adamkowalski has quit [Read error: Connection reset by peer]
adamkowalski has joined #zig
WendigoJaeger has joined #zig
<daurnimator>
leeward: what's wrong with LoggingAllocator?
dddddd has quit [Remote host closed the connection]
mahmudov has quit [Ping timeout: 258 seconds]
<leeward>
daurnimator: Bugs
<leeward>
For one thing, the prints haven't been updated with the no-variadic-function stuff.
<leeward>
I don't have the sample code I was playing with, but it took some doing just to get something to compile and then it segfaulted. Not certain the segfault isn't my fault, but the compile errors weren't.
adamkowalski has quit [Quit: Lost terminal]
<daurnimator>
ah okay
lunamn has quit [Ping timeout: 265 seconds]
lunamn has joined #zig
darithorn has quit [Quit: Leaving]
dingenskirchen has quit [Ping timeout: 248 seconds]
<mikdusan>
daurnimator: re: #3979 is concat operator for tuples going to be a thing, or are we expecting a compiler error?
charlesshenton has joined #zig
<charlesshenton>
How do I resolve this compile error? `const copts: []const []const u8 = .{"-std=c99"};` errors with "array literal requires address-of operator to coerce to slice type '[]const []const u8'"
charlesshenton is now known as cshenton
<cshenton>
It's not immediately clear where I should stick the & operator.
darithorn has joined #zig
return0e has quit [Read error: Connection reset by peer]
return0e has joined #zig
lygaret has quit [Quit: Leaving.]
lygaret has joined #zig
owl_000 has quit [Remote host closed the connection]
mahmudov has joined #zig
dingenskirchen has joined #zig
dingenskirchen has quit [Quit: dingenskirchen]
cshenton has quit [Remote host closed the connection]
ave_ has joined #zig
darithorn has quit [Remote host closed the connection]
_whitelogger has joined #zig
return0e_ has joined #zig
ave_ has quit [Quit: Connection closed for inactivity]
return0__ has joined #zig
return0e has quit [Ping timeout: 268 seconds]
_whitelogger has joined #zig
return0__ has quit [Read error: Connection reset by peer]
return0e has joined #zig
return0e_ has quit [Remote host closed the connection]
<Snektron>
Hm, for some reason i can't get zig to build a static binary inside a docker image
<Snektron>
also, i think there might be something wrong with the build system because sometimes it builds without detecting all required libraries
<Snektron>
Anyway it just fails on the part where it builds the stage1 compiler
<Snektron>
and it doesn't tell me why
<Snektron>
Ah, it's because zig0 segfaults
<Snektron>
did this commit even pass ci
<Snektron>
apparently so
<Snektron>
because zig0's entry point is literally an unmapped address
<Snektron>
hm, it works with just static llvm
doublex_ has quit [Read error: Connection reset by peer]
doublex_ has joined #zig
doublex_ has quit [Read error: Connection reset by peer]
doublex_ has joined #zig
lunamn has quit [Ping timeout: 240 seconds]
lunamn has joined #zig
dddddd has joined #zig
lupine_ has joined #zig
lupine has quit [Ping timeout: 250 seconds]
lygaret has joined #zig
lygaret has quit [Client Quit]
doublex_ has quit [Read error: Connection reset by peer]
doublex_ has joined #zig
lygaret has joined #zig
lygaret has quit [Quit: Leaving.]
<leeward>
Snektron: Do you have ideas why `zig build test` would fail with a bunch of undefined symbols in clang::?
lupine_ is now known as lupine
<Snektron>
i have no idea
<leeward>
:/
<Snektron>
im not really an expert on Zig's build process
<Snektron>
what are you building? The compiler?
<leeward>
Yeah, `make` seems to work fine, but `zig build test` fails hard.
<leeward>
I thought it might have been part of the thing you were talking about earlier.
