00:00:42 enne: "The player likes to be on top." 00:08:11 -!- upsy has joined ##crawl-dev 01:00:09 -!- Kurper has quit [Remote host closed the connection] 01:32:48 -!- Moriasc has joined ##crawl-dev 01:45:30 -!- syllogism has joined ##crawl-dev 02:41:39 -!- elliptic has quit [Ping timeout: 265 seconds] 02:50:24 !seen ixtli 02:50:25 I last saw ixtli at Mon Nov 1 20:55:02 2010 UTC (6w 4d 11h 55m 23s ago) quitting with message Quit: ixtli. 03:31:07 ugh 03:31:15 wtf. 03:35:36 ok, that was bad 03:35:44 but I can't reproduce it 03:37:32 crazy yiuf's cottage was generated disconnected 03:37:41 :| 03:37:50 that shouldn't ever happen. 03:37:52 it wasn't my build_rooms change, though, as that is not involved 03:39:13 hrm 03:46:44 -!- paxed has quit [Read error: Operation timed out] 03:47:15 it could be that minivault connectivity strategy relies too heavily on luck 03:47:50 it works by picking a random floor spot that's not in any vault and connecting it with an exit point of the vault, not overwriting vaults in progress 03:48:04 granted, it works in a lot of cases... 03:51:07 -!- paxed has joined ##crawl-dev 03:52:52 hrm 03:52:56 that might actually be the explanation 03:53:19 even though I don't want to sound like I'm just trying to say "it wasn't me who's borked it, honest!" 04:03:18 hm 04:05:26 oh bloody hell 04:05:30 well 04:06:13 the more random things it relies on being in place, the more likely that something will be generated that doesn't fit the schema, and as there is no error checking, you'll get a buggy placement? 04:07:05 honestly, though, what I'm sure in now is that the code that connects primary vaults and the code that connects minivaults needs to be unified 04:07:20 but I'm not yet sure how that should be achieved 04:07:37 because no present mechanism seems to be really good 04:08:24 dig_vault_loose makes no guarantees, either, but it happens to work in most normal (i.e., not caves) levels 04:08:48 _connect_vault has a slim chance of just not working 04:08:50 -!- valrus has joined ##crawl-dev 04:09:44 what is the 'mini_float' tag supposed to do? 04:11:28 or well, when has anybody actually confirmed that it does anything? 04:12:08 in all cases I can observe right now, it simply doesn't matter 04:13:06 Zaba: The dungeon builder is one of those files that drives some people to madness. 04:13:12 LOOK AWAY AND LIVE, ZABA! 04:13:51 -!- valrus has quit [Ping timeout: 272 seconds] 04:20:18 -!- Pialein has joined ##crawl-dev 04:32:42 -!- Pialein has quit [Ping timeout: 250 seconds] 04:33:56 -!- Mu_ has joined ##crawl-dev 04:41:48 -!- Textmode has quit [Ping timeout: 250 seconds] 04:53:24 or well. I lied about mini_float. 04:53:41 but.. 04:53:54 -!- eith has joined ##crawl-dev 04:54:09 it makes no sense for vaults that have at least one '@' 04:54:35 _pick_float_exits is only used when the vault has no predefined exits 04:54:54 Zaba: remember that hte unique vaults are placed directly by dgn_place_map 04:55:37 dgn_place_map is a wrapper for _build_seconday_vault 04:55:45 which is a wrapper for _build_vault_impl 04:55:58 secondary, even 04:56:20 connecting minivaults is done in _build_seconday_vault 04:57:16 -!- Moriasc has quit [Ping timeout: 260 seconds] 04:57:53 secondary. Damn. 04:58:18 my mini_float ramblings are not related to unique map placement 04:59:21 ok 04:59:49 I disown as much of dungeon.cc that I didn't write, and any of it that I did, as I can. 05:00:14 anyway... _pick_float_exits is not used for vaults that specify at least one exit manually. So, a minivault that has no predefined exits and no 'mini_float'.. is not connected at all, or what? 05:00:59 I guess placement code tries to ensure that it's connected somehow 05:01:03 or something 05:08:23 argh wtf. 05:12:48 -!- MarvinPA has quit [] 05:12:52 :) 05:12:55 this is amusing. 