02:06:11 <Cheibriados> New branch created: remove-old-monsters (3 commits) 13https://github.com/crawl/crawl/tree/remove-old-monsters
02:06:11 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-856-g09762d5: Re-sync axed / removed monsters 10(2 days ago, 2 files, 9+ 0-) 13https://github.com/crawl/crawl/commit/09762d521b79
02:06:11 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-857-g12243b5: Remove AXED_MON macro 10(2 days ago, 3 files, 23+ 112-) 13https://github.com/crawl/crawl/commit/12243b5fb0c8
02:06:11 <Cheibriados> 03Aidan Holm02 07[remove-old-monsters] * 0.25-a0-858-g7572d42: Remove dead monster code 10(2 days ago, 8 files, 4+ 43-) 13https://github.com/crawl/crawl/commit/7572d424ab7a
02:11:03 <gammafunk> hrm, wonder how you go about determining what should be tested and what shouldn't
02:11:59 <aidanh> for unit tests, the general philosophy is that you test externally visible behaviour, not internal implementation details
02:12:25 <aidanh> but i guess that's more 'how' than 'what'
02:12:55 <aidanh> i added that test case to verify that removing all that AXED_MON stuff made no difference
02:14:19 <gammafunk> good to know
02:14:23 <aidanh> although, now that I think about it, removed monsters will probably show up in char dumps, same as spells
02:15:05 <aidanh> so in this case I will probably revert that, and instead add a test case asserting that you can get the name for a removed monster (which is what the char dump code needs)
02:17:29 <aidanh> another example: that code adding support for minor versions > 255 *definitely* needs some test cases
02:31:20 <gammafunk> heh, yeah
02:34:48 <gammafunk> totally unrelated note, wow, C++20 has support for coroutines
02:39:13 <aidanh> some fine day in 2050, we will be able to use them
02:40:41 <aidanh> though if we can just upgrade cao and cbro, we can vendor abseil, which backports c++17 stuff
02:42:55 <Cheibriados> 03Aidan Holm02 07[wide-minor-tag] * 0.25-a0-856-gf876b0d: Add support for minor tags > 255 10(17 hours ago, 9 files, 147+ 43-) 13https://github.com/crawl/crawl/commit/f876b0dd9c35
02:45:06 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.25-a0-651-g8c222c7a2b
02:47:41 <aidanh> i've created a PR for the wide minors change; would appreciate it if others would take a quick glance
02:48:26 <Cheibriados> New branch created: pull/1375 (1 commit) 13https://github.com/crawl/crawl/pull/1375
02:48:26 <Cheibriados> Branch pull/1375 updated to be equal with wide-minor-tag: 13https://github.com/crawl/crawl/pull/1375
02:49:07 <Pinkbeast> I'll take a look, it is pertinent to our interests :-/
02:49:31 <aidanh> Pinkbeast: great, thank you!
03:16:21 <gammafunk> interesting that your use of UINT8_MAX is apparently the first outside of code to support MSVC
03:17:03 <gammafunk> we just like writing those constants out by hand I guess!
03:18:29 <aidanh> heh, yeah i was debating about that
03:18:43 <aidanh> !source is_ancient
03:18:44 <Sequell> Can't find is_ancient.
03:18:56 <aidanh> !source tag-version.h:337
03:18:56 <Sequell> https://github.com/crawl/crawl/blob/master/crawl-ref/source/tag-version.h#L337
03:19:34 <aidanh> looks like crawl apparently supports loading saves from major version 33?
