00:20:20 Unstable branch on cbro.berotato.org updated to: 0.34-a0-1292-g7f56f43775 (34) 00:55:28 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-1292-g7f56f43775 01:19:38 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4917 * 0.34-a0-1266-g22b6a94834: Make spell name lookup prefer spells that exist 10(10 hours ago, 3 files, 24+ 3-) 13https://github.com/crawl/crawl/commit/22b6a948343c 01:59:07 -!- indigaz26 is now known as indigaz2 04:32:47 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 05:10:10 Unstable branch on crawl.akrasiac.org updated to: 0.34-a0-1292-g7f56f43 (34) 08:59:37 <08n​icolae> i assume that any remaining fountains from the pre-slime-pit civilization would have long been eroded into smooth lumps with a hole and would only barely look like an artefact of sentient beings. also "smooth lump with a hole" was my nickname in college 08:59:50 <08n​icolae> some kind of trickling fissure might work as a notional fountain 09:14:43 -!- TAS-2012v is now known as TAS_2012v 09:56:26 <04d​racoomega> This may sound like a strange question, but how much would people here be opposed if "X comes into view." was changed to "You encounter X."? Let me explain the line of reasoning that has caused be to arrive here. So, for a couple days now I've been investigating various wierdnesses with monster_seen interrupts and comes into view messages. There's... actually quite a lot of things about them that are weird and/or broken (beyond just 09:56:26 being incredibly tangled code, which they also are). Of course, there's the periodic reports of monsters not interrupting delays until they actually hit you (of which there are several causes), but also there are fairly normal circumstances where you will never get a 'comes into view' message (including a gear notification which might have triggered force_mores) for a given monster at all! (For instance, if the first time you encounter a monster is 09:56:27 because of it opening a door, you never get a gear warning. Often it doesn't even print a message about it opening the door, either; the event is just completely silent.) A lot of this is a consequence of trying to make these messages 'read better' by having a variety of alternations for 'X comes into view' depending on the way something comes into view. There's an 'X opens the door', 'X blinks into view', 'X is thrown into view' and quite a few 09:56:27 more. But, due to other code structure, most of these have been mostly broken from a long time. Worse, they also require splitting message-printing code for given actions between the action itself and handling of an activity_interrupt elsewhere. This makes things more confusing to read (imo) and is also done very inconsistently. 09:56:28 <04d​racoomega> While dissecting this, I've also seen a truly impressive amount of commits over the decades going extra lengths to ensure that any codepath that might result in printing a 'comes into view' message does so before other messages related to the monster, and despite all of that it still doesn't always happen (and rarely can still print a 'comes into view' message twice, even). In theory, I could fix some more of these, but I'm of the 09:56:28 opinion that any approach which inherantly splits message-printing for a single event into multiple places is brittle and uncomfortably opaque. Several of these 'alternate comes into view' messages don't even use the same wording as the normal message they're trying to replace. I think some of the problem is that we're essentially trying to conflate several different things together: 1) Interrupting delays due to monsters coming into view 2) Printing 09:56:29 a message about a thing that is actually happening (ie: a monster opening a door) 3) Printing a notification that you have encountered a new monster for the first time (possibly including its gear) (1) shouldn't be strictly dependent on (2) or (3), but beyond this (2) and (3) aren't even the same thing. When you think about it 'X comes into view.' isn't a message merely informing you that a monster walked into view; monsters walk into view constantly 09:56:30 without producing messages. What it is actually saying is "You have encountered a new monster for the first time." And, after having reviewed this code and commit history for a bunch of days now, I think that a bunch of problems stem from trying to squish (2) and (3) together, in part because of the 'comes into view' wording. It feels semantically wrong to say that something 'comes into view' after it did anything else that indicated it was already 09:56:30 in view (including opening a door or being recalled or shouting or whatever). And yet, despite so many commits trying to enforce this, it still has problems. 09:56:42 <04d​racoomega> Which has led me to the conclusion that it would be better if we were a little more formal about this message being a 'monster is first encountered' notification that wouldn't read wrong if used alongside other messages. "The orc knight is recalled. The orc knight comes into view." reads dubiously in a way that "The orc knight is recalled. You encounter an orc knight." - at least it does to me. And by making the message structure 09:56:43 feel a little more UI-oriented, it frees a bunch of code to not have to worry about this; door code can just print messages for opening doors without concerning itself with maybe-or-maybe-not-communicating that this was the first time a monster opened a door (there are so many ways for monsters to enter the player's LoS in this game!) This is a lot of words for a relatively small wording change, but the phrasing is very iconically Crawl, which 09:56:43 caused me to hesitate a bunch. But the more I work with this code, the more strongly I feel that this would allow some big improvements to some very messy code and read better once people get used to it. (Note: regardless of this, I also have plans for a UI option that will properly interrupt on first spotting given enemies without having to match messages printed - with shorthand for 'interrupt on first spotting any unique', for instance.) 