there should be. I'll check that out soon! 16:43:25 it might just be me doing something wrong in both vaults :) 16:59:58 -!- VoidFox is now known as Guest92685 17:03:07 fwiw i built tiles and then save/loaded a game in your forest clearing a few times, and the trees seemed to stay yellow 17:04:46 that's great to know, thanks! I'm currently building tiles as well to check another vault in an older version on a hunch 17:08:05 though every time i start crawl on latest master, i get a bunch of errors reading "Lua error: global_prelude: bad header in precompiled chunk". anyone know what that's about? this is after a `make clean` just in case 17:08:47 the game appears to play fine, aside from printing that message like a hundred times 17:11:33 I checked some other vaults and whenever I load a character, the halos around the shops are missing. this is a version from before gammafunk's halo-related commit, to rule that out at least 17:19:24 amalloy: where can you see these error messages? do they appear in the terminal you started from? 17:19:37 yeah 17:20:09 I'm not getting any of those when starting tiles at least (latest trunk) 17:21:13 <|amethyst> amalloy: some kind of problem with you .des cache, try removing saves/des/* 17:22:37 ah, thanks |amethyst. i thought it was something to do with the des cache but didn't know where that was. that fixes it 17:23:31 -!- VoidF0x is now known as Guest19391 17:26:00 -!- jefus- is now known as jefus 17:26:34 kenran: it looks like setting the halo calls dgn_floor_halo which calls tile_floor_halo which seems primarily to set env.tile_flv. i guessed that tile_flv is perhaps just not serialised, but tag_construct_level_tiles seems to serialise it. that's about as far as i can get in debugging tiles 17:28:32 I'm not sure I understand all of that, but thanks for looking! I'll probably dive into that a bit as well, because I could confirm some issue with my trees at least (apart from the halo thing, which seems universal). The tree issue is the same as the one in my new vault; it has to do with a combination of COLOUR and TILE I think 17:31:51 while testing this, I stumbled upon a couple of (presumably older) bazaar vaults where initial design decision get overridden by the bazaar randomization in fixup_bazaar(). I don't think that was intended, but either making the colour randomization toggleable (as gf did with the halo) to revert these vaults back to their intended state, or just removing the respective headers are options 17:57:12 is there any way to figure out what vault was generated on a floor, online? There is runed doored room that that the monsters can just walk out of...seems like a design flaw. 17:57:33 <|amethyst> only once the game is over 17:57:47 hmmm... 17:58:06 could you peak at my game and see if i am seeing this right? maybe i am wrong. 17:58:23 <|amethyst> which server? 17:58:26 CJR 17:58:45 <|amethyst> can you press X? 17:58:54 <|amethyst> and scroll down a bit 17:59:25 it looks like those walls should be closed in...otherwise why have a runed door 17:59:58 <|amethyst> my guess is this is a vault with some randomness, and one of the variations opens the walls but doesn't remove the rune 18:00:01 this is a goof off run, so i dont mind just quitting to figure out what vault it is 18:00:14 <|amethyst> I think there was a minmay vault like that, this may be the same one 18:00:15 yeah, some the_grid variants can do that 18:00:39 you could cut the runed door from the subvault in question entirely and it would probably be fine 18:01:11 when im forcing the vaults to spawn with &L, it just makes a shitload of < instead of floor tiles. Is this proper? 18:01:13 there are some other vaults where this can happen easily due to monster digging though 18:01:44 seems to be crystal and stone, so in this case probably not digging 18:01:55 <|amethyst> nikheizen: what vaults? 18:02:03 ones i created 18:02:12 i have entry points 18:02:23 should a file a ticket? and if so what do you want in there, not sure what to put for htis. 18:02:33 but it also has an upstair sometimes, is that bugging it out? 18:02:49 <|amethyst> that does look kind of like one of the the_grid vaults, perhaps minmay can look at your map and confirm 18:02:53 !vault minmay_the_grid_triangle_9x9_subtriangle 18:02:53 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/variable/the_grid.des#L3950 18:03:08 <|amethyst> nikheizen: does it work normally when generated normally? 