05:13:02 in a ... well, being nice to zaba kind of way, I guess? 05:19:06 due, hm? 05:19:17 Zaba: I was expressing amusement at your pain. 05:19:29 ah 05:23:03 Zaba: Minivaults are always placed overlapping at least one floor square 05:23:27 And they're supposed to be surrounded by a layer of floor or use mini_float or explicit exits 05:26:48 greensnark: do you have a clue how to stop luaL_loadbuffer() from decorating the input's name? 05:26:49 Lua error: [string "/home/kilobyte/crawl/crawl-ref/source/meow"]:2: moo 05:26:51 rather than 05:26:54 Lua error: /home/kilobyte/crawl-0.7/crawl-ref/source/meow:2: moo 05:27:25 meow! 05:28:06 (I had to replace luaL_loadfile() with luaL_loadbuffer() to handle non-ASCII on Windows) 05:28:35 or due, for that matter... or anyone with a clue about lua (ie, not me) 05:29:13 just practicing some voodoo coding which fortunately works so far save for this minor issue 05:29:27 http://pgl.yoyo.org/luai/i/luaL_loadbuffer 05:29:36 greensnark, ok.. 05:29:47 according to that, it gets the chunk name from parameter 4. 05:29:48 what exactly is flood_find? 05:31:36 I don't think you can tell loadbuffer not to decorate the name 05:31:38 kilobyte: what are you passing for name? 05:31:39 "/home/kilobyte/crawl/crawl-ref/source/meow" 05:31:39 huh. 05:32:15 greensnark, I wonder how should vault connection be done properly 05:32:15 Zaba: in general, or in Crawl? I know only the former. 05:32:48 well, nothing of what is currently present in crawl seems proper to me 05:33:54 but I wonder how to make crawl do it properly 05:34:12 kilobyte: Try prefixing the filename with "@" ? 05:34:59 I can't even find where luaL_loadbuffer is defined :| 05:36:32 Or try prefixing it with "=" 05:36:34 "=filename" seems to do the right thing 05:36:34 at the moment.. normal vaults get dig_vault_loose treatment, which basically digs away from all the exit points in hope to reach something. Minivaults get _connect_vaults, which either does join_the_dots of exit points with random floor squares, or _connect_spotty 05:36:34 _connect_vault, even 05:36:37 I thought the room builder explicitly created a corridor to the connection points? 05:36:37 greensnark, yes, but given the nature of the rooms builder, it doesn't really make a large difference 05:36:37 Normal vaults used to use the room builder to place explicit connections as well 05:36:47 Just picking a random floor square and using join_the_dots sounds fine 05:37:48 greensnark, picking random square and doing join_the_dots can fail to connect the vault 05:37:49 You mean if it has to cross a vault to get there? 05:37:49 yes 05:37:49 greensnark: yay! Thanks! 05:38:00 that can leave the vault disconnected and nothing will care 05:38:22 likewise for _dig_vault_loose---but the way it works almost always ensures connectivity due to the nature of levels that _build_rooms generates 05:38:22 gave it an @ and now it's: 05:38:22 Lua error: /home/kilobyte/crawl/crawl-ref/source/meow:2: '=' expected near '' 05:38:44 _connect_spotty is only used in Orc 05:38:54 Zaba: Ok, use a floodfind out from the vault exit, respecting map masks until you find a floor square and connect to that 05:39:00 Okay, so name is just passed straight ot lua_load. However... lua_load is defined nowhere. :/ 05:39:11 greensnark: both @ and = work, is there a difference? 05:39:23 greensnark, and that should be used instead of both _dig_vault_loose and _connect_vault, right? 05:39:38 kilobyte: http://sprunge.us/HVRK 05:39:47 Zaba: Yep! 05:39:52 greensnark: I was getting there! 05:40:07 greensnark, sounds nice.. but what about _connect_spotty? Or can we live without a special way of connecting vaults in Orc. 05:41:45 _connect_spotty sounds pretty nice in theory 05:42:24 I'd say keep connect_spotty 05:42:50 If it fails it falls through to the standard connection code anyway 05:42:58 Maybe connect_spotty could also be used for other spotty branches 05:43:05 okay. 05:43:06 * due spots greensnark. 