<Snektron>
Weird, i haven't tried that
casaca has joined #zig
<Snektron>
I don't think thats related
<Snektron>
thats probably a misconfiguration on my part
<leeward>
ah
<Snektron>
For some reason zig0 is just an invalid executable when i try to compile Zig statically
<leeward>
weird
<Snektron>
So im now trying to see if -static-gcc will work
<Snektron>
because i don't really want to depend on libgcc
<Snektron>
Maybe i don't have the static version of some library
return0e_ has joined #zig
return0e_ has quit [Client Quit]
<Snektron>
its also weird how it doesn't fail to compile or anything
<Snektron>
it just produces a faulty executable
<leeward>
That is pretty odd.
return0e has quit [Ping timeout: 260 seconds]
return0e has joined #zig
<Snektron>
I made my own docker file instead of using the docker-zig one because i didn't want to wait on compiling llvm
<Snektron>
If the master builds were hosted any longer i would have used those
return0e has quit []
darithorn has joined #zig
doublex_ has quit [Read error: Connection reset by peer]
doublex_ has joined #zig
return0e has joined #zig
lygaret has joined #zig
darithorn has quit [Remote host closed the connection]
<frmdstryr>
Can zig convert c header files to zig?
<mq32>
frmdstryr, yeah "zig translate-c" does this :)
<frmdstryr>
What compiler does it use? Eg, #elif defined ( __GNUC__ )
<frmdstryr>
Nvm seems to use that
<Snektron>
frmdstryr: clang
<Snektron>
(which also defines gnuc)
lygaret has quit [Quit: Leaving.]
<frmdstryr>
23.6k loc haha, awesome
doublex_ has quit [Read error: Connection reset by peer]
doublex_ has joined #zig
lygaret has joined #zig
lygaret has quit [Quit: Leaving.]
lygaret has joined #zig
<leeward>
zig fmt has way too high a column limit.
<gonz_>
I don't think it has a high column limit as much as it just plain doesn't put indentation in there unless you have a trailing comma in an argument list or one of your own newlines in a long binary expression, etc.
<fengb>
Infinite is technically too high 🙃
<gonz_>
Hah, yeah, good point
<gonz_>
I'm happy no one chose 80 for everyone, at least.
<gonz_>
Though I guess I'd take any standard, really.
<gonz_>
Better to have a common standard that I dislike than no standard at all.
<leeward>
80 would make me the most happy.
<leeward>
My problem is that I formatted a line with reasonable wrapping, and zig fmt undid it
lygaret has quit [Quit: Leaving.]
tyler569 has quit [Quit: leaving]
<gonz_>
With binary expressions I'll usually put a newline after an operator and it'll accept it
<gonz_>
I wouldn't mind if it was more aggressive with auto-formatting, though. Less for me to care about in the end.
<Snektron>
I never use zig fmt anyway
<gonz_>
I use it all the time, but not actively. I write horribly mis-shapen expressions with spaces missing or even with the wrong indentation and it just takes care of it.
<gonz_>
Which is the point, I feel, to not sit there doing these braindead things yourself.
<leeward>
I'm just sad that tabs for indentation/spaces for alignment never caught on.
casaca has quit [Ping timeout: 265 seconds]
casaca has joined #zig
hryx has joined #zig
doublex_ has quit [Read error: Connection reset by peer]
doublex_ has joined #zig
<Snektron>
gonz_, i usually do that correctly when im writing anyway
<gonz_>
Seems pointless to me
<Snektron>
I dont spend effort on it
<gonz_>
I would argue that to get as consistent a result as with auto-formatting, you'd have to actually spend effort on it.
<gonz_>
To me it's an easy win because it doesn't generally look like ass and I didn't have to do anything for it.
<gonz_>
And with a language like zig it should be easy for everyone to accept an auto-formatter because they don't have years of believing their style is the best style.
<Snektron>
Oh well thats just because zig-style happens to be similar to what i usually format my preferred style
<hryx>
ahoj friends! I've been away from Zig code & chat for several months and have missed it terribly. I'm catching up on what's new and look forward to getting involved again