00:31:21 -!- greatzebu has quit [Quit: greatzebu] 00:37:12 -!- greatzebu has joined ##crawl-dev 00:40:13 Flying monster will not fly over deep water to pursue player. (https://crawl.develz.org/mantis/view.php?id=4548) by mong 00:49:47 -!- greatzebu has quit [Quit: greatzebu] 00:56:49 -!- valrus has quit [Remote host closed the connection] 01:02:58 -!- Fail_Bot has quit [Ping timeout: 276 seconds] 01:03:57 -!- ixtli has quit [Quit: Computer has gone to sleep.] 01:08:07 -!- elliptic has quit [Read error: Connection reset by peer] 01:08:30 -!- Zaba has quit [Ping timeout: 252 seconds] 01:09:49 -!- Zaba has joined ##crawl-dev 01:09:52 -!- Zaba has quit [Changing host] 01:09:52 -!- Zaba has joined ##crawl-dev 01:10:32 -!- dtsund has quit [Quit: dtsund] 01:23:38 -!- elliptic has joined ##crawl-dev 01:24:04 -!- Zaba has quit [Ping timeout: 252 seconds] 01:26:29 -!- evilmike has quit [Quit: something happened] 01:34:54 -!- Keskitalo has joined ##crawl-dev 01:35:05 -!- Zaba has joined ##crawl-dev 02:14:36 -!- galehar has joined ##crawl-dev 02:21:56 With the recent changes, the Felid physical class should be Monk, not Fighter. 02:40:51 Unstable branch on crawl.akrasiac.org updated to: 0.10-a0-709-g32602a7 (32) 02:55:57 -!- MarvinPA has quit [Read error: Connection reset by peer] 02:56:26 -!- MarvinPA has joined ##crawl-dev 03:12:32 -!- ainsophyao has joined ##crawl-dev 03:34:58 -!- purge has quit [Quit: .] 03:45:04 -!- Keskitalo has quit [Ping timeout: 244 seconds] 03:55:18 -!- mumra has joined ##crawl-dev 04:03:08 -!- Keskitalo has joined ##crawl-dev 04:31:59 -!- Textmode has quit [Quit: Ex-Chat] 04:35:49 so apparently xcode is no longer free? 04:36:42 -!- SamB_ has joined ##crawl-dev 04:39:21 -!- SamB has quit [Ping timeout: 252 seconds] 04:46:42 yay... 04:47:43 do we still support it? it seems like such an annoying chore 04:49:08 hey kilobyte? 04:49:11 moin guys 04:50:56 hiii napking! 04:51:24 kilobyte, due: how about replacing the documents in docs, which have been superseeded by the version in wiki, with a text file containing only the link to the wiki page? Additionally, a new target in the makefile "docs" to have them downloaded for a release? or maybe a CONFIG=release parameter? 04:53:09 (of course, that would need the text-output plugin for dokuwiki) 04:56:06 I do read these files often... 04:56:38 -!- Pingas has joined ##crawl-dev 04:58:44 03naesten * r7d3f6712e87d 10/crawl-ref/source/describe.cc: Mention "Make Ponderous" on religion (^) screen. 04:58:46 03naesten * re4caa5c1f068 10/crawl-ref/source/makefile: Add "docs" target to makefile. 04:58:46 03kilobyte * rb60b0fac8eed 10/crawl-ref/source/mon-stuff.cc: JoachimSchipper's semicolonicide. 04:58:46 03kilobyte * r34dec5fd0281 10/crawl-ref/source/dat/descript/monsters.txt: Change deep elf knights' description (JoachimShipper). 04:58:46 03naesten * r824651224f99 10/crawl-ref/source/stuff.h: Support GCC's printf warnings for functions declared in "stuff.h". 04:58:47 03kilobyte * ra698d0c01d6a 10/crawl-ref/docs/develop/monster_speech.txt: Fix a comment (JoachimShipper) 04:58:49 03kilobyte * rca3ed9893f44 10/crawl-ref/source/invent.cc: Make the menu header for wands "Wands" rather than "Magical Devices". 04:58:52 03naesten * r2dc5431ae592 10/crawl-ref/.gitignore: Add Doxygen output directory to .gitignore. 04:58:53 03kilobyte * rf71940ac76b6 10/crawl-ref/source/ctest.cc: Fix a mangled displayed int on big-endian 64 bit architectures. 05:00:19 Hmm, I wonder if it'd make sense to make gods (Ely for example) only accept sacrifices of things found *after* converting to the god. 05:00:38 so, you should not have any problem with running make docs or even read them online, which additionally gives you a chapter list, etc 05:01:57 having to issue a command is fine, reading them online in a clunky browser is not 05:16:02 Since there are now devs around: How about re-enabling FeMo now that the combo serves an actual purpose? 05:16:06 as in, the free piety? 05:16:06 Yeah. In fact FeFi has no advantages over it 05:16:06 -!- Ashenzari has quit [Ping timeout: 260 seconds] 05:16:51 -!- Ashenzari has joined ##crawl-dev 05:16:51 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 05:19:47 -!- MarvinPA_ has joined ##crawl-dev 05:22:17 -!- galehar_ has joined ##crawl-dev 05:22:42 -!- SamB__ has joined ##crawl-dev 05:23:20 -!- MarvinPA has quit [Ping timeout: 260 seconds] 05:25:33 -!- SamB_ has quit [Ping timeout: 252 seconds] 05:26:08 -!- mumra has quit [Quit: "Help! I've been g:lined from my mIRC!!" Bersirc 2.2: less n00bs [ http://www.bersirc.org/ - Open Source IRC ]] 05:32:24 -!- ais523 has joined ##crawl-dev 05:39:14 -!