05:43:11 greensnark, other spotty branches can get minivaults? :P 05:43:12 !coffee due 05:43:13 * Henzell hands due a barrel of soy latte, brewed by Gastronok. 05:43:28 Zaba: I suppose Slime and such could, in theory :P 05:43:49 Oh you cheated, you just repped for \[string. :( 05:44:01 due: Hehe, of course :P 05:44:14 greensnark: I have an excuse: my pills have worn off :0 05:45:01 03kilobyte 07unicode * r1e52b0fc5290 10/crawl-ref/source/initfile.cc: Tell DGL the encoding is always UTF-8. 05:45:34 -!- monky has quit [Quit: hello] 05:46:37 03kilobyte 07unicode * rbc9e9f380e90 10/crawl-ref/ (246 files in 30 dirs): Merge branch 'master' into unicode 05:46:49 03kilobyte 07unicode * r7642fc171f80 10/crawl-ref/source/ (view.cc view.h): Fix the "screenshot" in the char dump getting mangled to 7 bits. 05:46:49 03kilobyte 07unicode * r203b11eb2599 10/crawl-ref/source/macro.cc: Convert ofstream to fopen() for macro files. 05:46:49 03kilobyte 07unicode * r322329c0e7c0 10/crawl-ref/source/chardump.cc: Fix glyphs in the map dump getting mangled to 7 bits. 05:47:01 03kilobyte 07unicode * r28b815a46941 10/crawl-ref/source/ (stash.cc stash.h unicode.h): Convert ofstream to fprintf() for stash dumps. Translate charsets. 05:47:04 03kilobyte 07unicode * r851cdeb3d204 10/crawl-ref/source/chardump.cc: Translate charsets for everything written to the char dump file. 05:47:16 03kilobyte 07unicode * r64ca3ade4932 10/crawl-ref/source/ (unicode.cc unicode.h): A reader for Windows-style BOMmed files. 05:47:27 03kilobyte 07unicode * rc182b846613a 10/crawl-ref/source/macro.cc: Fix an use of fstream, allow macro files to be in dirs with non-ASCII paths. 05:47:35 kilobyte: you doing dgl patching too? 05:47:39 03kilobyte 07unicode * rad3bbac308f3 10/crawl-ref/source/ (6 files): Rename TextFileReader to FileLineInput I didn't notice, use it for config files. 05:47:51 03kilobyte 07unicode * rcb38198de7c8 10/crawl-ref/source/clua.cc: Allow lua files with a BOM -- including Windowsish 16 bit encodings. 05:48:02 03kilobyte 07unicode * r8d74ffd849f0 10/crawl-ref/licence.txt: Set the license info to what we'll probably end up with. 05:48:04 03kilobyte 07unicode * r4732151a9c0f 10/crawl-ref/source/ (makefile.obj unicode.h wcwidth.cc): Add Markus Kuhn's wcwidth(). All libcs but Windows ship it already... 05:48:05 03kilobyte 07unicode * r5600077b0b13 10/crawl-ref/source/wcwidth.cc: Update the table of combining characters for Unicode 6.0 05:48:05 03kilobyte 07unicode * re5e08161e7e4 10/crawl-ref/source/fontwrapper-ft.cc: Handle fixed-width characters of width 2 in tiles. 05:48:06 03kilobyte 07unicode * rb6d532090f71 10/crawl-ref/source/ (libutil.cc unicode.cc): Make strwidth() actually do what it's supposed to. 05:48:07 03kilobyte 07unicode * rbbc1968de451 10/crawl-ref/source/ (ng-input.cc startup.cc): Let Chapayev spell his character's name "????????????". 05:48:08 03kilobyte 07unicode * rd562ab002f5a 10/crawl-ref/source/ (database.cc unicode.cc unicode.h): Get rid of one last use of fstream. 05:48:09 03kilobyte 07unicode * re5001fba7786 10/crawl-ref/source/clua.cc: Stop lua from adding [string "%s"] to error messages. (greensnark) 05:48:50 greensnark, just find_first_from(some_point, env.level_map_mask) will not return a point from within a vault, right? 05:48:56 paxed: yeah, but I only analyzed the code for now... To actually do and test it, I'd have to get a DGL rig running. 05:49:42 kilobyte: http://nethackwiki.com/wiki/User:Paxed/HowTo_setup_dgamelaunch ;) 05:50:04 paxed: yeah, it might be not that hard, I just didn't got there yet. 05:50:04 Step 1: Find a virgin, perform a sacrifice. 