03:20:01 <gammafunk> interesting
03:21:19 <aidanh> %git ee7bbb8
03:21:19 <Cheibriados> 07advil02 * 0.23-a0-154-gee7bbb8: Improve the new save version api a bit and use it more 10(1 year, 8 months ago, 2 files, 44+ 31-) 13https://github.com/crawl/crawl/commit/ee7bbb803370
03:24:37 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.25-a0-857-g60a0bcd (34)
03:25:28 <aidanh> %git 35673f84
03:25:28 <Cheibriados> 07kilobyte02 * 0.12-a0-1828-g35673f8: Restore save compatibility with 0.11 final. 10(7 years ago, 9 files, 242+ 26-) 13https://github.com/crawl/crawl/commit/35673f843b19
03:25:48 <aidanh> i see, so presumably 17 is the last minor for major 33
03:25:56 <aidanh> i reckon that can probably be dropped though
03:55:36 <Kelbell> Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2605-g194d903447
04:42:54 <Rotatell> ziggurnaut (L4 GrGl) Crash caused by signal #15: Terminated (Sewer)
05:36:29 <Lantell> Napalmbreeze (L10 DsGl) Crash caused by signal #15: Terminated (D:9)
08:07:33 <advil> aidanh have you seen https://github.com/crawl/crawl/blob/master/crawl-ref/docs/develop/history/major-tag-history.md?
08:26:25 <advil> ahh we are using both a mix of lua-allocated and c++ allocated userdata, that's why some of this code is so weird
09:53:17 <Cheibriados> 03advil02 07[lua-gc-fix] * 0.25-a0-858-gc7df3f3: Handle another potential multiple delete case 10(48 minutes ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/c7df3f38a31f
09:53:17 <Cheibriados> 03advil02 07[lua-gc-fix] * 0.25-a0-859-g30a6dbd: Pointer checks for lua userdata objects 10(3 minutes ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/30a6dbd1211e
09:58:10 <advil> it kind of looks to me like all these luaL_checkudata should also be checking for a direct nullptr return instead of immediately *ing that pointer? Though I'm not sure it would be possible to trigger that error case from lua
10:20:51 <advil> it sure would be nice to wrap all userdata in a single templated class along the lines of item_wrapper, and then not have all these one-off macros
10:20:51 <advil> maybe I will resist that for the immediate fix though
10:35:29 <Gretell> Unstable branch on crawl.develz.org updated to: 0.25-a0-857-g60a0bcd8b5 (34)
10:44:51 <Gretell> Stable branch on crawl.develz.org updated to: 0.23.3-0-ge53cab806c (34)
10:47:53 <Cheibriados> New branch created: pull/1376 (3 commits) 13https://github.com/crawl/crawl/pull/1376
10:47:53 <Cheibriados> Branch pull/1376 updated to be equal with lua-gc-fix: 13https://github.com/crawl/crawl/pull/1376
11:16:25 <Gretell> Stable branch on crawl.develz.org updated to: 0.22.3-0-g6000bec5d9 (34)
11:26:21 <Gretell> Stable branch on crawl.develz.org updated to: 0.21.3-0-ge33f2bf4e8 (34)
11:29:09 <advil> oh heh I bet napkin read that security advisory I hit publish on
11:53:54 <Rotatell> ziggurnaut (L7 GrFi) Crash caused by signal #15: Terminated (D:4)
12:17:03 <aidanh> oh uh, did that advisory have the correct cve?
12:18:12 <aidanh> ok, it does, nvm
12:19:03 <advil> I added it
12:19:14 <advil> its formatting is messed up by the version list
12:22:18 <aidanh> the archaeology doc was a good read
12:23:15 <aidanh> i especially enjoyed the analysis section
12:23:52 <advil> thanks
12:30:35 <aidanh> advil: btw that branch is totally public
12:30:57 <advil> the one with the fixes? yeah that's intentional
12:31:34 <advil> or at least, it didn't seem harmful to me to have it be public
12:31:51 <Pinkbeast> Should we be cherry-picking some of this?
12:32:06 <advil> once it is merged, I'm hoping someone else will look at it too
12:32:31 <aidanh> well, a RCE was mentioned earlier in the chat, and the reporter's name is explicitly mentioned in the commit title
12:33:07 <advil> it would take a fair amount of skill + trial and error to get it to work in practice on a server from what is in that branch
12:33:28 <advil> should I be being more paranoid?
