[16:40:49] hi [16:41:17] guya [16:41:21] guys [16:44:05] he [16:44:05] he [16:44:09] hey [16:44:22] !stewards [16:48:01] oi [16:48:03] oi [16:48:04] oi [16:48:48] fuck [16:48:51] r [16:48:52] rf [16:48:53] ff [16:48:54] f [16:48:54] f [16:48:55] f [16:48:55] f [16:48:56] f [16:48:57] f [16:49:09] hi [16:49:22] hi [16:49:25] im sorry [16:50:04] any way we can prevent the messages of unregistered Libera accounts from going over the bridge? [16:57:52] mahir256: We could prevent any unregistered users on libera from speaking entirely, but I've noticed a *lot* more spam in the telegram->IRC direction than the IRC->telegram direction so I'd suggest more protections on the Telegram side might be a higher priority. [17:06:09] That might be true. The difference is that each of them rarely (ever?) manage to do more than one post because of higher activity of the admins in the Telegram group. (re @wmtelegram_bot: mahir256: We could prevent any unregistered users on libera from speaking entirely, but I've noticed a *lot* more spam in ...) [17:07:24] *he said before looking which group he was in* [17:09:02] My question was targeted for this specific group, rather than for other groups which are similarly bridged (such as the hackathon group) [17:12:33] I don’t recall much Telegram spam in this group specifically, but the moderation log only goes back until yesterday so I can’t say for certain [17:29:11] The matterbridge software that powers the bridge does not have any configuration options for this, so no. (re @mahir256: any way we can prevent the messages of unregistered Libera accounts from going over the bridge?) [21:46:29] In an attempt to align the code in Ninai/Udiron with this information, and in the hope that this will ease the process of migrating the code onto the wiki, most of it has now been adjusted so that (some notion of) persistency applies to both syntax trees and constructors. With respect to Udiron, this transition is mostly complete; there are some aspects of Ninai, however, that could be better aligned as yet. (re @wmtelegr [21:51:40] Some side effects of this transition are the replacement of the 'Clause' as a unit of syntax with the 'Catena' (Q1287636), the migration of (lots of) functions outside of class definitions, better type hinting for most (if not all) functions, and the start of more harmonized function documentation. (Here's hoping no further transformation as drastic as this is needed, though Denny may well have other plans.)