- st_ has quit [] 05:44:15 03galehar * rc2157fdbde39 10/crawl-ref/source/ (effects.cc player.cc): Smoother and less agressive maprot in abyss and labyrinths. 05:44:15 03galehar * r16d6930bddfe 10/crawl-ref/source/mon-stuff.cc: Fix some pathfinding issues. 05:44:16 03galehar * rbca85593ae91 10/crawl-ref/source/mon-pathfind.cc: Minor cosmetic fix. 05:44:42 -!- syllogism has joined ##crawl-dev 05:52:47 -!- ais523 has quit [Read error: Connection reset by peer] 05:53:37 -!- ais523 has joined ##crawl-dev 05:56:06 03kilobyte * rb66bc687d5ba 10/crawl-ref/source/item_use.cc: Fix broken affixation of the chaos brand. 05:59:53 * kilobyte weeps. Losing such a great opportunity to die in the future... and that was the only vorpalize scroll in this game :p 05:59:57 -!- Ashenzari has quit [Ping timeout: 260 seconds] 06:00:40 -!- Ashenzari has joined ##crawl-dev 06:00:41 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:10:36 -!- Keskital1 has joined ##crawl-dev 06:10:36 almost died due to chaos-induced explosion+transformation+berserk too, so it works like it should otherwise 06:10:36 -!- Ashenzari has quit [Ping timeout: 252 seconds] 06:11:26 -!- Ashenzari has joined ##crawl-dev 06:11:27 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:16:10 -!- Mu_ has joined ##crawl-dev 06:31:39 -!- Ashenzari has quit [Ping timeout: 276 seconds] 06:32:43 -!- Ashenzari has joined ##crawl-dev 06:32:44 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:34:27 -!- RichardHawk has quit [Quit: Recalculating TATARI at "the crossing crossing"] 06:40:42 -!- Ashenzari has quit [Ping timeout: 252 seconds] 06:41:26 -!- Ashenzari has joined ##crawl-dev 06:41:26 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:50:57 -!- Ashenzari has quit [Ping timeout: 240 seconds] 06:51:52 -!- Ashenzari has joined ##crawl-dev 06:51:52 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 06:54:56 -!- MarvinPA_ has quit [Read error: Connection reset by peer] 06:55:27 -!- MarvinPA_ has joined ##crawl-dev 07:02:20 -!- galehar_ has quit [Remote host closed the connection] 07:02:52 -!- Ashenzari has quit [Ping timeout: 258 seconds] 07:04:15 -!- Ashenzari has joined ##crawl-dev 07:04:16 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 07:12:52 -!- monqy has quit [Quit: hello] 07:29:54 -!- dtsund has joined ##crawl-dev 07:43:11 -!- hoody_ has joined ##crawl-dev 07:49:31 -!- mumra has joined ##crawl-dev 07:50:21 hi 07:50:31 regarding conversation last night about vault testing 07:50:59 what would make things infinitely easier is if it were possible to recompile a vault file without exiting crawl 07:51:34 so you can quickly preview changes within crawl itself ... doesn't sound *too* hard perhaps 07:52:41 heh, does restarting crawl really take that long? 07:53:43 -!- ais523 has quit [Remote host closed the connection] 07:53:57 -!- ais523 has joined ##crawl-dev 07:56:36 Zaba: since it recompiles all the .des files every time you make even a small change, it takes a non-trivial amount of time 07:56:46 ehm, it doesn't 07:56:51 it only recompiles the files that have changed 07:57:19 ok, didn't know that; but it still has to load all the resources and stuff 07:57:31 and, if you're working on one of the large des files, that file will still take time to recompile 07:57:54 (of course i test in tiles quite a lot, which takes longer to load than console) 07:58:01 then test in console :P 07:58:07 hehe 07:58:19 and I don't think des file compilation really takes as much time as you're making it seem :P 07:58:27 the other thing that takes a while is you have to spawn your vault in wizard mode, and then find where it's appeared on the map 07:58:49 you can do &L*your_map_name to have it placed on your position 07:58:52 so including starting crawl, loading your game, to finally reaching your vault, that's a dozen or two keystrokes 07:59:01 ok, didn't know about * 07:59:05 nobody does :P 07:59:19 needs to be in the documentation 08:00:11 that only works for mini- or floating vaults, of course 08:00:23 also, you can just make your vault PLACE: D:1 08:00:30 (temporarily, of course) 08:00:40 yeah but then you have to start a new game to test it 08:00:57 so you end up with dozens of XL:1 characters that you have to go and delete from the save dir 08:01:00 *shrug* that's what I always do :P 08:01:10 I ^Q everytime :P 08:01:17 i know none of these issues are huge on their own; but it all adds up to make the testing process fairly inconvenient 08:02:04 maybe a "no confirm on ^Q option" 08:02:32 ok i've added about * to the testing section on the wiki docs 08:02:39 that's pretty useful 08:05:40 -!