12:35:09 <aidanh> personally, i would err on the side of being a bit more paranoid, although i do think the stakes are not especially high in this scenario
12:35:39 <advil> well, if you look at the PR and tell me it is ok then I will merge it and we'll be good ;-)
12:35:50 <aidanh> tbf i haven't looked at this code in detail, and i'm not good at evaluating RCEs
12:35:59 <aidanh> anything to get your PR looked at, right :)
12:36:02 <advil> haha
12:44:32 <aidanh> looks good to me
12:45:30 <aidanh> well, i don't like the macro gunk in cluautil.h, but your 'TODO: ugh' comment already summed up how i feel there :)
12:45:40 <Gretell> Stable branch on crawl.develz.org updated to: 0.24.1-5-g77f818bfca (34)
12:45:46 <advil> classic crawlcomment
12:46:01 <advil> I do think this could be done much more nicely with a single templated wrapper class
12:46:50 <aidanh> add it to the crawl tech debt todo list
12:46:53 <advil> that handled all these validity checks in one place
12:46:54 <advil> yeah
12:49:03 <gammafunk> does a lot of that clua logic predate stone soup? was looking through the git log of that file and it wasn't easy to tell
12:49:15 <Cheibriados> 03advil02 07* 0.25-a0-858-g7f8d8a8: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/7f8d8a8652f7
12:49:15 <Cheibriados> 03advil02 07* 0.25-a0-859-g39757dd: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/39757dd64f1e
12:49:15 <Cheibriados> 03advil02 07* 0.25-a0-860-gf695395: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/f695395b5c0b
12:49:35 <advil> I tracked down the commits yesterday and most of it was added around 0.6 or so
12:49:35 <gammafunk> a lot of the lua infrastructure in general, I guess I should ask, since clua and dlua interfact stuff seems made similarly
12:49:41 <gammafunk> oh ok
12:49:48 <advil> though it was based on some earlier code
12:49:56 <advil> the 0.6 stuff generalized things a bit
12:50:25 <aidanh> how does everyone feel about dropping support for minor versions from before 2017?
12:50:51 <aidanh> that would give ~3 year save compat, and let us get rid of all but ~40 minor version tags
12:51:40 <Cheibriados> 03advil02 07[stone_soup-0.24] * 0.24.1-6-g7419f40: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/7419f40c810d
12:51:40 <Cheibriados> 03advil02 07[stone_soup-0.24] * 0.24.1-7-ge4f3edd: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/e4f3edded6fa
12:51:40 <Cheibriados> 03advil02 07[stone_soup-0.24] * 0.24.1-8-ga250c9d: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/a250c9d538d3
12:51:57 <aidanh> personally i feel that 3 years probably strikes the right balance, but i could be convinced either way
12:52:11 <Cheibriados> 03advil02 07[stone_soup-0.23] * 0.23.3-1-gde67f2c: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/de67f2c63e95
12:52:11 <Cheibriados> 03advil02 07[stone_soup-0.23] * 0.23.3-2-g41485c7: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/41485c70a80b
12:52:11 <Cheibriados> 03advil02 07[stone_soup-0.23] * 0.23.3-3-gaaff5ab: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/aaff5abdbfdb
12:53:08 <gammafunk> aidanh: what happens on a dgl setup for a game older than that cutoff?
12:53:35 <gammafunk> the prompting script doesn't have logic to check that a game can transfer, does it?
12:53:58 <aidanh> does it have any logic for major versions?
12:54:18 <Cheibriados> 03advil02 07[stone_soup-0.22] * 0.22.3-1-g560fcc9: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/560fcc9ae5f8
12:54:18 <Cheibriados> 03advil02 07[stone_soup-0.22] * 0.22.3-2-ga208a78: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/a208a78027f9
12:54:18 <Cheibriados> 03advil02 07[stone_soup-0.22] * 0.22.3-3-g34f245f: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/34f245f15755
12:54:35 <gammafunk> I want to say that it looks at that, but I'd have to check
12:55:15 <aidanh> if i understand correctly, dgl save transferring is just moving a save file around, right?