05:50:13 Step 2: Tears. Lots of them. 05:50:20 some stuff is pretty broken yet, including all input 05:50:45 Moin guys! 05:51:01 impressive - 67 people logged into the forum already 05:51:06 da King of Naps! 05:51:12 hi napking 05:51:19 hey :)) 05:51:25 What's up? :) 05:53:18 unicode is coming closer \o/ 05:53:49 Napkin: that light is not the end of the tunnel, it is the lights of the unicode train, come to run you down :D 05:54:01 bring it on! ;D 05:55:12 yeah... it will be a strict improvement for local games, both on Unix and Windows, with no regressions or troubles planned, but there will be some pain for DGL 05:55:17 ie, for you :p 05:55:33 hehe 05:55:48 i believe it's worth it :) 05:57:55 It means we can have hiragana pan lords! 06:02:26 hey greensnark? do you have some time to add the css sheets back to /scoring-alpha on CAO? 06:02:54 Napkin: That page is defunct, it's only scoring and scoring-stable now 06:03:07 ah 06:03:49 roger 06:04:11 greensnark, I don't understand how one would use flood_find for this. 06:04:30 you may want to mention the difference at the top? I'll change the links in the wordpress 06:04:45 Zaba: You probably can't use it unmodified 06:05:49 Napkin: At the top? 06:06:13 yeah, because both sites say "CAO/CDO Scoring Overview" at the top 06:06:41 maybe.. "(including games from unreleased versions)" in small behind that? 06:06:52 greensnark, I'd have to modify it so path_flood returns false for squares that are inside a vault, right? 06:07:05 if some flag is set somewhere 06:07:14 Napkin: Ah 06:07:19 or rather, doesn't call good_square 06:07:24 Zaba: Right 06:09:20 greensnark, what does path_flood's return value signify? 06:12:12 Zaba: I think true to signal that it's found the target 06:12:19 ah, ok. 06:12:24 alright... 06:12:50 Oh, and good_square just means the square will be part of the floodfind, not that it is the target 06:12:56 yes 06:13:04 I want to avoid looking into squares that belong to a vault 06:13:19 Aye 06:13:33 hm, it doesn't work... 06:13:45 The starting point of target_connections may be part of the vault 06:13:55 yes, it should work with such cases 06:14:03 as long as it's on the edge of a vault 06:14:16 Use _external_connection_points to get the nearest point that's not in the vault 06:14:34 greensnark, if I only check 'dc' in path_flood for being in a vault, it should work when 'c' is on the edge of the vault, too... 06:15:10 I don't really remember that code, you should trace through it and see why it's failing :P 06:15:47 yeah, I'm looking into it 06:16:17 ah, and I even know 06:18:00 hmm, now it just gives up and returns coord_def(0,0) 06:18:11 well, at least it avoids squares that are inside vaults now :P 06:18:29 ok, didn't know how where to put the links to the scores... created a new category "Scoring Overview" 06:18:59 greensnark, ah hell... flood_find checks whether features are traversable and such 06:19:16 that's certainly not what I need in this case :P 06:19:37 Zaba: Roll your own, flood find is trivial 06:19:54 I don't want to duplicate it, though 06:20:04 that'll make things no better than the current vault connection code, with regard to duplication :P 06:20:12 Maybe create a flood_find_iterator that allows you to specify which squares to continue the floodfind from 06:20:38 perhaps I could have a function that determines whether a feat is passable for flood_find purposes passed to it in a manner similar to in_bounds? 06:20:50 I think an iterator would be cleaner 06:21:03 Right now flood_find is pretty difficult to adapt to new tasks 06:21:11 yeah 06:21:19 An iterator would allow you to handle traversability etc. outside the core floodfind 06:21:31 I don't understand what exactly you mean by such an iterator, though 06:21:40 not enough C++ programming experience, I daresay :P 06:22:06 It'd maintain a list of points on the perimeter of the flooded area and ++ would pick the next point each time 06:22:42 Zaba.next()! 