Kabouik has quit [Remote host closed the connection]
Kabouik has joined #foot
Kabouik has joined #foot
florhizome[m] has quit [Ping timeout: 246 seconds]
IvarHill[m] has quit [Ping timeout: 246 seconds]
IvarHill[m] has joined #foot
florhizome[m] has joined #foot
talismanick has joined #foot
Ordoviz has joined #foot
talismanick has quit [Ping timeout: 255 seconds]
Ordoviz has quit [Ping timeout: 255 seconds]
Ordoviz has joined #foot
fitrh has joined #foot
fitrh has quit [Read error: Connection reset by peer]
fitrh has joined #foot
fitrh has quit [Quit: fitrh]
Ordoviz has quit [Ping timeout: 255 seconds]
Ordoviz has joined #foot
sentriz has quit [Read error: Connection reset by peer]
rodrgz has joined #foot
rodrgz has quit [Ping timeout: 246 seconds]
rodrgz has joined #foot
betaboon[m] has joined #foot
<testuser[m]>
sort of a general question, why are the colors of curses applications such a hit or miss thing? like this is what ncdu looks like: https://0x0.st/oYRM.png (foot config https://termbin.com/zw1p)
<testuser[m]>
its only good in applications where u can theme them directly like cmus or htop
<testuser[m]>
so they work good with your colorscheme
<testuser[m]>
err: render.c:3968: BUG in maybe_resize(): assertion failed: 'false'
radu24275 has joined #foot
<dnkl>
betaboon[m]: oops, my bad. That's a debug assert that should have been removed.
<dnkl>
it'll only trigger on debug build though. If you do a release build it'll go away
<betaboon[m]>
hm the segfault happens in a production-build. then that assert was just a red hering
<dnkl>
testuser[m]: maybe because they're mixing colors from the 256-color table (typically not configurable by the user) with the 16 base colors (your color theme)
ayushnix has quit [Remote host closed the connection]
<dnkl>
betaboon[m]: please don't post too long messages; this is IRC
<betaboon[m]>
dnkl: yeah sorry. i forgot. i promise to do better :P
<dnkl>
betaboon[m]: can you get a backtrace from a debug build (with the bogus assert removed)? If so, it'd be great if you could open a bug report
<betaboon[m]>
with the bogus-assert removed on a debug build the segfault doesnt happen
<betaboon[m]>
trying with a release-build with the assert removed now
<betaboon[m]>
release-build with bogus-assert removed also doesnt segfault any more
<dnkl>
depending on how you configure the build, it's possible to get a release builds with assertions.
<dnkl>
foot --version should tell you if assertions are enabled or not
<betaboon[m]>
I'm on nixos using nixpkgs-wayland. maybe the build defined there does that
<dnkl>
in any case, I'll remove the assert as soon as i get back home
<betaboon[m]>
ty. :)
rodrgz has quit [Ping timeout: 252 seconds]
<dnkl>
does the packaged version track the latest release, or git HEAD?
rodrgz has joined #foot
<betaboon[m]>
we're on HEAD
<betaboon[m]>
HEAD for wlroots, swaywm, foot ;)
<betaboon[m]>
but that's not the regular package tree for nixos. it's just for those ppl that want nightlies
<betaboon[m]>
and someone has to test the bleeding edge, right ? :P
<dnkl>
thanks, then I don't know what, other than the assert, that could trigger the crash... :|
<dnkl>
yeah, the more the better... :)
<betaboon[m]>
i just double checked: the removal of that assert is sufficient to get rid of the segfault for debug- and release-build. so yeah °\o/°
<testuser[m]>
betaboon: was it actually a segfault or just sigabrt ?
<betaboon[m]>
fault
<testuser[m]>
Sus
<dnkl>
betaboon[m]: it actually makes sense
<dnkl>
xassert is a bit different from a regular assert
<betaboon[m]>
with coredump and all. just didnt want to start fiddling with getting debug-symbols to get a proper bt. that's why i asked here first if its known
<dnkl>
in release builds, it expands to an "unreachable" hint to the compiler
<dnkl>
allowing it to optimize better
<dnkl>
in this case it probably meant throwing away all code after the assert
<betaboon[m]>
ahhh that makes sense
rodrgz has quit [Quit: WeeChat 3.7.1]
<dnkl>
betaboon[m]: fix pushed, thanks for reporting!
<betaboon[m]>
ty for the quick turnaround :)
<testuser[m]>
dnkl so what's the proper way to handle colors, say with ncurses APIs?
<dnkl>
I'd say either use the 16 base colors only, or, make sure your define all the colors you use yourself. How exactly to map that to ncurses, I don't know
ayushnix has joined #foot
Ordoviz has quit [Quit: WeeChat 3.7.1]
sterni has quit [Quit: WeeChat 3.6]
sterni has joined #foot
ayushnix has quit [Remote host closed the connection]
ayushnix has joined #foot
talismanick has joined #foot
alismanickt has joined #foot
alismanickt has quit [Remote host closed the connection]
talismanick has quit [Quit: Using Circe, the loveliest of all IRC clients]
talismanick has joined #foot
talismanick has quit [Read error: Connection reset by peer]