12:55:42 <advil> I think it will give an error if it's an incompatible version, but the behavior didn't look very streamlined and I didn't test carefully
12:56:04 <Cheibriados> 03advil02 07[stone_soup-0.21] * 0.21.3-1-g3172b64: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/3172b642cfb1
12:56:04 <Cheibriados> 03advil02 07[stone_soup-0.21] * 0.21.3-2-ga4d52fb: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/a4d52fb68638
12:56:04 <Cheibriados> 03advil02 07[stone_soup-0.21] * 0.21.3-3-g218456f: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/218456f21eee
12:56:21 <aidanh> so either you need to duplicate the compat logic outside the crawl binary, you use the crawl binary to detect compatibility, or you ignore the issue since it probably won't come up in practice
12:57:01 <Cheibriados> 03advil02 07[stone_soup-0.20] * 0.20.2-1-g4293f6b: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/4293f6ba0934
12:57:01 <Cheibriados> 03advil02 07[stone_soup-0.20] * 0.20.2-2-gab437c0: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/ab437c0e8b89
12:57:01 <Cheibriados> 03advil02 07[stone_soup-0.20] * 0.20.2-3-g29329e0: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/29329e05615b
12:57:33 <Cheibriados> 03advil02 07[stone_soup-0.18] * 0.18.2-1-g5919b2c: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/5919b2ceb7f6
12:57:33 <Cheibriados> 03advil02 07[stone_soup-0.18] * 0.18.2-2-ge543f87: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/e543f87803b6
12:57:33 <Cheibriados> 03advil02 07[stone_soup-0.18] * 0.18.2-3-g5ac21b8: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/5ac21b844394
12:57:33 <Cheibriados> 03advil02 07[stone_soup-0.19] * 0.19.6-1-g164314f: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/164314f6f29d
12:57:33 <Cheibriados> 03advil02 07[stone_soup-0.19] * 0.19.6-2-g76e50e7: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/76e50e748e15
12:57:33 <Cheibriados> 03advil02 07[stone_soup-0.19] * 0.19.6-3-g7d5932f: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 125+ 58-) 13https://github.com/crawl/crawl/commit/7d5932f76a58
12:57:50 <advil> sry
12:57:50 <advil> just trying to get my commit count up
12:57:50 <aidanh> heh
13:00:25 <advil> having to do a bit of manual merging as I go back further for this one
13:01:50 <Cheibriados> 03advil02 07[stone_soup-0.17] * 0.17.3-1-g905e30a: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/905e30a4a77f
13:01:50 <Cheibriados> 03advil02 07[stone_soup-0.17] * 0.17.3-2-g40c55e1: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/40c55e11f6bf
13:01:50 <Cheibriados> 03advil02 07[stone_soup-0.17] * 0.17.3-3-gb604f17: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 128+ 57-) 13https://github.com/crawl/crawl/commit/b604f171e5c8
13:02:21 <Cheibriados> 03advil02 07[stone_soup-0.16] * 0.16.3-1-g0378e38: Guard against repeated calls to lua_object_gc (ruderubik) 10(14 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/0378e38ea7ce
13:02:21 <Cheibriados> 03advil02 07[stone_soup-0.16] * 0.16.3-2-g6c48498: Handle another potential multiple delete case 10(4 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/6c48498a68bf
13:02:21 <Cheibriados> 03advil02 07[stone_soup-0.16] * 0.16.3-3-g345d91b: Pointer checks for lua userdata objects 10(3 hours ago, 9 files, 128+ 57-) 13https://github.com/crawl/crawl/commit/345d91bb4e6b
13:03:08 <advil> that might be about as far as I go
13:04:23 <aidanh> it might also be worth explicitly setting some sort of bugfix timeline for releases, especially security ones
13:05:05 <Henzell> Stable (0.24) branch on crawl.akrasiac.org updated to: 0.24.1-8-ga250c9d
13:05:11 <advil> in the security policy thing I wrote last night, I said we only support current stable and trunk
13:05:20 <advil> the practical problem is that some servers go way back
13:05:24 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.25-a0-860-gf695395b5c (34)
13:05:33 <aidanh> i must have missed that
13:05:36 <advil> Pinkbeast *now* I would recommend cherry-picking
13:06:19 <aidanh> well, clearly the servers are more in control of compatibility than we are
13:06:29 <aidanh> but i think it'd be good to reach some sort of compromise
13:06:41 <gammafunk> aidanh: we definitely do have users try to transfer old (3 years+) games in practice online; obviously doesn't happen very often
13:06:50 <advil> I'm kind of thinking of disabling custom rc files on cao before 0.16?