- dtsund has quit [Quit: dtsund] 08:08:09 My Serpent of Hell has the description "Well now, isn't this buggy?" 08:08:21 In Gehenna. 08:10:09 well no wonder 08:10:29 Why's that? 08:10:34 they have no description :P 08:10:43 Okay. 08:11:09 oh.. actually. 08:11:21 they have a hardcoded description for red/white/cyan/magenta. 08:11:31 (and the one that you're seeing for any other color) 08:13:02 heh nice. 08:13:47 their actual colors are ETC_FIRE/ICE/IRON/DEATH 08:20:32 -!- galehar has quit [Ping timeout: 252 seconds] 08:22:45 -!- Twilight13 has quit [Read error: Connection reset by peer] 08:24:12 -!- Twilight13 has joined ##crawl-dev 09:17:14 -!- hoody_ has quit [Remote host closed the connection] 09:17:38 -!- hoody_ has joined ##crawl-dev 09:22:49 -!- hoody_ has quit [Ping timeout: 276 seconds] 09:32:08 -!- hoody_ has joined ##crawl-dev 09:33:16 guess what royally screwed me in Zot, both on diving and on the Orb run? Not ancient liches or other shit. Ghost moths. On a strict non-caster. 09:33:30 a berserk moth hits for 60-100 every turn, for example 09:33:36 Spider is going to be fun :) 09:35:01 -!- Fail_Bot has joined ##crawl-dev 09:38:01 Hehe, cool. 09:38:31 !lg kilobyte won 09:38:33 16. KiloByte the Phalangite (L24 CeDK), worshipper of Yredelemnul, escaped with the Orb and 3 runes on 2011-09-15, with 1337161 points after 96064 turns and 9:39:56. 09:38:38 !log kilobyte won 09:38:39 16. KiloByte, XL24 CeDK, T:96064: http://crawl.develz.org/morgues/trunk/KiloByte/morgue-KiloByte-20110915-142816.txt 09:40:13 it's interesting how sharply Yred's power falls down 09:40:30 did you have any servants alive by zot:5? 09:41:01 the whole game: several servitors at a time, stashed extras, no need to bother about bone dragons. Zot:1 -- one fight and all servitors are gone 09:41:08 had 4 bone dragons in Zot:5 09:41:42 I think the most abusive yred chars are the casters with haste 09:41:44 some junk pan lord offed them, they heal too slowly to be of much use during the orb run 09:42:49 normally, you just rest if servitors/bone dragons are wounded and you expect opposition 09:46:44 Keskital1: beh, no moths in wrath in Spider, so no berserk. But then, people in early rune branches are so much weaker. 09:47:17 I wonder how hard tomb is with a yred host 09:47:53 dpeg suggested that to avoid problems with no rPois if the rune lock happens, we could choose one poisonous one wet branch. But that means you could get Spider and Shoals. 09:48:18 is there something wrong with getting spider and shoals? 09:48:55 elliptic: Yred has no use in non-living branches. The only power that works is Injury Mirror, and you can use it 1-2 times before you run out of piety, which you can't refill. 09:49:14 kilobyte: but the bone dragons do have use, no? 09:49:48 -!- Keskital1 is now known as Keskitalo 09:50:11 some people like to do tomb with TSO summons... I'm not sure exactly how hasted bone dragons compare but they are probably pretty good (and you can have more of them) 09:50:17 wolf spiders, redbacks, tarantellas are harder than nagas, ghost moths kick people's butt even with no rPois. And that without the Hellspider. 09:51:04 probably not durably enough to survive two greater mummies at a time, I guess 09:51:04 Well, nagas are too easy. 09:51:17 nagas are getting constriction, no? 09:52:22 Looking forward to that! 09:52:46 would it do much damage? remember that most of Snake is corridor combat 09:53:01 It needs wider corridors, yeah. 09:53:31 And Naga Warpers that blink next to you to constrict you in place. 09:54:15 -!- valrus has joined ##crawl-dev 10:15:07 -!- [1]J has joined ##crawl-dev 10:18:11 -!- hoody_ has quit [Remote host closed the connection] 10:18:36 -!- hoody_ has joined ##crawl-dev 10:23:21 -!- hoody_ has quit [Ping timeout: 260 seconds] 10:39:34 can someone with ops in crawl set the topic back to whatever it's supposed to be? somebody messed it up in the server upgrade madness yesterday 10:40:30 it should be "Dungeon Crawl Stone Soup | Play Crawl online now: type ??cao, ??cdo or ??webtiles for instructions | http://crawl.develz.org | FooTV: termcast.develz.org | ##crawl-offtopic: You got your gender in my peanut butter! Xom thinks this is hilarious!" 10:43:10 -!- ainsophyao has quit [Remote host closed the connection] 10:47:21 -!- edlothiol has joined ##crawl-dev 10:50:37 -!- Galefury has joined ##crawl-dev 11:01:06 -!- hoody_ has joined ##crawl-dev 11:03:11 -!- dtsund has joined ##crawl-dev 11:08:56 -!- [1]J has quit [Quit: HydraIRC -> http://www.hydrairc.com <- IRC with a difference] 11:10:57 -!- RichardHawk has joined ##crawl-dev 11:17:00 -!