18:03:17 this is the one that can generate a "pointless" runed door 18:03:28 Unstable branch on underhound.eu updated to: 0.21-a0-81-gc5150db (34) 18:03:33 |amethyst: yeah seems to 18:03:55 minmay: you want me to file a ticket about it? or leave it to you? 18:03:58 It's intentional just because there's no good way to get rid of the runed door only when it gets mashed, but should probably be changed nonetheless (probably by just removing the door) 18:04:13 pointless would describe most runed doors i think 18:04:50 yes 18:05:07 <|amethyst> "runed" tries to do too many things 18:05:43 it slices, it dices, it makes julienned players 18:06:49 oh wait i diagnosed it, fuckup on my part 18:06:53 <|amethyst> nikheizen: hm... I can't think of a reason it would do that unless there's a bad subst or something... could you post your map somewhere 18:06:55 turned out to be an amazing vault with 4 unrandart pieces of jewellery in it 18:06:57 <|amethyst> oh :) 18:09:22 *unrandart*? that has an impressively low chance of happening 18:09:24 yeah i was shuffling {/. instead of doing a weighted subst... oops 18:09:40 yeah 2 rings, 2 necklaces 18:09:46 maybe he means randart 18:09:48 you probably mean randart 18:09:50 anyway I support removing that runed door but won't submit a PR because it's a 1 character change 18:09:51 yeah sorry 18:09:52 and not unrandart 18:09:54 not unrand 18:10:02 yes, that's more likely :P 18:10:07 !vault minmay_the_grid_triangle_9x9_prize_special 18:10:08 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/variable/the_grid.des#L4472 18:10:22 minmay: you can use the github gui to generate a PR 18:10:40 or one of us can just change it 18:13:24 while you're at it 18:13:35 please remove the runed door from cheibrodos hotspot 18:13:49 a fire dragon is a perfectly reasonable lair:$ monster 18:16:03 yeah I felt that way as well 18:16:07 but that vault is fairly intense 18:16:50 !lairendkills current 18:16:52 Lair End kills (current): 64x minmay_lair_end_frog_pond (14.32%), 61x gammafunk_lair_ancient_temple (13.65%), 50x cheibrodos_lair_end_hotspot (11.19%), 43x grunt_lair_end_beastmaster (9.62%), 37x hangedman_lair_in_review (8.28%), 36x hangedman_lair_tendril_chambers (8.05%), 34x minmay_lair_end_enchanted_forest (7.61%), 31x evilmike_catoblepas_cave (6.94%), 31x wormcave (6.94%), 26x guppyfry_lair_e... 18:28:07 @??fire dragon 18:28:07 fire dragon (04D) | Spd: 10 | HD: 12 | HP: 73-106 | AC/EV: 10/8 | Dam: 20, 13, 1307(trample) | fly | Res: 06magic(60), 05fire++, 03poison, 12drown | Vul: 12cold | XP: 1073 | Sp: fire breath (3d24) [11!AM, 06!sil, 08breath] | Sz: Giant | Int: animal. 18:28:26 just leave regular doors to accomplish pretty much the same 18:31:04 well that would be fine if the vault didn't place 18:31:07 @?molten gargoyle 18:31:07 molten gargoyle (059) | Spd: 10 | HD: 7 | HP: 24-37 | AC/EV: 14/7 | Dam: 2004(fire:7-13) | 11non-living, 10items, 10doors, fly, unbreathing | Res: 06magic(60), 04fire+++, 10elec, 09poison+++, 12drown, 04rot+++, 13neg+++, 13torm | XP: 430 | Sp: b.magma (3d15) [11!AM, 06!sil] | Sz: Medium | Int: human. 18:33:12 a molten gargoyle is more dangerous than a fire dragon IMO 18:33:25 so it's kind of silly that you get 3 molten gargoyles and then the dragon is behind a runed door 18:35:34 i don't really agree with the first point, but 3 molten gargoyles are definitely more frightening than a single D 18:35:39 why are those more dangerous 18:35:46 *one of those 18:35:54 you can autoexplore into a fire/ice dragon earlier in lair anyhow 18:37:11 @?fire crab 18:37:11 fire crab (04t) | Spd: 10 | HD: 8 | HP: 41-61 | AC/EV: 9/6 | Dam: 1504(fire:8-15), 1504(fire:8-15) | Res: 06magic(40), 04fire+++ | Vul: 09poison | Chunks: 14noxious | XP: 778 | Sp: flaming cloud (d12) [11!AM, 06!sil] | Sz: small | Int: animal. 18:37:13 wow, vul: poison 18:37:16 I had no idea 18:41:31 @?apocalypse crab 18:41:31 apocalypse crab (00t) | Spd: 10 | HD: 8 | HP: 41-61 | AC/EV: 11/6 | Dam: 1509(chaos), 1509(chaos) | 05demonic, amphibious, evil, see invisible | Res: 06magic(60), 05fire++, 02cold++, 12drown, 04rot, 13neg+++, 13torm | Vul: 09poison, 08holy, 11silver | Chunks: 06mutagenic | XP: 447 | Sp: chaos breath (d12) [11!AM, 06!sil] | Sz: small | Int: animal. 