06:22:48 If the point is traversable, you'd call iterator.point_traversable() to add its neighbours to the next iteration 06:23:04 If the point is the point you want, you'd break out of the loop 06:23:13 Easy! 06:23:19 hm. 06:23:40 greensnark, and that would replace the current flood_find? 06:23:57 Yes, it can 06:24:09 it seems pretty ugly.. Duplicates a bunch of travel.cc code, in a way 06:24:26 Some amount of duplication is unavoidable given we're dealing with C++ 06:24:52 well.. at the moment, flood_find is a subclass of travel_pathfind, which seems to implement a lot of smart logic 06:25:09 wouldn't all that smart logic be gone if it were reimplemented in the manner described above? 06:28:02 The "smart" logic there is travel specific :P 06:28:34 A generic iterator class could also be used for travel itself 06:29:37 Mmmmduplication. 06:44:16 hrm. 06:59:25 greensnark, somehow, dirtily hacking this into flood_find seems easier for nwo :P 06:59:26 now* 07:05:17 filthy :( 07:05:34 I know 07:10:05 greensnark, how would other uses of flood_find be migrated to use an iterator? 07:10:25 I mean, points_connected_from/any_point_connected_from/has_exit_from 07:10:43 whatever they are 07:19:17 greensnark, it works 07:19:51 Zaba: An iterator would be flexible enough to be easily used for those other things too 07:20:18 greensnark, indeed.. but somehow I don't feel like figuring out how to transfer other flood_find users to a flood_iterator 07:20:41 Heh 07:20:56 it seems to be mostly used for various connectivity checks. Maybe travel.cc's smartness is suitable for those? 07:24:54 greensnark, moreover, I don't really know how to implement it properly :P 07:31:53 Zaba: No sweat, we can get to it later :P 07:32:02 due: Someone suggested a rat king vault!1!! 07:32:32 I can't believe we have a thief unique called Maurice when Maurice should plainly be a cat 07:32:49 greensnark, I guess that I could let the hackery in flood_find persist for now, then? It's this bad: http://sprunge.us/HhcL 07:33:04 greensnark: ... felid thief. <3 07:33:12 greensnark, and I daresay it works 07:33:27 Naga/Centaur ghosts use humanoid base tiles (https://crawl.develz.org/mantis/view.php?id=3002) by jpeg 07:33:28 Zaba: Well, ok for now, but we should definitely iteratorize it at some point :P 07:33:36 greensnark, yeah, I know. 07:43:25 03zaba * r8b7ed2da511f 10/crawl-ref/source/mapdef.cc: Check bounds in map_def::in_map. 07:43:26 03zaba * raceb0287119e 10/crawl-ref/source/dungeon.cc: Simplify finding of exits for float vaults. 07:43:28 03zaba * r6db22d69a422 10/crawl-ref/source/flood_find.h: flood_find: Add flags that tell it to avoid searching in vaults, and to check traversable features. 07:43:31 03zaba * reab1f320bc04 10/crawl-ref/source/dungeon.cc: Use flood_find to connect primary and mini vaults. 07:43:33 03zaba * rece20205d211 10/crawl-ref/source/dungeon.cc: Remove some no longer used functions. 07:43:45 greensnark, now fix vault connecting for me, kthx :P 07:43:48 brb 07:52:38 03j-p-e-g * rd2ab2ef026eb 10/crawl-ref/source/ (tiledgnbuf.cc tilepick.cc): Fix #3001: Player drawn on top of large monster tiles. 07:55:22 greensnark: I just read Maurice & His Educated Rodents (again) 07:56:11 Zaba: Wait, I thought you just fixed it? :P 07:56:22 Rat King monster could be hydra-like, and a severing attack would remove a rat from the rat king.. producing a rat monster. 07:56:26 Keskitalo: Awesome, isn't it? It's totally his best YA book 07:57:57 I really like Tiffany books too, but yeah, when I first read it it became my favourite Discworld book in general. Not sure what would be my favourite right now though. 