00:01:57 03advil02 07* 0.25-a0-1021-g227e4c2: Fourth(!) try to fix the build on CAO/CBRO 10(75 seconds ago, 1 file, 4+ 3-) 13https://github.com/crawl/crawl/commit/227e4c2ee8d2 00:02:17 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1020-ga314768f3c (34) 00:02:51 I was going to ask how you're able to test recompilation so quickly on CAO, but I guess if the build doesn't make it very far... 00:03:05 heh yeah 00:03:12 and cao uses ccache I think 00:03:16 cbro is slower 00:08:03 hm, I think version 3 if I had used stdint.h would have worked 00:16:42 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1021-g227e4c2ee8 (34) 00:21:20 are you serious 00:24:16 03advil02 07* 0.25-a0-1022-gb6211d5: Try 3b to fix compilation on CAO/CBRO 10(16 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/b6211d5327e1 00:24:16 03advil02 07* 0.25-a0-1023-g2f40044: Fix another c++11ism that is not working on CAO/CBRO 10(49 seconds ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/2f40044526e8 00:24:59 gammafunk you will get your wish 00:25:08 !!! 00:25:28 03advil02 07* 0.25-a0-1024-gbfbf38f: Revert "Try 3b to fix compilation on CAO/CBRO" 10(26 seconds ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/bfbf38f2cb72 00:26:41 I don't know why that didn't work, I actually *found* the header file on cao and looked at the constants 00:26:52 hm, I guess probably it was already included somewhere else, and didn't rerun 00:27:12 but if this compiles I might be done with this project for the night 00:28:17 I guess we know which servers most need to update their compilers 00:31:52 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1024-gbfbf38f2cb (34) 00:32:14 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1024-gbfbf38f (34) 00:32:26 finally 00:40:03 whoa, a tavern spammer 00:40:09 is registration open again?? 00:44:13 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-1024-gbfbf38f (34) 02:51:40 once again I volunteer to run a discourse server. Can get it up and running in time for the next stable release 02:54:44 that would be cool 03:30:41 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3054-g1a545eca1e 03:44:12 Fork (gooncrawl) on crawl.kelbi.org updated to: 0.22-a0-995-g3b949e0392 04:40:48 Since the next major update was/is going to break save compat anyways; I'd stripped most the == 34 stuff out; just to clean out bulk...now having a weird issue where saves fail to read (or they could be written wrong, hard to tell) if the player has seen certain features. Only ones I know cause this for sure are Dry Fountains, Zot Traps and 04:40:49 Dispersal Traps. I'm rather weirded out since everything else seems to marshall/unmarshall fine. 04:41:23 And it's only if has seen as well; just marshalling/unmarshalling a floor with the features on it is fine. 04:59:06 what do you mean, "stripped most of the == 34 stuff out"? that is not how i would describe a correct transformation 05:05:15 Bcadren: I think you have made a rod for your own back by trying to do this before vanilla does it and would seriously recommend you revert and leave it in. Even speaking as someone whose laptop takes half an hour to build the game from scratch. 05:07:49 i imagine it has ~0 impact on build time anyway 05:08:20 ifdef has gotta be one of the cheapest things a c++ compiler does 05:08:23 I wouldn't be surprised but even if it shaved ten minutes off it, I wouldn't try and do it myself ;-) 05:17:39 Yea, it's not; I suppose, the correct way to do it, but...if I don't care about old saves because of breaking it anyways completely cleaning out dead code...idk...yea are probably right I just being stubborn because of sunk hours into it at this point...I would have caught it earlier if it had reliably ALWAYS failed to save/load, but whent he 05:17:39 player has seen certain features, it crashes while attempting to unmarshall an item...uhh? Best I can figure for some reason certain features are acting like they have an extra item that doesn't exist for some reason...but why it marshalled fine before and not now is...uhh? sorry IDK...I feel pretty stupid and powerless and depressed about not 05:17:39 being able to solve this quickly 05:26:03 -!- amalloy is now known as amalloy_ 05:30:16 |amethyst I do know with my dungeon shortening for my fork; I should fix up those pregen/saved ghosts at least; do you know where they are? I've never actually FOUND those ghosts. (at the very least the ones in the dungeon itself will be underleveled unless I rescale their old floors to the new floor difficulty). 