13:06:55 <advil> if I can figure out a good way to do that
13:07:08 <advil> I don't think I can just disable lua without a rebuild of some kind
13:07:13 <aidanh> in the sense of asking server ops to disable versions older than ~3 years
13:07:26 <gammafunk> hrm, we don't bind a lua prompt by default do we
13:07:35 <gammafunk> so if we did that there'd for sure be no way to get a lua prompt?
13:07:36 <advil> no
13:08:05 <advil> hm, it might be bindable
13:08:05 <advil> via ~ somehow
13:08:28 <gammafunk> this is all assuming all servers actually rebuild any changes on those old branches, of course
13:08:47 <advil> yeah, some don't do it automatically
13:09:02 <advil> I think I could disable rc files by changing how crawl is started though
13:09:10 <advil> basically just point it to a dummy rc file instead of the user's
13:09:37 <advil> I *would* like to remove some of the old versions from CAO, but I'd also like to announce that a bit in advance
13:09:56 <aidanh> gammafunk: i guess if you limit the online versions to those with forward compatibility, you can force-upgrade saves as old versions are removed; or just leave it as an occasional admin task
13:09:56 <aidanh> advil: btw speaking of cao, were there remaining issues i needed to sort out with the chroot stuff?
13:10:24 <advil> well, it's somewhere in the queue of server admin stuff I need to get to
13:10:44 <aidanh> i don't mean to bug you about this, we're all busy of course, but i'd really like to get abseil in
13:10:56 <advil> step 1 was understanding dgamelaunch-config better so I knew what the scripts were even doing ;-)
13:11:12 <gammafunk> it's a google conspiracy to get more of their code in our codebase!
13:11:17 <aidanh> yeah, those are pretty mindbending
13:11:28 <advil> I still think there may be a huge problem in dealing with the 1T+ of data that is currently in the cao chroot
13:11:41 <advil> so right now I do have "figure out how to offload ttyrecs" higher in my queue
13:11:51 <aidanh> makes sense
13:12:15 <aidanh> part of the meta reason i'd like to integrate abseil is that it only has a five-year support window
13:12:57 <aidanh> i.e. if crawl starts to use it, it will effectively force us to have up-to-date server deployments
13:13:32 <aidanh> gammafunk: could be.. :)
13:19:31 <gammafunk> ??commits
13:19:31 <Sequell> commits[1/2]: http://s-z.org/neil/git/?p=crawl.git;a=shortlog
13:19:41 <gammafunk> ??commits[2
13:19:41 <Sequell> commits[2/2]: http://crawl.develz.org/trunk/changes.txt
13:20:50 <advil> aidanh re the updating code, you can see it here: https://github.com/crawl/dgamelaunch-config/blob/master/chroot/bin/crawl-git-launcher.sh#L140
13:20:59 <advil> (or a typical version)
13:21:26 <advil> it checks the major version number...I think it is not checking it online but major save version is stored in the versions db on build
13:21:31 <aidanh> typical, as in, differing by branch?