08:00:04 The Tiffany Aching books are good, but not in the same class as the Amazing Maurice! 08:02:51 The Finnish translation is great, and the translator thanked her own rats in the end of the book. It was like "they always help me in my work, but I think this time they were extra helpful" :) 08:05:51 * greensnark is deeply suspicious of translations :P 08:06:03 Although I have to admit the Asterix english translations are <3 08:09:46 NetHack is much better in the original German! 08:12:09 I found Soul Music more enjoyable in Finnish, even though the "Elvish" joke was untranslatable enough for the translator to simply explain it in a footnote :) 08:13:08 -!- Kurper has joined ##crawl-dev 08:19:38 -!- Fangorn_ has quit [Read error: Connection reset by peer] 08:40:04 If I wanted to find the current release version and release date of DCSS programmatically, where could I find that on the web? 08:41:50 git maybe? 08:44:49 thanks, I didn't want to download the whole git repo for that info :) but that hint lead me to docs/changelog.txt. That might do 08:45:19 Maybe this rat king thing could be expanded a little. Rat packs are boring enemies - maybe they should act like slime creatures and combine into rat kings. (Altough, just copying slime creatures' trick would be boring) 08:46:43 anyway, there could be "rat king", "grey rat king", "green rat king", and best of all "orange rat king" 09:08:32 -!- eith has quit [Read error: Connection reset by peer] 09:08:42 greensnark, well... :P 09:09:28 greensnark, I surely hope I didn't break it, at least :P 09:12:50 -!- Pialein has joined ##crawl-dev 09:14:40 -!- Unrodney has joined ##crawl-dev 09:14:48 Unrodney: last_release dcss 09:14:49 bhaak: Dungeon Crawl Stone Soup 0.7.1 has been released 4 months and 24 days ago! 09:15:08 Keskitalo: thanks, works now 09:15:23 -!- Unrodney has left ##crawl-dev 09:18:56 another option, instead of using changelog.txt, would have been to use dcss wikipeda article. I'm sure it's almost as current as the git repository :) 09:19:00 bhaak: "was released" works better than "has been released" 09:21:01 greensnark: thanks, I changed it. I almost used 'was' but it was late when I made the first version last night :) 09:26:27 now I wonder whether using _external_connection_points is really necessary 09:26:48 it works without it for primary vaults.. but I'm not too sure about _connect_spotty 09:29:14 it works due to the nature of flood_find 09:29:23 I'm not so confident in connect_spotty. I guess I should do some testing. 09:32:15 huh.. 09:34:06 greensnark, it seems like _connect_spotty likes to ruin vaults 09:34:26 greensnark, I have no idea how, but placing, say, enter_bailey_7 in the mines seems to consistently show it to me 09:35:31 even though it seems to do a fair amount of MMT_VAULT checks.. making the code not use it makes the vault appear normally 09:42:37 -!- Twilight-1 has joined ##crawl-dev 09:46:59 -!- Twilight13 has quit [Ping timeout: 272 seconds] 09:48:56 -!- valrus has joined ##crawl-dev 10:05:26 Zaba: One mask check seems to be inverted in connect_spotty 10:05:41 Zaba: While you're there, it's probably time to rename unforbidden :P 10:06:52 to 'forbidden'? I agree 10:07:15 Maybe map_masked or something like that 10:07:22 "forbidden" doesn't really say much 10:07:24 well, something that doesn't have inverted logic 10:07:54 Just do that in a separate commit though 10:07:57 It probably touches a lot of code 10:08:06 yeah 10:29:37 greensnark, am I understanding it correctly that the code in _build_vault_impl modifies place.exits by modifying target_connections? 10:33:57 I guess it does, otherwise _build_secondary_vault wouldn't be able to connect the vault. 10:35:16 greensnark, is there any reason why can't all that code just use place.exits? 