05:34:47 Execuda (L27 OgBe) ASSERT(item.quantity == 1) in 'items.cc' at line 2324 failed. (Abyss:3) 06:15:57 03gammafunk02 07* 0.25-a0-1025-g0c75a98: Don't partly override cell cache with global data 10(5 hours ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/0c75a98c51e9 06:15:57 03gammafunk02 07* 0.25-a0-1026-ge6b3a59: Update floor for features next to damaging walls (12233) 10(32 hours ago, 1 file, 23+ 0-) 13https://github.com/crawl/crawl/commit/e6b3a590811a 06:22:22 Bcadren: we're not going to be breaking compatibility with this release 06:23:41 advil and I are still figuring out the best way to implement rolling support, and ebering suggested breaking compat before switching, but there's no time frame for any of that 06:24:54 yea, but I probably am with the dungeon shortening. Had one look at the bizarre behavior that happened when I shortened the dungeon; tried to see what mainline did last time y'all shortened a branch (no support was necessary; probably a change due to procedural generation causing it to be necessary now, idk) and went ehh...fine I'll just break it. 06:24:55 re unmarshalling woes, your best bet is to get a save file that can reproduce the issue consistently and quickly (ideally crashing as soon as you open it) and then getting real friendly with your debugger 06:26:42 As for what the weird behavior was...going down stairs on D:9 (when D:8 is the last floor of the dungeon) took me to...Ziggurat:27. 06:27:51 nice 06:28:50 aidanh: I think 0c75a98 should be fine, but let me know if that's not the case somewhow. Needed that change for the subsequent commit 06:31:42 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1026-ge6b3a59081 (34) 06:32:40 your guess is as good as mine :) 06:32:59 I could fix that case; but there are a lot of ones and it would be a lot of new tagging code. Maybe I should forget about intentionally breaking it and go back to trying to support and add new supports at each branch change...for the D:9 problem, add a daction on loading a save from old version that next time you load a floor, if it's the lowest 06:32:59 you've reached in D (and the entirety of D isn't pregenerated) delete the bad stairs, and add entries to the necessary places to ensure complete-ability. But... see, is there even a way to check which floors are pregened? 06:36:34 Alright. I'm going to take a break, eat something and think about this. I still think it might be fair to strip out SOME old tagging stuff from my fork, given that saves older than 0.22 literally can't exist for it. 06:38:07 !pr 1375 06:38:07 https://github.com/crawl/crawl/pull/1375 06:38:29 Bcadren: ^ you may be interested 06:50:30 I will say one thing I don't like about not having major version bumps is...how ugly some enums are getting. (some listing the same thing in two places with #if MAJOR_VERSION == 34 in one place and #if MAJOR_VERSION > 34, others getting where related entries are very far from each other.) You -can- clean that up on minor bump, but it's...I've done 06:50:30 it once on a few enums...you get this weird if tagminor < {foo} switch (enumVal) case 64: enumVal = 16; break; ...etc. course there's no point to that other than wanting pretty enums, so...*slow shrug* 06:59:00 well it's not about making a guess; it's related to a change you made in d8832a9 07:04:45 it looks right to me, but I can't say I understand that part of the webtiles code very well 07:07:14 yeah, that's where I'm at too; that relationship of cell buffers is a bit confusing with all the intermediate cell representations 07:10:04 Bcadren: re ifdef duplication with enums, see https://github.com/crawl/crawl/issues/1364#issuecomment-617278487 07:11:06 an obvious alternative is to have per enum tables that map marshalled ints to enums 07:11:53 that's in effect, having an enum that casts into another enum... 07:12:35 no: we'd just have a vector of enum values with some gaps 07:13:15 which sounds silly; especially since it's mostly to keep the enum that's actually actively used "cleaner" ...and maybe smaller? (total possible enum values including legacy over 256, currently active enum less than 256)...course that's...really...really not that important these days in terms of practical efficiency. 