18:41:42 wow, that's a lot of vulnerabilities 18:44:09 <|amethyst> my suspicion is that runed door was inspired by the one in wormcave 18:44:20 <|amethyst> which is there because it used to be a secret door 18:44:45 <|amethyst> s/is that/is that that/ 18:44:45 well, you're probably right that chei thought of the idea from there, but we did discuss it 18:44:50 <|amethyst> ah 18:44:55 he didn't want the vault to be too overwhelming 18:45:02 but I think based on this discussion I'll do a slight rework 18:45:23 and maybe thin out some of the threat in the vault a bit, use a normal door, and perhaps randomize what's in that chamber 18:46:19 frequently the dragon will get released which will make the vault more dangerous, but this can be mitigated by there being maybe one less molten on average and there sometimes being something like a fire crab or lindwurm in its place 18:46:54 fire crabs might actually get more kills, but it's not too bad if it stays behind that door 18:46:57 @?lindwurm 18:46:57 lindwurm (04k) | Spd: 10 | HD: 9 | HP: 40-58 | AC/EV: 8/6 | Dam: 20, 10, 10 | Res: 06magic(40) | XP: 644 | Sp: fire breath (3d18) [11!AM, 06!sil, 08breath] | Sz: Large | Int: animal. 18:47:04 thankfully those are not smart lizards 18:48:12 also we should probably do 0.20.1 soon 18:48:22 so if people would like to cherry-pick anything that hasn't been done already 18:48:25 please feel free 18:49:06 I have at least two commits for objstat+mapstat I need to make still, but some things still need to be cherry-picked 18:53:21 it's time to fire up the aws cpu instance 18:53:45 for anyone interested in looking through the list, i believe https://gist.github.com/amalloy/1aece79cfc349a0ef1d1677abdb2bc9e is all the candidates for cherry-picking 18:53:46 its brain is a neural net processor, a learning computer 18:54:52 er 18:54:54 that list 18:54:56 hm, i guess that list also includes things that have already been cherry-picked 18:55:13 yes 18:55:22 you'd probably want to just go back to the last cherry-pick 18:55:30 I don't think anything has been cherry-picked since 18:56:01 er, I guess that has to be the case, but I'm saying that there should be no missing "in between" potential cherry-picks 18:56:09 %git 0.20.0 18:56:09 07gammafunk02 * 0.20.0: Update the Debian changelog for 0.20 10(4 weeks ago, 1 file, 25+ 0-) 13https://github.com/crawl/crawl/commit/35d994404934 18:56:15 %git stone_soup-0.20 18:56:15 07amalloy02 * 0.20.0-15-g7f486f2: Don't let equipped-weapon accuracy affect throwing accuracy (minmay) 10(2 weeks ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/7f486f25589b 18:56:31 right, that is what i did. git log origin/stone_soup-0.20..origin/master 18:57:28 but of course it also includes commits in master that have mirroring commits on 0.20 18:57:32 yeah 18:57:56 I think I usually just look at the list of commits from the last cherry-picked trunk commit to trunk HEAD 18:58:22 not sure if there's a fancier way to do it from the 0.20 branch that can understand cherry-picked commits 18:59:07 <|amethyst> git cherry 18:59:21 pop your 18:59:35 yes, i have been trying to figure out how git-cherry works 19:00:58 `git cherry -v origin/master origin/stone_soup-0.20` seems to be the intended use, but it produces nearly the same list, notably including commits that have been cherry-picked already, like 431dd88b 19:05:12 <|amethyst> flip those 19:07:50 <|amethyst> git cherry -v origin/stone_soup-0.20 origin/master | grep ^+ 19:08:18 <|amethyst> since you're looking for "things in master that haven't been applied to 0.20 yet" 19:09:48 <|amethyst> the way things are named ("upstream", "head") in the documentation is based on a workflow where things get cherry-picked into trunk, not the other way around 19:10:34 <|amethyst> the way you wrote it looks for commits in 0.20 that didn't come from trunk 19:10:55 nice 19:10:57 <|amethyst> of course, it's not smart enough to handle commits that have been cherry-picked but required conflict resolution 19:11:20 <|amethyst> it just looks for equivalent diffs (modulo line numbers) 19:12:09 New branch created: altarstats (1 commit) 13https://github.com/crawl/crawl/tree/altarstats 19:12:09 03gammafunk02 07[altarstats] * 0.21-a0-80-g34f8187: Initialise temple and overflow altar vaults with each mapstat iteration 10(13 hours ago, 2 files, 13+ 4-) 13https://github.com/crawl/crawl/commit/34f81875a7ce 19:59:09 The build passed. (altarstats - 34f8187 #8470 : gammafunk): https://travis-ci.org/crawl/crawl/builds/246003714 20:28:06 momplestiltskin (L24 DrFE) ASSERT(in_non_diamond_int(r.start)) in 'ray.cc' at line 217 failed. (Orc:2) 20:45:18 *rage doesn't trigger on riposte, does it? 20:50:32 random thought 20:50:54 what if you could use telepathy with such precision that you could interface with an unsocketed computer chip sitting face down on a desk 20:51:05 er, telekenesis 20:51:38 ProzacElf: i believe it does 20:51:50 sweet 20:52:14 -!