10:11:19 <06d​olorous_84348> Sounds interesting. 10:18:16 <06d​olorous_84348> I figure "You encounter [monster]." is fine. The old phrasing may be somewhat iconic, but so was the old pre-Beogh message sequence of "The orc priest prays. Something smites you! You die..."; in other words, it's not so iconic that it's unchangeable. Besides, with the relatively new Blind status, I'm surprised we haven't gotten any nitpicky bug reports of "how can this monster 'come into view' while I'm blind?" 10:31:29 <08n​icolae> yeah i don't have much to say, other than you make some very good points and also if a minor phrasing change will result in a cleaner codebase well that sounds like a win to me 10:32:29 <08n​icolae> you could even get fancier with it The orc knight is recalled. There's an orc knight here! with an exclamation point or not based on threat 10:32:39 <08n​icolae> Captain, there be whales here! 10:33:31 <08n​icolae> anyway, like i said, i don't think whatever iconicity "X comes into view." has in the crawl playerbase is worth whatever side effects it's having on the program structure 10:33:51 <08n​icolae> go forth, and change the phrase 10:34:34 <08n​icolae> we've axed way more iconic things than that 10:50:45 <09g​ammafunk> like mimics 10:54:46 <08n​icolae> my personal favorite example is that we cut eyes of draining but their splash screen remains 11:07:46 <09g​ammafunk> hrm, I wonder why we have to use vsnprintf for putting a formatted message into a buffer 11:07:57 <09g​ammafunk> even call malloc ourselves 11:08:32 <09h​ellmonk> eyes of draining will come back 11:08:44 <09h​ellmonk> 🔜 11:09:21 <09g​ammafunk> this is a hellcrawl2 thing isn't it 11:13:39 <08n​icolae> no. put it in regular crawl. the people demand eyes of draining, and by people i mean mostly me, and by eyes of draining i mean mostly just the tile 11:15:07 <09h​ellmonk> I mean there will be a new eye of draining mons design for mainline 11:15:11 <09h​ellmonk> either from me or someone else 11:15:16 <09h​ellmonk> the people demand big eyeball 11:19:14 <08n​icolae> it's true 11:19:18 <08n​icolae> thank you, hellmonk 11:19:24 <08n​icolae> thank you for loving big eyeball 11:53:38 <09g​ammafunk> You're all in the pocket of big eyeball 13:09:17 <06r​egret-⸸nde※> (Making it an eye of max hp draining and just shoving it into mid / late D and Slime wouldn't take too much effort, and would help with the continued small efforts to make rN more relevant with spiky individual scary monsters, but it would technically require me finding a new icon and new vfx for Draining Gaze. It'd probably also require different colours than the splash screen's eye, but that one always was technically off 13:09:18 on its colours anyway?) 13:10:24 <11O​dds> I’ve always thought “X comes into view” is rather confusing. It took me a while to realise that it only happens once per monster. 13:11:26 <06r​egret-⸸nde※> I've never really worried about the correspondence between splash screen art and what's in the actual game, myself, though, since pretty much nothing in Denzi's set really matches what's in game. 13:14:17 <06r​egret-⸸nde※> title_denzi_kitchen_duty does remind me we have a passable cauldron tile in rltiles/unused, though I worried before deploying such that I have no idea how to stop the endless complaints of people not being able to partake from it unless it's a cache-of-food-equivalent instead of a statue equivalent. 13:14:17 <06r​egret-⸸nde※> https://cdn.discordapp.com/attachments/747522859361894521/1440797348753309898/tavern.png?ex=691f7698&is=691e2518&hm=d68ec0a4ce83ca3e460290546a5be2e1b89973e340e77d8baaf00db47c7ccc1a& 13:31:20 03DracoOmega02 07* 0.34-a0-1293-gc5498a261d: Remove some broken logic regarding monsters picking up launchers (Acrobat) 10(7 minutes ago, 3 files, 2+ 46-) 13https://github.com/crawl/crawl/commit/c5498a261d27 13:54:31 <09h​ellmonk> cauldron wizlab where you have to drink all the soups (for random potion effects) in order to enable ctele 14:07:51 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4917 * 0.34-a0-1266-ga9ca1739e1: Make spell name lookup prefer spells that exist 10(23 hours ago, 3 files, 26+ 3-) 13https://github.com/crawl/crawl/commit/a9ca1739e149 14:11:06 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4914 * 0.34-a0-1266-ga4e675f501: Change solo tentacles only to attack targets in our LoS 10(3 days ago, 1 file, 9+ 0-) 13https://github.com/crawl/crawl/commit/a4e675f50170 16:40:27 Unstable branch on underhound.eu updated to: 0.34-a0-1293-gc5498a261d (34) 19:32:54 03CrawlOdds02 {WizardIke} 07* 0.34-a0-1294-ge5381b52e2: Make spell name lookup prefer spells that exist 10(28 hours ago, 3 files, 26+ 3-) 13https://github.com/crawl/crawl/commit/e5381b52e207 22:12:38 <06p​leasingfungus> eye of draining and it sends you down the sewer 22:19:49 <09h​ellmonk> Need nose and ear monsters for my upcoming mr potato head vault 23:35:34 Unstable branch on crawl.develz.org updated to: 0.34-a0-1294-ge5381b52e2 (34) 23:58:50 Windows builds of master branch on crawl.develz.org updated to: 0.34-a0-1294-ge5381b52e2