07:19:26 sure if 23 is saved 23rd position in the vector is 63, it's still similar to an enum casting into another; two int values that refer to the same specific item in different contexts. 07:22:27 i would say that trying to squeeze out bits like that is just misguided; there's gzip compression, so it accomplishes nothing; time would be better spent on actual architectural issues 07:35:49 03Aidan Holm02 07[ui-refactor] * 0.25-a0-1013-g9e6dd43: Extract ScissorStack to its own class 10(21 hours ago, 3 files, 93+ 46-) 13https://github.com/crawl/crawl/commit/9e6dd43dcc11 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1027-g7cf53f7: Allow any floor to be annotated with ! 10(3 weeks ago, 7 files, 194+ 53-) 13https://github.com/crawl/crawl/commit/7cf53f78967f 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1028-g79cd046: Fix whitespace after merge conflict 10(10 days ago, 2 files, 1+ 3-) 13https://github.com/crawl/crawl/commit/79cd0465326e 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1029-g032e3e1: Unbrace. 10(10 days ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/032e3e1ea65c 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1030-g563265e: simplify code, and add space to prompt 10(10 days ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/563265ee5ce7 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1031-g84695a4: Implement reviewer comments 10(8 days ago, 2 files, 8+ 16-) 13https://github.com/crawl/crawl/commit/84695a4810cc 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1032-gfd433de: Added exception handling for certain keypresses in certain prompts 10(4 days ago, 2 files, 15+ 3-) 13https://github.com/crawl/crawl/commit/fd433deefce0 08:30:26 03Peter Gerlagh02 07https://github.com/crawl/crawl/pull/1369 * 0.25-a0-1033-gbc5d042: change prompt_for_quantity into prompt_for_int, and remove bugfix 10(7 minutes ago, 2 files, 1+ 4-) 13https://github.com/crawl/crawl/commit/bc5d042f109f 08:39:11 thanks for your guidance aidanh and pinkbeast , I am going to -try- to mainline save compat through my rearrangement now...including still cleaning up enums, but using minor version tag to rearrange the once. (switch block; case (old value): newvalue), etc. :) 08:51:42 New branch created: enum-marshalling (2 commits) 13https://github.com/crawl/crawl/tree/enum-marshalling 08:51:43 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1027-g7639ba2: Reorganize code 10(22 minutes ago, 1 file, 10+ 5-) 13https://github.com/crawl/crawl/commit/7639ba2c1dd4 08:51:43 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1028-gb128cbb: Remove future code for marshalling monster types as shorts 10(23 minutes ago, 1 file, 0+ 13-) 13https://github.com/crawl/crawl/commit/b128cbb7bac6 08:56:10 no problemo 09:16:16 Build failed for enum-marshalling @ b128cbb7 https://github.com/crawl/crawl/actions/runs/102502028 (failed: Linux Build) 09:16:24 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1028-g61bbca5: Remove future code for marshalling monster types as shorts 10(48 minutes ago, 1 file, 2+ 13-) 13https://github.com/crawl/crawl/commit/61bbca52f7b7 09:19:00 03Aidan Holm02 07* 0.25-a0-1027-g4c3fe21: Move union/intersect methods to ui::Region class 10(23 hours ago, 4 files, 107+ 19-) 13https://github.com/crawl/crawl/commit/4c3fe21bcbb7 09:19:00 03Aidan Holm02 07* 0.25-a0-1028-gf27dea0: Extract ScissorStack to its own class 10(23 hours ago, 3 files, 93+ 46-) 13https://github.com/crawl/crawl/commit/f27dea0dada7 09:19:50 I won't be happy with these commits until we've queried the current scissors 09:20:59 lol 09:29:09 The build has errored. (enum-marshalling - b128cbb #13325 : Aidan Holm): https://travis-ci.org/crawl/crawl/builds/686093073 09:37:53 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1028-gf27dea0dad (34) 09:54:33 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1029-gadcc015: Reorganize code 10(85 minutes ago, 1 file, 10+ 5-) 13https://github.com/crawl/crawl/commit/adcc01533090 09:54:33 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1030-g4cc3ac3: Remove future code for marshalling monster types as shorts 10(86 minutes ago, 1 file, 2+ 13-) 13https://github.com/crawl/crawl/commit/4cc3ac345159 09:54:33 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1031-gafc2a92: Marshall all monster_type enums as uints 10(59 minutes ago, 3 files, 17+ 7-) 13https://github.com/crawl/crawl/commit/afc2a9255b80 09:54:33 