00:03:08 or...nevermind; I was sure that was right *shrug* 00:35:36 dpmdpm: I assume you meant "remote server" there? 00:36:45 I wonder if the clua persistant file could be used in some way 00:37:49 that file contains lua data as its contents and is loaded by clua; it can be created from clua by writing to a special table 00:38:53 What's the hazard of these arbitrary code injection bugs? 00:40:03 not sure if loadfile is actually useable from clua though 00:47:06 Bcadren: worst case is remote shell 00:47:09 gammafunk yes 00:48:51 dpmdpm: not sure if the c_persist lua file is relevant to your question about loadfile, but it's a file that gets created and stored on a server if you define the lua table c_persist 00:51:23 example file created by my clua macros: http://crawl.berotato.org/crawl/rcfiles/crawl-git/gammafunk.rc.persist 00:55:59 <|amethyst> can't you also create a pretty much arbitrary file by making a new account and editing its rc? or editing your own macros file 00:56:09 Thanks. That's helpful. It doesn't seem like that file could ever start with the Lua bytecode header 1b 4c 75 61 51 00 etc so it seems safe 00:58:23 amethyst you're suggegsting making crawlrc a Lua bytecode file directly? I think the crawl rc file parser would block that. It looks for Lua code inside { } brackets. 00:58:48 <|amethyst> not in the account you're exploiting from, but a different account 00:59:06 <|amethyst> that doesn't ever need to run crawl, just the webtiles or dgamelaunch rc editor 01:00:11 Huh. Interesting. I guess if you're at all worried that loadfile could be exploited then best to block all three functions: load, loadfile, and loadstring 01:23:43 There isn't any way to make tile-flags.h go past 0x8000000000000000ULL without a version break is there? 01:25:40 Bcadren, Now I feel bad for using some of the last-few-available of those slots for the threat indicator graphics :) 01:27:34 ByrelMitchell? 01:28:34 I overwrote one that's not used anymore for my fork...I also think it may be safe to be like (this flag is POSSESSIBLE if you're with Yred; but it's IDEALIZED if you're with Hepliaklqana). 01:28:59 (Turned Death's Door (used to be cast by one Unique) into...I think the stab icon). 01:30:01 (Split confused and just distracted into two icons for how often that's relevant.) (Confused kept the question mark; distracted gets an image of a bloody dagger; to indicate "can backstab"). 01:31:03 !next 01:32:16 The RNG declares that svendre shall win DESu next. 01:32:24 Suppose could roll paralyzed into that and have 3 on two bits...but yea that's a CROWDED enum. 01:34:31 Bcadren, when I was working in there, I wondered whether it even really needed to be like that. Is the space saved from packing all those tile traits into 1-bit flags really worth it? (Maybe it is, I don't know, serious question) 01:36:04 I honestly don't know. If the flags are set serverside; then probably. If the flags are set player side, definitely not. 01:37:16 aidanh is the better one for that question though 01:38:37 Wait...those flags are local tiles only; the webtiles is much smaller; yea no...it's not that necessary. IDK why it's that way other than tradition. 01:40:56 It's also traditional to compress sprites to an extreme it's like crawl was a gameboy game hurting for cartridge space or something. 01:43:18 yeah that was what I was thinking about it too 01:43:44 you could probably refactor the local one straight from that bitfield to an enum_class with very little performance difference...that might be pushing your luck for webtiles though 01:44:34 Probably right with the "tradition" guess...I know the game has been around a long time, maybe back in the day, the game was big enough even then relative to the hardware of the time that it *was* worthwhile to space-save that much, and there just hasn't been any pressing reason to restructure. 