| 2022-06-10 10:45:21 +0200 | <noze> | okay, I did that, but it wasn't so trivial; I had to set my wayland layout to US first |
| 2022-06-10 10:47:08 +0200 | <noze> | what's the right way to get richer unicode input? |
| 2022-06-10 10:55:24 +0200 | <Solid> | unicode entry on GNU/Linux is done via compose key sequences |
| 2022-06-10 10:55:34 +0200 | <Solid> | I don't know if wayland supports those |
| 2022-06-10 10:55:44 +0200 | <Solid> | but on X11 you just instert the unicode character verbatim and it should just work |
| 2022-06-10 10:56:54 +0200 | <noze> | xcompose sort of works under wayland, but I can't seem to figure out how to reload it without restarting my session. oh well... |
| 2022-06-10 10:58:43 +0200 | <noze> | do you know if there's a way to enter a character by typing out its code? E.g. if I want to type a single 'A' (silly example), I'd type 0x41, or 65 |
| 2022-06-10 11:00:10 +0200 | pils[m] | (~pilscatca@2001:470:69fc:105::2:b9c) (Quit: You have been kicked for being idle) |
| 2022-06-10 11:00:23 +0200 | <noze> | as in, is there a `read_unicode_sequence` such that I can bind <Multi_key><U><U> to read_unicode_sequence, and then type out the sequence? |
| 2022-06-10 11:00:42 +0200 | pils[m] | (~pilscatca@2001:470:69fc:105::2:b9c) |
| 2022-06-10 11:03:43 +0200 | pils[m] | (~pilscatca@2001:470:69fc:105::2:b9c) () |
| 2022-06-10 11:07:10 +0200 | <noze> | Is it possible to emit keys that are not listed in Keycode.hs? |
| 2022-06-10 11:10:34 +0200 | <noze> | a lot of keys are defined in /usr/include/X11/keysymdef.h |
| 2022-06-10 11:10:34 +0200 | <noze> | can kmonad emit them? |
| 2022-06-10 11:34:34 +0200 | <Solid> | the answer to the first question is no; kmonad is relatively low-level and so a priori only knows about keycodes that the linux kernel itself specifies; definitely no unicode, as how exactly input works there is the job of the environment |
| 2022-06-10 11:35:17 +0200 | <Solid> | There is KMonad.Keyboard.ComposeSeq for all kinds of compose key sequences that are accepted by default |
| 2022-06-10 11:35:27 +0200 | <Solid> | i.e., the unicode character itself should work |
| 2022-06-10 11:35:54 +0200 | <Solid> | for anyting else you would have to specify a macro manually (like #(menu " s) or something) |
| 2022-06-10 11:37:29 +0200 | <noze> | thank you Solid for all the explanations :) |
| 2022-06-10 11:39:21 +0200 | <noze> | okay so I guess I'll just have to bite the bullet and work with an additional level of indirection; it would have been nice to just specify the unicode code point in kmonad, but unless there's something like read_unicode_sequence I don't see how that could be made to work |
| 2022-06-10 11:40:11 +0200 | <Solid> | yeah it's not completely out of the question that this will ever be supported, but it's definitely in at least the medium future |
| 2022-06-10 11:41:02 +0200 | <Solid> | you'd have to recognise the environment in which kmonad was started somehow, which is hard in itself, but then people who start it as a service surely also want unicode entry, so it's more like the "best current environment" |
| 2022-06-10 11:41:08 +0200 | <Solid> | which tbh sounds like a bit of a nightmare |
| 2022-06-10 11:41:25 +0200 | <Solid> | and I don't even know if there is a reliable way to input unicode in e.g a TTY |
| 2022-06-10 11:41:36 +0200 | <Solid> | (without going via keymaps and all that) |
| 2022-06-10 18:26:27 +0200 | <noze> | okay, so xcompose only seems to support input sequences up to length 8, but with that I was able to code any 6 hex digit unicode char "FOO" with code '0x9ABCDE' as <Multi_key> <u> <x> <9> <a> <b> <c> <d> <e> : "FOO" |