13:21:43 <aidanh> er, by server, loosely corresponding to branch
13:21:57 <advil> I'm just not sure when that code might have last changed
13:22:02 <advil> not majorly in a while
13:22:17 <advil> (and cpo probably does something else entirely)
13:22:26 <gammafunk> cpo just force transfers, yeah
13:22:40 <gammafunk> alex might have rolled his own script for that
13:23:03 <aidanh> this just looks like it mv's the file
13:23:38 <advil> you want to look a bit lower for the version logic
13:23:44 <aidanh> oh, nvm, rightt
13:24:33 <advil> yeah, the version stuff is all from sql calls above
13:25:45 <aidanh> i still think the 'proper' way to do this is to check with the binary itself (i.e. something like `crawl --check-compat ./save.cs`)
13:26:01 <aidanh> but maybe i'm just thinking that because servers are so difficult to change that anything else is virtually impossible
13:27:03 <aidanh> in any case, if this script updates a save that's not compatible, due to coming from a too-old minor version, the end result is that the player will get an error message
13:27:26 <aidanh> i.e. no data loss or anything like that
13:27:43 <advil> yes, but I think in that scenario it might take manual intervention to make the save playable again
13:28:14 <advil> I believe they'll get prompted to upgrade the save, and if they hit yes, it'll move it to a new directory where it will never be able to run
13:28:39 <gammafunk> do we need commit updates being sent to CRD list?
13:28:55 <aidanh> 'never be able to run' == when they try to play, they'll get a save incompatibility error?
13:29:00 <advil> yeah
13:29:02 <advil> I think so
13:29:07 <advil> this is speculation though, I haven't tried it
13:29:52 <aidanh> sure, that sounds like exactly what'd happen, since we're discussing adding a new version restriction
13:29:52 <advil> oh heh https://sourceforge.net/p/crawl-ref/mailman/crawl-ref-commits/ still exists
13:30:01 <aidanh> so the default behaviour (moving to the latest trunk folder) will (should?) still be in play
13:30:22 <gammafunk> yeah, that must be what Napkin is asking about
13:30:28 <advil> gammafunk I bet it would make dpeg (wherever he is) sad if you stop that
13:30:39 <gammafunk> right, he's leaving it in place for now
13:31:52 <advil> can't figure out a way to tell if that list has any subscribers
13:32:42 <gammafunk> we figured out that git can fetch fine from github over git://
13:32:54 <gammafunk> so we'll use that as a solution on CDO for getting tourney and website
13:33:16 <advil> was cdo having some issues today?
13:33:24 <gammafunk> and 0.24 is installed on CDO and builds are working again
13:33:48 <aidanh> btw, is there any reason, apart from the usual accretion of tech debt, why online players can only play one game per version?
13:34:14 <gammafunk> aidanh: I think even locally, you can only have one game per username
13:34:34 <aidanh> true, but you get to pick your username there
13:35:19 <Tuxedo[Qyou]> hi guys, any idea on how to request a fork to be hosted on a server?
13:35:19 <Tuxedo[Qyou]> do you know who's the leader of x-crawl.de?
13:35:57 <aidanh> er, rather, when webtiles / dgl starts a game, the crawl binary is given the path to the save file, so duplicate names shouldn't matter
13:36:44 <aidanh> Tuxedo[Qyou]: it's not an official server
13:37:06 <Tuxedo[Qyou]> yeah, but do you know who's the leader?
13:37:18 <Tuxedo[Qyou]> can it become an official server like cpo?
