00:55:26 03gammafunk02 07* 0.25-a0-590-ga39ba6b: A clua interface for scrolls of acquirement 10(2 days ago, 3 files, 63+ 7-) 13https://github.com/crawl/crawl/commit/a39ba6bebb3a 00:56:30 if any lua guru's can answer the question at the end of that commit message about a better way to pass that lua item array, would appreciate it 01:02:54 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-590-ga39ba6bebb (34) 01:03:28 03gammafunk02 07* 0.25-a0-591-g2ad0d86: Complete the documentation of c_choose_acquirement 10(86 seconds ago, 1 file, 5+ 3-) 13https://github.com/crawl/crawl/commit/2ad0d861f170 01:17:30 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-591-g2ad0d861f1 (34) 02:20:52 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-2989-g2cb1824106 03:11:01 Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-591-g2ad0d861f1 03:11:54 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-591-g2ad0d86 (34) 03:48:34 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-443-g80245de385 10:22:32 Stable (0.22) branch on underhound.eu updated to: 0.22.1-61-g4f546fdb5a 12:12:26 -!- krator44-- is now known as krator44 12:19:07 I am trying to toy around with the code, to learn it. Where do we decide what type of body armour we generate unrands for? 12:19:08 crawl-ref/source/artefact.cc 12:19:42 That seems like the likely place, but other than robes, it doesn't appear to include logic like 'pick plate' or 'pick scale mail' as the base armour type. 12:20:13 Also, I am a VIM kid, but does anyone here use VisualStudio or Code::Blocks? 12:23:09 Stable (0.23) branch on underhound.eu updated to: 0.23.1-93-ge536e68a2c 12:23:58 AFAIK you're looking at _try_make_item_unrand in makeitem.cc which invokes find_okay_unrandart 12:24:45 But by the time you get there, the base type and perhaps subtype are already known. 12:25:07 Beyond that it's at random, so "pick plate" is just down to what proportion of armour unrands are plate. 12:28:47 Thanks, Pinkbeast. I think I am having problems with code navigation. My C++ is certainly rusty, following the code flow to reason my way around the system isn't natural yet. 12:37:20 Ah, _get_random_armour_type in makeitem.cc. 12:50:49 theJollySin: I'm not sure that's involved. 12:51:43 I think there are two possibilities here - the stuff mentioned above is invoked just to try and make unrand armour, in which case (say) the odds of getting the OCPA are 1/ 12:53:30 Or it may be invoked to "try and make unrand armour, and it's a CPA" in which case _get_random_armour_type would be involved 12:53:36 And IDK what the typical path is. 12:55:16 Damn. I am sorry, my initial statement was off. Damn it. I am looking for *randart* generating code, not unrandart. 12:55:32 No wonder this conversation went sideways on me. That's an important typo. 12:55:32 yeah, was gonna say, the unrand armour types are hard-coded 12:56:14 theJollySin: OIC. Then indeed it's "_get_random_armour_type" and maybe we try and make a randart out of that 12:56:19 I was just going to play around with the *randart* code, to learn more about the item system and the acquirement system. *facepalm* 12:56:24 (that would be in art-data.txt) 12:56:54 advil: Indeed, but I thought the question was about which one you get 12:59:05 theJollySin: You might find https://github.com/damerell/crawl/commit/7981e0b6151dd680665f4d6fb092ca09228a45c7 interesting as a commit that adds a new randart property, a sort-of-new kind of randart, and futzes around with the default way properties are generated. 13:00:08 Ah, neat. Thanks. 13:00:19 there is code to make randarts out of unrands, and it uses either OBJ or FALLBACK (if defined) in art-data.txt 13:00:29 for regular randarts, you probably want to have a like at `make_item_randart` in artefact.cc 13:00:41 especially _init_artefact_properties 13:00:51 (NB it's not in vanilla's coding style because I don't have to follow it, and also it's written by a bad C++ programmer) 13:01:51 then you would want to follow up on how make_item_randart is called (e.g. by `git grep make_item_randart`) 13:02:18 *have a look at 13:02:40 advil: JOOI, do you know the answer to the question I thought was asked? (ie, "given I have found an unrand, are the odds it's the OCPA a) 1/ b) about the same as the odds a random randart is a CPA c) something else d) dependent on where/how it was found?") 