- valrus has quit [Remote host closed the connection] 11:22:24 -!- edlothiol has quit [Ping timeout: 252 seconds] 11:44:25 -!- SamB__ is now known as SamB 11:47:28 -!- dtsund has quit [Quit: dtsund] 11:49:01 kilobyte: yeah, I actually find merge requests to be just about the easiest way to suggest changes 11:55:58 I mean, pushing a branch and then fiddling with a web UI is a lot less trouble than having to figure out how to create a series of patches and then either email them to the list or post them to mantiss from my Debian box... 11:58:12 -!- greatzebu has joined ##crawl-dev 12:00:30 kilobyte: commit 6c9450d (unordered map) breaks the mac build 12:01:13 because to build for os 10.4 and support ppc users we have to use gcc 4.0, which has some weird unordered_map problem 12:01:39 works fine with newer gcc versions, but those versions won't build a ppc binary 12:06:50 Hunter Weapon Choice (https://crawl.develz.org/mantis/view.php?id=4549) by LorrdErnie 12:07:23 !tell kilobyte commit 6c9450d (unordered map) breaks the mac build, because to build for os 10.4 and support ppc users we have to use gcc 4.0, which has some weird unordered_map problem. It works fine with newer gcc versions, but those versions won't build a ppc binary 12:07:25 greatzebu: OK, I'll let kilobyte know. 12:08:26 -!- valrus has joined ##crawl-dev 12:09:09 !tell Napkin mac builds are broken at the moment (as of commit 6c9450d), so if there are no new binaries for a little while until we get it sorted out, that's why 12:09:10 greatzebu: OK, I'll let Napkin know. 12:09:30 roger, thanks for letting me know 12:09:30 Napkin: You have 3 messages. Use !messages to read them. 12:10:02 greatzebu, what is the weird unordered_map problem? 12:10:54 there is a compiler error when using const_iterators on unordered_maps that seems to only show up on gcc 4.0 12:11:42 I am not sure of the exact cause 12:11:46 but the error message is: 12:11:52 directn.cc: In function ‘void setup_feature_descs_short()’: 12:11:52 directn.cc:3233: error: no matching function for call to ‘Internal::hashtable_iterator, false, false>::hashtable_iterator()’ 12:11:53 make: *** [directn.o] Error 1 12:12:17 i have no clue either - kilobyte can probably help you out 12:12:38 yeah, probably true 12:12:44 MarvinPA_: you can use "/msg chanserv access ##crawl list" to find out who has ops 12:12:51 my concern is that it's probably a bug in gcc 4.0 that has since been fixed 12:13:03 so there might not be a great solution 12:13:30 SamB: yeah, i think only a couple of people have ops in there that are around frequently 12:13:32 at which point the options are basically either abandon ppc builds or don't use unordered_map 12:13:48 oh ok Napkin does, didn't realise that 12:13:49 on it 12:13:56 thanks :) 12:15:42 done :) 12:16:07 gcc >4.0 is not able to do ppc builds? 12:16:24 greatzebu: or use a typedef to either std::unordered_map or std::map, depending on the target 12:16:38 SamB: good point 12:17:05 I'm not sure on using gcc > 4.0 to do ppc builds. I think that it doesn't work, but I'm not completely positive 12:19:08 we'll probably get a philosophic explaination ;> 12:19:16 it might be that the makefile requires gcc 4.0 on 10.4 builds for other reasons 12:19:31 i'll do some experimenting 12:20:04 great :) 12:20:25 for the record: i'm using gcc 4.4 to cross-compile the windows binaries on CDO 12:20:58 so, if you experiment, use one >=4.4 12:23:16 looks like the mac 10.4 sdk won't build with anything newer than 4.0 12:23:28 supporting old os versions sucks 12:26:12 i would say nobody is still using a version released in 2005, but we got multiple requests for support when i was making binaries using the 10.6 sdk 12:26:15 so, i don't know 12:26:46 don't worry yet 12:27:56 I'm guessing there are Apple patches on that 4.0 that are more-or-less needed for OS X/ppc, but aren't available for later versions of GCC 12:28:10 yes, I think that's the case 12:28:26 Apple has been trying to move away from gcc for a long time 12:28:42 I just meant on account of being PPC-specific, actually 12:28:56 i doubt apple itself provided any help nor even patches to the gcc project ;) 12:29:03 they provide patches 12:29:10 but they're not easy to merge 12:29:26 okay, maybe not in diff form 12:29:40 help, not so much... 12:29:52 hasn't apple moved to clang? 