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1032-g4cfacba: Deduplicate weird monster type compatibility stuff 10(40 minutes ago, 1 file, 17+ 23-) 13https://github.com/crawl/crawl/commit/4cfacba26791 10:14:15 The build passed. (enum-marshalling - 61bbca5 #13326 : Aidan Holm): https://travis-ci.org/crawl/crawl/builds/686101536 10:16:44 Build failed for enum-marshalling @ 4cfacba2 https://github.com/crawl/crawl/actions/runs/102556816 (failed: Linux Build,MacOS Build,Linux MingW Build,Catch2/codecov) 10:19:40 03advil02 07* 0.25-a0-1029-ga90a8e0: One more attempt to get UINT8_MAX from a header on cao/cbro... 10(61 seconds ago, 1 file, 7+ 5-) 13https://github.com/crawl/crawl/commit/a90a8e02b826 10:31:54 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1029-ga90a8e02b8 (34) 10:34:55 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1029-ga90a8e0 (34) 10:35:23 one alternative.. you could upgrade the compiler :) 10:35:32 well, i know cbro presents serious challenges on that front 10:35:47 (which is not to say that cao is any easier) 10:36:21 I mean, if either of these were my own server I would have done it quite a while ago 10:36:48 right 10:37:25 I do plan on trying to get the compilation chroot working on cao sooner or later 10:38:04 while i realize a lot of the difficulty is (as you say) social not technical, it's made me reconsider the conventional wisdom of optimizing for stability with e.g. lts releases 10:39:13 maybe it makes more sense in a non-volunteer setting 10:39:15 i'd almost prefer running servers on something like arch, but with packages kept back two weeks, and with a canary server running at the latest versions 10:50:48 btw for cao's chroot, did you directly move the data into the chroot? or mount it in. 10:52:43 so far I've just used the chroot for running cloud uploading scripts that needed python 3 10:52:43 anticlimactic I know 10:54:21 I think right now bind-mounting the data is the only viable option, but I'm still trying to figure out ways of gettting data off of there 10:55:56 I started to upload it to my b2 account and kind of lost that game of chicken for now; it's cheap to upload/store but it's so many little files that it will soon literally cost me money to get a directory listing 10:57:00 and then I circled back around to, why the hell do we save this stuff 10:58:08 My bucket is at 1,768,681 files, with some misc cszo data and ttyrecs from just the two worst accounts on cao (tstbtto, qw) 10:58:38 that's only 18G, and the total amount to be uploaded if it is all saved is in the neighborhood of 800G I would estimate 10:59:21 so something like 80 million files total in its current format? 11:01:24 I don't mind paying for the storage, it would end up only being ~ $5USD per month, but what I realized is that if I go this route it then becomes this intractable mess of future organizational debt with no obvious solution 11:02:01 and for something that probably 95% literally never has been accessed since creation 11:02:08 (too bad it's hard to get stats on that) 11:03:35 and here I was thinking that cao webtiles was running inside my rube goldberg contraption 11:03:35 no rush though 11:03:35 oh sorry :-/ 11:04:13 The build failed. (enum-marshalling - 4cfacba #13328 : Aidan Holm): https://travis-ci.org/crawl/crawl/builds/686127778 11:04:28 yeah, I ran through some of the scripts to test things out and have been using it on the server, but haven't tried to get it going for real yet 11:04:29 no worries, I know that server is a lot of work 11:05:16 the problem with bind mounting is that there's a lot of different folders that would need to be mounted in 11:05:38 I guess how I would put it is that it absorbs as much work as I give it, but largely proceeds chugging along with 0 work still 11:05:54 are there any other programs (dgl maybe?) that depend on the folder structure? 11:06:15 I think everything except the morgue/ttyrec directory can be copied in 11:06:37 that is the one that is 800GB or so 11:09:02 I was more thinking about the next time the chroot will need to be updated 11:09:17 it'd be nicer to have everything bind mounted in 11:09:37 ah I see, so it is more container-like? 11:10:19 do you know if there's any hidden costs to bind mounting a lot? or is it pretty much free? 11:11:40 basically free 11:13:39 and yeah, more container like 11:16:35 ideally the chroot upgrade process would be 1) unmount binds, 2) move a symlink, and 3) remount 11:31:07 03advil02 07* 0.25-a0-1030-g21ee38b: Fix some warnings on diverse compilers 10(79 seconds ago, 4 files, 18+ 18-) 13https://github.com/crawl/crawl/commit/21ee38bd7443 11:31:31 hm, are ttyrecs from accounts like qw worth keeping? 