00:00:41 Stable (0.22) branch on CRAWL.XTAHUA.COM updated to: 0.22.0-27-ga576289 00:02:13 Stable (0.21) branch on CRAWL.XTAHUA.COM updated to: 0.21.1-77-g46dbe93 00:10:07 =nemelex SpIE 00:10:09 Defined keyword: nchoice => VpFE|BaEn|CeIE|DDCj|HaAE|KoVM|CeEE|BaFE|DrAM|TeWr|SpSu|DEAK|BaNe|DgEn|MuWr|FoTM|GnAK|MfEE|HuAE|DEHu|GnCK|VSSu|HaCj|OpAM|OgVM|FoWz|HOEn|SpIE 01:12:26 Stable branch on crawl.develz.org updated to: 0.21.1-77-g46dbe93 (34) 02:57:03 hey devs. How about breaking save compat this version? Clean up #ifdef #endif in the codebase 03:03:45 looks like about 14k lines, or 3% of codebase 03:11:09 Stable (0.22) branch on crawl.beRotato.org updated to: 0.22.0-27-ga576289 03:13:10 Unstable branch on crawl.beRotato.org updated to: 0.23-a0-108-gf7ee2ed (34) 04:32:21 -!- amalloy is now known as amalloy_ 04:51:51 hi guys, is bolt spell reach decreased with each target hit, as before? 08:42:55 alexjurkiewicz: how are you calculating that figure? it'd probably be more useful to know how often existing compat code affects development (with recent commits as a proxy?) 08:43:56 I'd suspect that compat code, once written, would not be changed very regularly or have much noticable impact on development 08:48:27 I think he just means that about 3% of the codebase is wrapped in a #if TAG_MAJOR_VERSION == 34 08:48:44 areas affected by save compat (tags.cc) are super messy, so there would be value in dropping save compat 08:51:01 and there are some ridiculous things we are still handling that are just not well tested any more, in principle saves from like 2011-12 can load 08:51:33 so I'd support dropping save compat (but I think I've been saying that almost since I became a dev) 08:52:24 it's not entirely a trivial project though, since those ifdefs have been around for so long in some cases 08:53:23 I believe |amethyst has checked every once in a while to see what happens if you remove it and fixed some things, but I don't know how long it has been 08:57:53 I don't suppose there'd be a way to keep save compat for games from the previous release, but not from earlier releases? 08:59:43 I think that'd strike a healthy balance between player and dev convenience 09:09:00 probably 09:09:39 save versioning doesn't line up with release versioning, though, so there'd be some work in figuring out how to implement it 09:10:22 in practice I think the biggest impact would be for online trunk games, though, so that solution wouldn't impact those 09:10:56 well, the idea is that they would be lined up 09:24:28 Stable (0.22) branch on underhound.eu updated to: 0.22.0-27-ga576289601 09:36:00 hm, dith shard triggered by a smite targeted spell will fire through friendly creatures 09:36:11 that seems like non-ideal behavior? 09:36:34 would rather have it just not fire 09:48:33 it looks like before some recent change it even angered them, guess the current state is an improvement on that 10:14:18 ??peacekeepr 10:14:18 peacekeepr ~ peacekeeper[1/1]: A fast, armored {manticore} found in {desolation}. 10:14:20 ??peacekeeper 10:14:20 peacekeeper[1/1]: A fast, armored {manticore} found in {desolation}. 10:14:26 should that spawn in vaults? 10:15:34 oh I guess a desolation entrance vault might place them 10:37:11 !messages 10:37:12 No messages for ebering. 10:53:30 the webspace that I host the api docs is down for the remainder of the month because the building is having its power distribution infrastructure rebuilt 10:54:13 I got an e-mail today asking for a mirror or something 10:54:36 so uh anyone interested in doing that? 11:09:35 I could look into that, those docs are definitely worth having up 11:10:13 we still haven't merged the PR that generates those docs, right? 11:10:38 correct 11:12:20 I have some more local fixups 11:12:27 let me force push those 11:13:10 this looks ready to merge to me but it's big enough that waiting until after t is possibly a good idea? 11:13:35 ya, I figured it would have to wait until after t 11:14:21 at least for the merge. but if someone with hosting and initiative could check out the branch and compile and host the docs 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-109-g5f0d7bb: Switch from LuaDoc to LDoc 10(7 weeks ago, 5 files, 55+ 769-) 13https://github.com/crawl/crawl/commit/5f0d7bbe806d 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-110-g254c616: De-Doxygen-ify 10(7 weeks ago, 18 files, 2+ 92-) 13https://github.com/crawl/crawl/commit/254c61607cb9 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-111-g43133e5: Switch over and expand old luadoc 10(7 weeks ago, 2 files, 432+ 146-) 13https://github.com/crawl/crawl/commit/43133e554bd3 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-112-ge706a03: Document dlua additions to the crawl module 10(7 weeks ago, 1 file, 49+ 1-) 13https://github.com/crawl/crawl/commit/e706a0383947 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-113-gef307ca: Stubs for dlua only modules. 