12:29:57 yes, more or less 12:29:58 (generalizing from GDB here) 12:30:15 despite the fact clang's C++ support is... lacking 12:30:44 it's gotten a lot better recently 12:30:53 omg.. creating software that runs on an apple platform without paying royalties? that must have kept apple crying all night 12:30:56 and the error/warning messages are great 12:31:21 Napkin: huh? 12:31:36 haha, i think the issue was more that they can't integrate internal gcc stuff into their dev tools without open sourcing the tools 12:31:55 gcc was deliberately made monolithic to force that to happen (possibly a bad move, IMO) 12:32:16 it's what drove the development of clang, as a result gcc ended up getting less monolithic as soon as clang looked like it was in danger of eventually winning 12:32:22 bad engineering, but i guess understandable from a political perspective 12:32:31 GCC definitely goes out of its way to make integration difficult 12:32:39 intentionally 12:32:48 I know 12:32:54 -!- Galefury has quit [Read error: Connection reset by peer] 12:32:56 GDB, thankfully, has broken free from this mindset to some extent 12:33:05 with the python scripting 12:36:10 i didn't know gdb supported scripting 12:36:16 i'll have to look into that 12:36:18 RMS is a bit paranoid about that kind of thing... 12:36:24 yeah, it's recent 12:36:39 last few years, I mean 12:36:44 right 12:37:17 and it's still kind of rough around the edges 12:38:05 When I try to use the libstdc++ pretty-printer scripts with Crawl, they have a tendancy to crash GDB... 12:38:49 still, sounds promising. i love new debugger features :-) 12:39:47 yeah, I'm sure they'll get it ironed out soon 12:40:22 I'm wondering how long until they have symbol/source server support, like windbg has 12:44:44 03MarvinPA * r434304d99f22 10/crawl-ref/source/initfile.cc: Remember Hunter/AM weapon choice between games 12:45:05 It seems like they have all the file-format features needed, and that the symbol server support would be mostly a matter of putting in something (possibly using some additional Python scripting hooks) to let GDB grab symbols over the HTTP and writing some scripts to arrange the symbol files in an appropriate manner 12:46:45 Source would be slightly trickier; MS's approach of embedding the needed VCS command lines in the symbol files would probably end up being the way to go, though 12:47:42 interesting. do you think distros would host symbol servers if the feature was available? 12:48:14 -!- Galefury has joined ##crawl-dev 12:48:24 I do 12:49:11 that would be very nice 12:51:11 -!- Galefury has quit [Read error: Connection reset by peer] 12:51:13 I'm not sure quite what package releases would be covered (would Debian skip that for "experimental"? "unstable"?), or that everyone would do it, but I'm pretty sure some would 12:52:03 I mean, mozilla.org does it for Windows builds of Firefox already, and at least some builds of Chrome do it too.. 12:54:09 nice, I didn't know that 12:57:13 I'm kind of expecting Redhat to do it first, seeing as some of the most active GDB developers seem to work for them... 13:00:32 hmm, mumra is right: there really should be a manual for Wizard mode 13:02:19 just read the source :P 13:02:48 Right... 13:03:00 I mean, sure, I do, but there's a lot of it scattered all over 13:03:20 and not all .des contributors are even going to know any C++ 13:03:41 -!- ais523 has quit [Remote host closed the connection] 13:03:46 could add information about * into the &L prompt 13:04:35 but, it should be kept in mind that that feature really sounds much better than it is. If the vault ends up overwriting the player with something solid, the player gets randomly teleported and has to teleport back to the vault anyway. 13:04:43 It'd still be nice to have some easily-searched text file that described each command in slightly higher verbosity than the help screen 13:05:03 Which could mention such caveats 13:07:10 there's another thing that's really begging for better docs, and that is the difference between normal vaults and minivaults 13:07:23 unfortunately it's pretty hard to describe properly 13:08:50 I gathered 13:09:29 in particular, it is important to not mess that up with primary vs. secondary vaults. Normal vaults can be both, minivaults can only be secondary. 13:09:37 it *sounds* like vaults are placed first, then most of the level generation gets done, then minivaults 13:09:52 I don't even know what primary/secondary vaults are 13:09:57 only _one_ vault can be placed before the level is generated, and it is the primary vault 13:10:01 ah 13:10:10 vaults placed on an already existing level are all secondary 13:10:45 it's different in D and elsewhere, actually 13:10:55 elsewhere doesn't get random primary vaults at all 13:11:01 It's at least obvious that if the entire level is taken up by a vault, that's not a minivault :-) 13:11:13 (that is, only PLACE: vaults like branch endings are placed as primary) 13:11:31 but normal vaults can still be placed in non-D, except they're always placed as secondary 13:11:49 How big are D's primary vaults? Are the starting-location vaults primary? 