11:34:34 not sure how you'd define interesting vs non-interesting tvs; I guess you could look at game data from the corresponding games, but that seems likely to be slow 11:36:57 i was thinking more like 'bots aren't likely to be interested in their own ttyrecs' 11:37:34 then again, i'm not too sure that the story is with these ttyrecs anyway 11:38:14 any opinion on my suggestion of prompting users whether to save their ttyrec after game over, and defaulting to 'no'? 11:40:42 03Aidan Holm02 07[enum-marshalling] * 0.25-a0-1033-g22dd63c: fixup! Deduplicate weird monster type compatibility stuff 10(22 seconds ago, 1 file, 3+ 4-) 13https://github.com/crawl/crawl/commit/22dd63cde3ca 11:50:18 aidanh: don't like that, since the recordings are pretty helpful for us as developers/game designers 11:50:25 Build failed for master @ 21ee38bd https://github.com/crawl/crawl/actions/runs/102648971 (failed: Linux Build,MacOS Build,Linux MingW Build) 11:51:27 if someone dies to something somewhere, pretty useful for us to see what the player did exactly 11:52:07 what i had in mind was that it would be recorded, and deleted afterwards 11:52:32 sorry, after what? 11:52:43 after their game finishes (unless they choose to keep it) 11:53:08 although i guess that falls into the 'if someone dies to something somewhere' case 11:53:22 I think something like a 2 year (or whatever) lifespan policy would be what I'd do if I were setting up something new 11:53:23 right, that doesn't help us that much; we'd need to be able to query for deaths and watch the recordings 11:53:28 yeah 11:53:46 literally all of the qw / tstbtto ttryecs I archived are beyond that in fact 11:53:52 even a 3-6 mo lifespan is fine 11:53:55 tstbtto is a bot also btw 11:54:14 even then it'd be a shame if there were no way a player might designate their ttyrecs for longterm storage 11:54:30 they could download them? 11:54:31 some tvs are really good, so a facility for that would be helpful if there's to be a limit 11:54:41 they could, but we couldn't 11:54:57 oh i see, yeah 11:55:13 I'm spitballing a bit here, but I imagine if there were a way for people to explicity designate recordings for longterm 11:55:30 obviously that is its own can of worms 11:55:46 if we're forced to put a limit on them, that's what we have to do, I guess 11:57:26 someone just needs to put a few hours in and make a twitch-like ttyrec viewer that lets you browse your ttyrecs and mark them for archiving, etc 11:57:35 should be a quite project done over a weekend 11:57:36 thanks for volunteering! :) 11:57:39 s/quite/quick/ 11:58:07 nope, I used so legally I'm fine 11:59:16 I guess in truth if we can archive all of 2 years or a year or even half a year as ebering says 11:59:29 that's going to cover day-to-day footv watching just fine 11:59:51 it's more that when people go back to watch the ttyrec of that one time they got .owned 11:59:54 .owned 11:59:55 3. [tdam=93;mhp=88] gammafunk the Reanimator (L13 MuNe of Sif Muna), blasted by an orc zombie (exploding inner flame) (animated by the player character (hexed by the player character)) on D:11 on 2015-06-27 20:58:47, with 40342 points after 23764 turns and 2:13:42. 12:00:09 re being able to query for deaths and watch recordings, is that for crash debugging? that does sound like something a significantly shorter window would cover 12:00:30 aidanh: it's for game design and just for fun :) 12:00:47 yeah, I would guess that 95% of ttyrecs I've watched for debugging purposes are very recent at the time 12:00:53 basically, it's very helpful for us to see what players die to in the stuff we make (monsters, vaults) 12:01:17 and yeah, even for non-debugging purposes we mostly need recent 12:01:19 .gfgk * day 12:01:20 2. Penny the Cryomancer (L26 NaIE of Vehumet), demolished by a caustic shrike on Depths:1 (gammafunk_ghost_berserking_beasts) on 2020-05-12 00:42:00, with 587939 points after 55384 turns and 4:41:21. 12:01:24 a command I use a lot 12:01:39 if I see one that looks interesting, I watch it 12:01:45 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1030-g21ee38bd74 (34) 12:02:14 gotcha 12:02:53 6 months would definitely cover e.g. trunk development; more time in the past is nice, since stable is also played a lot 12:03:03 aren't they independent? 