01:47:15 That's honestly worth trying for standalone tiles; see if there's a significant performance hit for just using an enum-class instead of that old enum. Webtiles I'm less eager to say "yea refactor" because A LOT of stuff gts done serverside and streamed and I just don't know 01:54:30 Psymania found your commit; can't take it for my fork; I have wretched and mute in those bits... 02:00:35 It seems like there's an awful lot of redundancy there too. IE a monster gets hasted: add_ench(ENCH_HASTE); then if ENCH_HASTE add is(MB_HASTED) [description flag]; then if local tiles and MB_HASTED set the tile-flag TILE_FLAG_HASTED...from a memory perspective it really really doesn't make sense to have the information "this monster is hasted" set 02:00:35 three places. 02:02:03 enchilada_haste 02:04:29 can try refactoring it out; but I honestly don't know if it'd cause performance issues; it could be writing to the byte arrays because it's faster lookup; than checking enchant values all the tie. 02:09:57 Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-510-ge065f6d 02:26:13 03gammafunk02 07* 0.25-a0-511-gfc656d3: Improve UI for attempting to recast Dragon's Call 10(11 hours ago, 2 files, 9+ 8-) 13https://github.com/crawl/crawl/commit/fc656d3889d8 02:26:13 03gammafunk02 07* 0.25-a0-512-ga8e59b5: Properly handle recasting of Frozen Ramparts (shummie) 10(9 hours ago, 1 file, 5+ 0-) 13https://github.com/crawl/crawl/commit/a8e59b5b3bad 02:26:13 03gammafunk02 07* 0.25-a0-513-ge77a692: Clean up icy walls upon level load 10(5 hours ago, 1 file, 24+ 1-) 13https://github.com/crawl/crawl/commit/e77a69270c2c 02:32:57 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-513-ge77a69270c (34) 02:48:30 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-443-g80245de 03:22:23 Stable (0.21) branch on crawl.beRotato.org updated to: 0.21.2-2-g132acf7 03:31:49 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-2961-g3345d5eee1 03:37:52 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-513-ge77a692 (34) 03:58:53 Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2195-gd5260a0a46 04:01:36 -!- OICU812 is now known as EuphOria 04:01:48 -!- EuphOria is now known as OICU812 07:10:02 -!- Tiobot is now known as Guest38152 07:11:45 ??dispater 07:11:46 dispater[1/2]: Lord of Dis (duh), and guardian of the iron rune. Has summon greater demon, crystal spear, iron shot, hellfire, and lots of AC to back it up. Drops his {staff of Dispater} if killed. 07:14:17 in 0.24, Dispater seems to have "Hurl Damnation" rather than Hellfire. 07:39:07 03theJollySin02 07https://github.com/crawl/crawl/pull/1284 * 0.25-a0-514-g1b4ad0e: Fixed arrival vaults without sufficient tactics 10(3 minutes ago, 1 file, 16+ 11-) 13https://github.com/crawl/crawl/commit/1b4ad0e31895 08:04:13 !learn edit dispater s/hellfire/damnation/ 08:04:13 Use: !learn edit dispater[NUM] s/// 08:04:17 !learn edit dispater[1 s/hellfire/damnation/ 08:04:18 dispater[1/2]: Lord of Dis (duh), and guardian of the iron rune. Has summon greater demon, crystal spear, iron shot, damnation, and lots of AC to back it up. Drops his {staff of Dispater} if killed. 11:24:43 Stable (0.23) branch on underhound.eu updated to: 0.23.1-92-g5aa5405057 14:27:48 -!- aidanh_ is now known as aidanh 14:54:19 -!- Psymania_ is now known as Psymania 18:23:32 Unstable branch on underhound.eu updated to: 0.25-a0-513-ge77a69270c (34) 23:36:51 lokapujya (L1 VSTm) Crash caused by signal #15: Terminated (D:1) 23:38:35 gerbil (L9 TrMo) Crash caused by signal #15: Terminated (D:6) 23:38:41 lokapujya (L2 VSTm) Crash caused by signal #15: Terminated (D:1) 23:40:21 jtro (L1 FoAK) Crash caused by signal #15: Terminated (Abyss:1) 23:40:22 Tossi (L24 FoEE) Crash caused by signal #15: Terminated (Shoals:3) 23:40:30 lokapujya (L1 VSTm) Crash caused by signal #15: Terminated (D:1) 23:41:01 svendre (L16 DESu) Crash caused by signal #15: Terminated (D:15) 23:42:03 lokapujya (L1 VSTm) Crash caused by signal #15: Terminated (D:1) 23:42:44 Luxidream (L17 VSAs) Crash caused by signal #15: Terminated (Depths:4) 23:52:03 !tell Bcadren yeah, that commit's a big one, I'm not surprised it's taking that long :/ 23:52:04 aidanh: OK, I'll let bcadren know. 23:52:38 !tell Bcadren on the plus side, unused params is a really good indicator of bugs; I was able to weed out a lot of weird behaviour 23:52:39 aidanh: OK, I'll let bcadren know. 23:54:13 !tell Bcadren try cherrypicking the fix off an older release branch; I had to alter it for 0.24 and below 23:54:13 aidanh: OK, I'll let bcadren know.