13:12:01 yes, entry vault is a most obvious example of a primary vault 13:12:22 Do any of those cover the whole level? 13:12:29 yeah, like, say, river lethe 13:13:14 (Are we both talking about the vaults that the player starts the game in?) 13:13:17 yes 13:13:32 oh wait, no, there aren't ORIENT: encompass entry vaults, sorry 13:13:40 (but technically that's entirely possible) 13:14:56 Where should the primary/secondary distinction be covered? 13:15:12 I have no idea 13:15:43 it's quite messy, to be honest, especially the surprising differences between D and non-D :P 13:16:47 SamB, ah, actually, non-D -can- get random primary vaults, but only when they're ORIENT: encompass 13:16:57 it just never really comes up on practice, since there are no such vaults 13:17:15 SamB, the distinction between D and non-D stems from the differences in layouts, I think. 13:17:59 there are only, uh, three layout generators that can build a level 'around' a primary vault 13:18:36 Hmm. introduction.txt mentions a number of vault types, but doesn't say anything about primary/secondary per se. 13:19:28 -!- Galefury has joined ##crawl-dev 13:19:32 well, it's an implementation detail 13:19:39 It does mention "special level" vaults, like Temple and branch ends 13:20:09 vault makers shouldn't care about that unless they run into problems caused by that 13:20:26 yeah 13:21:30 -!- Galefury has quit [Read error: Connection reset by peer] 13:23:59 maybe it should be mentioned in the ORIENT subsection of syntax.txt? 13:29:06 -!- Galefury has joined ##crawl-dev 13:30:28 another thing that could use documentation, afaict, is the various lua scripting interfaces 13:31:34 -!- Galefury has quit [Read error: Connection reset by peer] 13:32:06 -!- st_ has joined ##crawl-dev 13:34:42 !tell dpeg http://roguelikeradio.blogspot.com/2011/09/episode-3-brogue.html 13:34:42 Napkin: OK, I'll let dpeg know. 13:39:32 -!- Galefury has joined ##crawl-dev 13:41:32 -!- Galefury has quit [Read error: Connection reset by peer] 13:45:25 -!- dtsund has joined ##crawl-dev 13:46:50 -!- Galefury has joined ##crawl-dev 13:47:19 Zabe: Different kinds of vaults are covered in the introduction - https://crawl.develz.org/wiki/doku.php?id=dcss:help:maps:introduction#kinds_of_vaults 13:47:29 uh, Zaba I mean ;) 13:48:58 SamB: I was going to start improving the Lua documentation in the wiki docs - I've been spending a lot of time looking thru source to find out how functions work. But kilobyte and due were talking about the possibility of doc comments (e.g. doxygen) which would be all-round better 13:49:48 If luadoc was of any use with C++ sources, that would certainly be nice... 13:50:49 SamB: I saw a way that could be done, customising luadoc to interpret C-style comments 13:51:08 but how would it know what it was you were documenting? 13:51:36 I guess you'd have to say so at the beginning of the comment? 13:53:18 SamB: http://lua-users.org/lists/lua-l/2011-03/msg00586.html 13:54:05 luadoc only looks for the -- markers; it doesn't actually parse any code 13:54:14 oh 13:54:32 that would certainly make things simpler 13:56:34 pretty neat, it can happily live alongside any c++ documentation 13:57:46 so, any idea what the modifications were, or whether they were actually related to the use of C++? 13:58:39 i don't think you need any modifications - just point luadoc at the c sources, it'll extract those method definitions 14:00:41 although - the OP describes a method of getting doxygen to produce Lua documentation: http://lua-users.org/lists/lua-l/2011-03/msg00537.html 14:01:03 but then someone else posted that luadoc can be abused to do it very easily 14:01:19 hmm, one other thing that seems important to me is explaining a bit about why lua conditionals can be used to conditionally set settings in crawlrc 14:02:01 yeah, [ab]using luadoc seems much saner 14:02:09 i'm guessing crawlrc uses a similar compilation system to des files? 14:02:21 I'm not sure 14:02:41 I guess so, it does use : prefixes for lua the same way... 14:02:56 i had this idea that it'd be nice if the speech files allowed the same 14:03:19 so you could actually pass parameters into the speech system and modify the text around them; 14:03:31 oh? is that different from what the description files already allow? 