13:37:53 <aidanh> their homepage states that the server is maintained by krfreak
13:38:06 <Tuxedo[Qyou]> alright, ty
13:38:27 <aidanh> i think that's unlikely, due to 1) philosophical differences between devs, and 2) the huge tech debt problem we already have with the existing servers
13:38:30 <aidanh> np
13:39:00 <gammafunk> x-crawl.de isn't an official server and wouldn't become one, yeah
13:39:39 <aidanh> apart from the philosophical / interpersonal issues, i personally would be strongly against supporting another server in any way until we have our server tech debt more under control
13:42:55 <gammafunk> yeah, that's sort of how we've operated, with the one caveat that if an existing server goes offline and a replacement is available, we support it
13:43:17 <gammafunk> because for a long time it wasn't clear that we'd make much headway aobut said tech debt
13:43:27 <gammafunk> s/aobut/about/
13:47:08 <aidanh> well, as the author of a probably-not-even-half-complete and still-pretty-buggy refactoring of crawl's ui code, i do think headway is possible
13:48:19 <aidanh> i mean, look at our test coverage! we went from 0% to 3% over the last few months
13:48:22 <aidanh> now that's real progress
13:50:28 <aidanh> although it occurs to me that i'm probably more ardent about this than most, because the ui subsystem was (is) a part of crawl particularly susceptible to layers upon layers of hacks
13:56:38 <gammafunk> aidanh: and yeah, I agree about changes, but I think if we do lose, say, a US server, we kind of do want to replace it quicky if one becomes available
13:57:07 <gammafunk> since that helps the other servers not get so overloaded
13:57:32 <aidanh> that seems reasonable
14:00:01 <advil> yeah, for example if cbro were to go away, not replacing it soon would have a definite cost
14:14:27 <Henzell> Stable (0.19) branch on crawl.akrasiac.org updated to: 0.19.6-3-g7d5932f
14:17:19 <Gretell> Monster database of master branch on crawl.develz.org updated to: $(VERSION)
14:17:37 <gammafunk> oops
14:24:56 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.25-a0-861-g0cf3fa53f1
14:26:09 -!- Tiobot is now known as Guest21591
14:32:03 <Cheibriados> 03advil02 07[stone_soup-0.15] * 0.15.3-1-gb09fd63: Guard against repeated calls to lua_object_gc (ruderubik) 10(16 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/b09fd6357c16
14:32:03 <Cheibriados> 03advil02 07[stone_soup-0.15] * 0.15.3-2-ge506468: Handle another potential multiple delete case 10(5 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/e50646807ff2
14:32:34 <Cheibriados> 03advil02 07[stone_soup-0.14] * 0.14.3-1-ga692938: Guard against repeated calls to lua_object_gc (ruderubik) 10(16 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/a6929384fac8
14:32:34 <Cheibriados> 03advil02 07[stone_soup-0.14] * 0.14.3-2-g4a9013c: Handle another potential multiple delete case 10(5 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/4a9013c4ae9e
14:33:06 <Cheibriados> 03advil02 07[stone_soup-0.12] * 0.12.4-1-g931bde3: Guard against repeated calls to lua_object_gc (ruderubik) 10(16 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/931bde3d45e3
14:33:06 <Cheibriados> 03advil02 07[stone_soup-0.12] * 0.12.4-2-g62eed22: Handle another potential multiple delete case 10(5 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/62eed2202bb9
14:33:06 <Cheibriados> 03advil02 07[stone_soup-0.13] * 0.13.3-1-g682833f: Guard against repeated calls to lua_object_gc (ruderubik) 10(16 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/682833fe0216
14:33:06 <Cheibriados> 03advil02 07[stone_soup-0.13] * 0.13.3-2-g293ddf4: Handle another potential multiple delete case 10(5 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/293ddf48c20a
14:33:38 <Cheibriados> 03advil02 07[stone_soup-0.11] * 0.11.4-1-g92545cb: Guard against repeated calls to lua_object_gc (ruderubik) 10(16 hours ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/92545cb749b4
14:33:38 <Cheibriados> 03advil02 07[stone_soup-0.11] * 0.11.4-2-gff741ca: Handle another potential multiple delete case 10(5 hours ago, 1 file, 13+ 4-) 13https://github.com/crawl/crawl/commit/ff741cad8138
14:34:17 <advil> ok for real this time I'm not going back any further
14:35:18 <advil> that's not the full patch but it is the commits that handle the immediate problem as reported
14:37:13 <Henzell> Stable (0.18) branch on crawl.akrasiac.org updated to: 0.18.2-3-g5ac21b8
14:50:15 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.25-a0-861-g2dbde65baa
14:56:27 <advil> gammafunk as far as I can tell you can't bind an unbound command by CMD_... name without the rc file, do you think that's right?