10:39:16 03zaba * refd600849837 10/crawl-ref/source/dungeon.cc: Make _connect_vault_exit check for whether the target point is in bounds. 10:39:18 03zaba * r62cfaebcd789 10/crawl-ref/source/ (dgn-shoals.cc dungeon.cc dungeon.h): Remove dgn_Vault_Excavatable_Feats, as it is no longer used. 10:39:22 03zaba * r761170151261 10/crawl-ref/source/ (7 files): Rename 'unforbidden' to 'map_masked' and invert logic accordingly. 10:39:34 03zaba * r0aa92142c393 10/crawl-ref/source/dungeon.cc: Fix a case of inverted map mask check in _connect_spotty. 10:47:05 -!- Kurper has quit [Remote host closed the connection] 11:03:23 -!- casmith789 has quit [Ping timeout: 240 seconds] 11:06:16 -!- wafflepants has joined ##crawl-dev 11:07:42 -!- casmith789 has joined ##crawl-dev 11:29:15 phew.. this phpbb3 thingy is a bit buggy sometimes 11:29:31 and installing mods is a real pain if you have local changes 11:29:50 and.. a different theme is already considered a local change *sigh* 12:17:28 what's the current save-game extention again? 12:17:31 .sc? 12:18:31 got it, .cs 12:46:53 bah 12:47:10 _connect_spotty does not ensure connectivity. 12:47:50 or well. 12:48:05 -!- elliptic has joined ##crawl-dev 12:50:04 I don't think it does. 12:50:07 meh. 13:22:45 -!- Siber has quit [Quit: Whoa! Partial hit!] 14:07:08 -!- monky has joined ##crawl-dev 14:08:25 "You hear the urgent ticking of a clock, a long way away." -- 2 squares from the stairs. 14:21:57 kilobyte: the 'long' distances are really weird anyways 14:35:01 kilobyte: was that a message upon entering the level? 14:41:00 elliptic: yeah (not for the first time) 14:41:43 the "distant snort" is correct since there's the whole lab between you and the minotaur, but for the clock, most of the messages use the portal's position 14:42:11 yeah, the distance description is broken when you enter the level I think 14:42:24 seems fine for messages you receive while traveling around the level, though 14:42:35 i kind of think it would be nice if sounds in crawl were somewhat directional 14:43:40 in zotdef alarm traps tell you what direction they are in when they go off 14:43:53 i think it would be quite feasible to display that by using tiles on the edge of LOS to cut down on text spam 14:44:21 tiles on the edge of LOS? 14:44:37 elliptic: the out-of-LOS tiles adjacent to the in-LOS tiles 14:44:52 eronarn: there aren't any vertically 14:45:13 (at least if you use 80x24) 14:45:16 not due north/south, no 14:45:50 but even slightly north|south east|west works 14:46:02 i am not sure if in practice it would look nice, but i could definitely imagine it working 14:47:30 in RL, human directional hearing is pretty weak 14:53:20 yeah, i wouldn't want it to be guaranteed or precise; but it is pretty weird when, say, a monster around a corner shouts 14:54:29 being able to know roughly how loud it is, and roughly where it's coming from (say, compass direction) would be an improvement 14:56:16 oh, did the earth spell changes make it into trunk yet? 14:57:53 ah, looks not, good; i want to patch marvinPA's patch before his patch is used to patch trunk with patched earth spells 14:59:10 remove one of the ac checks for BEAM_FRAG? 15:00:41 -!- Elvashi has joined ##crawl-dev 15:09:53 monky: no, i want to make it use max rather than summed 15:09:58 summed makes no sense 15:10:22 oh right 15:13:40 three checks would be fine if it just used the max, but three rolled checks is just goofy because cockroaches and stuff 15:13:46 @??giant cockroach 15:13:46 giant cockroach (07s) | Speed: 12 | HD: 1 | Health: 3-7 | AC/EV: 3/10 | Damage: 2 | Res: 06magic(1) | Vul: 09poison | Chunks: 07contaminated | XP: 1. 15:16:17 hmm guess i'll go shower then write that patch when i get back 15:24:32 -!- OG17 has quit [Read error: Connection reset by peer] 15:30:52 -!- OG17 has joined ##crawl-dev 15:35:18 -!