10(7 weeks ago, 21 files, 57+ 18-) 13https://github.com/crawl/crawl/commit/ef307ca181dd 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-114-g26c2a9e: Document the rest of clua in ldoc 10(7 weeks ago, 16 files, 1676+ 106-) 13https://github.com/crawl/crawl/commit/26c2a9eff716 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-115-gf97c5bc: Sort lua api table of contents. 10(7 weeks ago, 1 file, 6+ 5-) 13https://github.com/crawl/crawl/commit/f97c5bce03d7 11:15:11 03ebering02 07https://github.com/crawl/crawl/pull/809 * 0.23-a0-116-gc4dd3b4: Clean up internal comments (aidanh) 10(5 weeks ago, 13 files, 14+ 56-) 13https://github.com/crawl/crawl/commit/c4dd3b426a61 11:17:37 is it completely static? 11:18:43 yes 11:26:23 03Floodkiller02 {Aidan Holm} 07* 0.23-a0-109-g5e4b187: Add another set of Gnoll playerdoll tiles 10(6 weeks ago, 3 files, 2+ 0-) 13https://github.com/crawl/crawl/commit/5e4b18702807 11:27:49 ebering: done, http://lua.dcss.xyz/ 11:28:07 one thing I can see is that it would be nice to have a version number somewhere there 11:28:33 ebering may need time for dns to propagate depending on where one is, I can see it already though 11:29:47 (for anyone who clicks on that link now and sees the comment above, I manually added a version number) 11:30:25 (to the main page, at least) 11:36:33 advil: good idea re version number 11:36:45 I'll try to stick that in some time after t 11:37:05 also you might be interested in my reply to one of the Lua threads on tavern just now 11:37:26 you can use the add macro api features to add an arbitrary Lua coroutine as a macro, though I'm not 100% clear on how it works 11:37:53 but in principle crawl will keep calling back into your coroutine until it terminates 11:37:54 whoa do you think that coroutine thing would work? 11:38:00 yeah 11:38:51 as far as I can tell from the code when you type in a ^D macro you essentially get func () { crawl.do_command(Cmd) yield() ... } plus the delays are set up 11:39:04 I guess I don't fully understand why crawl would reenter the routine 11:39:08 the macro interrupt tables are a bit tricky but 11:39:14 though if we can get that to work it would be *great* 11:41:41 I don't have time to test now but looking at delay.cc, specifically run_macro and MacroDelay::handle() it looks like it will Just Work 11:41:53 I'm also pretty sure we have lua compiled in c mode, not c++ mode, so I'd be a bit worried that a state change like that (with setjmp/longjmp) would mess something up on the crawl/c++ side 11:44:42 I'm not even sure that compiling it in c++ mode will make coroutines safe, that may be more about error handling 11:45:23 -!- aidanh_ is now known as aidanh 11:46:08 from some googling I think we'd need something like http://coco.luajit.org/index.html for coroutines to not mess up the c++ stack 11:46:29 I'm not sure how you mean? I don't think Lua coroutines set/longjmp? 11:47:18 it seems to vary by version but they mostly are 11:47:52 https://stackoverflow.com/questions/34303507/lua-coroutines-setjmp-longjmp-clobbering has a lot of useful info 11:52:30 My vague memory from digging through the docs is that the clua api does not call back into lua 11:52:52 erm digging through to write the docs 11:54:38 I'm not sure it never does, there are a few callbacks as well as stuff like stash search and autofight 11:54:55 stash search may be using dlua 12:10:28 Unstable branch on crawl.jorgrun.rocks updated to: 0.23-a0-109-g5e4b187028 (34) 12:22:07 Stable (0.21) branch on underhound.eu updated to: 0.21.1-77-g46dbe93d77 13:01:57 Stable (0.22) branch on crawl.akrasiac.org updated to: 0.22.0-27-ga576289 13:13:09 -!- amalloy_ is now known as amalloy 13:18:57 Unstable branch on crawl.akrasiac.org updated to: 0.23-a0-109-g5e4b187 (34) 15:40:27 -!- The topic of ##crawl-dev is: Crawl Development | Logs: http://s-z.org/crawl-dev/ | People with +v have commit access. | Please keep general Crawl-related chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to a pastebin service, please. 17:33:01 =nemelex GhEn 17:33:04 Defined keyword: nchoice => VpFE|BaEn|CeIE|DDCj|HaAE|KoVM|CeEE|BaFE|DrAM|TeWr|SpSu|DEAK|BaNe|DgEn|MuWr|FoTM|GnAK|MfEE|HuAE|DEHu|GnCK|VSSu|HaCj|OpAM|OgVM|FoWz|HOEn|SpIE|GhEn 18:22:06 Unstable branch on underhound.eu updated to: 0.23-a0-109-g5e4b187028 (34) 18:27:44 Whitespace on some info screens is trimmed so columns don't line up/there is inadequate spacing. 13https://crawl.develz.org/mantis/view.php?id=11632 by NormalPerson7 19:05:26 -!