12:03:56 they are, but previous version has a longer history (mostly in its alpha, of course) 12:07:05 I'm definitely not trying to dictate to server owners, to be clear; it's a decision they get to make as to how much ttyrec they store 12:17:06 The build was broken. (master - 21ee38b #13330 : advil): https://travis-ci.org/crawl/crawl/builds/686170393 12:20:01 03advil02 07* 0.25-a0-1031-g05ec187: Fix a local tiles error + a few more warnings 10(38 seconds ago, 2 files, 6+ 3-) 13https://github.com/crawl/crawl/commit/05ec187f372a 12:32:12 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1031-g05ec187f37 (34) 12:50:19 03advil02 07* 0.25-a0-1032-gd75fab4: Fix more warnings for tag_upgrade=true 10(44 seconds ago, 4 files, 7+ 0-) 13https://github.com/crawl/crawl/commit/d75fab455af9 13:02:15 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1032-gd75fab455a (34) 13:03:27 03advil02 07* 0.25-a0-1033-gb4f6ace: Endless warnings 10(32 seconds ago, 2 files, 4+ 0-) 13https://github.com/crawl/crawl/commit/b4f6ace0b662 13:10:36 I actually did find one genuine probable bug amidst all those warnings, so it was more or less worth it...zombie kills may not have been getting recorded correctly in some cases 13:17:09 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1033-gb4f6ace0b6 (34) 13:29:11 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1032-gd75fab4 (34) 13:53:10 03ebering02 07* 0.25-a0-1034-g6171570: Adjust the price of tremorstones 10(37 seconds ago, 1 file, 1+ 4-) 13https://github.com/crawl/crawl/commit/61715709bc10 14:01:58 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1034-g61715709bc (34) 14:40:49 The build passed. (master - 6171570 #13335 : Edgar A. Bering IV): https://travis-ci.org/crawl/crawl/builds/686226129 14:43:02 03advil02 07* 0.25-a0-1035-g7f99b00: Disable travis notifications 10(65 seconds ago, 1 file, 0+ 18-) 13https://github.com/crawl/crawl/commit/7f99b00658a4 14:43:35 we could maybe even turn off travis at this point? 14:46:57 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1035-g7f99b00658 (34) 14:58:41 I think so 15:27:09 03ebering02 07* 0.25-a0-1036-g6f112fb: Further tune down tremorstones 10(4 minutes ago, 3 files, 12+ 4-) 13https://github.com/crawl/crawl/commit/6f112fb28d0e 15:32:24 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1036-g6f112fb28d (34) 18:13:03 -!- Mike_ is now known as Guest71906 18:23:55 Unstable branch on underhound.eu updated to: 0.25-a0-1036-g6f112fb28d (34) 18:49:59 with travis, you would remove the notification configuration downt he bottom 18:50:30 *could 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1037-gf75c35c: Remove scarves of cloud immunity 10(2 days ago, 10 files, 21+ 35-) 13https://github.com/crawl/crawl/commit/f75c35ca80db 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1038-g97dead9: Move invisibility from cloaks to scarves 10(2 days ago, 1 file, 6+ 4-) 13https://github.com/crawl/crawl/commit/97dead9f7433 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1039-g8388af8: Add preservation (rCorr) and stealth to cloaks 10(2 days ago, 8 files, 15+ 14-) 13https://github.com/crawl/crawl/commit/8388af883bdd 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1040-g6a14d1d: Remove scarves of spirit shield 10(2 days ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/6a14d1d4ee8f 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1041-g19c8d55: Add Harm to scarves 10(2 days ago, 8 files, 14+ 2-) 13https://github.com/crawl/crawl/commit/19c8d5534f14 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1042-g0691371: Remove amulets of harm 10(2 days ago, 17 files, 13+ 46-) 13https://github.com/crawl/crawl/commit/0691371e69f8 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1043-g8a86455: Allow Harm on randart armour 10(2 days ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/8a86455914d8 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1044-gca3883d: Remove amulets of the gourmand 10(2 days ago, 28 files, 26+ 134-) 13https://github.com/crawl/crawl/commit/ca3883d1df94 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1045-gba0266b: Remove amulets of reflection 10(2 days ago, 26 files, 27+ 67-) 13https://github.com/crawl/crawl/commit/ba0266bf1e40 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1046-g459fc58: Remove amulets of Rage 10(30 hours ago, 33 files, 27+ 169-) 13https://github.com/crawl/crawl/commit/459fc583a17f 19:00:22 03ebering02 07[amu-cloak-scarf-shuffle] * 0.25-a0-1047-gac2cbc2: Remove outdated amulet descriptions from translations 10(29 hours ago, 13 files, 0+ 283-) 13https://github.com/crawl/crawl/commit/ac2cbc2f562a 19:00:35 Branch pull/1397 updated to be equal with amu-cloak-scarf-shuffle: 13https://github.com/crawl/crawl/pull/1397 19:01:17 Branch master updated to be equal with amu-cloak-scarf-shuffle: 13https://github.com/crawl/crawl/tree/master 19:02:14 huh, never seen it do that before 19:03:42 I guess that's whaat happens if I force push the rebased version before merging 19:03:56 in a close enough timeframe 19:04:06 yeah, it only checks every 60 seconds or something 19:05:33 I know everyone isn't totally happy with the state of harm, but the discussion seemed to be a mix of "its current risk/reward ratio isn't right" and "no such effect could work" 19:05:43 I'd believe the former but I don't believe the latter 19:32:00 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1047-gac2cbc2f56 (34) 19:39:36 New branch created: pull/1400 (1 commit) 13https://github.com/crawl/crawl/pull/1400 19:39:36 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1400 * 0.25-a0-1018-g6f2fa94: Show GDR in body armour descriptions 10(12 hours ago, 4 files, 39+ 13-) 13https://github.com/crawl/crawl/commit/6f2fa944d895 19:40:57 03ebering02 07* 0.25-a0-1048-g1529702: Fix a leftover amulet of reflection 10(26 seconds ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/1529702659a5 19:41:22 does sqrt come from #include ? 19:42:30 03ebering02 07* 0.25-a0-1049-gc3b2b15: And a few more 10(26 seconds ago, 3 files, 2+ 6-) 13https://github.com/crawl/crawl/commit/c3b2b15ad081 19:42:40 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1400 * 0.25-a0-1018-gc35144a: Show GDR in body armour descriptions 10(12 hours ago, 4 files, 40+ 13-) 13https://github.com/crawl/crawl/commit/c35144a8ddcc 19:42:51 ??rebuild 19:42:51 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.kelbi.org/rebuild/ Bug gammafunk, advil, |amethyst, or Nap.Kin for CDO. Use your powers wisely. 19:45:41 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1049-gc3b2b15ad0 (34) 20:39:56 afualwayswins (L27 GrFi) ASSERT(is_armour_brand_ok(item.sub_type, ego, true)) in 'makeitem.cc' at line 886 failed. (Depths:1) 20:47:43 !crashlog 20:47:44 21961. afualwayswins, XL27 GrFi, T:519144 (milestone): https://crawl.kelbi.org/crawl/morgue/afualwayswins/crash-afualwayswins-20200513-003946.txt 20:50:11 oh heh you reused preservation didn't you 20:50:45 though that looks like it should be ok 20:50:58 except for the tag major part 20:55:11 I think it must be stealth 20:56:54 ??spirit shield 20:56:54 guardian spirit[1/1]: While wearing this amulet or scarf, damage you take is divided proportionally between MP and HP. Reduces magic to 0 on wearing. Note that a Deep Dwarf wearing this does not regain MP naturally. Other races that mutate slow healing 3 still do, however. 20:57:20 gammafunk: player ghost zig death https://www.reddit.com/r/dcss/comments/gikehj/yasd_megazigging_muwz_of_gozag/ 20:57:35 nice 20:57:43 yeah there haven't been too may of those so far 20:57:54 !learn edit guardian_spirit s/ Other races that mutate slow healing 3 still do, however.// 20:57:54 !lg * zig current alpha cikiller=player_ghost s=name,char 20:57:54 guardian spirit[1/1]: While wearing this amulet or scarf, damage you take is divided proportionally between MP and HP. Reduces magic to 0 on wearing. Note that a Deep Dwarf wearing this does not regain MP naturally. 20:58:42 5 games for * (zig current alpha cikiller=player_ghost): Hartz (GrEE), ragingrage (GrWn), Spect (DDFi), tv1990 (MuWz), will825x (OpEE) 20:59:15 just three years after I changed that behaviour 21:04:10 03advil02 07* 0.25-a0-1050-g88e6ef3: Update makeitem.cc 10(74 seconds ago, 1 file, 4+ 3-) 13https://github.com/crawl/crawl/commit/88e6ef3a42fa 21:05:59 did we decide on a feature freeze? 21:06:02 should we? 21:07:07 if the tournament were to start May 29 that would put it at this friday for two weeks 21:08:01 granted we haven't started implementing new rules, but I *think* that's doable in the time involved? 