ayushnix has quit [Read error: Connection reset by peer]
ayushnix has joined #foot
erectus has joined #foot
primalmotion has quit [Read error: Connection reset by peer]
primalmotion has joined #foot
alexander has quit [Quit: ZNC 1.8.2+deb3.1 - https://znc.in]
alexander has joined #foot
h-erectus has joined #foot
boomboxnation has quit [Ping timeout: 252 seconds]
boomboxnation has joined #foot
erectus has quit [Ping timeout: 260 seconds]
<dnkl>
rockorager: interresting observation; for me, shift+space *does* set shift as consumed. Both in plain us and se layouts
<dnkl>
unfortunately, ignoring modifiers wont work - xkb returns text for e.g. ctrl+space
h-erectus has quit [Ping timeout: 260 seconds]
<dnkl>
not really related, but I think we should drop the hardcoded caps-lock, and instead mask out "locked" modifiers, as reported by xkb
<dnkl>
rockorager: funny thing: shift+space treats shift as consumed when NumLock is disabled, but when NumLock is *enabled*, shift isn
<dnkl>
isn't consumed
<dnkl>
if that isn't a keymap bug, I don't know what it is...
<dnkl>
hmm, the mode flag in xkb_state_key_get_consumed_mods2() affects this
<dnkl>
I've been using the GTK mode for the kitty keyboard protocol, because "it better matches kitty" (my own comment)
<dnkl>
switching to "XKB" mode "fixes" shift+space - it is reported as text regardless of NumLock
<dnkl>
so, I guess we'll have to compare a bunch of key combos with kitty, and if there are differences, see if they are backed by the specification or not.
<dnkl>
code would be simpler if we could drop the GTK mode for kitty
erectus has joined #foot
erectus has quit [Read error: Connection reset by peer]
<dnkl>
kitty has a similar issue, where backspace is reported as text when NumLock is off, but as an escape sequence when it's on
<rockorager>
dnkl: do I have this correct: with numlock enabled and gtk mode, this shift + space didn’t report any text?
<rockorager>
using gtk, this reports text and *also* that shift wasn’t consumed
<rockorager>
curious if gtk is adding the text reporting bit on top of what xkb reports
<dnkl>
rockorager: first, GTK isn't really a part of this. It's just a mode XKB implements when translating keyboard state into "consumed" modifiers. Just wanted to make that clear
<dnkl>
but yeah, in that mode, the set of consumed modifiers differs for shift+space depending on whether numlock is enabled or not
<dnkl>
xkb gives you text back in both cases, but foot decided it wasn't text *only* when shift wasn't included in the "consumed" set
<dnkl>
because then the consumed set didn't cancel out the "effective" set.
<sewn>
does foot artificially dim the colors when SGR 2 is provided?
<sewn>
or does it use the 'dim' color set (90-100 afaik)