- bh_ is now known as bh 19:56:14 aidanh: I just counted lines between #if TAG_MAJOR_VERSION == 34 and the next #endif. I will look into stats about recent commits 20:00:42 Out of the 970 master commits since Jan 5 (0.21.0 tag date), 19 modified a #if TAG_MAJOR_VERSION == 34 line, 69 had it in diff context (so, vaguely potentially adding save compat to an existing block) 20:01:12 well, that's number of matching lines, not matching commits actually, so numbers might be a little high 20:07:57 -!- amalloy is now known as amalloy_ 20:39:22 that's not measuring the pain when someone (lately me mostly, I think) has to edit a heavily save compat'd part of some file 22:27:07 =nemelex DESk 22:27:10 Defined keyword: nchoice => VpFE|BaEn|CeIE|DDCj|HaAE|KoVM|CeEE|BaFE|DrAM|TeWr|SpSu|DEAK|BaNe|DgEn|MuWr|FoTM|GnAK|MfEE|HuAE|DEHu|GnCK|VSSu|HaCj|OpAM|OgVM|FoWz|HOEn|SpIE|GhEn|DESk 23:05:02 -!- amalloy_ is now known as amalloy 23:06:02 alexjurkiewicz: well, I don't have any objections to cleaning up save compat, but it'd be nice to have some way to upgrade old games; I don't think a direct break is the way to go 23:06:35 not sure how others feel about my support-previous-version-only idea 23:22:26 as in break compat from 0.21 and earlier? 23:23:02 that'd be fairly drastic but I guess it could help the codebase a lot 23:23:14 one thing is that we have several servers without dgl 23:23:30 so those servers need a reasonable setup 23:23:53 I guess if nothing else they could just wipe games 23:45:15 is it possible to query gretell to give a unique a wand and see the damage? 23:46:14 @??ancient lich hd:9 23:46:14 ancient lich (00L) | Spd: 10 | HD: 9 | HP: 30-40 | AC/EV: 20/10 | Dam: 2013(drain) | 07undead, 10doors, evil, spellcaster, see invisible, unbreathing | Res: 13magic(immune), 05fire, 02cold++, 10elec, 09poison+++, 12drown, 04rot+++, 13neg+++, 13torm | Vul: 08holy | XP: 861 | Sp: b.cold (3d17), force lance (3d13), sum.greater demon, slow, invisibility / b.corrosive (3d15), crystal spear (3d2.. 23:46:41 or maybe is wand of acid the same as corrosive bolt 23:46:46 I think if you drop TAG_MAJOR_VERSION == 34 stuff, there will still be support for 0.21.0 23:47:15 when was the last minor version? 23:47:35 that's not the main issue, exactly, the issue is what happens to those trunk games 23:47:38 in terms of server admin 23:47:51 a lot of broken games for some of the non-dgl servers 23:47:55 for dgl servers, it will just keep you playing on the old version, right? 23:48:06 which servers are non-dgl? I only know of CPO 23:48:12 yes, for dgl ones it will (hopefully) just not attempt transfer 23:48:39 but lld, cwz might have some problems 23:48:40 I think for DGL, it asks the user to transfer, they'll say yes out of habit, it will fail, and then give them the option to roll back to the previous version 23:49:02 well dgl servers aren't the problem, it's those non-dgl ones 23:49:16 at least that's one of the issues of just breaking games 23:50:09 You could announce a flah day on those servers "on Aug 8, Crawl will update and old trunk games will be unplayable. Finish your important games before then!" 23:50:15 flag 23:50:37 <|amethyst> well, at least on the non-DGL servers 23:51:03 well server admin can announce it, yeah, but the main issue is how do those admins want to handle it 23:51:26 <|amethyst> oh, alexjurkiewicz even said "on those servers" 23:51:28 we can tell hong/dplusplus that they'll either have to delete those saves automatically for trunk N-2 games or have something to prevent updating those games 23:51:29 <|amethyst> ignore me :P 23:51:48 I wouldn't be surprised if they just elect to delete them 23:52:20 but they have their own setup and also merge in their own local patches for localization and webtiles customization, so I'm not sure what they'd want 23:52:28 <|amethyst> could put together a script using crawl -save-version 23:52:40 <|amethyst> to automatically determine which saves should be deleted 23:52:47 <|amethyst> (or moved to backup or whatever) 23:53:18 when crawl starts, prompt the user "can't load save, delete it? y/N" 23:53:28 <|amethyst> it might already do that? 23:53:55 <|amethyst> hm, or maybe it doesn't ask to delete 23:55:26 <|amethyst> yeah, not sure if it's different if DGL is defined, but without it you get " This game comes from an incompatible version of Crawl (0.11-a0-2905-g7a5bf1d). 23:55:29 <|amethyst> Unless you reinstall that version, you can't load it. 23:55:32 <|amethyst> Do you want to DELETE that game and start a new one? 23:55:34 <|amethyst> " 23:58:17 I guess admins don't even have to delete old saves then, unless they want to save space 23:58:52 if it's hard to coordinate with CWZ and LLD, you can even add code to Crawl which will prompt users on those servers about the upcoming flag day 23:59:46 I think any notion of coordination might only be the first time we start doing this