13:03:53 for armour in particular a key place to look is also in `_try_make_armour_artefact` in makeitem.cc 13:04:19 Pinkbeast not sure of the odds immediately 13:04:32 there's lots of vaults that specify particular unrands 13:04:49 key call is `_try_make_item_unrand` in makeitem.cc 13:07:05 Oh, not counting the vaults, fair point 13:07:17 where the odds are mostly going to be determined by find_okay_unrandart, which is doing an even odds thing as long as the constraints it is provided are met 13:07:42 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-591-g2ad0d86 (34) 13:07:52 i.e. that function is just doing reservoir sampling 13:08:05 Mmm, that's as far as I got but I hadn't yet worked out if the constraint is usually "it's armour" or "it's armour of type foo" or "don't care" 13:08:10 but I'm not sure how often / what kinds of constraints are usually provided 13:09:10 if the call path is from _try_make_item_unrand it's going to depend on the probability of the base item 13:14:19 where that could happen randomly, or happen in response to some item placement that could have other restrictions (e.g. I'm pretty sure something like `quick blade good_item` could place the captain's cutlass) 13:17:26 _get_random_armour_type in makeitem.cc is fairly surprising to me. To my memory I've never seen a randart scarf, fire/ice/quicksilver/swamp dragon armour. 13:17:50 theJollySin: You won't see randart scarves because vanilla only had them in er 0.19 or so 13:17:57 ah 13:18:56 +1 ice dragon armour "Ynnofu" {rF- rC++ MR++ Regen+ Int+3} # vanilla 0.18 13:19:16 +11 ice dragon scales "Pomudig" {rF- rC++ MR+ Stlth+} # vanilla 0.19 13:19:27 They're out there. 13:19:44 But they're not common because the overwhelming majority of dragon scales you see are got by killing dragons. 13:25:51 !source _armour_disallows_randart 13:25:51 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/makeitem.cc#L697 13:33:39 Oh, huh. The opposite of that commit I was looking at. Okay. 13:34:47 theJollySin: Er, so the commit I linked you, I linked because it messes around with randart generation, not because it's in vanilla (which it's not, but I'm not aware of any recent vanilla changes to show you) 13:34:47 Pinkbeast Yeah, I see they are *possible*. Still, those things you linked are great fun and I'm a little sad I don't see them in game. 14:20:30 %git 14:20:30 07gammafunk02 * 0.25-a0-591-g2ad0d86: Complete the documentation of c_choose_acquirement 10(13 hours ago, 1 file, 5+ 3-) 13https://github.com/crawl/crawl/commit/2ad0d861f170 14:21:00 advil: not sure if you saw this, but a pose a sort of lua question at the end of this commit message, so if you have any insight about whether what I did was the right approach or could be improved, let me know 14:21:12 er, sorry 14:21:15 %git HEAD~1 14:21:15 07gammafunk02 * 0.25-a0-590-ga39ba6b: A clua interface for scrolls of acquirement 10(3 days ago, 3 files, 63+ 7-) 13https://github.com/crawl/crawl/commit/a39ba6bebb3a 14:21:20 that one 14:22:19 feels like I should be able to pass an argument to c_choose_acquirement that is the array of item.Item but the clua argument passing stuff is a bit weird 14:23:22 in other news, qw can now win 0.25 games unattended again and can properly consider 0.25's acquirement scrolls: https://github.com/elliptic/qw/pull/30 14:24:12 was thinking maybe we could get a copy of qw repo into the crawl org so we can merge some things like this to keep qw viable for testing 14:24:31 maybe it'd be better to just update the local copy of qw in the source tree, although I guess we could use a submodule for that 14:39:22 New branch created: pull/1313 (1 commit) 13https://github.com/crawl/crawl/pull/1313 14:39:23 03Byrel Mitchell02 07https://github.com/crawl/crawl/pull/1313 * 0.25-a0-592-g9aa1b82: Apply AC to Ramparts damage before compensating for monster speed (Flugkiller) 10(7 minutes ago, 1 file, 3+ 2-) 13https://github.com/crawl/crawl/commit/9aa1b82bd395 14:58:46 -!- Psymania_ is now known as Psymania 19:22:42 Unstable branch on underhound.eu updated to: 0.25-a0-591-g2ad0d861f1 (34) 20:34:40 -!- TAS-2012v is now known as TAS_2012v 22:04:42 -!- Yermak_ is now known as Yermak 23:24:04 -!- bh is now known as Guest99181 23:29:03 -!- behalebabo_ is now known as behalebabo