14:03:42 which would remove a certain amount of code logic relating to messages (and pave the way more for multilingual versions) 14:03:57 hmm. multilingual versions seems pretty futile to me! 14:04:26 yeah, you can't properly pass parameters into the speech/description files; sometimes they have replacements run manually afterwards 14:04:57 e.g. @The_monster@ and @to_player@ etc. are just manually replaced by whichever function is using those strings 14:05:32 i think crawlrc has a simple c++ parser, des files use lex/bison 14:06:05 multilingual is certainly a massively improbably thing, having seen how much text exists in crawl and how much of it is hard-coded :) 14:06:30 greatzebu: but the thing that lets lua conditionals encompass non-lua stuff might be pretty similar 14:06:59 one thing that you could do is have non-english description databases that people could use 14:07:02 ah, they definitely might be, i don't know much about the lua bridge 14:07:20 but keep the majority of the game output in english 14:09:10 but I guess any English-assuming substititions performed on the descriptions would foul that up... 14:11:02 mumra: so, I guess the luadoc comments would need to go in the .cc files, right? 14:18:31 -!- Galefury has quit [Quit: Stay sane inside insanity!] 14:42:40 -!- Keskitalo has quit [Quit: leaving] 14:47:31 mumra: hmm, slight snag; it looks like luadoc's standard taglet only processes files with extensions ".lua" and ".luadoc" :-( 14:48:08 SamB: one option is a script that copied all the *.cc files to *.lua and deleted them after 14:48:52 also, for the .lua case it seems to have some regexes to detect function starts 14:49:40 yes; the examples in that thread had the lua function signature within the /* */ comments# 14:50:05 so it's assuming that only the function start is detected by luadoc and nothing else is required 14:50:59 i think it'd be pretty easy to modify the doclet to look at cc files 14:51:31 uh, taglet i mean 15:10:38 03MarvinPA * rd447a99e2d15 10/crawl-ref/source/mon-place.cc: Properly upgrade shadow demons to 2s 15:16:01 -!- evilmike has joined ##crawl-dev 15:19:21 -!- MarvinPA has joined ##crawl-dev 15:22:55 -!- MarvinPA_ has quit [Ping timeout: 276 seconds] 15:37:53 -!- syllogism has quit [] 15:40:08 "goooood! i hate those statues!" :D 15:40:24 hey mumra? 15:40:49 could you message me, which files from the docs you have wikified? 15:41:42 lo 15:47:13 -!- ussdefiant has quit [Ping timeout: 260 seconds] 16:11:49 -!- petete has joined ##crawl-dev 16:16:17 -!- hoody_ has quit [Remote host closed the connection] 16:23:40 greatzebu: I intentionally used std::unordered_map precisely because I was afraid of problems with ancient compilers :p 16:24:46 there's an option to use hash_map, if that won't work and there is no easy way to get std::unordered_map to work, I'll just revert 16:26:01 -!- monqy has joined ##crawl-dev 16:27:08 kilobyte: ok, i'll take a look at using hash_map 16:33:14 hrm, it doesn't seem one can just #define something with std:: in its name 16:34:59 a single invocation of sed can work around this, though 16:36:25 or alternatively, we can stick with std::map -- there won't be a speed increase but it's not like there are any big mappings anywhere in Crawl; and the one case where big mappings can occur can use a hand-made hash 16:36:35 for the moment, why not just #ifdef APPLE typedef std::map<...> hash_map_type #else typedef std::unordered_map<...> hash_map_type #endif 16:38:09 probably the only place where a speed increase would be at all relevant is on servers, which won't be macs anyway 16:39:56 a generic hash might be even slower than a red-black tree for small data sizes 16:41:00 it is certainly less memory efficient, but i don't know if that's a concern or not 16:43:21 -!- galehar has joined ##crawl-dev 16:44:40 -!- galehar changed the topic of ##crawl-dev to: Dungeon Crawl Stone Soup | Play Crawl online now: type ??cao, ??cdo or ??webtiles for instructions | http://crawl.develz.org | FooTV: termcast.develz.org | ##crawl-offtopic: You got your gender in my peanut butter! Xom thinks this is hilarious! 16:45:03 wrong channel, galehar ;) 16:45:19 oh sorry 16:45:29 I misread MarvinPA's request 16:45:41 np 16:45:50 -!- galehar changed the topic of ##crawl-dev to: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 16:45:52 but i don't remember the old topic here 16:45:56 ah, great :) 16:48:09 !tell edlothiol WebTiles run stable now! \o/ 16:48:09 Napkin: OK, I'll let edlothiol know. 16:48:20 mumra: yeah, I think copying the .cc files to .cc.luadoc is probably the way to go 16:48:59 I've got it sort-of working, though obviously there isn't much in the way of actual documentation ;-) 16:49:49 kilobyte: i figured out the cause of the unordered_map problem 16:50:09 old versions of unordered_map::iterator have no default constructor 16:50:19 apparently it was not required by the standard 16:51:14 so there is an easy but annoying workaround to just use explicit initialization for any hash table iterator 16:52:08 how about declaring a defaut constructor ourselves? 