02:47:43 New branch created: pull/1969 (1 commit) 13https://github.com/crawl/crawl/pull/1969 02:47:43 03nikheizen02 07https://github.com/crawl/crawl/pull/1969 * 0.27-a0-1201-geab98b5: Rework and Reflavour: Gnoll Shaman -> Bouda 10(32 hours ago, 42 files, 270+ 70-) 13https://github.com/crawl/crawl/commit/eab98b5512cb 09:52:16 -!- advil changed the topic of #crawl-dev to: Crawl Development, moving from freenode | Logs: http://s-z.org/crawl-dev/ | People with +v have commit access. | Please keep general Crawl-related chat to #crawl. | Long stuff to a pastebin service, please. 10:04:03 -!- advil changed the topic of #crawl to: dungeon crawl / migrating from freenode `##crawl` (note ## -> #!) / watch this space / bot moves still in progress 10:10:55 lol 10:11:04 I guess chei is registered under |amethyst too? 10:11:09 not sure how to unvoice chei 10:11:17 if you're autovoice 10:52:20 03PleasingFungus02 07* 0.27-a0-1201-gcae1a64: Improve Mindburst lore (dinky) 10(2 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/cae1a64fa401 10:56:15 03PleasingFungus02 07* 0.27-a0-1202-ge72e491: Fix mindburst/2h ogre lore 10(39 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/e72e49191cae 11:57:58 does sizzell support sasl sign in? 11:58:52 hm maybe not? 12:02:42 it seems my vps provider belongs to an ip range required to authenticate over sasl 12:10:07 hmm 12:10:15 yeah it seems that BasicBot doesn't support sasl 12:10:32 is this a DigitalOcean IP? 12:12:54 yes 12:14:38 one idea might be to run a bouncer and route it through that, not sure if there's a simpler alternative 12:17:05 |amethyst if/when you're around any ideas? I see that Cheibriados is connected over SASL 12:32:19 my full name is badteeth23, a dynamic tooth counter, but i left shmup in the dust awhile ago advil 12:32:25 just a relic in some chans 12:46:49 -!- shhpicyCat is now known as crankyCat 12:50:09 floraline: it looks to me like BasicBot *should* support SSL if POE::Component::SSLify is installed and you pass `ssl => 1,` to the `new` call, that seems to be all that Cheibriados is doing 12:50:16 it's not as if I've tested this though 12:55:16 and cao of course does not have the right perl modules 13:06:25 Unstable branch on crawl.akrasiac.org updated to: 0.27-a0-1202-ge72e491 (34) 13:10:30 advil: it looks like i can now connect with ssl, but it stills kicks me out with the message 'SASL access only' 13:19:23 New branch created: pull/1971 (1 commit) 13https://github.com/crawl/crawl/pull/1971 13:19:23 03hellmonk02 07https://github.com/crawl/crawl/pull/1971 * 0.27-a0-1201-g74d5156: New Zot:5 shop vault 10(12 minutes ago, 1 file, 18+ 0-) 13https://github.com/crawl/crawl/commit/74d5156d8b5a 13:23:25 hmm yeah they aren't really the same, are they 13:29:44 looks like it's a no go: https://rt.cpan.org/Public/Bug/Display.html?id=124292 13:36:16 back to my other idea which would be running a bouncer that supports sasl and sending the bot through that; I guess there's a real question as to how much effort it is worth though 13:48:16 I wonder if this will be an issue for Sequell, I can't see any evidence that it implements SASL and it is still BasicBot at its core 14:36:00 Two randarts with the same name 13https://crawl.develz.org/mantis/view.php?id=12594 by ogaz 15:18:44 <|amethyst> hmm... just using port => 6697, ssl => 1 15:19:00 <|amethyst> not sasl 16:11:19 yeah I was showing my ignorance about the difference between those two things 16:11:37 not all IP addresses will require SASL, in fact most won't 16:37:25 yeah, I had to implement sasl myself for cerebot/beem/lomlobot because using AWS ip 16:37:33 it was not fun implementing sasl myself 16:38:09 want to move those over to an asynchronous irc lib, but none at the time seemed to have working sasl support and I was less confident in modifying those 16:39:12 oh I have voice now 16:53:00 I keep seeing things that make me think "SASL PLAIN looks easy" and then I realize that entails working on an IRC bot in perl again 16:53:43 but it definitely seems like libera has expanded the IP ranges covered by that 16:55:33 -!- kate-- is now known as kate- 17:09:43 Lethal poison message not properly accounting for Lignification 13https://crawl.develz.org/mantis/view.php?id=12595 by ogaz 17:10:53 also I keep feeling like if I just do the right google search I should be able to find pastable code that does this, but I'm starting to wonder if we are the only old people still using POE::Component::IRC 17:13:19 "IRC uses PLAIN, which is just username + "\0" + password encoded as base64. To implement SASL, you have to send CAP LS before the initial USER command, and then a CAP REQ :sasl when you receive CAP * LS :... sasl ... line, after that you send AUTHENTICATE base64(username+\0+password) and finally a CAP END." 17:13:35 is what makes me think it could be very simple to implement (from https://stackoverflow.com/questions/48816235/no-ident-response-and-notice-you-need-to-identify-via-sasl-to-use-this-ser) 17:33:36 -!- allbery_b is now known as geekosaur 17:41:35 advil: I actually had to do: 17:41:36 authdata = base64.b64encode(f'{self.username}\x00{self.username}' 17:41:37 f'\x00{self.password}'.encode()) 17:42:03 as I recall, there were some subtle differences between what you see described in the standard and what Freenode expected 17:42:21 which I only found out by looking at irssi+SASL over non-ssl and using tcpdump 17:43:08 also my USER command: 17:43:08 self.send_raw(f"USER {username} {username} {self.server} :{realname}") 17:43:30 tbf I haven't actually tested the bots on Libera, but I am using irssi just fine with sasl and libera 17:44:17 can confirm about the sasl capability request 17:49:14 looks like the only difference between what you describe and what I'm currently doing is I'm just looking for a CAP event with an 'ACK' argument instead of a specific capability response, before sending AUTHENTICATE 17:49:44 that might just be a cheap hack compared to inspecting the response to confirm sasl is available, though 17:49:48 I think if the server is guaranteed to have the capability there's no need to check 17:50:01 yeah 17:50:24 if only I'd implemented this instead of all the join after auth stuff I did, it would've solved the same problem 17:50:40 and still worked now 17:51:56 apparently I expect an AUTHENTICATE event with a target of '+' before sending the AUTHENTICATE credentials 17:52:39 apparently I also need a 900 irc event before sending CAP END 17:53:21 fun 17:53:26 %git a8e22f6d9 17:53:26 07kate-02 * 0.17-a0-495-ga8e22f6: Remove the Tukima and Eringya wizlabs 10(6 years ago, 4 files, 11+ 316-) 13https://github.com/crawl/crawl/commit/a8e22f6d9ab1 19:42:29 -!- allbery_b is now known as geekosaur