21:08:07 I will be able to spend time on it at least 21:11:36 yeah, same 21:11:56 I suppose this friday would be ok 21:12:15 I have some arrival vault reworks that could go in either before or after freeze 21:12:27 and some other vault tweaks 21:17:08 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1050-g88e6ef3a42 (34) 21:48:54 <|amethyst> Hm, controversial maybe but: I feel like the change to Searing Ray to auto-target goes a little against the newer "positioning-for-spells" philosophy 21:55:54 <|amethyst> I know ordinal directions are a little more artificial than "distance" or "number of transparent cells", but I feel like, in the most common case of only a single monster in range, the spell lost an element of player skill 21:56:41 <|amethyst> s/a little // 21:57:04 <|amethyst> or maybe just setting up a good shot made me feel clever 22:19:10 -!- amalloy_ is now known as amalloy 22:20:53 how do you set a monster attitude in arena? 22:24:46 advil: Friday is good for me for feature freeze; I was going to float it but you beat me to the punch 22:26:33 |amethyst: in the commit message I acknowledge this made it less positional, but requiring the player to understand the intersection of our beam aliasing and monster pathfinding algorithms seems a bit cruel 22:27:13 afualwayswins (L27 GrFi) ASSERT(is_armour_brand_ok(item.sub_type, ego, true)) in 'makeitem.cc' at line 886 failed. (Pan) 22:27:39 !crashlog afualwayswins 22:27:59 1kb left a comment warning me that being too strict in there would lead to asserts and I did not heed it well enough 22:28:15 &versions 22:28:24 1. afualwayswins, XL27 GrFi, T:529964 (milestone): https://crawl.kelbi.org/crawl/morgue/afualwayswins/crash-afualwayswins-20200513-022653.txt 22:28:26 CAO: 0.25-a0-1032-gd75fab4, CBRO: 0.25-a0-1024-gbfbf38f, CDO: 0.25-a0-1007-g10a8c3caa8, CKO: 0.25-a0-1050-g88e6ef3a42, CPO: 0.25-a0-1050-g88e6ef3, CUE: 0.25-a0-1036-g6f112fb28d, CWZ: 0.25-a0-734-g73332ae, CXC: 0.25-a0-1024-gbfbf38f2cb, LLD: 0.25-a0-1013-g3a3e078c59 22:29:32 oh, uhoh 22:29:46 this seems to be in the acquire uncancel. 22:30:02 hm no no it doesn't 22:43:54 03ebering02 07* 0.25-a0-1051-gcc65eb4: Remove another too clever fallthrough in makeitem.cc 10(68 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/cc65eb46a3d9 22:44:15 oh that's one thing I do need to finish before freeze, I guess 22:44:15 well, debatably 22:44:15 def. a release goal 22:47:02 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1051-gcc65eb46a3 (34) 22:50:10 I'll work on that now that the accursed frozen ramparts tile changes are figured out 22:51:08 btw, for console players, a poll: should slimy and icy walls (from frozen ramparts) just go ahead and recolour non-solid features adjacent to the damaging wall? 22:51:31 I was going to update console to do this, but it seems disallowing those to be recolored is explicitely mentioned 22:51:52 but it seems fine to me to recolor open doors, fountains, arches as lightgreen/lightcyan when near slimy/icy walls 22:52:35 for shallow water we could do likewise, but there is one wrinkle for amphibious species and dark water 22:52:44 s/dark/deep/ 22:53:17 we could use green for deep water near slimy walls, but for deep water near icy walls there's not really an appropriate color, unfortunately 22:54:27 an alternate/related idea is to show an indicator under `x' like "near a(n) slime-covered(ice-covered) wall" 22:54:38 which could be done either with or instead of recoloring 23:26:02 IDK if you care what I think, as a console player unfamiliar with the look of Frozen Ramparts; but if you do I'd be happy to look at A/B images. 23:32:16 hmm travis was actually generating test failures for these is_armour_item_ok issues, I wonder why gh actions didn't? 23:32:21 and right after I disabled IRC too 23:32:27 *IRC notifications 23:33:26 is the answer that we're not actually running tests on gh actions?? 23:40:34 it also looks like this is not testing monster 23:40:47 ok, so my claims of superseding travis were largely false 23:44:26 03advil02 07* 0.25-a0-1052-g5e67357: Try to make actions run `make test` 10(63 seconds ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/5e67357dd7e7 23:45:36 haha denied 23:46:53 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1052-g5e67357dd7 (34) 23:47:38 03advil02 07* 0.25-a0-1053-ge6544ee: ... 10(27 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/e6544eeaff64 23:59:18 uh, `make: *** No rule to make target 'test'. Stop.` 23:59:21 what am I missing