- Pialein has quit [Read error: Connection reset by peer] 15:59:41 -!- Pialein has joined ##crawl-dev 16:16:41 03j-p-e-g * r613a21cb2a86 10/crawl-ref/source/ (7 files in 3 dirs): Add Centaur/Naga variant lich form tiles, as suggested by psyshvl. 16:53:09 -!- galehar has joined ##crawl-dev 18:11:44 Jesus Christ... phpBB3 needs a coding convention BADLY. 18:12:04 And installing a simple addon like "Thanks for Post" is hell^2! 18:12:17 now I definitely need a coffee... 18:12:21 :)) 18:12:43 https://crawl.develz.org/tavern/viewtopic.php?p=159#p159 18:12:46 hey due :) 18:15:01 thank god I have that forum code in git.. otherwise i'd go crazy on update 18:15:46 and if I hear any complaint about my beautiful "Thank" button.... grrrr.. wuff! :> 18:16:49 well... 18:16:53 where is it? 18:22:40 either I can't see it because I'm too tired 18:22:48 or it is because I'm a moderator 18:23:06 I can't thanks people, I can only punish them 18:23:26 I didn't deleted their posts, should be good enough... suckers... 18:23:26 :) 18:23:31 -!- galehar has quit [Quit: bedtime] 18:31:11 unbelievable... 18:32:03 due? could you check please? :) 18:32:57 hm? 18:33:51 Napkin: What do you want me to look at? 18:34:28 if you can see the "Thank" button in the forum at bottom right in the post body, beside the "Quote" button 18:35:01 after logging in, if you haven't done so yet ;) 18:35:08 I see a thank button 18:35:13 I pressed it 18:35:13 *phew* 18:35:14 It works. 18:35:17 thanks :) 18:35:20 Perhaps galehar needed to log out and then back in again? 18:35:48 nah, either he wasn't logged in at all or just didn't see it 18:36:14 i never needed to re-log when i was testing 18:56:12 -!- casmith_789 has joined ##crawl-dev 18:57:21 -!- Lollipop has joined ##crawl-dev 18:57:23 -!- casmith789 has quit [Ping timeout: 240 seconds] 18:57:34 -!- Pialein has quit [Read error: Operation timed out] 19:02:08 -!- casmith_789 is now known as casmith789 19:09:51 -!- Lollipop has quit [Remote host closed the connection] 19:29:48 -!- syllogism has quit [] 19:32:37 Butchering still isn't fixed (https://crawl.develz.org/mantis/view.php?id=3004) by casmith789 19:32:37 Deep Dwarf Death Knights exist (https://crawl.develz.org/mantis/view.php?id=3003) by casmith789 19:37:39 TSO should warn when readIDing scrolls of summoning (https://crawl.develz.org/mantis/view.php?id=3005) by Nexos 20:07:18 yay a glowing drain on D:27... 20:07:30 ??sewer 20:07:30 sewer[1/5]: The Sewers, contributed by Eino Keskitalo, who apparently has some to spare. They are timed and will disappear eventually if you do not enter them. 20:07:33 ??sewer[2] 20:07:33 sewer[2/5]: < LordSlothN> Why ist here an altar to zin in the sewers? < pointless> Zin purifies the waste that flows through the sewers 20:07:35 ??sewer[3] 20:07:35 sewer[3/5]: is an altar to a good god guaranteed in sewers? ogaz: no, sometimes you get zin (zing) 20:07:37 ??sewer[4] 20:07:37 sewer[4/5]: contains fish and rats, at the least. "rats" includes up to green, and some vaults have giant frogs. 20:07:38 ??sewer[5] 20:07:38 sewer[5/5]: beware the sickly merfolk 20:08:13 D:3-6 20:15:04 -!- Elvashi is now known as Textmode 20:23:25 -!- Fangorn_ has joined ##crawl-dev 20:24:46 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 20:31:27 -!- Pialein has joined ##crawl-dev 20:37:50 -!- Pialein has quit [Ping timeout: 276 seconds] 21:04:53 -!- Twilight-1 has quit [Read error: Connection reset by peer] 21:06:04 -!- Twilight-1 has joined ##crawl-dev 22:19:51 -!- stabwound has quit [Quit: ZNC - http://znc.sourceforge.net] 22:20:56 -!- stabwound has joined ##crawl-dev 22:32:39 -!- stabwound has quit [Ping timeout: 240 seconds] 22:44:18 -!- stabwound has joined ##crawl-dev 22:54:41 -!- Zaba has quit [Ping timeout: 264 seconds] 22:58:16 -!- Zaba has joined ##crawl-dev