- amalloy is now known as amalloy_ 20:52:41 well, i'm not giving up the zealot sword on my deee, and i don't want to suffer dith wrath to switch to ash, so i might find out the hard way 20:52:42 xD 21:34:23 hey all, i recently tried pulling the latest git repo / installing from apt (ubuntu 16.04) and noticed an odd issue 21:34:24 espais: You have 1 message. Use !messages to read it. 21:34:28 !messages 21:34:28 (1/1) advil said (4d 10h 41m 30s ago): there've been scattered reports of something like that happening when num lock is on, though the reports aren't with tiles. If that's not it I have no idea. All keypresses? 21:35:05 keypresses are getting double-triggered for me 21:35:41 not all keypresses it seems though 21:35:45 for instance ctrl+a will double-trigger 21:35:51 amalloy_: yes, riposte does in fact trigger *rage 21:35:54 hitting enter will double trigger 21:36:50 seems like its the non character keys 21:36:59 !tell advil seems like its the non-character keys 21:37:00 espais: OK, I'll let advil know. 21:37:10 !tell advil ctrl+a, enter, tab, etc. 21:37:10 espais: OK, I'll let advil know. 21:37:32 Lamp's of fire trail is disrupted 13https://crawl.develz.org/mantis/view.php?id=11119 by Yermak 21:50:04 espais: did you try changing numlock if your keyboard has it? 21:50:04 advil: You have 2 messages. Use !messages to read them. 21:50:40 so actually it doesn't have a num lock 21:50:59 surface pro with a non-extended keyboard 21:51:08 ok, I guess it probably isn't that 21:51:19 though i do have it hooked up to a dock at work that has a normal keyboard...i'll check that situation tomorrow 21:51:32 but most times its just the non-extended 21:53:34 -!- amalloy_ is now known as amalloy 21:59:00 I doubt numlock plays a role but it would be useful to know if it's specific to the surface pro keyboard 22:01:03 there was another report of a similar problem involving a surface, https://www.reddit.com/r/dcss/comments/6i3pt4/keyboard_issue_with_ubuntu_build/ 22:01:14 no helpful information though 22:01:39 unfortunately this is the kind of thing that will be v. hard for us to debug without a surface (I don't think any current devs have one) 22:02:09 you know, speaking of numlock, is there a way to make shift+numlock keys do that semi-auto-explore thing that I can do with shift+arrows or shift+vikeys? 22:02:47 heh, that's actually an alt for me 22:03:53 oh heh 22:04:46 johnstein: https://crawl.develz.org/mantis/view.php?id=2486 22:08:03 hm 22:08:09 if i hooked up an external keyboard do you think that would help? 22:08:15 or is the surface aspect the troubling part 22:09:57 ok hooked up a usb keyboard with a numpad 22:10:09 numlock on or off it still happens 22:10:17 advil: so we can blame console? 22:10:56 just for ref: http://imgur.com/a/TEig1 (the off/on is a single ctrl+a hit ... key press confirmed) 22:14:51 espais: have you checked out https://crawl.develz.org/mantis/view.php?id=10817 ? May be worth a sanity check 22:16:02 not doubling here fwiw 22:16:10 (build is about 2 hours old) 22:16:24 what build do you have? 22:16:47 i've read that linux handles sdl keypresses differently than other OSs, but i'd imagine it would've been caught by now 22:16:51 sorry, OS 22:17:17 hm, I saw 'console' above. 22:17:19 i'll give that make chain a command 22:17:32 "This is Dungeon Crawl Stone Soup 0.21-a0-81-gc5150db" 22:17:49 I don't generally do local tiles build; this is a webtiles build I'm running in console 22:18:18 gotcha 22:33:51 ok, reporting in 22:33:58 that make string definitely fixed the issue 22:34:11 IIRC, i installed SDL from the apt repo 22:40:18 if this is an issue from apt's side, should i post about it somewhere official? 22:40:36 not sure how many ubuntu users you guys have/know of 22:41:13 i wanted to give local a shot because i keep having lag issues where i'm at with some of the nearby webtiles servers 22:51:35