14:58:14 <advil> another thing I can imagine doing for versions where a rebuild isn't possible is to use some simple regexes to detect lua code (probably not worrying about false positives) and refuse to use the rc file if there is any
14:58:32 <advil> but none of this makes sense if there's a way to get to the lua console I haven't thought of
15:01:20 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-860-gf695395 (34)
15:07:26 <gammafunk> advil: hrm, seems like that is the case
15:09:32 <gammafunk> had no idea that ctrl-I was an alternate for TAB
15:10:56 <gammafunk> advil: I don't think anyone would go back and enable wizard mode for these old versions for everyone, but if they did, that'd be a problem :)
15:11:17 <gammafunk> you can even pull up a dlua console from that
15:12:02 <gammafunk> heh, &` is an interesting wizmode command I never knew about
15:12:36 <gammafunk> also, according to wizard mode, the c in clua stands for "client" not "crawl"
15:20:56 <Lantell> Stable (0.20) branch on underhound.eu updated to: 0.20.2-3-g29329e0561
15:49:04 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.25-a0-861-g2dbde65baa
16:31:17 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-860-gf695395b5c
16:49:05 <alexjurkiewicz> advil: thanks for backporting the fixes to past versions (CPO cares about back to 0.15, at least theoretically)
16:50:45 <gammafunk> fyi, CDO has been updated so that the windows trunk builds are working again
16:51:50 <gammafunk> by fyi, I mean, letting everyone in channel know
17:21:04 <Cheibriados> 03advil02 07[stone_soup-0.15] * 0.15.3-3-g8e88d0d: Tweak to get 0.15 compiling again 10(42 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/8e88d0d95c33
17:35:46 <Henzell> Stable (0.15) branch on crawl.akrasiac.org updated to: 0.15.3-3-g8e88d0d
17:38:26 <Cheibriados> 03advil02 07[stone_soup-0.16] * 0.16.3-4-gc506b73: Resolve an incomplete merge... 10(68 seconds ago, 1 file, 0+ 4-) 13https://github.com/crawl/crawl/commit/c506b73efe28
17:53:30 <Henzell> Stable (0.16) branch on crawl.akrasiac.org updated to: 0.16.3-4-gc506b73
17:55:57 <Rotatell> Rem (L16 DrCj) Crash caused by signal #15: Terminated (Snake:3)
18:17:25 <Cheibriados> 03advil02 07[stone_soup-0.14] * 0.14.3-3-g02348ce: Tweak to get 0.15 compiling again 10(57 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/02348ce18852
18:24:00 <Lantell> Unstable branch on underhound.eu updated to: 0.25-a0-860-gf695395b5c (34)
18:30:41 <Henzell> Stable (0.14) branch on crawl.akrasiac.org updated to: 0.14.3-3-g02348ce
18:31:56 <Cheibriados> 03advil02 07[stone_soup-0.13] * 0.13.3-3-g4e9f7b7: Tweak to get 0.15 compiling again 10(72 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/4e9f7b769a49
18:46:13 <Henzell> Stable (0.13) branch on crawl.akrasiac.org updated to: 0.13.3-3-g4e9f7b7
18:48:06 <Cheibriados> 03advil02 07[stone_soup-0.12] * 0.12.4-3-g8567966: Tweak to get 0.15 compiling again 10(88 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/856796605cf1
19:00:46 <Henzell> Stable (0.12) branch on crawl.akrasiac.org updated to: 0.12.4-3-g8567966
19:06:17 <Cheibriados> 03advil02 07[stone_soup-0.11] * 0.11.4-3-g4668e67: Remove some modern c++ code that was cherry-picked 10(59 seconds ago, 3 files, 3+ 3-) 13https://github.com/crawl/crawl/commit/4668e67d316d
19:13:14 -!- Tuxedo[Qyou] is now known as werecarpet
19:19:03 <Henzell> Stable (0.11) branch on crawl.akrasiac.org updated to: 0.11.4-3-g4668e67
19:21:27 <Lantell> Stable (0.24) branch on underhound.eu updated to: 0.24.1-8-ga250c9d538
22:18:25 <Lantell> Stable (0.16) branch on underhound.eu updated to: 0.16.3-4-gc506b73efe