16:52:41 03galehar * rfb23133a762f 10/crawl-ref/source/dat/des/branches/pan.des: Increase the chance to find an exit in a Demonic Rune vault. 16:52:43 03galehar * r2d69262cb804 10/crawl-ref/source/mon-movetarget.cc: Fix pathfinding distance inconsistency. 16:52:43 03galehar * rb8dc175f63fe 10/crawl-ref/source/player.cc: Little improvement to the map rot algorithm. 16:52:43 03galehar * r5e0bcc91de93 10/crawl-ref/source/mon-movetarget.cc: Fix a crash with DEBUG_PATHFIND. 16:52:43 03galehar * r64d129586ba2 10/crawl-ref/source/mon-movetarget.cc: Fix broken logic. 16:52:52 03galehar * rda42bcd28637 10/crawl-ref/source/ (spl-cast.cc spl-tornado.cc): Tornado adjustments. 16:53:09 i'm not sure how we can provide a constructor for unordered_map<...>::iterator 16:53:32 SamB: Nice. Maybe there's a way to have a build step that auto-publishes the docs to wiki? ;) ...Are you sure you can't just alter the taglet to pick up .cc files as well? 16:54:10 i think ext/hash_map does solve the problem as well, though 16:54:40 Well, it turns out that the doclet also more-or-less assumes that input files will be either .lua or .luadoc 16:55:06 I suppose we could monkey-patch them somehow, but that seems worse 16:56:39 -!- ussdefiant has joined ##crawl-dev 19:46:38 -!- Ashenzari has joined ##crawl-dev 19:46:38 19:46:38 -!- The topic of ##crawl-dev is: DCSS Development | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 19:46:41 weapon choice is inconsistently remembered at game start (not the same as 4537) (https://crawl.develz.org/mantis/view.php?id=4550) by reid 19:51:44 Devastator: Monsters shudder even if they're unaffected by Shatter (https://crawl.develz.org/mantis/view.php?id=4552) by reid 19:51:44 Devastator: "DEAD MONSTER shudders." (https://crawl.develz.org/mantis/view.php?id=4551) by reid 19:53:44 shudders? isn't that the same message for when abjuration doesn't completely work on a monster? 19:54:42 -!- Euphoria has joined ##crawl-dev 19:54:42 19:54:53 -!- Euphoria has quit [Changing host] 19:54:53 -!- Euphoria has joined ##crawl-dev 19:54:53 19:55:02 ? 19:56:39 also for lrd-ing wood golems i think 19:56:48 huh 19:56:49 oh, that's "shudders violently" 19:59:18 -!- purge has joined ##crawl-dev 19:59:18 20:00:02 !messages 20:00:07 !message 20:00:27 "purge:" ? 20:00:37 is greeting everyone 20:00:50 wheres the greeting part? :) 20:00:54 -!- ixtli has joined ##crawl-dev 20:00:54 20:01:21 oh, including itself i guess 20:01:22 how strange 20:01:25 -!- dtsund has quit [Quit: dtsund] 20:02:31 ?? 20:03:37 -!- hoody has joined ##crawl-dev 20:03:37 20:03:49 um yeah 20:06:48 experience display does not always update correctly when reaching XL 27 (https://crawl.develz.org/mantis/view.php?id=4553) by reid 20:11:50 -!- Pingas has quit [Read error: Connection reset by peer] 20:12:57 great 20:21:20 hahaha, why is ashenzari suddenly so friendly 20:21:28 including itself, apparently 20:21:37 17:46:46 < Ashenzari> Ashenzari: 20:22:37 -!- Twilight13 has quit [Read error: Connection reset by peer] 20:22:56 -!- Twilight13 has joined ##crawl-dev 20:22:56 20:23:07 -!- rawrmage has left ##crawl-dev 20:23:07 20:23:08 -!- rawrmage has joined ##crawl-dev 20:23:08 20:23:13 lol 20:23:31 poor ashenzari 20:26:39 he's cursed 20:30:57 parts and joins, apparently 20:31:34 -!- hoody has quit [Remote host closed the connection] 20:31:59 -!- hoody has joined ##crawl-dev 20:31:59 20:36:50 -!- hoody has quit [Ping timeout: 260 seconds] 20:37:26 20:51:02 -!- hoody has joined ##crawl-dev 20:51:02 21:16:29 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 22:07:56 why do gods gift scythes, they are barely even legitimate weapons 22:08:56 oka and trog never seem to actually be useful for polearm guys 22:19:18 -!- Wensley is now known as Roque 22:24:13 -!- Roque is now known as Wensley 22:25:08 !tell elliptic buff scythes next! 22:25:09 Wensley: OK, I'll let elliptic know. 22:53:01 -!- MarvinPA_ has quit [Ping timeout: 240 seconds] 22:59:20 -!- ainsophyao has quit [Remote host closed the connection] 23:07:09 -!- MarvinPA has joined ##crawl-dev 23:07:09 23:17:47 -!- evilmike has quit [Quit: something happened] 23:26:34 -!- Wensley has quit [Quit: ChatZilla 0.9.87-rdmsoft [XULRunner 1.9.0.17/2009122204]] 23:41:21 -!- valrus has quit [Read error: Connection timed out] 23:41:46 -!- valrus has joined ##crawl-dev 23:41:46